Aging & Apetite

Aging &Appetite
Socialandphysiological approaches intheelderly
Marie-FrangoiseA.M.Mathey
Promoter: Dr.W.A.van Staveren
Hoogleraar indevoedingvan ouderemens
Co-promotor: Dr. CdeGraaf

Universitairdocent
Afdeling HumaneVoeding &Epidemiologie

Wageningen Universiteit
•Z%/al
Aging&Appetite
Socialandphysiologicalapproaches intheelderly
Marie-FrancoiseA.M.Mathey
Proefschrift
Terverkrijging vandegraadvandoctor
opgezagvande Rector Magnificus
vanWageningen Universiteit,
Dr. Ir. L.Speelman,
inhetopenbaar teverdedigen
opwoensdag 6September 2000
des namiddagstevier uur indeAula
vanWageningen Universiteit
CENTRALE LANDBOUWCATALOGUS
0000 0807 0928

The studies described in this thesis were partly funded by the 'Stichting Voeding
Gezondheid enOuderen',Wageningen,the Netherlands.
Thestudieswere partofthe research program ofthegraduate schoolVLAG.
Financial support for the publication of this thesis by the Wageningen University is
gratefullyacknowledged.
EIBLIOTHHHX
LANDBOUWUNIVV.RSTTuT
WAGENHNGt'N
Mathey, Marie-FrangoiseA.M.
Aging &Appetite. Socialandphysiological approaches intheelderly.

ThesisWageningen University-With ref.- Withsummary in DutchandFrench.
ISBN 90-5808-270-9
Coverdesign:"etje doisencore penser atoutca?!", Marie-FrancoiseMathey
Printing: Grafisch Service CentrumVan Gils B.V., Wageningen,the Netherlands
©2000,M.-F.A.M.Mathey
^0^1O\ ,2ZZP\
Stellingen -Theorems -Theoremes

1. Individualised nutritional care does improve nutritional status in institutionalised
elderly residents and therefore should be a right for all care-depending elderly
people (o.a.ditproefschrift).
2. Ambience of food consumption is an important determinant of dietary intake in

institutionalised elderly residents (o.a.ditproefschrift).
3. Sufficient individualised care in elderly institutions is not a finance-related but a

management-related issue(o.a.ditproefschrift).
4. Monitoring body weight is a reliable indicator of nutritional status in

institutionalised elderly andshould beusedasascreeningtoolto preventweight
loss(o.a.ditproefschrift).
5. Scientific truth, which I formerly thought of as fixed, as though it could be

weighed and measured,ischangeable.Add afact, changethe outlook, and you
have a new truth. Truth is a constant variable. We seek it, we find it, our
viewpointchanges,andthetruthchangestomeetit
WilliamJ.Mayo (1861-1939).
6. How far you go in life depends on your being tender with the young,
compassionate with the aged, sympathetic with the striving, and tolerant of the
weakandstrong,becausesomeday inlife,youwillhavebeenallofthese things
GeorgeWashington Carver(1864-1943).
7. Colette avait raison :ilest incroyable deconstater quedans une societe aisee il
n'y a pas de nourriture pour ses personnes agees et pas de travail pour ses
apprentis.
8. Senescencebegins,andmiddleageends,
Thedayyourdescendantsoutnumberyourfriends
Odgen Nash (1964).
9. "Celuiquiesttoujourssage,meneunevielamentable"
CANDIDE, Voltaire (1757).
10.Aubouquet onjugelevin ;aI'odeur,lafleur;aulangage, I'homme.
11."Jemepressederiredetout,depeurd'etreobliged'enpleurer"
Beaumarchais (1732-1799).
12.Werken meteenbuitenlandsetaal isaccepteren om niet honderd procentvanje

capaciteitentekunnengebruiken.
Stellingen behorende bijhet proefschrift

Appetite&Aging
Socialandphysiologicalapproaches intheelderly
Marie-FrancoiseA.M.Mathey.Wageningen,6September2000
Alamemoire deJean Mathey
Abstract
Aging&Appetite:Socialand physiological approaches intheelderly
Ph.D.-thesisbyMarie-FrangoiseA.M.Mathey,DivisionofHuman Nutrition &

Epidemiology,Wageningen University, theNetherlands. September6, 2000.
Aging isoften accompanied by anorexia of aging,described asadecline inappetite,

a lower dietary intake and followed by unexplained weight loss. The present thesis
describes research on anorexia of aging. Focus was given to social and
physiological determinants of appetite andthe relationship with dietary intake and/or
bodyweightwasexamined inobservation aswellasininterventionstudies.
Results showed first that appetite and the extent to which food is enjoyed, varied
greatly between elderly people. These differences may be explained by differences
in the health characteristics of the elderly groups studied. Second, social and
environmental factors remained important determinants of appetite, more especially
in elderly with an unstable or poor health condition. Further, the incapacity to adjust
energy intake to physiological challenges on long-term seems to be a non-reversible
process. In daily practice, this lack of regulation suggests that the consumption of
energy and nutrient dense supplements between meal could help to prevent weight
loss inolder adults atrisk.
In conclusion, the properties of foods and the context in which the foods are
consumed remain important determinants of dietary intake in the elderly and are
thereby major riskfactorsfor anorexia ofaging and itssubsequent weight loss. From
a public health perspective, the lack of regulation in appetite and dietary intake
should encourage the use and consumption of nutritional interventions in elderly at
risk.
Contents
Chapter 1 General introduction 11
Chapter 2 AssessingappetiteinDutchelderlywiththeAppetite,Hunger 23
feelingsandSensoryPerception(AHSP)questionnaire
Chapter 3 Socialandphysiologicalfactors affecting food intake in 33

elderly subjects:anexperimental comparative study
Chapter 4 Healtheffect ofimproved mealambiance inaDutch 47

nursing home:aone-year intervention study
Chapter 5 Effect ofaneveningsupplement providedto nursing-home 63

elderly onbodyweight anddietary intake
Chapter 6 Flavorenhancement offood improves dietary intakeand 75

nutritional status ofnursing homeelderly
Chapter 7 Generaldiscussion 89
Summary 105
Resume 109
Samenvatting 113
Remerciements 117
Abouttheauthor 120
1
General Introduction
Chapter 1
The worldwide raising number of people aged of 65 and over is well documented
(1;2). This growing number is the consequence of two phenomena, an increased
absolute number and an average longer life expectancy (Table 1). As a
consequence the elderly represent a substantial segment of the European
population.
Tablel: Evolution oftheEuropean elderlypopulation inthelast10years

Lifeexpectancyatbirth(y) % ofthepopulationaged>65y old
In1990 In1999 In1990 In1999
Country M F M F M+F M+F
France 72.7 80.9 74.6 82.3 14 15.4
The Netherlands 73.8 80.1 74.6 80.4 13 13.4
European Union 73.0 80.0 74.0 81.0 14 15.8
Basedonreferences(1;2)(M=male,F=female)
Aging is known as "a process that converts healthy individuals into frail ones, with
diminished reserves in most physical systems and an exponentially increasing
vulnerability to most diseases and death"(3). Thus the growing number of elderly
inevitably leadstoan increased demand of health care andattentionforthisgroupof
thepopulation.
Theelderly population
Different human beingsageatadifferent rate,implyingthat asochronological ageis

not necessarily equal to biological age. Aging has been described as a complex
system (3)influenced byalarge numberofinternalandexternalfactors (Fig1).
Therefore apart from a chronological description the elderly population has often
beenclassified accordingtotheir healthstatus(4-7):
-successful agers are independently living and present little or almost no loss of
functioningthat coulddefinedasaging perse,
-usualagersare independently livingwithavariety of medicalconditionsand
-accelerated agers carrying an heavy burden of chronic diseases and disabilities
most of them residing in institutions. This classification highlights the heterogeneity
oftheelderlypopulation.
12
General introduction
In all agers an adequate dietary intake has well been recognized as a necessary
factor in improving longevity (8), maintaining good health (9) and quality of life (10).
Aging is associated with many social and physiological changes (11;12),which may
negatively influence energy and nutrient use. Besides, simply consuming enough
food maybecome amajor issueforolder people (13).
Conception Old age

Figure 1:Diversityoffactorsinfluencingagingfromyoungtooldage
Both European andAmerican health surveys showed that at the age of 65-70 y and
beyond body weight tends to decrease, even in healthy individuals (5;14-18). This
involuntary unexplained weight loss in later life increases the risk of protein-energy
malnutrition, micronutrient deficiencies and nutrition-related illnesses and is
associated withfrailty and increased morbidity (19).
Intervention studies inthe elderly showed that this loss of body weight would result
from a dysregulation of the ability to regulate food intake i.e., a decline in appetite
control (20-22) and energy balance (23). This loss of ability makes it difficult to
compensate for the day to day fluctuation in dietary intake and subsequently may
leadtomalnutritionand unintentionalweightloss.
Appetite
Appetite is defined as the process, which directs eating and guides the moment-to-
moment selection and intake offoods (24).The control offood intake also occurs at
different levels:-social,psychological andphysiological.
13
Chapter1_
Eating is not only a biological action; it has also social, cultural and symbolic
meanings (25). Therefore social, physical and environmental (=non-physiological)
factors are also important determinants of appetite and food intake.As an example,
physical and sensory aspects of food such as color, form,texture, crispiness, smell
are relevant parameters for food choice and consumption (26). These parameters
promote signals preceding the ingestion of food and will stimulate the 'cephalic
phase of appetite'. The cephalic phase generates responses inthe gastro-intestinal
tracttoanticipatethe ingestion anddigestion offood(27).
Meals mayoften betheonlytimewhen afamily istogether. Eatingwithfamily and/or
friends has beenfoundto have apositive effect onappetite (28-30). Studies showed
that social facilitation i.e., eating more in the presence of others occurs in the
presence offamily andfriends but notwithstrangers and ismainly mediated bymeal
duration(31).
During and immediately after eating, afferent information from the ingested food
provides the major control on appetite. Several physiological determinants intervene
intheprocess ofenergy and nutrient intake.They primarily provide positive feedback
foreatingwhenfood isacting inthe mouth (sensory and cognitive phases). Secondly
they provide negative feedback information when food reaches the stomach, the
small intestine andthe bloodflow (postprandial and post-absorptive phases) leading
first to satiation, the process which terminates eating within a meal and then to
satiety,thestatewhich inhibitsfurther eating (24;32).
As anexample,taste and smell perception is a key determinant ofthe palatability of
foods (33) and represents a major factor of the sensory-specific satiety. The latter
has been defined as a progressive decrease in pleasantness of a particular food
following consumption (34).Sensory-specific satiety isassociated with adecrease in
consumptionofthe previously eatenfoodandashifttootherfoods hereby promoting
intakeofawider variety offoods andanutritionally balanced diet (33;35).
Thegastro-intestinal tract haschemo-and mechano-receptors,whichwill monitor the
physiological activity following the food intake and then pass the information to the
brain through the vagus nerve (24). This information is translated into brain
neurochemical activity andforms partofthe post-ingestive controlofappetite.
The resulting brain activity involving a large number of neurotransmitters,
neuromodulators, pathways and receptors determines the strength of motivation,the
patternofbehavioralevents (36-38) andthewillingness tostartorstopeating(24).
14
With aging, this complex system may be affected, leading to an appetite

dysregulation, which would promote an inadequate dietary intake and subsequent
involuntaryweightloss.
Appetite dysregulation:Anorexia ofaging

The loss of appetite occurring with age or so-called anorexia of aging has been
defined by Morley and Silver (39) as "the physiological decrease in food intake
occurring to counterbalance reduced physical activity and lower metabolic rate, not
compensated in the long term". The main consequence of this loss of appetite is
unintentional weight loss, which is an important predictor for mortality (14; 17;40-
42).Anorexia ofaging has many potential causative parameters that may bedivided
in three groups: social and environmental, psychological, and physiological and
medicalparameters.
Socialandenvironmentalparameters
Retirement leads to changes in life habits and daily rhythm but also to a reduced
household income. Poverty or low income has been cited as one ofthe major social
causesfor alowerfood intake(43).
With advancing age, remarkable loss of functioning such as decreased physical
ability,visualand hearing impairments occurred (44). Consequently itbecomes more
difficult for an elderly person to perform basic daily activities such as shopping/
buying the desired food, cooking andfeeding oneself (45). Besides, limited physical
ability and/or the need of assistance for daily activities may restrain the social
network. The lattermayresult hereby resulting inisolationwithinaneighborhood and
nocompanyduring meals (46;47).
Psychologicalparameters
Mental disorders have been estimated to be responsible for the major parts of
psychological causes of anorexia of aging and unintentional weight loss, especially
innursing homeelderly(48).
An important psychological parameter, bereavement is induced by the loss of
spouse or widowhood,which often eventuates insocial isolation and loneliness (43;
49).This latter mayeventually makethe elderly personfeel an overwhelming burden
of life (43)and leadtodepression.
15
Chapter1_
Depression has beenfound to bethe most common cause ofweight loss in medical
outpatients and nursing home elderly residents (50). In this regard food might be
usedasaweapon,akindofsubconscious wishofdeath (44;51).
Dementia orcognitive impairment isalsoafrequentcauseofdecline indietary intake
and weight loss since subjects can simply forget to eat (49; 52). It seems difficult to
determine the reasons for a decrease in appetite. Potentially it may be caused by
indifference or lack of concern about eating, memory loss and impairment of
judgment i.e., inability to recognize the need to eat. Recently it has been suggested
that dementia might also be paired with an increased hunger and absence of
satiation. This disturbance ofthe internal appetite regulation might result in hypo-or
hyperphagia (53;54).
Physiologicalandmedicalparameters
Aging ispairedwiththeoccurrence ofpathological andphysiologicalvariations.
Age-related physiological changes influencing food intake are a decreased basal
metabolic ratedueto adecline infat free mass or sarcopenia (55) and avariation in
gut hormonal responses and secretion (11; 12; 21). These changes may lower
gastric emptying and appetite response to meal intake challenges resulting in an
earlysatiety and asubsequent overalldeclineoffood intake.
Dysfunction of taste and smell senses, caused by normal aging as well as certain
disease state may affect the palatability and hedonic responses to foods (56-62).
This lossofsensory perception pairedwithadiminished sensory-specific satiety may
limitbothvariety andquantity offoodsconsumed.
Numerous infections and medical disorders, both acute and chronic can produce
anorexia (49; 63). A number of these disorders lead to inflammation and cytokines
production resulting in a lower appetite (64). Cancer, AIDS, cardiac and geriatric
cachexia arerelatedto severe loss ofappetite andfood aversions generally resulting
inexcessivewasting (43;49).
Disease-related changes such as decreased mobility secondary to cerebrovascular
disorders or meal-induced dyspnea in chronic obstructive pulmonary disease also
promote lossofappetite(44).
Intake of medication may influence the endogenous opiate regulation of appetite
anorexigenic effect. Drug intakecan alsoadversely affect appetite and dietary intake
bycausing nauseaandvomiting aswellasalteringtasteandsmellsenses (43;59).
16
Anorexia ofaging makes itdifficult for anelderly personto respondto environmental

changes and maintaining an internal state. It therefore contributes to unintentional
weightloss.
Involuntary weight loss

At the age of 65-70 and beyond, both loss of lean body mass (sarcopenia), and
decline offat body mass result ina decline intotal body mass (40).These changes
have been documented in both cross-sectional and longitudinal surveys (14; 17;40;
65). Unintentionalweight loss hasbeenfoundto happen inboth institutionalized and
non-institutionalized elderly and is associated with physiological, psychological and
immunologic consequences, regardless of the underlying causes (48; 64; 66-68).
This weight loss increases the risk of protein-energy malnutrition and dangerous
underweight inthe elderly, more especially in nursing home residents. Among free-
living frail elders, weight loss was shown to be a predictor of early mortality after
controlling for smoking, and functional and health status indicators (69). These
findings confirmed the important role of losing weight as a major risk factor for the
downward spiral leading to frailty and mortality. Involuntary weight loss of 4-10 %of
original body weight has been found to be an important predictor of morbidity and
mortality (41;42; 70;71).
Rationale
The etiology of weight loss occurring with old age is, as explained above,
multifactorial and in this regard, anorexia of aging seems to be a key issue.
Thereforethe aim ofthis thesis isto examine the potential link existing between the
determinants ofappetite,dietary intakeandbodyweight inolderadults.
Outline ofthethesis
Inthe presentthesis bothsocialand physiologicalfactorswere investigated as major
determinants of the regulation of appetite. This was realized in both short-term and
long-term studies inelderly subjects.
Chapter 2 presents an in depth cross-sectional observation study on possible
differences in self -assessment of appetite feelings in three health categories of
elderly.
Chapter3presents astudy on short-term regulation ofappetite and dietary intake in
free-living elderly. Inthis study a social (environmental) manipulation was combined
17
Chapter1_
with a physiological manipulation consisting of preload-test meal design in order to

investigate the role of social facilitation and of energetic and nutrient challenges on
appetite controlandfoodintake.
Long-term regulation of appetite i.e. body weight maintenance (or weight loss
prevention) was assessed in three intervention studies conducted in nursing home.
This category of people was chosen for these three studies since we expected that
thedietary intakeofthese peoplewould be lowasaconsequence ofapoorappetite.
In Chapter4, the long-term effects of a changed environment and atmosphere on
appetite and health were explored in a Dutch nursing home. Since social support
and physical environment have been reported as some of the major reasons for a
decrease indietary intake innursing home residents, aone-year intervention with an
improved ambianceoffoodconsumptionwasconducted.
Chapter5describes theeffects ofa4-week supplementation with anevening energy
richdrink onthedietary intakeandbodyweight ofnursing homeelderly residents.
In chapter 6, the influence of a 16-week intervention with or without added flavor
enhancers onappetite,dietary intakeandbodyweight ispresented.
Finally, in chapter 7, the main findings are summarized; methodological problems,
conclusions and implications arediscussed.
References
1. INED. Population &societes. bulletin mensueld'information deI'INED (282). 1993.
2. INED. population &societes. bulletin mensuel d'information de I'INED(348). 1999.
3. Miller R. The biology of aging and longevity. In: Hazzard W, Bierman E, Blass J, Ettinger W, Halter
J, Andres R, eds. Principles of geriatric medicine and gerontology. 3rd Ed. New York: Mc Graw-Hill,
Inc., 1994:3-18.
4. Rowe JW, Kahn RL. Human aging: usual and successful. Science 1987;237:143-149.
5. Diehr P, Bild DE, Harris TB, Duxbury A, Siscovick D, Rossi M. Body mass index and mortality in
nonsmoking older adults:the Cardiovascular Health Study. Am.J.Public Health 1998;88:623-629.
6. van derWielen, RPJ.Vitamin intake and status in elderly Europeans. 1-167. 1995. Department of
Human Nutrition,Wageningen Agricultural University, the Netherlands.
7. de Jong, N. Sensible aging, nutrient dense foods and physical exercise for the vulnerable elderly.
1-191. 1999. Department of Human Nutrition & Epidemiology, Wageningen University, The
Netherlands.
8. de Groot L, Van Staveren W, Burema J. Survival beyond age 70 in relation to diet. Nutr Rev
1996;54:211-212.
9. Sullivan DH. Impact of nutritional status on health outcomes of nursing home residents. J Am
Geriatr.Soc 1995;43:195-196.
10. Vetta F, Ronzoni S, Taglieri G, Bollea MR. The impact of malnutrition on the quality of life in the
elderly. Clin.Nutr. 1999;18:259-267.
18
11. Cook CG,Andrews JM, Jones KL,et al. Effects of small intestinal nutrient infusion onappetite and
pyloric motility are modified byage. Am J Physiol. 1997;273:R755-R761
12. Lovat LB.Age related changes ingut physiology and nutritional status. Gut 1996;38:306-309.
13. Stitt S,O'Connell C, Grant D. Old,poor &malnourished. Nutr Health 1995;10:135-154.
14. Comoni-Huntley JC, Harris TB, Everett DF, et al. An overview of body weight of older persons,
including the impact on mortality. The National Health and Nutrition Examination Survey I -
Epidemiologic Follow-up Study. J.Clin.Epidemiol. 1991;44:743-753.
15. Allison DB, Gallagher D , Heo M, Pi SF, Heymsfield SB. Body mass index and all-cause mortality
among people age 70 and over: the Longitudinal Study of Aging. Int J Obes.Relat.Metab Disord.
1997;21:424-431.
16. de Groot C, Perdigao AL , Deurenberg P. Longitudinal changes in anthropometric characteristics
of elderly Europeans. SENECA Investigators. Eur J Clin Nutr 1996;50 Suppl2:S9-15.
17. Lehmann AB, Bassey EJ. Longitudinal weight changes over four years and associated health
factors in629 menandwomen aged over65. Eur JClin Nutr 1996;50:6-11.
18. Fischer J, Johnson MA. Low body weight and weight loss in the aged. J Am Diet Assoc.
1990;90:1697-1706.
19. Mowe M, Bohmer T. Nutrition problems among home-living elderly people may lead to disease
and hospitalization. Nutr Rev 1996; 54:S22-S24
20. Rolls BJ, Dimeo KA, Shide DJ. Age-related impairments in the regulation of food intake. Am J
Clin Nutr1995;62:923-931.
21. Clarkston WK, Pantano MM, Morley JE, Horowitz M, Littlefield JM, Burton FR. Evidence for the
anorexia of aging: gastrointestinal transit and hunger in healthy elderly vs. young adults. Am J
Physiol. 1997;272:R243-R248
22. Zandstra E, Mathey M-F, de Graaf C, Van Staveren W. Short-term regulation of food intake in
children, young adults andtheelderly. European Journal of Clinical Nutrition 2000;54:239-246.
23. Roberts SB, Fuss P, Heyman MB, et al. Control of food intake in older men. JAMA
1994;272:1601-1606.
24. Blundell JE.the biology of appetite. ClinAppl Nutr 1991;1:21-31.
25. Rogers PJ. Eating habits and appetite control: a psychobiological perspective. Proc.Nutr.Soc.
1999;58:59-67.
26. Drewnowski A. Taste preferences andfood intake. Annu.Rev.Nutr. 1997; 17:237-253.
27. Blundell JE, Halford JC . Regulation of nutrient supply: the brain and appetite control. Proc Nutr
Soc 1994;53:407-418.
28. De Castro JM. Social facilitation of the spontaneous meal size of humans occurs on both
weekdays andweekends. Physiol.Behav.1991;49:1289-1291.
29. De Castro JM, Brewer EM . The amount eaten in meals by humans is a power function of the
number of people present. Physiol.Behav. 1992;51:121-125.
30. De Castro JM. Family and friends produce greater social facilitation of food intake than other
companions. Physiol.Behav. 1994;56:445-445.
31. Feunekes Gl, De Graaf C , Van Staveren WA. Social facilitation of food intake is mediated by
mealduration. Physiol.Behav. 1995;58:551-558.
32. Stubbs RJ. Nutrition Society Medal Lecture. Appetite, feeding behaviour and energy balance in
human subjects. Proc Nutr Soc 1998;57:341-356.
19
Chapter1_
33. Guinard JX, Brun P. Sensory-specific satiety: comparison of taste and texture effects. Appetite.
1998;31:141-157.
34. Rolls BJ.Sensory-specific satiety. Nutr Rev1986;44:93-101.
35. Lozano Dl, Crites SL, Aikman SN. Changes in food attitudes as a function of hunger. Appetite
1999;32:207-218.
36. Westerterp-Plantenga M. Eating behavior in humans, characterized by cumulative food intake
curves -a review. Neurosci.Biobehav.Rev 2000;24:239-248.
37. JamesWP, McNeill G, Ralph A. Metabolism and nutritional adaptation to altered intakes of energy
substrates. Am.J.Clin.Nutr. 1990;51:264-269.
38. Hirsch J,Hudgins LC, Leibel RL, Rosenbaum M. Diet composition and energy balance in humans.
Am.J.CIin.Nutr. 1998;67:551S-555S.
39. Morley JE,Silver AJ. Anorexia inthe elderly. Neurobiol.Aging 1988;9:9-16.
40. Mattila K, Haavisto M, Rajala S. Body mass index and mortality in the elderly.
Br.Med.J.CIin.Res.Ed. 1986;292:867-868.
41. Murden RA, Ainslie NK. Recent weight loss is related to short-term mortality in nursing homes. J
Gen.Intern Med 1994;9:648-650.
42. Berkhout AM, van HJ, Cools HJ. [Increased chance of dying among nursing home patients with
lower bodyweight](in Dutch). NedTijdschr Geneeskd 1997;141:2184-2188.
43. Morley JE. Anorexia in older persons: epidemiology and optimal treatment. Drugs Aging
1996;8:134-155.
44. Morley JE, Silver AJ, Fiatarone M, Mooradian AD. Geriatric grand rounds: nutrition and the
elderly. University of California, Los Angeles [clinical conference]. J Am Geriatr.Soc 1986;34:823-
832.
45. Keller HH, Ostbye T, Bright SE. Dietary habits of seniors with some activity limitations. Can J Diet
PractRes. 1999;60:214-221.
46. Goodwin JS. Geriatric ideology: the myth of the myth of senility. J Am Geriatr.Soc 1991;39:627-
631.
47. Goodwin JS. social, psychological and physical factors affecting the nutritional status of elderly
subjects: separating cause andeffect. Am JClin Nutr 1989;50:1201-1209.
48. Morley JE, Kraenzle D. Causes of weight loss in a community nursing home. J Am Geriatr.Soc
1994;42:583-585.
49. Morley JE.Anorexia ofaging: physiologic and pathologic. Am J Clin Nutr 1997;66:760-773.
50. Wilson MM, Vaswani S, Liu D, Morley JE, Miller DK. Prevalence and causes of undernutrition in
medical outpatients. Am.J.Med. 1998;104:56-63.
51. Silver AJ, Morley JE, Strome LS, Jones D, Vickers L. Nutritional status in an academic nursing
home. JAm Geriatr.Soc1988;36:487-491.
52. Marcus EL, Berry EM.Refusal to eat inthe elderly. Nutr Rev 1998;56:163-171.
53. Keene J, Hope T, Rogers PJ, Elliman NA. An investigation of satiety in ageing, dementia, and
hyperphagia. IntJ Eat.Disord. 1998;23:409-418.
54. Keene JM, HopeT. Hyperphagia indementia: 2. Food choices and their macronutrientcontents in
hyperphagia, dementia and ageing. Appetite 1997;28:167-175.
55. Visser, M. body composition and energy metabolism in elderly people. 6-9-1995. Department of
Human Nutrition,Wageningen Agricultural University, the Netherlands.
20
56. Wysocki CJ, Pelchat ML.The effects of aging on the human sense of smell and its relationship to
food choice. Crit.Rev Food Sci Nutr 1993;33:63-82.
57. Clydesdale FM. Changes in color and flavor and their effect on sensory perception in the elderly.
Nutr Rev 1994;52:S19-S20
58. Schiffman SS. Perception of taste and smell in elderly persons. Crit.Rev.Food Sci.Nutr.
1993;33:17-26.
59. Schiffman SS.Tasteand smell losses innormalaging anddisease. JAMA 1997;278:1357-1362.
60. Rolls BJ,McDermottTM. Effects of age on sensory-specific satiety. Am J Clin Nutr 1991;54:988-
996.
61. Rolls BJ. Do chemosensory changes influence food intake in the elderly? Physiol.Behav.
1999;66:193-197.
62. De Graaf C, Polet P, Van Staveren W. Sensory perception and pleasantness of food flavors in
elderly subjects. JGerontol 1994;49:93-99.
63. Lesourd B. Protein undernutrition asthe major causeof decreased immunefunction inthe elderly:
clinical andfunctional implications. Nutr Rev 1995;53:S86-S91
64. Yeh SS, Schuster MW. Geriatric cachexia: the role of cytokines. Am.J.Clin.Nutr. 1999;70:183-
197.
65. Matthews LE. Monitoring weight changes in nursing home residents. J Nutr Elder. 1989;9:67-72.
66. Blaum CS, Fries BE, Fiatarone MA. Factors associated with low body mass index and weight loss
in nursing home residents. JGerontolA Biol Sci Med Sci 1995;50:M162-M168
67. Fabiny AR, Kiel DP. Assessing and treating weight loss in nursing home patients. Clin
Geriatr.Med1997;13:737-751.
68. Morley JE, Silver AJ. Nutritional issues in nursing home care. Ann.Intern Med 1995;123:850-859.
69. Payette H, Coulombe C, Boutier V, Gray-Donald K. weight loss and mortality among frail living
elders: aprospective study. J GerontolA Biol Sci Med Sci 1999;54:M440-M445
70. Allison DB, Zannolli R, Faith MS, et al. Weight loss increases and fat loss decreases all-cause
mortality rate: results from two independent cohort studies. IntJ Obes 1999;23:603-611.
71. Wallace Jl, Schwartz RS , LaCroix AZ, Uhlmann RF, Pearlman RA. Involuntary weight loss in
older outpatients: incidence and clinical significance. J.Am.Geriatr.Soc. 1995;43:329-337.
21
2
Assessing appetite in Dutch elderly with the Appetite, Hunger and
Sensory Perception(AHSP)questionnaire*
Marie-FrancoiseA.M. Mathey, Nynke de Jong, Lisette C.P.G.M. de Groot,

Cees de Graaf and Wija A. van Staveren
Abstract
Background: In investigating anorexia of ageing attention is often given to physical and
psychological determinants of appetite in elderly but little has been done to provide
information onself-assessment ofappetite andsensory perceptions inthe elderly.
Objectives: to provide data and detect possible differences in self-assessment of appetite,
hunger feelings and sensory perception in different health group of elderly using theAppetite,
HungerfeelingsandSensory Perception(AHSP)questionnaire.
Methods:three healthcategories ofelderly subjectswere usedforthe present study: free-living
with no help, free-living with help and nursing home elderly. For each group, collected data
were general characteristics, anthropometry and answers to the AHSP. The AHSP
questionnaire includes 29-items focusing on feelings of hunger and appetite as well as taste
andsmellperceptionaddressing boththepresentsituationandtheperiodbeforeretirement.
Results: Significant differences were observed between the 3 health groups for appetite,
hunger feelings, present taste perception and present smell perception (P<0.05). Appetite and
hungerwerefoundtoberelatedtobodyweight inthehealthiest butnotintheothers.
Conclusion:Adecline inhealthstatus ispairedwith adecrease inappetite,hungerfeelings and
sensory perceptions ofelderlysubjects.
'submitted
Chapter2
Introduction
Poor dietary intake and weight loss as well as nutrition-related acute or chronic
illnesses in elderly people have been described in many cross-sectional as well as
longitudinal studies (1-6). Still the question remains what comes first: either the
disease or inadequate nutrition. Loss of appetite may be an early indicator of
'anorexia of ageing' leading to malnourishment, especially in the frail elderly (7) (8).
Therefore itseems importanttohaveagoodtooltoassess appetite intheelderly.
In investigating this 'anorexia of ageing' attention is often given to physical and
psychological determinants of appetite in elderly subjects (7) but only a few studies
havefocused onthe validity and reliability ofthe elderly 'self-assessment of appetite
and sensory perceptions (9;10). Different questionnaires have been developed but
they mainlyfocused oneating pattern orattitude andfrequency of eating rather than
onfeelings of hunger andappetite combined with sensory perception (11). Until now
only the Mini Nutritional Assessment gave slightly insight in appetite feelings (10).
However its major aim was to determine nutritional risks rather than self-evaluation
of hungerandappetite.
DeJong (12) developed asurvey-convenient toolfor estimating appetite and hunger
feelings as well as taste and smell perception in the elderly: the Appetite, Hunger
and Sensory Perception (AHSP) questionnaire. This tool wasfirstly tested against a
taste Perception test and a smell identification test based on the Connecticut
Chemosensory Clinical research center (CCCRC) test (13) and secondly more in
depthagainst energy intake (9). Results showed arelatively high internal validity and
a good reliability ofthis tool compared to the other outcome measures and provided
accuratedescriptive dataonelderly self-assessment ofappetite.
In the present study we further examined this newly developed tool: the Appetite,
Hunger and Sensory Perception (AHSP) questionnaire. The objective ofthe present
study was to determine whether or not this questionnaire would be able to
distinguish differences of feelings of appetite and hunger and sensory perception
between three elderly categories differing in health status: -healthy independently
living,-frail independently livingand-nursing homeresidents.
Methods
Studypopulation
Data used for the present analyses were derived from three different studies: two
intervention studies and one observational study. For all subjects, general
characteristics were collected on age, gender, number of diseases, and dental
status.
24
Self-assessmentofappetiteintheelderly
SENECA population:free-livingwithout help

In 1999, the final part of the Survey in Europe on Nutrition and the Elderly, a
Concerted Action (SENECA) study took place (14). In7countries participating inthe
final study, the protocol included the AHSP questionnaire to give indications on
appetite self-assessment in successful agers. Data for the current analyses were
collected from a random age-stratified population of elderly men and women
including 84 Dutch subjects. All subjects were recruited in 1988-89 using the
following criteria: born between 1913and 1918,free living inatraditionaltownwitha
stable population of 10000to20000 inhabitants. Fromthe results ofthefirst andthe
second Seneca study itappears thatthese subjects could beconsidered as 'healthy'
elderly (2;14-16).
Frailelderly population:free-livingwith help

For the frail elderly group, baseline data of 190 elderly subjects who started a
randomised placebo controlled intervention trial described elsewhere were used in
the present study(17;18). Frailty selection criteria to recruit these subjects were
partly based on criteria defined by (19)) i.e., Including the need of requirement of
help services at home such as home care and meals onwheels, a BMI (body mass
index)<25kg/m2 or presenting recent involuntary weight loss, no regular practise of
physical activities of moderate to high intensity, no use of vitamins and mineral
supplements and no dementia. Subjects were recruited in Wageningen and
surrounding municipalities.
Nursing homeelderly residents: nursing home

For the nursing home elderly, subjects included were recruited for a nutritional
intervention study at the nursing home Rustenburg, Wageningen, the Netherlands.
Dutch elderly people are referred to a somatic ward of a nursing home when,
because ofdiseases, a person cannot take care of his/herself and isthereby unable
to function at home (20;21). Selection criteria were: being older than 65 y of age, no
knowndementia,noknowndepression (assessedwithascore below5ontheGeriatric
Depression Scale(22)),nodiseaseinterminalphase.
Anthropometry
Bodyweight:
For all subjects' body weight, as index of the nutritional status, and height were
measured in the early morning after voiding with subjects dressed in light clothing
and without shoes. For SENECA and nursing home subjects, it was realized to the
nearest 0.5 kg using a calibrated mechanical weighing scale (Seca, Hamburg,
25
Chapter2
Germany). Body weight of the frail free-living elderly was measured to the nearest
0.01kg using a calibrated digital weighing scale (ED6-T; Berkel, Rotterdam, The
Netherlands).
Body height
Standing heightwas measuredfor allfree-living subjects tothe nearest 1mmusinga
wall mounted stadiometer. For nursing home subjects body height was estimated
from knee-to -floor height. The knee-to -floor height (KFH) was measured twice by
a single trained observer with a stadiometer in a sitting position, from the anterior
surface of the thigh to the floor with the ankle and the knee each flexed at a 90°
angle against the metallic help. Body height was derived usingthe following formula
(23):
Height (incm) =3.16*KFH (incm)
Appetite, HungerandSensoryPerceptionquestionnaire (AHSP)

The AHSP questionnaire includes 29-items focusing on the main determinants of
energy and macronutrient intake:feelings of hunger and appetite as well astaste and
smell perception addressing both the present situation and the period before
retirement.After reading each itemtogether with atrained interviewer, subjects hadto
score on a 5-point Likert scale with verbally labelled categories (see appendix). Five
variableswereinitiallycalculated:
- Presenttasteperception:8items,range8-40
- Presentsmellperception:3items,range3-15
- Presentsmellperceptioncomparedtothepast:3items,range3-15
- Appetite:6items,range6-30
- Dailyfeelingsofhunger:9itemsrange9-45
The lowerscores onthesevariables indicated a lowself-perception i.e. deteriorationof
the item beingjudged. For the variable appetite, presenttaste perception and present
smell perception compared to the past a lower score also indicated that these items
had deteriorated compared with the time before retirement. A higher score
corresponded to a positive self-perception of these attributes such as for instance a
betterappetiteormorefeelingsofhunger.
Data analyses
To determine if healthwould be amajor influencing factor, subjects were grouped in
3 health categories according to residence and requirement of health care: nursing
home,free-living with help andfree-living without help. Means ±standard deviations
(SD)were calculated pergroupforthe mainoutcomevariables.
26
.Self-assessmentofappetiteinthe elderly
Differences in general characteristics between the three groups were assessed

using a general linear model procedure
The internal validity and consistency of the AHSP questionnaire was tested through
the use of Cronbach's<x.
To evaluate the effect of health on the AHSP variables a multiple analysis of
variance was conducted using a model in which health was adjusted for age and
gender.
Spearman correlation coefficients were calculated to quantify the association
between body weight as main indicator of nutritional status and the variables of the
AHSP questionnaire. Gender differences in the AHSP variables were assessed in
the combined groups by using an unpaired t-test.
A p-value < 0.05 was considered statistically significant. Data were analyzed using
the SAS program (24).
Results
General characteristics of the study population
Table 1shows the general characteristics of the study population. Age, disease and
body weight differed per group (P< 0.05) while use of dentures was similar. Gender
distribution also differed between groups (P=0.001). Percentages of men were 47%,
25% and 20% of men for the 'free living without help', the 'free-living with help' and
the nursing home categories, respectively.
Table1:Generalcharacteristicsofthe3categoriesofDutchelderlysubjects
Total Free-living Free-living Nursing
Variable population without help withhelp home
(n=316) (n=60) (n=190) (n=66)
Age [mean(SD)y] 80.2(5.3) 83.1(1.6) 78.3(5.5) 83.0(4.5)
Gender [men/women] 88/228 28/32 47/143 13/53
Bodyweight [mean(SD)kg] 68.2(11.2) 72.8(11.3) 66.2(9.6) 70.1(13.8)
Disease (%) 89 79 91 100
1
Denture: Partialorcomplete 87 - 85 92
None 13 - 15 8
1
. Data not available fortheSENECA study
AHSP questionnaire results

The internal consistency of the questionnaire was relatively high with Cronbach's <x
varying from 0.71 for the variable appetite and present taste perception to 0.76 for
both smell perceptions (Table 2).
27
Chapter2
Results of the multiple analyse of variance showed significant different health effects
for the present taste perception [F (315,30)=1.8, P=0.008], daily feelings of hunger [F
(315,30)=1.93, P=0.003], appetite [F (315,30)=1.59, P=0.03] but not for the present
smell perception compared to the past [F (315,30)=1.03, P=0.43], nor for the present
smell perception [F (315,30)=1.30, P=0.14].
Table2: Meanscores of theAppetite, Hunger and SensoryPerception questionnaire in the

threecategoriesofelderlysubjects
Cronbach Total Free-living Free-living Nursing
Variable
coefficient population without help with help home
(PossibleRange)
a (n=316) (n=60) (n=190) (n=66)
Appetite 0.71 20.7±4.6 22.1±3.9 A
20.7±4.3 B
19.4±5.6B
Daily feeling of hunger 0.75 34.3±6.1 36.1±5.8 A
34.2±5.7 AB
33.2±7.2B
Presenttaste perception 0.71 26.0±5.0 27.4±4.5A 26.3±4.4A 24.0±6.4B
B
Presentsmellperception 0.76 11.5±2.4 12.2±2.5A 11.4±2.4 11.3±2.3B
Presentsmellperception
comparedtothepast 0.76 8.5±2.2 9.0±2.3A 8.3±2.2A 8.5±2.2A
Different letters (A, B)represent significant difference betweengroups assessed witha
multiple analysis ofvariance, P<0.05.
Table 3 shows the association between body weight as indicator of the nutritional
status and the variables of the AHSP questionnaire. These associations differed per
group. Relatively high correlation coefficients were observed in the free-living group
without help for appetite, daily feelings of hunger and present taste perception.
Table3: Spearman correlation coefficients between body weight as indicator of nutritional

statusandtheAHSP variablesforthe3categoriesofelderlysubjects
Total Free-living Free-living
Nursinghome
population receivingno receiving help
Variable (n=63)
(n=316) help(n=55) (n=190)
r P r P r P [ P.
Appetite 0.157 0.006 0.409 0.002 0.143 0.04 -0.078 0.54
Dailyfeelingofhunger 0.134 0.01 0.315 0.02 0.031 0.67 0.103 0.42
Presenttaste perception 0.158 0.005 0.351 0.008 0.221 0.002 -0.135 0.29
Present smellperception 0.145 0.01 0.151 0.27 0.110 0.13 0.147 0.25
Present smell perception
comparedtothepast 0.066 0.25 0.146 0.28 0.079 0.27 -0.118 0.35
28
.Self-assessmentofappetiteintheelderly
In the free-living group with help lower correlation coefficients were found. Only
appetite and present taste perception had a significant correlation with body weight.
Nosignificant correlationwasfound inthe nursing homecategory.
Women generally scored lower (P<0.05) than men for appetite, present taste
perceptionanddailyfeelings ofhunger butscored similarlyforsmellperception.
Discussion
Results of the present study revealed that health status influences self-assessment
ofappetite,hungerfeelings andsensory perception inDutchelderly.
The internal consistency of the AHSP questionnaire explored by Crohnbach's alpha
was satisfactory and similar to that previously observed (9). Due to the lack of
biological markers for appetite in the elderly, external validity of this questionnaire
could only be performed against external measures such as taste score or dietary
intake (9;12). Inthe present study, none ofthe subjects haddifficulties to understand
andanswerthequestions. Ingeneral,thetime neededto completethe questionnaire
was much lessthan ten minutes,whichwas acceptable inall categories of subjects.
Results of the present study show that the AHSP questionnaire, especially
developed for elderly subjects has a good feasibility and reliability to detect
differences infeelings of hunger and appetite inelderly subjects with different health
status included inourstudies.
According to the cross-sectional findings obtained in 1989, SENECA participants
were considered to be ina better health status than average (14). Longitudinal data
collected in 1993confirmed thatthesurviving sample ofthe population was relatively
healthy (15;25). In this paper their results on the AHSP questionnaire have been
used as a reference. These results show that in a healthy Dutch elderly population
appetite and hunger feelings are still indicators of body weight, one of the principal
'markers'ofelderly dietary intake. Howevertheir olfactory perceptionwas not related
to body weight. Since it is known that a poor smell perception (12) is related to a
poorappetite,apoorer smellperception could beoneofthefirst indicators predicting
decline inpleasantness offood andperhaps subsequent alteration indietary intake.
Scores of the frail free-living group requiring health care were intermediate between
the other two groups. Data of this group 'at risk' selected for its (unconscious) frailty
i.e. diminished physiological reserves and sub-optimal health status but no obvious
signsofmalnutrition showed noclear relationship between appetite andbodyweight.
One of the possible explanations might bethat, inthis group,feelings of hunger and
appetite were less important than symptoms of disease as determinants of body
weight andmaybenutritionalstatus.
29
Chapter2
Inthe nursing home group no association could be established between feelings of

hunger or appetite and nutritional status. Nursing home subjects, known as
accelerated agers (26), have a low dietary intake, a low physical activity level and a
relative poor health puttingthem highly at risk for malnutrition. Since depression has
been cited as one of the major cause of failure to thrive and weight loss in nursing
home (27), we selected a population who did not have depression in order not to
bias our results. Still the discrepancy between appetite and body weight confirmed
the fact that a poor health together with a lack of autonomy would negatively
influence appetite and thereby dietary intake.Another possible explanation is that in
this population health /disease conditions more than appetite is an indicator of body
weight. Besides, the rhythm of a nursing home day is often based on the meal
service. In this situation, this elderly group highly probably eats because they are
served but not because they are hungry or enjoy it anymore. This points out the
importance of non-physiologic parameters such as environment, meal ambience for
appetite and hunger feelings and subsequent dietary intake inthis elderly population
(28).
Until this study, it was often postulated, but actually never determined, that nursing
home elderly residents or frail elderly would have a more reduced appetite than
healthy free-living would. In this regard the present study confirmed the findings of
DeJong (1999) inaDutchelderly population aswellasthose of Hininger (2000) (29)
observed in a French elderly population that a deteriorating health together with a
reduced autonomy negatively affect self-assessment ofappetite and hungerfeelings.
Further long-term studies are now required toconfirm andfurther explore this lossof
appetite feelings in relation to body weight changes and specific markers of
nutritionalstatus.
In summary, our results show that appetite is related to the health status of elderly
subjects. In case of a good health, appetite was shown to be an indicator of body
weightwhilewithaweakening healthstatus,thestateofdisease ratherthan appetite
influences bodyweight.
References
factors in629 menandwomen aged over 65. Eur JClin Nutr 1996;50:6-11.
2. de Groot LC, Hautvast JG ,Van-Staveren WA. Nutrition and health of elderly people in Europe: the
EURONUT-SENECA Study. Nutr.Rev. 1992;50:185-194.
3. Moreiras O, Van Staveren W, Cruz JA, Nes M, Lund-Larsen K. Intake of energy and nutrients.
Euronut SENECA investigators. Eur J Clin Nutr 1991;45 Suppl 3:105-119.
4. Payette H, Gray DK, Cyr R, Boutier V. Predictors of dietary intake in a functionally dependent
elderly population inthe community. Am J Public Health 1995;85:677-683.
5. Roberts SB, Fuss P, Heyman MB,et al. Control offood intake inolder men. JAMA1994;272:1601-
1606.
30
.Self-assessment of appetite in the elderly
Br.Med.J.CIin.Res.Ed. 1986; 292:867-868.
7. Morley JE.Anorexia ofaging: physiologic and pathologic. Am JClin Nutr 1997;66:760-773.
8. Mowe M, Bohmer T. Nutrition problems among home-living elderly people may lead to disease and
hospitalization. Nutr Rev 1996;54:S22-S24
9. de Jong N, Chin a Paw J,de Graaf C, Van Staveren W. Effect of dietary supplements and physical
exercise on sensory perception, appetite, intake and body weight in frail elderly? Br J Nutr 2000;(in
Press)
10. Vellas B, Garry P, Guigoz Y. Mini nutritional assessment (MNA): research and practice in the
elderly. 1ed. Karger, 1999.
11. Shahar D, Shahar A. CEBQ: Composite eating behavior questionnaire -development and
performance. J Nutrition, Health &Aging .1999;3:11-18.
12. de Jong N, Mulder I, de Graaf C, Van Staveren WA. Impaired sensory functioning in elders: the
relation with its potential determinants and nutritional intake [see comments].
J.Gerontol.A.Biol.Sci.Med.Sci. 1999;54:B324-B331
13. Cain WS, Gent JF, Goodspeed RB, Leonard G. Evaluation of olfactory dysfunction in the
Connecticut Chemosensory Clinical Research Center. Laryngoscope 1988;98:83-88.
14. de Groot L, Van Staveren W. Description of survey towns and populations. Euronut SENECA
investigators. Eur J Clin Nutr 1991;45 Suppl 3:23-29.
15. de Groot C, Van Staveren W, Dirren H, Hautvast JG. Summary and conclusions of the report on
the second data collection period and longitudinal analyses of the SENECA Study. Eur J Clin Nutr
1996;50Suppl 2:S123-S124
16. de Groot LC, Beck AM, Schroll M, Van-Staveren WA. Evaluating the DETERMINE Your
Nutritional Health Checklist and the Mini Nutritional Assessment as tools to identify nutritional
problems inelderly Europeans. Eur.J.Clin.Nutr. 1998;52:877-883.
1-191. 1999. PhD thesis, Department of Human Nutrition & Epidemiology, Wageningen University,
The Netherlands
18. Chin a Paw, JMM. Aging in Balance, physical exercise and nutrient dense foods for the
vulnerable elderly. 1-176. 10-19-1999. PhD thesis, Division of Human Nutrition and Epidemiology;
Wageningen University, The Netherlands.
19. Chin a Paw J, Dekker J, Feskens E, Schouten E, Kromhout D. How to select a frail elderly
population? a comparison ofthree working definitions. JClin Epidemiol 1999;52:1015-1021.
20. Ribbe MW, van-Mens JT, Frijters DH. [Characteristics of patients during their stay in a nursing
home andatdischarge]. Ned.Tijdschr.Geneeskd. 1995;139:123-127.
21. Ribbe MW, Ljunggren G, Steel K, et al. Nursing homes in 10 nations: a comparison between
countries and settings. Age Ageing 1997;26 Suppl2:3-12.
22. Yesavage JA. Geriatric Depression Scale. Psychopharmacol.Bull. 1988;24:709-711.
23. Berkhout AM, Cools HJ, Mulder JD. [Measurement or estimation of body length in older nursing
home patients]. Tijdschr.Gerontol.Geriatr. 1989;20:211-214.
24. SAS Institute Inc. SAS/Stat user's guide version 6. fourth ed.Cary, USA: SAS Institute Inc., 1989.
of elderly Europeans. SENECA Investigators. Eur JClin Nutr 1996;50 Suppl2:S9-15.
26. van der Wielen RP, van Heereveld HA, de Groot CP, Van Staveren WA. Nutritional status of
elderly female nursing home residents; the effect of supplementation with a physiological dose of
water-soluble vitamins. Eur J Clin Nutr 1995;49:665-674.
1994;42:583-585.
31
Chapter2_
28. Elmstahl S, Blabolil V, Fex G, Kuller R, Steen B. Hospital nutrition in geriatric long-term care
medicine. I. Effects ofachanged meal environment. Compr.GerontolA 1987;1:29-33.
29. Hininger I, Mathey M-F, Maugourd M-F, Sidobre B, Ferry M. Les modifications du comportement
alimentairedes personnes agees en fonction de I'environnement et de I'age: legout, I'odorat, I'appetit.
Dietecom 2000, Paris, France.2000.
32
3
Social and physiological factors affecting food intake in elderly
subjects:anexperimentalcomparativestudy*
Marie-FrancoiseA. M. Mathey, Elizabeth H. Zandstra, Cees de Graaf and

Wya A. van Staveren
Abstract
The decline in average food intake in elderly people is attributed to both physiological and
social factors. These factors are usually studied in isolation. The present study concerns an
experiment inwhich the effect of social setting on food intake is compared with the effect of
physiological challenges onfood intake in24elderly subjects (6men and 18women,age:75+
4.9 y, BMI=26.6+3.5 kg/m2).
Physiological effectswere assessed usingapreload-test-meal designwithanoload,0kJ;and
4 preload conditions: lowcarbohydrate/low fat, lowenergy, 0.4 MJ;highfat, lowcarbohydrate,
medium energy, 1.1 MJ; high carbohydrate, low fat, medium energy 1.1 MJ; high fat, high
carbohydrate, highenergy, 1.8 MJ.The preloads consisted of 300g of strawberry yogurt drink,
and were served at 10:00 A.M. The test-meal, served 90 minutes after the preload
consumption,was alunch ofwhich subjects couldeat ad libitum.Social effects onfood intake
were assessed by using two social settings at lunchtime: cozy and non-cozy. Dependent
variables were food intake at lunch and ratings of appetite assessed before the preload, and
between preload andtest-meal.
Results showed that energy intake at lunch was significantly decreased after the high
carbohydrate preload andthehighfat-high carbohydrate preload (intakecompensation of23%
and 15%, respectively),comparedtothenopreloadcondition.Theotherpreloadsdidnothave
a significant effect on food intake. Energy intake was of 2.5±0.5MJ in the cozy social setting
and of 2.5±0.6MJ the non-cozy one. Appetite feelings were generally lower after the preload
conditions compared tothe noloadcondition (P<0.05), buttherewere nosignificant effects of
the macronutrient orenergy content ofthe preloads on appetite feelings. It is concluded within
the context of this study in healthy elderly subjects that food intake responds more to
physiological challenges thantoshort-term changes insocialsettings.
*Food, Quality &Preferences 2000,11(5):397-403

Chapter3
Introduction
Aging isfrequently associatedwith lossofappetite,decline infoodandenergyintake.
Longitudinal studies inelderly confirmed this spontaneous decline in both energy and
nutrient intake with age, this effect being paralleled by weight loss (1-4). As
requirements intheelderly remain unchangedorincreasefor mostofthe nutrients(5),
elderly areathigher riskfor malnutrition andnutrition-related acuteorchronic illnesses
(6).
The diminished ability of older adults to control food intake, often described as
anorexiaofaging, islikelyto betheconsequence ofseveralinternal,physiologicaland
external, social factors as mentioned by Morley and Silver (7). With increasing age,
food intake is thought to be influenced by alterations of the gastrointestinal tract and
by functional disabilities (8-10). The results of two studies suggested that elderly
subjects present an impaired ability to regulate their intake when under- or overfed.
Thesestudies indicatedthatelderly subjectswere lesscapablethanyounger subjects
tocontroltheirfood intake(11;12).
Aging is also accompanied by various social changes. Psychological and socio-
economic problems such as depression as a result of certain life events and
loneliness,mayreducefood intakeandcontributetotheriskofundernutrition(13-15).
In young subjects, the social settings of meals or snacks have been shown to play a
role on eating behavior. Redd and de Castro (1992) showed that, in real life outside
thelaboratory,foodintakewas60%higherwhenmiddle-aged subjectsatewithothers
presentthanwhenthey ate alone.These results suggest that social facilitation during
meals has acausal influence on eating behavior and that the amount eaten in meals
by adults is a power function of the number of people present (16). Feunekes etal.
(1995) showed that social facilitation of spontaneous meal size in young adults was
mediated by meal duration. However, another recent study revealed that, in young
subjects,physiological cues,andthusthe control offood intake, remain unchanged in
different socialsettings(17).
Inelderly subjects,changes inmealenvironment havealso beenfoundto result inan
increase inbothenergyand nutrient intake (18). Dataof DeJongandcoll. (1996)also
showed that elderly subjects ingested 50-70% more energy when having breakfast
together inadining roomthanwhenhavingregular breakfastontheirown.
Although it isclear that both psychological and socialfactors play an important role in
food intake in elderly people, these factors have mostly been studied in isolation. In
the present study, we investigated the influence of both social and physiological
factorsonfood intakeinhealthyelderlysubjects.
34
Elderlyandfoodintake
Inorder to meet the aim of the study, both social and physiological manipulations were
combined. For the social manipulation, two types of settings at lunchtime were used.
To compare physiological factors in those different settings, preloads that varied in
macronutrients and energy contents were used. Preloads and lunch were separated
by a delay of 90 minutes in order to assess the macronutrient effect of the preload
(19;20).
Rolls et al. (1991) reported that the time interval between preload and subsequent test
meal may affect the degree of compensation; a nearly perfect complete compensation
was shown in young adults after 30 minutes, and a compensation of 6 1 % to 90% was
found after 90 minutes. In the present study we chose a 90-minute time interval
because we believe that a 20-minute interval does not measure the physiological
effects, but only volume and weight effects of the preload, whereas a longer time
interval includes both post-ingestive and post-absorptive physiological effects
especially in elderly subjects. Besides digestion of regular amounts of fat and
carbohydrate results in the largest difference after a time interval of 60 to 90 min in
appetite physiological responses (21).
Subjects and methods

Subjects
Twenty-five (25) free-living elderly men and women (Table 1) participated in the study.
They were recruited by advertisements and byword of mouth throughout the town and
the surroundings of Wageningen (The Netherlands). Excluded from participation were
those who were taking medication known to influence food intake, or those who had a
body mass index (BMI) below 21 kg/m2, since risks for frailty are higher in subjects
with a low BMI. Subjects signed an informed consent before entering the study.
Table1: Subjectsgeneralcharacteristics(n-24)
Men/Women 6/18
Age (y) 75.5(4.9)
Weight (kg) 72.4 (8.9)
Heic|ht(m) 1.65(0.1)
BMI (kg/m2) 26.6(3.5)
Restrainedscore 2.9 (0.8)
Results are given as Mean (SD); BMI (kg/m2): Body Mass Index. The restrained score was
measuredwiththe helpoftheDutch Eating Behavior Questionnaire (34).
35
Chapter3
The exact study purpose was blinded to the subjects: they were informed that the
study concerned the sensory properties ofthe yogurts and the test-meal items. They
were not aware that their food intake was measured. One subject dropped out
because ofsickness duringthe study. Intotal, 24subjects successfully completed the
experimentalperiod.
The study was conducted at the Division of Human Nutrition & Epidemiology of the
Wageningen University,withapprovalfromitsMedical EthicalCommittee.
Design
For the assessment of physiological effects, a 'preload-test-meal' design with covert
macronutrient manipulation was used. The social effect was measured by changing
the socialsetting ofthe 'test-meal' (lunch). Eachsubject participated in5preloads x2
social settings of test-meal = 10 conditions, plus an additional test day to familiarize
thesubjectswiththeexperimentalprocedurebutalsotogivethepossibilitytosubjects
to getto know each other. During this test-day, we noticed that some people already
knew each other: most of them were friends, neighbors or even family (husband and
wife,cousins).Allexperimentaldayswereseparatedbyat leasttwodays.
This study design resulted inwithin-subjects repeated measures with each subject as
hisowncontrol.
Table2: Macronutrientandenergycontentsofthepreloads
LF/LC HF/LC LF/HC HF/HC
Energy (MJ) 0.4 1.1 1.1 1.8
protein (g) 10 8 9 8
%totalenergy 45 12 14 7
fat(g) 0.2 22 0 20
carbohydrate (g) 13 12 57 57
LF/LC:lowfat- lowcarbohydrate preload,HF/LC:highfat- lowcarbohydrate preload,LF/HC:
lowfat-highcarbohydratepreload,HF/HC:highfatandhighcarbohydratepreload.
Preloads
The preloads consisted of four strawberry yogurt drinks (300g each) with covert
manipulation ofenergy,fatandcarbohydrate contents butmatchedforweight,volume
36
and sensory properties by using fat and sugar replacers. Sensory properties were
tested in a pilot-study and no significant differences between preloads were found.
The preloads were either rich or low infat and either rich or low in carbohydrate and
they had similar absolute protein contents (Table 2). They were prepared within one
dayofservingandwererefrigerated.
Table3: Energyandmacronutrientcontentsofthefooditemsservedatlunch
Items KCal(/100g) KJ(/100g) Protein(g) CHO* (g) Fat(g)
Brownbread 222 886 8.8 40.5 2.8
White bread 260 1101 8.1 52 2.1
Margarine 741 3048 0.1 0.7 82
Peanut butter 648 2686 26.2 16.2 53.1
Jam 241 1023 0.2 60.0 0.0
Chocolate paste 321 1358 4.0 65.0 5.0
Cheese 376 1558 24.2 0.0 30.8
Ham 137 573 17.4 1.6 6.8
Milk 46 194 3.6 4.6 1.5
Orangejuice 37 157 0.5 8.8 0.0
Tea 0 0 0 0 0
Coffee 1 3 0 0.2 0
Apple 50 211 0.4 12.0 0.0
Orange 47 198 1 10.6 0
coffee creamer 110 464 7.7 10.5 4.2
Sugar 400 1700 0.0 100 0.0
*: Cho=carbohydrate
Test-meal
Thetest-meal consistedofabread-based lunch buffet consumed ad-libitum servedon
trays (Table 3). All trays were identical with prepackaged and preweighted foods for
each of the ten conditions. Subjects could choose between several items and were
toldtoeatanddrinkwhateverandasmuchastheywanted.
Socialsettingofthe test-meal
Lunchintake(=test-meal)occurredintwodifferentenvironments.
37
Chapter3
Inthe cozy setting, subjects were eating together in a chic restaurant-style decoration:
tables were dressed with tablecloth, place mat and flowers. Subjects were encouraged
to converse with each other to obtain an amicable atmosphere. For the non-cozy
setting, the same eating room was specially equipped with individual delimited
locations: subjects were alone and eating on a non-decorated table in a limited space.
The eating room decoration was also kept as undecorated as possible and subjects
were not allowed to communicate with each other.
In both settings, research assistants were helping with, amongst others unpacking
foods, serving coffee to subjects who askedfor it.
Procedure
The schedule of each experimental morning is described in Table 4. On each
occasion, participants were instructed to eat their usual dinner the day before the
experimental session. They were also required to eat a regular breakfast at 8:00 am
and to write down their consumption in a diary. Breakfast diary was used to check
compliance with these instructions but also to ensure that subjects would start each
test with asimilar state of hunger.
Subjects arrived at the department at 9:45 am. At 10:00 am, they filled in the first
Visual Analogue Scale (VAS, baseline). Then, they received one of the 4 preloads to
be consumed within 15 minutes or no preload in a randomized order. They were not
allowed to talk to each other during the preload consumption. They had to fill in new
VAS 15, 45, 75, 105 and 150 minutes after the first rating.
Table 4:ExperimentalSchedule (VAS=VisualAnalogue Scale)

Time Procedure
08:00a.m. Subjects consumetheir breakfast athome
09:45 a.m. Arrivalatthe Department of Human Nutrition
1stVASappetite assessment
10:00a.m. Subjectsdrinkthe preloads
10:15a.m. 2ndVAS appetite assessment
10:45a.m. 3rdVASappetite assessment
11:15a.m. 4thVASappetite assessment
.„ .r 5thVASappetite assessment
11:45a.m. , . , Y.. , .
andstartofthelunch
Endofthe lunch 6thVASappetite assessment
38
Inbetween preload andtest-meal,subjectsweretogether playinggames, reading but

were notallowedto discuss overthe preload orthe experiment itself, these last being
checkedbyresearchassistants.
Ninety minutes after the end of the preload consumption, they were offered thead-
libitumlunch in one of the two settings. The order of the settings was also randomly
assigned.
Measurements
Foodandenergyintake atlunch
Intake at lunch was assessed by determination of food wrappings and by weighing
back the food leftover. Consumed amounts were then converted in energy and
macronutrientscontents accordingtothe Dutch Nutrient Database(22).
Mealduration
Mealdurationwas recorded by researchassistants inbothsettings. Inbothcases,the
lunchwasconsidered finishedwhensubjects hadcompleted the lastVisualAnalogue
Scale.
Subjective appetite feelings

Appetite feelings were measured since they are considered as an intermediate
variablefor the physiological effect. During each experimental morning,subjects were
askedtoratetheirappetiteonaVisualAnalogueScale(VAS) insixrespects:'appetite
for a meal', 'appetite for something sweet', 'appetite for something savory', 'satiety
(fullness)', 'feeble,weak with hunger', 'appetite for asnack' (23). Each ofthese terms
was placed above the center of a 150-mm line, anchored at the left-hand and the
right-hand sides with the Dutch terms for "weak" and "strong", respectively. Subjects
wereinstructed bothorallyandonpaper.
Data analyses
Energy intake regulatory responsesforthe preloadswere calculated byassessing the
response to the energy content of the preload. The percentage of compensation at
lunchwascalculated asfollows:
%compensation =[(ELI noload-ELIpreloadx)/EPx]*100
with ELI= energy intake at lunch (kJ), x= one of the preloads and EPx= energy
contentofthispreloadx(kJ).
39
Chapter3
Statistical analyses
Effects of preloads and settings on intake were evaluated by analysis of variance
(a=0.05). The one-way ANOVA was carried out in stages. First, a repeated measure
analysis was conducted on all ten conditions with subjects, settings and preloads as
independent variables. When a parameter appeared not to be significant, it was
eliminated from the model and results were reported with only subjects and preloads
or settings as independent variables.
A two-way analysis of variance, excluding the no load condition, and with both
carbohydrate and fat with two levels (low and high), were realized to determine the
specific macronutrient effect of preloads on lunch intake (a=0.05).
Appetite rating forms (VAS) were read automatically by an Optimal Mark Reader, and
were transformed into scores from 1for weak to 25 for strong. Absolute ratings were
used to carry out repeated measure analyses bytime periods ( a =0.05)
Setting effects on meal duration were also assessed by a repeated measures analysis
procedure (a=0.05).
All statistical analyses were carried out with SAS statistical software package
(Statistical Analyses System; SAS Institute Inc., Cary, USA).
Results
Intake at lunch
No influence of social setting on lunch intake was found:the mean intake was of 2510
± 546 kJ inthe cozy setting and of 2526 ± 655 kJ inthe non-cozy one (P=0.68).
Table 5:Energycompensation atlunchafterdifferentpreloadcondition

Type of preloads
noload LF/LC HF/LC LF/HC HF/HC

Lunch + preload intake
26531550 2957±590 3575±613 3513±533 4217±550
(kJ)[mean iSD]
Lunch intake (kJ) [mean
26531550 25631590 2429+613 2392l533§ 2388±550§
±SD]
Preloadintake(kJ) 0 394 1146 1121 1829
Reduced intake(kJ) 90 224 261 265
Compensation (%) 23 20 23 15
LF/LC: Lowfat-low carbohydrate preload (^control), HF/LC:highfat-low carbohydrate preload,
LF/HC: low fat-high carbohydrate preload, HF/HC: high fat-high carbohydrate preload. §:
Energy intakesignificantly different, PO.05, comparedwiththe noloadconditionintake.
40
ElderlyandfoodIntake
Energy intake at lunch is presented inTable 5. When the data of both settings were
combined,asignificant effect ofthe preloads onenergy intakeattest-mealwasfound
[F(4,210)=8.8, P<0.001]. Compared to the no load condition, intake was significantly
reduced after the carbohydrate-rich (HC) and the fat- and carbohydrate-rich (HF/HC)
preloads. Contrast analyses (excluding the no load condition) showed a significant
effect of carbohydrate [F(1,210)=7.5, P=0.01] and aborderline significant effect offat
[F(1,210)=3.4,P=0.07]onenergycompensationduringlunch.
Mealduration
Mean test meal duration was significantly longer in the cozy setting (35± 4 minutes)
than inthe non-cozy one(27±6minutes) [P<0.001],andthis regardlesstothetypeof
preload.
VAS
rating
15 45 105 end
time after the p r e l o a d c o n s u m p t i o n ( m i n u t e )
Figure 1: Absolute 'appetite forameal'ratings afterdifferentpreload conditions: no load (+),

LF/LCm, LF/HC (O),HF/LC (±), HF/HC (it). Ratings assessedbyusing aVisualAnalogue
Scale were measuredbefore (t= 0min) andat15, 45,75,105min afterthepreloadsandafter
the lunch. The scale was a 150-mm line, anchored atthe left-hand and the right-hand sides
with the Dutch terms for "weak"and "strong", respectively. Data are mean values.
41
Chapter3
Appetitefeelings
Resultsfor satiety (fullness),appetitefor asnack,forsomething savory,for something
sweet, and feeble, weak with hunger were similar to appetite for a meal. Therefore,
only appetite for a meal is presented (Figure 1). Subjects started the experimental
morningswith asimilar state of hunger [F(4,209)=0.4, P=0.78].As shown in Figure 1,
subjects ratedtheirappetitefeelings higher afterthe noload condition [F(4,209)=12.0,
P=0.005]. Nosignificant differences were found after the different conditions pertime
onappetitefeelings.
Discussion
Results of this study suggest that food intake of apparently healthy elderly subjects
was not influenced bythe social environment ofthe lunch. Incontrast to expectations
(18;24;25), subjects did noteat more inthe cozy setting than inthe non-cozy one. In
addition, a physiological ability to regulate food intake at the test-meal was shown,
although subjects only compensated to a small extent for the differences in energy
intakeofthepreloads.
Although the energy intake was lower after all preloads as compared to the no load
condition, compensation at lunch was found to be slightly more influenced by the
macronutrientthan bythe energy content ofthe preloads. If appetite regulation would
be based uponthe energy content of food as postulated in other studies (1;20),than
energy intake at lunch would have proportionally decreased according to the energy
level of the preloads. I.e. results would have been similar for both high fat and high
carbohydrate preloads andmuchlowerfortheveryenergy-rich preload.This suggests
then that in this study healthy elderly would have still been capable to detect
macronutrient challenges. In other words, macronutrient would have a post-ingestive
effectonthecontroloffood intakeofelderly andmight influence morethe subsequent
energyintakethantheenergycontentoffoods.
Results of the VAS data indicated that preloads reduce more appetite feelings than
the no load, but there were no differences in between preloads. These results in
appetite feelings were strangely not confirmed by the energy intake during the
following meal: despite the fact that subjects started the lunch with similar states of
hunger after the preloads, they did not show similar energy intakes. In young adults,
the VAS isa good indicator ofwhat isthe food intake going to be (23). One possible
explanation why we did not find this in our study is that this difference between
physiological and VAS results might indicate an impairment in the ability to detect
nutrient density. This impairment of the physiological control of intake may reflect an
42
alteration of the endocrine system between the gastrointestinal tract and the appetite
center. Forexample,cholecystokinin, hormone release bythegut inresponsetofood,
and especially fat and protein ingestion, has been shown to decrease appetite.
Besides, CCK levels have been found to be elevated in older humans (14;26;27) so
these high levels combined with an increased sensitivity to CCK may lead to early
satiety inelderly people or decreased precision inelderly appetite control.This might
beoneofthe underlyingparameters partakingintheso-calledanorexiaofaging.
Inaddition results ofthe present study are based onasingle exposure tothe preload
inanimposedsocialenvironment. Since itisknownthatfood acceptance patternscan
be modified by post-ingestive consequences of food intake, further study with
repeated exposures are required.The optimal length ofthetime interval inapreload-
test-meal design remains a lively debate whether it aims at assessing satiety or
satiation effects of energy or macronutrient challenges in a culturally adaptedsetting.
In this regard, we believe that a 90-minute time interval has a greater validity and
reliability than ashorter one regarding Dutch usual eating patterns indaily life,where
Dutchelderly peopleareusedtoconsumeamidmorningsnack(22).
Severalstudiesshowedthat peopletendtoeat moreincompany ofothersthanwhen
they eat alone (16;28;29). This has often be heeded as an example of social
facilitation, defined to be the enhancement of a certain behavior due to the sheer
presenceofothers. Infact, apositive association has beenfound betweentheenergy
intake at an eating and drinking time and the number of people present at that time,
independently from the day or period of measurement. Therewith, the number of
people present has been suggested to have a causal effect on food intake mediated
by an increase in the meal duration. Those results were observed with young and
middle-aged adults intheir usual environment such asfamily or close friends (25;29-
31). However, this phenomenon was not observed inour study despite the significant
difference in meal duration showing that changes in social settings did have an
important influence(17).
There are a few possible explanations regarding the absence of effect of the social
environment onfood intake.Thefirst possibleexplanation isthat, inthe presentstudy,
lunches in bothsocialsettings were consumed atthe research institutewhich cannot
beconsideredasa'natural'eatingenvironment (30;32).Furthermore,the presentation
ofthefoodswas identical inboththe cozyandthe non-cozy environment. Inthemore
naturalistic studies (16;33) it seems probable that changes insocial settings of meals
wereaccompanied bychanges inthe presentation andserving offoods. Itis nowwell
43
Chapter3
known that the presentation of foods can have a dramatic effect on the food intake
(33).
Another explanation referred to the fact that subjects participating in this study were
apparently healthy volunteers. They were obviously not suffering from denture or
digestive disorders but also did not present any external sign of depression or
sadness, which would have made them more responsive to environmental changes.
Besides, the Dutch restrained eaters' scale (34) showed that this group was quite
weight conscious. This together with the fact that the day to day variation of energy
intake at lunch is rather low inthe Dutch adult population, lunch being mainly bread-
based meals (22),mightexplainwhyfood choice and energy intake at lunchwere not
affected by the changes in meal environment. The positive findings of Elmstahl and
coll. (1987) on food intake occurred by modifying the meal environment in homes for
elderly people,thus inanaturalenvironment of ingeneralfrailelderly people andover
alongerperiodwith repeatedexposures.
In conclusion, in our study physiological parameters have a stronger effect than
changes in social environment on appetite and energy intake in apparently healthy,
free-livingelderly. However, mealduration inthiselderly populationwas morestrongly
affected bymealenvironment inlinewithotherstudies inyoungsubjects.
Acknowledgment
WewouldliketothankHennyRexwinkel,MarloesvanderKampenBasDuijkersfortheirhelp
in collecting the data and Dr KJ Melanson for her precious suggestions during the data
analysis. We also gratefully acknowledged the " Holland Sweetener Company" (Maastricht,
TheNetherlands)fortheirdonationofaspartame.
References
1. DeGraaf C, Hulshof T,Weststrate JA, Jas P. Short-term effects of different amounts of protein,fats,
andcarbohydrates onsatiety. AmJClin Nutr 1992;55:33-38.
investigators. EurJClin Nutr 1991;45Suppl3:23-29.
3. Roberts SB, Young VR, Fuss P, et al. What are the dietary energy needs of elderly adults? Int J
Obes.Relat.Metab Disord.1992;16:969-976.
4. LehmannAB, Bassey EJ.Longitudinalweightchanges overfouryearsandassociated healthfactors in
629 menandwomen agedover65. EurJClin Nutr 1996;50:6-11.
5.van der Wielen RP,deWild GM, de Groot CP, Hoefnagels WH, Van Staveren WA. Dietary intakes of
energy and water-soluble vitamins in different categories of aging. J Gerontol A Biol Sci Med Sci
1996;51:B100-B107
6. Mowe M, Bohmer T. Nutrition problems among home-living elderly people may lead to disease and
44
7. MorleyJE,SilverAJ. Anorexia intheelderly. Neurobiol.Aging 1988;9:9-16.

8. Clarkston WK, Pantano MM, Morley JE, Horowitz M, Littlefield JM, Burton FR. Evidence for the
anorexia of aging: gastrointestinal transit and hunger in healthy elderly vs. young adults. Am J Physiol.
1997;272:R243-R248
9. Rolls BJ,McDermottTM. Effects ofageonsensory-specific satiety. AmJClin Nutr 1991;54:988-996.
10. Stubbs RJ. Nutrition Society Medal Lecture. Appetite, feeding behaviour and energy balance in
human subjects. Proc NutrSoc 1998;57:341-356.
11. Pfau C. Methodology of a survey on meal patterns in private senior households. Appetite.
1999;32:38-45.
12. Rolls BJ, Dimeo KA, Shide DJ.Age-related impairments in the regulation of food intake. Am J Clin
Nutr1995;62:923-931.
13.Goodwin JS.Geriatric ideology:the mythofthemythofsenility. JAm Geriatr.Soc1991;39:627-631.
14. Morley JE.Anorexia inolder persons:epidemiology and optimal treatment. DrugsAging 1996;8:134-
155.
15. Mulley GP. Preparingforthe lateyears. Lancet 1995;345:1409-1413.
16. DeCastro JM,Brewer EM.Theamount eaten inmeals byhumans isapower function ofthe number
of peoplepresent. Physiol.Behav. 1992;51:121-125.
17. Kim JY, Kissileff HR. The effect of social setting on response to a preloading manipulation in non-
obesewomenand men. Appetite 1996;27:25-40.
medicine. I.Effects ofachanged mealenvironment. Compr.GerontolA 1987;1:29-33.
19. Rolls BJ, Kim S, McNelis AL, Fischman MW, Foltin RW, Moran TH. Time course of effects of
preloads high in fat or carbohydrate on food intake and hunger ratings in humans. Am J Physiol.
1991;260:R756-R763
20. Rolls BJ, Hammer VA. Fat, carbohydrate, and the regulation of energy intake. Am J Clin Nutr
1995;62:1086S-1095S.
21. Maas Ml, Hopman WP, van-der-Wijk T, Katan MB, Jansen JB. Sucrose polyester does not inhibit
gastricacidsecretion orstimulatecholecystokinin release inmen. Am.J.Clin.Nutr. 1997;65:761-765.
22. Anonymous. NEVO 1996. Stichting Nederlands voedingsstoffenbestand. Dutch Nutrient Database
1996.The Hague,The Netherlands:Voorlichtingsbureau voordevoeding, 1996.
23. Hulshof T, De Graaf C, Weststrate JA. The effects of preloads varying in physical state and fat
contenton satietyandenergy intake. Appetite 1993;21:273-286.
24. de Jong N, de Graaf C, Van Staveren W. Effect of sucrose in breakfast items on pleasantness and
food intake intheelderly. Physiol.Behav. 1996;60:1453-1462.
25. Feunekes Gl, De Graaf C , Van Staveren WA. Social facilitation of food intake is mediated by meal
duration. Physiol.Behav. 1995;58:551-558.
26. Berthelemy P, Bouisson M, Vellas B, Moreau J, Albarede JL, Ribet A. Postprandial cholecystokinin
secretion inelderlywith protein-energy malnutrition. JAm Geriatr.Soc 1992;40:365-369.
27. Morley JE.Neuropeptides, behavior, andaging. JAm Geriatr.Soc 1986;34:52-62.
28. De Castro JM. Social facilitation of the spontaneous meal size of humans occurs on both weekdays
andweekends. Physiol.Behav.1991;49:1289-1291.
29. DeCastroJM. Socialfacilitation offood intake inhumans. Appetite 1995;24:260-26s.
30. De Castro JM. Family and friends produce greater social facilitation of food intake than other
companions. Physiol.Behav. 1994;56:445-445.
31. De Castro JM. How can eating behavior be regulated in the complex environments of free-living
humans? Neurosci.Biobehav.Rev 1996;20:119-131.
45
Chapter3
32. DeCastro JM. How can energy balance beachieved byfree-living human subjects? Proc Nutr Soc
1997;56:1-14.
33. Meiselman HR.The contextual basisforfoodacceptance,foodchoice andfood intake:the food, the
situation and the individual. In: Meiselman HR, MacFie HJH, eds. Food choice, acceptance and
consumption. London:Blackie, 1996:239-263.
34. van Strien T. The concurrent validity of a classification of dieters with low versus high susceptibility
towardfailureof restraint. Addict.Behav. 1997;22:587-597.
46
4
Healtheffect of improved mealambiance ina Dutch nursing home:
aone-year interventionstudy*
Marie-FrancoiseA. M. Mathey, Vincent G. G.Vanneste, Cees de Graaf,

Lisette C.P.G.M. de Groot and Wija A. van Staveren
Abstract
Objective: To determine the effect of an improved ambiance of food consumption on health
and nutritional status of Dutch nursing home elderly residents (n=38) in a one-year
intervention study.
Design: parallelgroup intervention study.
Intervention program: Improvement of ambiance focused on 3 points: 1. Physical
environment and atmosphere of the dining room, 2. Food service 3. Organization of the
nursing staff assistance
Measurements: Dietary intake, biochemical indicators of nutritional and health status and
quality of life (Sickness Impact Profile and Philadelphia Geriatric Center Moral Scale) were
assessed at baseline and after one year of intervention. Body weight, used as an indicator
of compliance and nutritional status,was assessed every four months.
Results: 22 subjects completed the one-year intervention trial. Mean body weight
significantly increased (+3.3 kg, p<0.05) inthe experimental group (n=12), not inthe control
group (-0.4 kg, p=0.78; n=10). Health status biochemical indicators and the SIP score
remained stable in the experimental group indicating relatively stable health conditions. On
the contrary, negative changes in the control group suggested a decline in health status.
Dietary intake,whichwas insufficient at baseline, increased in bothgroups.
Conclusion: This study showed that improving the ambiance of food consumption is anon-
negligible issue to improve nutritional status and to stabilize health of nursing home
residents.
•provisionallyaccepted
Chapter4
Introduction
In western countries poor nutritional status is highly prevalent in nursing home
residents (1-3).This poor nutritional status, caused by lowfood and energy intake, is
often associated with unintentional weight loss, a higher risk of morbidity and a lower
qualityoflife(4-7).
Food intake in nursing homes depends to a large extent on the quality of the food
service system (8). In most of the nursing homes, meals are individually served on
trays to the residents either in a common dining room or in their own room. This
serving system combined with an inadequate physical environment does not seem to
bethemostappropriatewaytostimulateappetiteandmealtimeenjoyment(9).
Since poor appetite is one of the factors for an insufficient diet in institutionalized
elderly, astimulating ambiance should beconsidered in mealservices. Earlier studies
inyounger peopleshowedthatsocialfacilitationofeatingwas relatedtoabettersocial
atmosphere during meal consumption (10-12). Changing the social ambiance of food
consumption in nursing homes might therefore be a realistic and effective way to
improve mealtime enjoyment and nutritional health. Since we believe that nursing
home residents are not capable anymore to make change to improve their eating
environment, the nursing staff isthen responsible for it. The idea behind itwas that if
thenursingstaffwouldprovideabetter mealserviceand improvetheambianceduring
food consumption, the residents would feel more comfortable and secure in their
eating environment. Therefore their appetite and food intake might be stimulated and
could resultinhealthbenefitonthelong-term.
Inthe present studywe investigated the influence of a 1-year intervention, combining
selectedsocialandenvironmental manipulations offood consumption onnutrition and
healthstatusofnursinghomeresidents.
Patientsand methods
Designandsetting
A parallel design was used for this one year intervention study conducted at the
nursing homeAeneas, Breda,the Netherlands. Dutch elderly people are referred to
asomaticward ofanursing homewhen,because ofdiseases, aperson cannot take
careofherself and isthereby unabletofunction at home (13). Fourwards,eachwith
15 residents and comparable for diseases and treatment, were randomly assigned
to beineither thecontrol (twowards) or intheexperimental group (twowards). Data
werecollected at baseline andaftertheintervention.
48
_Health effectofmealambiance in nursinghome
Patients
Sixty Dutch nursing home residentsfrom four somaticwards were invitedtojoin the
study. Both residents and their contact person received detailed information about
the study. The main inclusion criteria were to be older than 65 y of age and to be
resident in this home for more than 3 months at the start of the study. General
exclusion criteria were parenteral nutrition, terminal phase of a disease. A specific
exclusion criterion for the analyses of biochemical indicators of health was applied
forthe patientswithsevereanemia.
Beforethe start ofthe intervention,allvolunteers ortheir contact person hadto sign
an informed consent. The study protocol was approved by the Ethical Committee of
the nursinghome.
Theinterventionprogram
The intervention program, developed after discussion with nursing staff and patients,
aimed at creating a better ambiance during food consumption and focused on three
issues: physical environment and atmosphere ofthe dining room,meal situation, and
organization of the nursing staff assistance (Table 1). During the intervention, the
same mealswere served inbothgroups andthe usual meal patternwasmaintained:
breakfast and supper were bread-based meals and at noon a cooked meal was
served.
In the control group the original situation was kept. The dining room was not
decorated andthe 15 residents of award would eat inthe dining roomwith about 2
to4 nurses present. Inthisway attention would bepaid onlyto residents with known
eating disorders and residents did not consume their meal at the same time. The
cooked mealwas served on atray and the carers did stay inthe dining room during
mealtime. For the bread-based meal, residents got ready-to-eat sandwiches that
wereprepared inadvance bythe nursesthereby restrictingthedailyfoodchoice.
Measurements
Bodyweight
A fixed protocol was defined and used to assess patients' body weight as index of
the nutritional status every four months. It was measured before breakfast after
voiding (tothe nearest 0.1 kg,Secaweighing scale inwhichthe patients could sit in)
withthe patient dressed inlightclothingwithoutshoes.
49
en c "co as
CD CD
CD
o i
E
o 5 o
1 CD
CD Q.
T3 -o
TO sz
CD
as
CO
CO 3
p1 _ o
.52
TO c o CO
"> ,: • > c CD
a. SI o
o c
^oas O)
CD
O
c
g to CD
g
Q.
Q. 15
01 a. 'o
JtO «*—
SZ E o O o *—
.S >.
I 3
CO
c
O
CO
CD
.n
to
Q.
*^
!5
T3
CO
"co
CO
as
.c
E •w CD
o
o i_ CO
o o o A
CO
c 3
o CO CD CO
"c as 3 o
a> l _
E "co c
o O)
(D
o
O CD Q. a.
E o CO Q. • • - •
T3
i 3t c
C 1
•D CD
'T3
c OS cCO
(4 co
o T3 o
C T3 co
0) 3 c OS CD c CO
"a 'c a> XI as •o
C CO OS
c c E TO -Q CD c
a) o OS 3 as
CD O
CD CO Q. o SZ CD as as
a o a)
TO a. 3 Q.
CD 73
D) Q. x:
3 0) CD
£• sz CD C Q. "fis CD"
(0 O)
E CD a.
c 0) 'tz o CD
o > CO
CD T3 c a)
E it
o
CD -Q
C
i_
CD
T3 **-
T3
15 CD
c "O '•«-' Q. CD £ o
r- CD
CO
o CD
(/) CO >
o CD
SZ CD
CO o
O
c CD .c CO
OS
m E
C
c "5
t_
CD
><
CD o o CD
(0
CD
E 73 a.
SI o Q.
Q. co « co > i _
CO C OS 3 as
CO CD \£ CD o CO CO
CO 3 as as
it
>
c
g
c
1 E
C
1 * 8 OS
o
o
T3
CD
£
CO "g
>
OS
o 3 o t » CD •D
| 11 |
CD
CD *•— O 03 CO
£ c CO
3
i— TD
o OS CO O
CD C 03 <U CD ^ co S 5
c CO 2 CD as "o
(0
c
0!
0.
O
CD
•a
CO
3
=5
T3
C O
^ E
E ><
HOS
-C
CO F
—
O
CD
CD
2-
c
as
.1
1 «C°Q.D
O
m 4 0J 2 b CO c CD co ,9
>.
CD
i
0)
• • • • • • • • • •
c 1M E
o V)
o
oc
CO
o c T3
S
<u
g O
'i_
CD CD
o N
o 01
73 L. "c CO
CD OS
U) o sz as
3 | o
o
CD
c o O E
E o c
CO T3
i_
D)
.§ o
L_ a)
c
' c
CD
CO
CD
CD
O <fl o
Q. OS
*-<
0) CO
a> T3 as CD
Q Q
CD
CD 01
O 5 GO
SZ
•*••
C c
CD
H.
o t
o E
c
o
8 o
p o
0) k.
o <a c •c
CO a! S & 0) Q>
a> 3 c 10 .c 10
Q a>
(0 o .5
"5 •5 o
_o n 0) o
01 5> •o u.
^ c 73
H •c «J
Q.
0) T3
t/) C
1_
CD
3
C CD
CD
^
• - — "
O co :
(D CU
co
E re E
CU
CJ
(0
co CU
T3 CU CJ
a> nj
E c l _
E E .c CU
CO
CU
CO sz CO
O) CU
TO c 3 l _ L_
CD £
a> 3 c c
"O CO
3 "55 CU
CO
c E T3 it
o O
CO
re
CO
o "to
o
c
"CD
CU
o c
c CO CO
ID 'c E
CD 73 CU
c
o 3
> a) '•*-» c
nj CO
CU
" * • '
o D o
c .Q
o •c cr
CD o
a> *-
re "co c XI><
.Q
aj CU
"Jo re a)
-»-» O) £ CU "S5
0 c
CU E CU
£ CO
1_ E
o "co
it c »*— c
CU <u
S CU
to 3 re CU
en *•- O Q. >
CU
C o TJ .c CD
CO "a CU .•ti c
"O
3 c
c
5 ' l _
C
o re CU o
In 1_ . c o 'c
•*-»
CD
CO
c o
CU
D
c
<D
cn
c
c
r> CU
E
CD 13 1* E
CD CU
CL re O CO -£± CD
o •— — C
CO CN 5 TO CO •— CJ)
a> i _
o <U CU o o
ir •2 Z 2 E z a.
• • • • •
cu
.c
**
H—
o
c
o
'•4-rf
c
CO
o CU N
O
TO
N c ^
at CD
C "O
co "co c
u> CO CO
CO
O CO
w
!fc
to
<w
10
Ol
CU
D ,c
jo
t.
C 3
o •2
o
CU
Chapter4 _
Dietaryintake
Dietary intake data were collected by project dietitians using both observation and
weighing-back methods (14) and lastedthree consecutive days including aweekend
day. For bread-based mealsfood intakewasfollowed bymeans of precise individual
weighing of foods before service and left-overs up to an accuracy of 0.1 g. For the
cooked meal, individual menus and recipes of the 3 days of dietary record were
obtained in advance from the kitchen. Food consumption at the cooked meal was
registered by keeping records of chosen foods and amounts served.After themeal,
waste was weighed.All foods and beverages consumed outside regular mealtimes
were also carefully recorded. Portion sizes were derived from a Dutch table of
regularfood portion sizesand household units (15). Dietary datawere converted into
nutrients usingacomputerized versionofthe Dutchfood compositiontable(16).
Biochemicalindicatorsofhealth status
Fasting blood samples were taken at baseline and at the end of the intervention for
hematological analyses by antecubital venipuncture with the patient in sitting
position (one 5 ml EDTAtube, one 5mlgeltube). Hemoglobin and leukocytes were
analyzed infresh EDTA blood on a Technicon hematology H1 analyzer (Technicon
Instruments Co.,Tarrytown, NY, USA).Thegeltubewascentrifuged andfrozen at
-75°Cto beanalyzed inonerun.
Qualityoflife
Quality of life was assessed by combining the Dutch versions of two validated
questionnairestodetect possiblechanges inhealth related behavioralandfunctional
status. The first questionnaire, a part of the Sickness Impact Profile 68 (SIP, 17
items) focusing on self perceived physical autonomy was used as a behaviorally
based measure of health status. Together with the SIP, a Dutch version of the
Philadelphia Geriatric Center Moral Scale (PGCMS, 17 items) focusing on self
perceived lifesatisfactionwas added (17;18).
Data analyses
For the residents who completed the trial, means ± standard deviations (SD) of
baseline and absolutes changes values were calculated per group for the outcome
variables. For the dropout group, only baseline characteristics as well as body
weight datawere analyzed.Changeswere compared by using an unpaired t-test for
differences between groups and by using a paired t-test for difference between
baseline and follow-up within groups. A p-value < 0.05 was considered statistically
significant. Datawereanalyzed usingthe SAS program (19)
52
_Healtheffectofmealambianceinnursinghome
Results
Patients baseline characteristics
Figure 1 present the study flow chart. From the sixty elderly invited to participate in
the trial, forty-two were volunteers and thirty-eight could be enrolled. After one year,
twenty-two could complete the whole study. Drop out cases (control, n=7 and
experimental, n=9) were patients who failed in completing the study mostly because
of death (75%), move (12.5%) or progressive diseases /personal reasons (12.5%)
Table 2 presents patients baseline characteristics. Gender distribution, age and
mean body weight were similar for control, experimental and drop out groups.
Patients were comparable with respect to quality of life, diseases and treatments. All
participants used medications (3.5±1.8 medications per individual and per day)
mainly prescribed for cardiovascular diseases, pain or digestive track disorders.
Table2:Baselinecharacteristicsofthestudypopulation
variable total populatio control group experimental drop out
n=38 n= 10 group, n= 12 n=16
Meanage (y)1 82.2(7.9) 78.2(7) 82.6(7.5) 84.8(14.9)
Gender (male/female) 13/25 3/7 4/8 6/10
1
Mean bodyweight (kg) 62.9(13.4) 63.9(12.5) 66.5(14.0) 59.2(10.8)
Dentures (% yes)1 80 80 84 82
1
Wheelchair (% yes) 90 80 92 95
Diagnoseddiseases (mean
persubjects)1 2.6(1.3) 2.3(1.3) 3(1.2) 2.6(1.4)
1
Drug use(mean number/day) 3.5(1.8) 2.9(1.7) 4.1 (1.7) 3.5(1.9)
2
SIP (range 0-100) - 42(19) 33(21) -
PGCMS (range 0-100)2 - 67(30) 58(16) -
Data areshown asgroup: (1) mean (SD) or percentage of 'yes' answered. (2) SIP [Sickness
Impact Profile] and PGCMS [Philadelphia Geriatric Center Moral Scale] were only assessed
for n=8incontrol group and n=8 inthe experimental group.
Body weight
Mean body weight significantly increased over one-year period in the experimental
group: +3.3 kg±5.0, p<0.05; n=12, while it stayed relatively stable in the control
group :-0.4±4.0 kg, p=0.78; n=10. These changes in body weight over one year
period differed significantly between groups (P<0.05).
53
Chapter4
Study population: 60 invitation letters
Inclusion criteria: Non-responders: 18

-Older than 65yof age
-Arrived inthewards more Volunteers: 42
than 3months ago
-No parenteral nutrition
-Noterminal phaseof
disease
Selection
1
Eligible: 38
1 ' •
Control group Baseline measurements Experimental group
N=17 N=21
Death: 1 +.
-•Death: 1
Dropout: 2-4-
N=14 N=20
After 4 months
Death: 1 Discharge:1
- > Death:3
N=13 N=16
After 8 months
^Discharge: 1
Death:3
->Death: 3
After 12 months
N=10 N=12
Completed trial
Figure 1: Flow chart of the nursing home residents for the one-year intervention study
54
_Healtheffectofmealambianceinnursinghome
Figure 2 shows the repartition per group for weight changes. Number of people
losing and gaining weight was similar in the control group. In the experimental group
a trend for weight gain was observed.
Dcontrol
10-
9-
M intervention
8- 7
7-
number
6- 5 5» B
4
of
subjects
5-
4-
P
3-
2- 1
1- o t^gM
0-
weight loss weight stable weight gain
Figure2:Numberofresidentslosingorgainingweightovertheone-yearinterventionperiod.
Body weight characteristics of the drop out subgroup are presented in table 3. Most
of the drop out residents presented weight loss inthe last 4 months.
Table3Characteristicsofthedropoutgroups: ageandbody weightatbaselineandabsolute

changesatagiventime (C;controlgroup,E:experimentalgroup)
Periodof
dropout 0-4 months 4-8months 8-12 months
Group C E C E C E
n 3 1 1 4 3 4
Age at baseline 85.7±12.2 91.0 95.0 84±11 82.7±4.5 83.3±9.8
Bodyweight:
Baseline 55.6±12.5 52.0 48.1 64.0±19.3 69.7±10.1 55.5±12.7
After 4 months - - 48.0 64.1±18.7 70.8±8.1 55.0±14.4
After 8months - - - 62.7±7.9 54.2±12.8
Dietary intake
Table 4 presents dietary intake and absolute changes after a one-year intervention
period for the residents who completed the trial. At baseline dietary energy intake
55
Chapter4
was low in both groups (5.4±1.5 MJ for the control group and 6.1±1.4 MJ for the
experimental group) and below the minimum Dutch requirement for nursing home
residents (6.8MJ/Day).
Table 4: Selected nutrient andenergy intake of nursing home elderly residents: baseline
valuesandabsolute changesafterone-yearintervention
Control group (n=10) Experimental g roup(n=12)
variable baseline Absolute baseline absolute
changes changes
Energy
kJ 5431 (1540) 767(1040)* 6055(1360) 819(1699)
kcal 1292(368) 185(247)* 1440 (324) 199(406)
Energy (kJ)/Weight (kg)a 87(26) 12(21) 96 (27) 4(27)
Carbohydrate (g) 167(44) 7(40) 189(46) -4(48)
%total energy 54(10) -6(5)* 53(6) -8(6)*
Fat(g) 48 (20) 14(12) 52(15) 21 (21)
%total energy 32(8) 5(5)* 32(5) 8(5)*
Protein (g) 47(16) 7(11) 52(15) 8(16)
%total energy 14(2) 0(2) 15(3) 0(3)
(g)/bodyweight (kg)a 0.7 (0.2) 0.1(0.2) 0.8 (0.3) 0.0 (0.3)
Alcohol 1(0) 0(1) 1(2) 0(2)
%total energy 0(1) 0(1) 0(1) -0(2)
Dietaryfiber (g) 10(4) 3(2) 12(4) 4(2)*
Calcium (mg) 756 (204) -115(147)* 893 (254) 32 (269)
Total iron (mg) 5.8 (2.4) 1.4(1.7)* 6.4 (2.5) 0.7 (3.0)
Vitamin B1(mg) 0.6 (0.2) 0.1 (0.2) 0.7 (0.3) 0.1 (0.4)
Vitamin C(mg) 55.5 (17.7) -9.3 (23.8)§ 45.0 (24.6) 22.3 (43.9)*
Totaltocopherol (mg) 9.5 (3.5) -2.4(2.1)*§ 9.5 (2.8) 3.5 (4.4)*
Resultsaregivenasmean(SD);a:n=11fortheexperimentalgroupandn=9forthecontrol
group. §: Significant difference between control and experimental group, (P<0.05), *:
Significantdifference inchangeswithingroupascomparedtobaseline(P<0.05)
As expressed per unit of body weight dietary intake was also lower than the
recommended adequate intake (120 kJ/kg body weight) with 87±26 kJ/kg body
weight and 96±26 kJ/kg body weight for the control group and the experimental
group, respectively. As aconsequence ofthe low energy intake, intake of vitaminA,
vitamin Bcomplex andvitamin Cwas belowthe Dutch minimum requirements.
At the end of the intervention period, no differences in energy and macronutrient
intakewereobserved between groups. Changes inintake ofvitamin CandvitaminE
56
_Health effectofmealambiance innursing home
differed significantly between the control and the experimental group. Despite a
significant increase,dietary intake inthecontrolgroupwasstill belowthe Dutch RDA
for bothenergy and nutrientintake.
Inthe experimental group, changes in energy intake showed a positive trend and a
significant increase in intake of vitamin Candvitamin Ebrought these values above
the RDAs. The increased intake in vitamin E was the consequence of an overall
increase in dietary intake and could not be related to a particular type of food.
Increase intake of vitamin C in the experimental group was mainly due to the
consumption of fruit juices and dairy products, which represented about 2.8 drinks
(or420g)per patient and provided about 85.7 %ofthevitamin Cintake.
Biochemicalindicators ofhealth status

Hemoglobin and its related factors stayed relatively stable inthe experimental group
while they decreased inthe control group (Table 5). Significant differences between
groups were observed for changes in hemoglobin and two hemoglobin-related
factors: MCH (Mean Corpuscular Hemoglobin) and MCHC (Mean Corpuscular
Hemoglobin Concentration).
Table 5:Biochemical indicators ofhealth status ofnursing home elderly residents: baseline
valuesandabsolute changesatthe endofone-yearintervention.
control( group experimental group
n baseline absolute n baseline absolute
changes changes
Hb(mmol/l) 8 8.15(1.20) -0.33 (0.43) *§ 10 8.24 (0.59) 0.18(0.68)
Ht(l/l) 8 0.40 (0.05) -0.02 (0.02) 10 0.42 (0.02) 0.00 (0.03)
RBC(*10E12/I) 8 4.45 (0.57) -0.00(0.41) 10 4.61 (0.45) 0.09 (0.29)
MCV(fl, 10E-15) 8 90.4 (3.6) -3.5 (4.2)* 10 90.8 (7.7) -1.7(1.9)*
MCH(amol) 8 1825(142) -66(113) § 10 1798(162) 8(48)
MCHC (mmol/l) 8 20.2 (0.4) 0.1 (0.4)§ 10 19.8(0.4) 0.5 (0.5)*
Leukocyte (10E9/I) 7 5.9(1.2) -0.8(1.3) 10 8.0(1.0) -0.9(1.2)*
Thrombocyte (10E9/I 7 234.9 (32.7) 25.4 (39.0) 9 280.8 (86.8) 12.7(50.7)
Results are presented as mean (SD); §: significant difference between control and
experimentalgroup,(P<0.05);*:significantdifference inchangeswithingroupascompared
tobaseline(P<0.05),Ht:hematocrit,hb:hemoglobin
Qualityoflife
Quality of life could only be assessed in 8of 10 and 8 of 12subjects for the control
andexperimental group, respectively. The mean SIP (Sickness Impact Profile) score
inthe control group significantly declined (-13±12%, p<0.05) while itstayed stable in
the experimental group (-2±11%). Mean changes in PGCMS (Philadelphia Geriatric
57
Chapter4
Center Moral Scale) scores were relatively stable with -2±19% for the control group
and-3±20%fortheexperimentalgroup.
Discussion
This study showed that in a group of nursing home residents with poor health,
improvingthesocialambiance offood consumption ledtoapositive change inmean
body weight and a relatively stable health condition. The nutritional and health
benefits of the intervention program were reinforced by a stable self-perception of
functional status in the experimental group. In the control group slight declines in
body weight and biochemical indicators of health status were paired with a
significant decrease oftheself-perceived functionalstatus.
These results are in agreement with former findings revealing that a higher body
weight appear to be afavorable factor in residents above 85 years old (20;21) and
that a low or decreasing mean body weight is associated with poor health in Dutch
nursing home residents (22). In other words weight gain does not seem to be
detrimental in such an elderly population. Furthermore, our results emphasize the
importance of long-term body weight monitoring as a screening tool to assess
change inhealthandsubsequently quality of life ina nursing homesetting.
A striking issue inthis study was the relatively high drop out rate with about 42% in
each group. However this observed rate is in accordance with the mean time of
residence in somatic wards in a Dutch nursing home of 392 days. According to the
latest survey about 64 % of the nursing home resident will move or die within 6
months of arrival (23). This drop out rate is more likely to be explained by the
preexisting morbid conditionsthan bystudy apathy.
Mean daily energy intake in the experimental group was increased and was above
the limit of 6.8 MJ (24). We assume that the positive changes in health and
nutritional status may result from achange indietary intake,which was probably the
effect of the intervention program. As an example, the higher choice offered
between meal with the availability of fruit juices and dairy drinks was partly
responsibleforthe increase invitamin intake.
Inthisstudy biochemical measures ofvitamin status were not included butthe blood
parameters used were related to health and less so to the diet. However, the
observed increase in dietary intake was not reflected in a higher measured energy
intake in the experimental group compared to the control group. This might be
explained bythefact thatthedietary intakewas measured forthree days at baseline
andfollow-up giving actually asnapshot of usual dietary intake. Dayto day variation
might be rather high especially in a nursing home resident population. Furthermore,
considering the very low baseline values -although similar to other studies in nursing
58
Healtheffectofmealambiance innursing home
home elderly resident (25-27)-, one could hardly expect a further decrease in food
intake in both groups. The measured differences in mean body weight give a better
estimate of the overall changes in cumulative differences in energy intake over the
one-year experimentalperiod.
Although it has often been acknowledged that psychosocial factors might have an
important contribution to alow nutritional status (27-32),there arefew data available
on this issue (33). As far as we know only one study also assessed the health
effects of a changed meal environment (33). In the latter study an improvement of
food intake and biochemical indicators of health status was also observed. However
this study was performed on a relatively short-term period and since they had no
control group it is difficult to say if the supplementary food intake resulted from the
intervention or from other factors. Results of our study shows that environmental
factors are important for nutritional health in a nursing home residents population
where internal factors like diseases, well-being or morbidity and external factors
such as nutritional intake and medications are decisive for compliance rate and the
ultimateresults.
Inthis study the chosen option was to re-organize the timetable of the nursing staff
inorderto haveenough nurses present at mealtimes andtoteachthem that careful
attention to the nutritional intake of nursing home residents was both a clinical anda
quality-of-life issue. This could be realized during the whole intervention program by
conducting an efficient re-organization of the whole nursing staff planning and
schedules. The rescheduling was realized with no extra financial expense. In
agreement with nursing staff and patients, it led to special attention for bath and
wash and meals activitieswhileaslight reduction occurredforotheractivities.
Byproviding individualized care at mealtime and ensuring that anadequate number
of staff were available to assist patients who needed help, meal time became an
individualized and pleasant social event for both staff and patients. Results of this
study were sufficiently beneficial to convince the board and management of the
nursing home to apply the developed intervention protocol at the end of the trial to
allothersomaticwards ofthe nursing home.
In conclusion, improvement of the social ambiance of food consumption in this
nursing home was a non-negligible way to stabilize health and nutritional status of
nursing homeelderly residents.
Acknowledgement
Wewould like to thank the residents andthe nurses of the nursing homeAeneas for their
cooperation in this study. We also would like to thank Rineke Meulendijks for her help in
collectingthe data.
59
Chapter4
References
1. Gants R. Detection and correction of underweight problems in nursing home residents. J Gerontol
Nurs.1997;23:26-31.
2. Henderson CT. Nutrition and malnutrition in the elderly nursing home patient. Clin Geriatr.Med
1988;4:527-547.
3. Lowik MR, van den Berg H., Schrijver J, Odink J, Wedel M, Van Houten P. Marginal nutritional
status among institutionalized elderly women as compared to those living more independently (Dutch
Nutrition Surveillance System). JAm Coll Nutr 1992;11:673-681.
1990;90:1697-1706.
factors in629 men andwomen agedover65. EurJ Clin Nutr 1996;50:6-11.
6. Mowe M, Bohmer T. Nutrition problems among home-living elderly people may leadto disease and
7. Silver AJ, Morley JE, Strome LS, Jones D, Vickers L. Nutritional status in an academic nursing
home. JAm Geriatr.Soc 1988;36:487-491.
8. Marcus EL, Berry EM.Refusaltoeat intheelderly. Nutr Rev1998;56:163-171.
9. Bower ME.The resident dining room-a CQI project. J Nutr Elder. 1996;15:47-54.
10. Feunekes Gl, DeGraaf C , Van StaverenWA. Social facilitation offood intake is mediated by meal
duration. Physiol.Behav. 1995;58:551-558.
11. De Castro JM. Social facilitation of the spontaneous meal size of humans occurs on both
weekdays andweekends. Physiol.Behav. 1991;49:1289-1291.
12. DeCastro JM.Social facilitation offood intake in humans. Appetite 1995;24:260-26s.
13. Ribbe MW, Ljunggren G, Steel K, et al. Nursing homes in 10 nations: a comparison between
countries and settings. Age Ageing 1997;26 Suppl2:3-12.
14. Cameron M, Van Staveren W. Manual on the methodology of food consumption studies. Oxford
University Press, 1988.
15. Donders-Engelen MR., Van Der Heijden L, Hulshof KFAM R. Maten, gewichten and
codenummers 1997 (food portion sizes and coding instructions 1997). Rapport TNO Zeist. 1997 ed.
The Netherlands: 1997.
16.Anonymous. NEVO 1997. Stichting Nederlands voedingsstoffenbestand. Dutch Nutrient Database
1997.The Hague,The Netherlands:Voorlichtingsbureau voor devoeding, 1997.
17.Van Campen C , Kerkstra A. [Perceived quality of life of elderly somatic nursing-home patients.
Construction of a measuring instrument]. Tijdschr Gerontol Geriatr. 1998;29:11-18.
18.Van Campen C , Kerkstra A. [Experienced quality of life of somatic nursing home patients: a
review of measuring instruments]. Tijdschr Gerontol Geriatr. 1996;27:20-28.
19. SAS Institute Inc. SAS/Stat user's guide version 6. fourth ed.Cary, USA: SAS Institute Inc., 1989.
20. Tayback M, Kumanyika S, Chee E. Body weight as a risk factor in the elderly. Arch.Intern.Med.
1990;150:1065-1072.
Br.Med.J.CIin.Res.Ed. 1986;292:867-868.
lower bodyweight](in Dutch). NedTijdschr Geneeskd 1997;141:2184-2188.
home and atdischarge]. Ned.Tijdschr.Geneeskd. 1995;139:123-127.
24. van der Wielen RP,deWild GM,de Groot CP, Hoefnagels WH,Van Staveren WA. Dietary intakes
of energy and water-soluble vitamins in different categories of aging. J Gerontol A Biol Sci Med Sci
1996;51 :B100-B107
25. Frisoni GB, Franzoni S, Rozzini R, Ferrucci L, Boffelli S, Trabucchi M. Food intake and mortality in
thefrail elderly. J GerontolA Biol Sci Med Sci 1995;50:M203-M210
60
_Health effectofmealambiance innursing home
26. Blaum CS, Fries BE, Fiatarone MA. Factors associated with low body mass index andweight loss
in nursing home residents. JGerontolA BiolSci Med Sci 1995;50:M162-M168
27. Turic A, Gordon KL, Craig LD, Ataya DG, Voss AC. Nutrition supplementation enables elderly
residents of long-term-care facilities to meet or exceed RDAs without displacing energy or nutrient
intakes from meals. JAm DietAssoc. 1998;98:1457-1459.
28. Potter J, Langhome P, Roberts M. Routine protein energy supplementation in adults: systematic
review. BMJ.1998;317:495-501.
29. van der Wielen RP, van Heereveld HA, de Groot CP, Van Staveren WA. Nutritional status of
elderly female nursing home residents; the effect of supplementation with a physiological dose of
water-soluble vitamins. EurJ Clin Nutr 1995;49:665-674.
30.Van Houten P, Lowik MR. [Nutrition and nutritional status of female somatic nursing home
patients]. NedTijdschr Geneeskd1995;139:227-231.
31. Tripp F. The use of dietary supplements in the elderly: current issues and recommendations. J
Am DietAssoc. 1997;97: S181-S183
32. Morley JE, Silver AJ. Nutritional issues innursing home care. Ann.Intern Med 1995;123:850-859.
medicine. I. Effects ofachanged mealenvironment. Compr.GerontolA 1987;1:29-33.
61
5
Effect of an evening supplement provided to nursing home elderly
onbodyweightanddietaryintake*
Marie-FrancoiseA. M. Mathey, Cees de Graaf, Wim Schimmel
and Wya A. van Staveren
Abstract
Objective:to investigate the effects of the introduction of an evening vitamin-and energy-rich
snack onaverage dietary intake andbodyweight ofnursing homeelderly.
Participants: men andwomen aged 65 and over participated either inthe experimental group
(N=26) orthecontrolgroup (N=26)
Methods: a parallel intervention study design was used. Every evening for 30 days, the
experimental group consumed an energy and vitamin-rich drink (200 ml; 1,2 MJ) around
19:30,whereas thecontrol group received usualcoffee andtea. Outcomes were bodyweight,
average dietary intake basedon3-days record measured before andafterthe experiment.
Results: a weight gain trend was observed in the experimental group (+0.8kg, P<0.05)
whereas the weight of the control group stayed stationary (+0.1 kg). Dietary intake increased
moreintheexperimental group comparedtothecontrolgroup.
Conclusions: inour nursing home elderly, a regular intake of energy and vitamin -rich drinks
inthe evening hadapositive influence onfood intake and preventedweight loss.
'Submitted
Chapter5
Introduction
Unintentional weight loss leading to low body weight is one of the first
indicators of reduced nutritional status in elderly patients (1;2). This is
important for nursing home elderly where adequate nutrition seems to be
essentialforwell-beingandoptimalmedicalcare(3-6).
Inadditiontoage-related changes,factors suchaslack offood choice, lackof
help at eating time, depression, medications and cognitive function
impairment may contribute to a low food intake and weight loss. Several
studies showed that, due to its multifactorial origin, decrease in appetite
remains adifficult problem ingeriatric care(7-10).
Nursing home elderly with reduced physical capacity often have a fixed meal
pattern that might not be physiologically adequate (8). An individual
spontaneous intake outwith regular mealtime is not possible within most
organizations. Inwestern country institutions, the last meal is often served in
the late afternoon. Thereafter there is a time gap during which no foods are
provided. When food is only served between 9 a.m. and 5 p.m., it seems
difficult to ensure sufficient intake of macro- and micronutrients. In this
situation, itmight be beneficial from anutritional but also social perspective to
provide asnack inthecourseoftheevening.
The aim of this study was to assess, in a one-month period, the effect of an
evening vitamin- and energy-rich snack in nursing home elderly on dietary
intakeand bodyweight.
Subjects and methods

Design
Forthe interventionwe usedaparalleldesignwith repeated exposures (30)to
an evening snack. Wards were randomly assigned to control or experimental
group.
Subjects
Sixty-one nursing home elderly from two somatic wards were invited to join
the study. The main inclusion criterion was to be older than 65 y of age.
Exclusion criteria were renal dialyze, or specific nutritional disease-related
care such as parenteral nutrition. Beforejoining the study, subjects and their
contact person received detailed information describing the study and signed
aninformed consent.
64
.Snackfornursinghomeelderly
The study was conducted at the institution Tilburg Zuid (Tilburg, The
Netherlands),with approvalfromthe Medical Ethical Committee ofthe Division
ofHumanNutrition&Epidemiology oftheWageningen University.
Supplement
We chose for an easy to drink milk-based product available in small volume
(200 ml),with an attractive packaging andtasty flavors (strawberry-raspberry,
vanilla, apricot, chocolate) but also a high nutrient density especially for
water-soluble vitamins, calcium and magnesium (Table 1). This type of drink
wasalsoselectedfor itsavailability inmost institutions.
Table 1:nutritionalcompositionofthe snackofferedtoinstitutionalisedelderly
Perportionof %RDAper
200ml 200ml
energy (kJ) [kCal) 1200[300] 15
carbohydrate (g) [En%] 42 [55] -
fat(g) [En%] 10[30] -
protein (g) [En%] 11[15] -
vitamin B1(mg) 0.30 21
vitamin B2 (mg) 0.36 23
vitamin B6(ng) 1.5 21
vitamin B12(ng) 0.90 90
vitamin C(mg) 16 27
vitaminA (ng) 150 19
vitamin D (IKJ) 0.76 15
vitamin E(mg) 3 30
folic acid (fig) 54 27
calcium (mg) 160 20
magnesium (mg) 60 20
iron (mg) 3 21
sodium (mg) 160 -
potassium (mg) 340 -
zinc (mg) 3 20
[En%]:percentageoftheenergyofthesnacksuppliedbythismacronutrient
Procedure
Every evening around 19:30, subjects from the experimental group were
personally offered the drink. They were asked to choose the flavor and to
consume the drink before 20:30. If needed they could receive help from the
staff. The drink was served either in their bedroom or in the living room.
Compliance, consumed quantity, chosen taste and place of consumption
were registered. Simultaneously, the control group received coffee and tea
onlyto paythesameattentionto bothgroups.
65
Chapter 5
Beforeandattheendofthe intervention,dietary intakeand bodyweightwere

recorded.
Measurements:
Dietaryintake
Dietary data were collected by trained research assistants using both
observation and weighing-back methods (11). For all subjects dietary
measurement lasted three consecutive days starting on Sunday. The usual
meal pattern inthe institution was continued duringthe intervention: breakfast
and supper were bread-based meals and at lunchtime around noon a cooked
meal was served. Bread-based meals were prepared and served by the
nursing staff in the dining room. Intake was recorded after weighing of these
foods before serviceand left-over uptoanaccuracy of0.1g.
For the cooked meal food consumption was registered by keeping record of
foods and portion sizes served; after the meal, waste was weighed. Portion
sizes were derived from a Dutch table of regular food portion sizes (12). All
foods and beverages consumed outside regular mealtimeswerealso carefully
recorded.
Dietary data were converted into nutrients using a computerized version of
the Dutchfoodcompositiontable 1996(13).
Body weight
A fixed protocol was used to assess subjects' body weight as index of the
nutritional status. Itwas measured before breakfast, after voiding andwiththe
subjects in light clothing without shoes on a sitting Seca weight scale(to the
nearest 0.1kg).
Dafa analysis
Baseline means ± standard deviations (SD) and absolutes changes ± SD
were calculated per group for alloutcomes. Theywere compared by using an
unpaired t-test for differences between experimental and control groups and
withapairedt-testfordifference withingroups.
A p-value < 5% was considered as statistically significant. All analyses were
conductedwithSASstatistical package(14;15).
66
.Snackfornursing home elderly
Results
Subjects characteristics and dietary intake at baseline
Table 2 presents subjects' compliance to the trial. Drop out cases were
subjects who failed in completing the study because of death, low snack
consumption (snack consumed less than 20 times in 30 days), parenteral
nutrition due to a daily energy intake below 600kCal (2500kJ) or absence
during part of the observation period.
Table 3 presents baseline characteristics. Gender distribution and age were
similar for both groups. Experimental group mean body weight was slightly but
not significantly higher than in the control group. All participants used
medications (4.2±1 per subject and per day) mainly prescribed for
cardiovascular diseases or nervous system disorders.
Table2:Complianceofinstitutionalisedelderlytoeveningsnack
Reason Control Experimental
total number of patients per group, 31 30

refusal
non-respondents 3 2
exclusion criteria 2 2
number atthe start ofthe study 26 26
dropout
deceased 1 2
snack consumption below 20 portions/month - 7
other reasons (•) 5 3
Total completing thetrial 20 14

(•): Absence during dietary record,acute illness during thetrial, death of afamily
member
Energy intake was rather low with 6.3±0.4MJ for the control group and
6.8±0.4MJ for the experimental group, and below the Dutch RDA
(7.8MJ/Day). Energy intake expressed in kJ/kg body weight was also lower
than the RDA (120 kJ/kg body weight) with 99 ± 5 kJ/kg body weight for the
control group and 100 ± 7 kJ/kg body weight for the experimental group.
Macronutrient intake was comparable in the two groups: percentage of energy
derived from carbohydrate was lower than the RDA (50-55% of total energy)
while percentage of fat intake was higher than the RDA (30-35% of total
energy).
67
Chapter5
Table3:baselinecharacteristics[mean (SD)]ofinstitutionalisedelderly
Control Experimental p-value
Anthopometricdata§
gender (M/F) 9/17 8/18 -
age (y) 80.5 (2) 79(2) 0.63
weight (kg) 63.5 (2.6) 69.5 (2.3) 0.09
Dietaryintakedata§§
energy (MJ)
mean(SD) 6.3 (0.4) 6.8 (0.4) 0.43
median 6.5 6.8 -
kJ/ bodywt (kg) 99(5) 100(7) 0.91
carbohydrate (g)
Mean (SD) 167(9.5) 188(12.8) 0.20
%total energy 46(1) 47(2) 0.41
fat(g)
Mean (SD) 70 (25.5) 71 (25.3) 0.87
%total energy 41(1) 39(1) 0.35
protein (g)
Mean(SD) 50.0 (3.8) 56(4.3) 0.33
%total energy 13(1) 14(1) 0.58
Protein/bodywt (kg) (0.1) 0.8(0.1) 0.61
§: N=26 in both groups; §§: N=20 for the control group and N=22 for the
experimental group.
Percentage of protein intake as well as absolute intake were in general
adequate at baseline for both groups. Intake of vitamin and minerals in both
groups did only fulfill 60% (vitamin B1, B6, calcium and magnesium) to 75%
(vitamin B2, C) of the Dutch RDA (data not shown).
After the intervention period

The experimental group was separated in two sub-groups: compilers (n=14)
and drop out subjects i.e. those who consumed the snack less than 20 times
(n=7). Therefore results on dietary intake and body weight changes are
presented for the control vs. compilers of the experimental group and for the
compilers vs. the drop out of the experimental group.
Dietary intake
Dietary intake after the intervention period and absolute changes are
presented in Table 4. Total energy intake was significantly different between
control and compilers. Changes in macronutrient intake occurred in both
groups but in an inverse direction.
68
"5.
E
o
o
co 1^ .
CO
CO - r , CD | 0 Q-
I a1^
"CD
io
?>
52. ^
"9 CNT
T-
CO CN
CN
vi 3
ID. e
.. c
c m ?
c 3 -t->
§ §• § S"5
CD
° §o-°
c Q
& CD > -
CD CO CD CD
•Q
C CO .—.
CO •
... —^
CD T-' 0 0 CN | ^ .2
•E a ji
o « w CO « S ^ Ct b CD C
O CD
c co oo c ?
o UJ 2 cS
u
c
CD
CD
CO
++ +- ++
0}
CD CD -v ++ CO
fO CD
•Q
c
o c
O)
2I
CD <D
C f 03
O fc
M
<N
co ' CN" CN" :t co
.co CO CN CO' , . 2 > CD

C t CO
c •c C " CO CO | C 0 D>
o o Ci Si "
CD o |oo 2
I
o
o di CD
"C> 'CD o
o c CD .— CN c
c CD CD £ . CN CD
CO ^ •* ^- o
E r^ o C CD
c £ »
.o Q.
o c
CN
CN ^
8 °° CN
0 0 CO
Hi
CN
c
CD
.c
JO CO
00' CO CO
CD 2# Ci N. U0 CO jo f j
c\T <o~
CD CD ^ in
O CO o co ^.CN O CN <N —'O
CD ci
P 2 S;d
fc
CO
o
o -° P VI
CD Q -
^, c ..
CD CO
CD •o <2
"25 l~-- ID
CO CD
•Q
CD ^ CN CN 2- to " ^ CD
.CO c CM !S. g) E
"co o co " o £ a>
o
c O ^ s
CD T-
^S
I"- •*
co ^ o %8
%\
! !
CD a) o
CD g>
>> CD i- o
T3
O ^^ >.
O) >>
u>
><l >%
D
£d
is
.Q
s
CD ,—,
L_
CD
C ,—^
i_
CD
C ,—^CD
L_
T3
O
^ VI
Q.Q.
"CD Q Q Q c CD j .
w
*-*
CO CO CD <Z> CD
^^ 03 CD X _-
5 0) C CD o
3 S cCO CO
C "D c «S
"~" "J? CO > ; c 0)
Q .C CO O CO o c o ! - -
C ' CD CD CD O CD "•""' ,— * CD • ^ ^
CD
*-" o II o
CD p c c JO
E ^•8 U>
E o E a.
3
C C
CD
C
CD CO
o 3l3 Q. z°
• . CO
Chapter5
While carbohydrate intake of the compliers significantly increased both in

absolute intake (24 ± 24 g) and percentage of total energy intake (3%), this
intake in the control group stayed relatively stable with a small significant
decrease in percentage of total energy intake (-3%). Besides, percentage of
fat intakesignificantly decreased inthecompliers butdidsignificantly increase
in the control group. Protein intake increased in both groups but, despite a
significant change in the control group (10 ± 13 g), it was still significantly
higher inthe compliers (72 ± 14g) compared tothe control group (60± 19g)
orthedropout (49±13g). Energyand macronutrient intakes inthe compliers
group were found to be generally significantly higher than in the drop out
group. Macronutrient intakeofthedropoutgroupwas relativelystable.
vitaminB1 SPgMST"^^^^^
vitaminB2
vitaminB6
-
vitaminC
vitaminA to
^^^^^m 1
vitaminD
vitaminE S
-2C)% 0% 20% 40% 60% 80% 100% 120%
Figure 1a: Relative changes in vitamin intake of elderly nursing home inhabitants:
controlP , n=20),experimentaldropout( I I ,n=7)andexperimentalcompliers( • ,
n=14).
Figure 1b: Relative changes in mineral intake of elderly nursing home inhabitants:
controlP , n=20),experimentaldropoutP ,n=7)andexperimentalcompliers
( • , n=14).
70
Snack fornursing home elderly
Figure 1shows relative changes in micronutrient intake after the intervention

period. As expected, intake of vitamin B6, C and Esignificantly increased in
compilers (66%, 96% and 78%, respectively) compared to the control group
(14%, -8%and-7%, respectively).Within the experimental group, comparison
of vitamin intake between drop out and compilers significantly differ in
riboflavin (9 % vs. 44%), vitamin D (-5% vs. 21%) and vitamin E (-2% vs.
78%). Only changes inzinc intake differed significantly between control (-4%)
and compilers (34%). Nodifferences in changes of micronutrient intake were
observed betweendrop outandcontrolgroup.
Bodyweightchanges
Table 5 presents body weight changes. Compliers showed a small but
significant increase of0.8 ± 1.7kg(P=0.05) in bodyweight.A relatively stable
bodyweight was observed inthe control group (0.1± 1.5kg)while a decrease
(-0.4±1.7kg)occurred inthedropoutgroup.
Table 5:body weight ofinstitutionalised elderly atbaseline andabsolutes changes
[mean (SD)]aftertheinterventionperiod
Group Baseline absolutechanges
comparedtobaseline
Control (n=20) 64.6(15) 0.1(1.5)
Experimental compliers(n=13) 69.8(15.0) 0.8(1.7)f

dropout(n=7) 70.1(6.5) -0.4(17)
%: P<0.05, significant difference in changes after the intervention period within
groups.
Discussion
Results of this study showed both qualitative and quantitative improvements
offood intakeinthegroupconsumingtheeveningsnack.
Asnack isdefined asafood oralight meal,whichcan beconsumed between
the normal meals i.e. between breakfast, dinner and supper. In free-living
elderly, snacking habits have been found to be important due to its relative
large part in the daily food intake (16;17). In institutions, snacking habits
mostly depend on food choice and availability and on physical possibilities
that people have to get extra foods. Inthis situation, providing a vitamin and
energy-rich drink as snack is a possibility to correct an unbalanced diet (18).
As far aswe know this study was the first to assess the effect of an evening
vitamin- and energy-rich snack on diet adequacy and body weight in nursing
home elderly. However, a drop out of twenty-seven percent shows that
compliance to a nutrient-dense snack in the evening is difficult for some
71
Chapter5
groups. Factors such as diseases, medications and environmental

parameters likefamilyvisitsdiddistractfromtaking adrink.
Although this group considered as accelerated agers, benefits from similar
health care, it is still heterogeneous regarding disease status and medicine
uses but also body weight. Therefore standardized protocols are difficult to
conduct within this population. Our subjects had a low energy and nutrient
intake. Since dietary data rely on observed intake andweighted left-overs we
believe that we have a rather good picture of their food intake. The fact that
these data are similar to previous findings regarding compliance and food
intakesuggestthatthisgroup indeed hadalowdietary intake (7;19-21).
The daily intake of an energy and nutrient rich snack had a positive
contribution on the general intake of macro and micronutrients. When the
snack was included inthediet intakeofcertain micronutrients such asvitamin
B1, B6, C and D, calcium and magnesium reached the Dutch RDA. This
means that the snack intakewas not compensated but consumed in addition
to the daily eating pattern. On one hand this indicates that there is indeed a
needfor an extra snack inthe evening. Onthe other hand, a general decline
in homeostatic capabilities has been observed in elderly and it might explain
why this extra food consumption was not regulated. The positive weight
change confirmed both facts: the need for extra energy and the lack of
regulation, inagreement withearlierfindings (22-25).
Although baseline body weight in the experimental group was slightly higher
than in the control group we think that the observed weight gain after one
month indicates a possible improvement ofthe general nutritional status.This
type of supplementation and the time may give an easy solution to help stop
or even reverse weight loss. Furthermore, depending on the snack
composition, a regular consumption can contribute to a better proportion in
terms of type of fat and carbohydrate converted to energy. An additional
advantage of this supplement rich in complex carbohydrate is that it helps to
shorten nocturnal fasting (26). Further long-term studies with assessment of
biochemical parameters are now necessary to confirm the positive effects of
snacking inthispopulation.
In summary, this study showed that the intake of an evening vitamin- and
energy- richsnack could improve nutritional status innursing homeelderly.
Acknowledgment
WewouldliketothankAntineBreimer,BiancaLooise,BrendavanUffelenand Hugo
Ngadijofortheirhelpincollectingthedata.Wealsowouldliketothanktheresidents
72
.Snackfornursing home elderly
and the nursing staff of the nursing home Tilburg Zuid for their cooperation in this
study.
References
1. Berkhout AM,van HJ,Cools HJ. [Increased chance of dying among nursing home patients
with lower bodyweight](in Dutch). NedTijdschr Geneeskd 1997;141:2184-2188.
2. Mulley GP. Preparing for the lateyears. Lancet 1995;345:1409-1413.
3. Lesourd B, Decarli B, Dirren H. Longitudinal changes in iron and protein status of elderly
Europeans. SENECA Investigators. EurJ Clin Nutr 1996;50 Suppl 2:S16-S24
4. Lesourd B. Protein undernutrition as the major cause of decreased immune function in the
elderly: clinical andfunctional implications. Nutr Rev 1995; 53:S86-S91
5. Mowe M, Bohmer T. Nutrition problems among home-living elderly people may lead to
disease and hospitalization. Nutr Rev 1996;54:S22-S24
6. Payette H, Gray DK, Cyr R, Boutier V. Predictors of dietary intake in a functionally
dependent elderly population inthecommunity. Am J Public Health 1995;85:677-683.
7. Fabiny AR, Kiel DP. Assessing and treating weight loss in nursing home patients. Clin
Geriatr.Med1997;13:737-751.
8. Morley JE, Kraenzle D. Causes of weight loss in a community nursing home. J Am
Geriatr.Soc 1994;42:583-585.
9. Morley JE. Anorexia in older persons: epidemiology and optimal treatment. Drugs Aging
1996;8:134-155.
10. Morley JE.Anorexiaof aging: physiologic and pathologic. Am JClin Nutr 1997;66:760-773.
11. Cameron M, Van Staveren W. Manual on the methodology of food consumption studies.
Oxford University Press, 1988.
12. Donders-Engelen MR., Van Der Heijden L, Hulshof KFAM R. Maten, gewichten and
codenummers 1997 (food portion sizes and coding instructions 1997). Rapport TNO Zeist.
1997 ed.The Netherlands: 1997.
13. Anonymous. NEVO 1996. Stichting Nederlands voedingsstoffenbestand. Dutch Nutrient
Database 1996.The Hague,The Netherlands:Voorlichtingsbureau voor devoeding, 1996.
14. SAS Institute Inc. SAS/Stat user's guide version 6. fourth ed. Cary, USA: SAS Institute
Inc., 1989.
15. SAS Institute Inc. SAS/Stat procedure guide version 6. fourth ed. Cary, USA: SAS
Institute Inc., 1989.
16. Summerbell CD, Moody RC , Shanks J, Stock MJ, Geissler C. Sources of energy from
meals versus snacks in220 people infour agegroups. Eur JClin Nutr 1995;49:33-41.
17. Vincent D, Lauque S, Nourashemi F, Faisant C, Lanzmann-Petithory, Vellas B. Nutritional
value and part of snacks in the daily food intake of 186 healthy elderly subjects in the
Toulouse study. J Nutrition, Health &Aging . 1998;1 (1):39-43.
18. Turic A, Gordon KL, Craig LD, Ataya DG, Voss AC. Nutrition supplementation enables
elderly residents of long-term-care facilities to meet or exceed RDAswithout displacing energy
or nutrient intakes from meals. JAm DietAssoc. 1998;98:1457-1459.
19. Van Staveren WA, de Groot CP, Blauw YH, van-der WR. Assessing diets of elderly
people: problems andapproaches. Am J Clin Nutr 1994;59:221S-223S.
20. van der Wielen RP, de Wild GM, de Groot CP, Hoefnagels WH , Van Staveren WA.
Dietary intakes of energy and water-soluble vitamins in different categories of aging. J
Gerontol A Biol Sci Med Sci 1996;51:B100-B107
21. van der Wielen RP,van Heereveld HA, de Groot CP,Van Staveren WA. Nutritional status
of elderly female nursing home residents; the effect of supplementation with a physiological
dose ofwater-soluble vitamins. Eur JClin Nutr 1995;49:665-674.
73
Chapter5
22. Pelchat ML. Foodcravings inyoung andelderly adults. Appetite 1997;28:103-113.

23. Roberts SB, Fuss P, Heyman MB, Young VR. Influence of age on energy requirements.
Am J Clin Nutr 1995;62:1053S-1058S.
24. Rolls BJ.Appetite andsatiety intheelderly. Nutr Rev 1994;52:S9-10.
25. Rolls BJ, Dimeo KA, Stride DJ. Age-related impairments in the regulation of food intake.
Am J Clin Nutr1995;62:923-931.
26. Chang WK, Chao YC, Tang HS, Lang HF, Hsu CT. Effects of extra-carbohydrate
supplementation inthe late evening on energy expenditure and substrate oxidation in patients
with liver cirrhosis. JPEN.J Parenter.Enteral Nutr 1997;21:96-99.
74
6
Flavorenhancement offood improvesdietary intakeand nutritional
status ofnursing homeelderly*
Marie-Francoise A.M. Mathey, Els Siebelink, Cees de Graaf

and Wya A. Van Staveren
Abstract
Background: taste and smell losses occur with aging. These changes are supposed to
decrease the enjoyment of food, subsequently reduce food consumption and negatively
influence the nutritional status ofelderly, especially thefrailones.
Objectives: to determine if the addition of flavor enhancers to the cooked meal for nursing
home elderly promotes food consumption and provides nutritional benefits.
Design: a 16weeks parallel group intervention consisting of sprinkling flavor enhancers over
the cooked meal ofthe 'flavor' group (n=36) and notover the controlgroup meal(n=31).
Measurements: Intake atthe cooked meal:before,after 8and 16weeks of intervention.
Feelings of appetite, daily dietary intake and anthropometry: before and after the
intervention.
Results: on average body weight of the 'flavor' group increased (+1.1±1.3 kg, p<0.05,) as
opposed (P<0.05) to that of the control group (-0.3±1.6 kg). Daily dietary intake decreased
in the control group (-485±1245kJ, p<0.05) but not in the flavor group (-208±1115kJ,
P=0.28). Intake at the cooked meal increased inthe 'flavor' group (133±367kJ, p<0.05) but
not inthe control group (85±392kJ). A similar trendwas observed for hunger feelings, which
increased only inthe 'flavor' group.
Conclusion: adding flavor enhancers to the cooked meal was an effective way to improve
dietary intake andbodyweight innursing homeelderly.
*submitted
Chapter6
Introduction
Inadequate dietary intake is often observed in nursing home elderly (1;2).
Accordingly, this population is highly at risk of developing undetected malnutrition
and nutritional deficiencies. This malnutrition contributes to a reduced quality of life,
animpaired healthstatus often calledfrailty(3-7).
Taste and smell losses occur with aging (8) and may influence the enjoyment of
food and thereby affect the nutritional intake of older adults. Most studies on taste
and agingfocused ontaste acuityand sensitivity rather than on hedonic preference.
Aging is associated with an increase in taste and smell thresholds, and elderly
subjects when blindfolded had about one-half the ability of young subjects to
recognize blended foods (9). Many studies on preferences indicate that elderly
subjects would prefer higher concentrations of stimuli for solutions of sucrose,
sodium chloride, and citric acid thanyounger subjects do(10;11). The studies,which
demonstrated that elderly subjects prefer a higher level of tastants but failed to
associatethisto ahigher consumption offoods containingthesetastants (12-14).
These age-related deficits in taste and smell are supposed to decrease food
consumption and probably contribute to negative changes in eating behavior (15-
18). Few studies explored the relationship between sensory impairment, hedonic
response,andalteredfood intake inthe elderly (12). Recently de Jong and coll. (19)
demonstrated that a poor appetite is related to loss of sensory perception but they
could notshowaneffect onintake.
Also Schiffman and Warwick (20) observed no changes in dietary intake of elderly
subjects after 3weeks consuming flavor enhanced foods,although they did observe
an improved immunefunction andgrip strength. Reasonsfor notfinding aneffect on
dietary intake might bethe reliability ofthedietary methods andthe short-term ofthe
observations.
Therefore ourobjectivewastodeterminewhether the addition offlavor enhancers to
the cooked meal during 16 weeks would lead to an increase in food consumption
andthereby provide nutritional benefitsto nursing homeelderly.
Subjectsandmethods
Subjectsandsetting
The study was conducted at the nursing home "Rustenburg", Wageningen (The
Netherlands). Selection criteria were: being older than 65 y of age, no known
dementia or residing in a somatic ward (21), no known depression, no disease in
terminal phase, no allergy to monosodium glutamate, already residing inthe nursing
home for more than 3 months and consuming at least five days a week the cooked
meal provided bythe nursing home kitchenat lunchtime. Seventy-one residents were
76
Flavorenhancementoffoodforelderly
enrolled in the study. The study protocol was approved by the Medical Ethical
Committee of the Division of Human Nutrition & Epidemiology, Wageningen
University.
Design and procedure

A parallel group intervention design was applied.The intervention consisted of adding
flavors to the main dish of the cooked meal of the 'flavor group' while the control group
received normal flavored meals. The study was carried out in a period of 17weeks: a
week of run-in period and an experimental period of 16 weeks. After the baseline
measurements, subjects were randomly assigned to be in the control group (n=34) or
to the 'flavor' group (n=37).
Table1:Descriptionoftheexperimentalschedulefollowedduringthestudy
Week 0 1 2 3 4 5 6 7 8 g 10 11 12 13 14 15 16
Intervention:Controlgroup (noflavors)
Experimental periods Runin and'flavor' group(with addedflavors)
Measurements:
Dietary intakeatthecooked meal X X X
Totaldailydietary intake X X
Anthropometry X X
AHSPquestionnaire* X X
GDS** X
Compliance X • x|""x| x] x] X X x| x| x| x| x[ x| x[ x[ x| X
*AHSP questionnaire: appetite hunger and sensory perception questionnaire (22); **GDS: Geriatric
depressionscale(23)
Anthropometry and appetite data were assessed before and at the end of the
intervention period. Dietary intake data were collected before, after 8 weeks and at
the end of the trial. Compliance i.e. consumption of the served meal was checked
daily during the 16 weeks experiment by keeping records of meal orders and
deliveries (Table 1).
77
Chapter6
Flavorenhancers
Four flavor powders were available to enhance the cooked meal (Table 2): chicken
flavor, beef bouillon flavor, turkey flavor and lemon butter (fish) flavor (IFF BV,
Hilversum, The Netherlands). Choice of the added flavor was determined by the
nature ofthe protein-rich mealcomponent and bythe cooking process. Flavors were
sprinkled just before meal delivery with the help of a spice-shaker over the whole
main dish including the carbohydrate rich components and the vegetables. The
amount sprinkled perdishwas 1±0.2 gofflavor powder.
Table2: Referencesingredients for100gofready-to-use flavorproduct

Chicken flavor Beef bouillon Turkey flavor Lemon butter
flavor (fish) flavor
(#1 594 5287) (#15 94 5401) (#13 606019) (#1360 5938)
Protein 1 1 1 2
Fat 8 4 10 7
Sugars/Starch 58 61 58 59
Salt <1 <1 <1 <1
MSG 30 30 28 31
Others <2 <3 <2 <1
MSG: monosodium glutamate; Sugars/starch ismainly lactose; others: free-flowing agents and acids.
Measurements
Anthropometry
Bodyweight:
Patients' body weight, as index of the nutritional status before and after the study,
was measured before breakfast after voiding (to the nearest 0.5 kg, Seca weighing
scale, Hamburg,Germany) withsubjects dressed inlightclothing andwithoutshoes.
Kneeheight:
The knee-to-floor height (KFH)was measured twice byasingletrained observerwith
a stadiometer in a sitting position, from the anterior surface of the thigh to the floor
with the ankle and the knee each flexed at a 90° angle against the metallic help.
Body heightwasderived usingthefollowing formula's (24):
Height (incm) =3.16*KFH(incm)
Dietaryintake
Totaldailydietary intake datawere collected using acombination of a 3-days record
andweighing-back methods before and at the end of the intervention. Bread-based
meal, snack and beverage consumption was recorded by means of individual food
diaries andchecked by interviews with atrained dietician. Portion sizeswere derived
froma Dutchtable of regularfood portion sizes and household units(25).
78
Flavor enhancement offoodforelderly
Dietary intake atthe cooked mealwas assessed with a3-day weighing-back method
beforethe intervention. Based onthe information onthedaytodayvariation,a
7-day weighing-back method was used after 8 and 16 weeks (26). This enabled us
to detect a mean difference of at least 70 kJ. Individual menus and recipes for the
measurement days were obtained from the kitchen. Food consumption was then
registered bykeeping records ofamounts served andweighingwasteafterthemeal.
Dietary data were converted into nutrients using the Dutch food composition table
(27).
Appetite, hungerfeelings andsensoryperception questionnaire (AHSP)

Subjects responded to a 29-items questionnaire about their feelings of hunger,
appetite andtheirtaste and smell perception (19).After reading the question together
with an interviewer, subjects had to score on a 5-point scale. A higher score
corresponded to a more positive feeling of their sensory perception, a better appetite
and morefeelings of hunger. Fivevariableswere calculated:presenttasteperception,
8 items; present smell perception, 3 items; present smell perception compared to the
past,3items;appetite,6items;anddailyfeelingsofhunger, 9items.
Geriatric depressionscale (GDS)

The GDS (23), used to assess depression status of the subjects, consisted of 15
items to be answered with 'yes' (1) or 'no' (0). The answers summed upto obtain a
score,witheach score above 5indicating adepressive status.
Data analyses
Only data of subjects completing the study were analyzed. Means ± standard
deviations (SD) of baseline and absolutes changes were calculated for the outcome
variables per group. Changes were compared with an unpaired t-test for differences
betweengroups orwith apairedt-testfordifferences within groups.A p-value<0.05
was considered statistically significant. Data were analyzed using the SAS program
(28).
Results
Subjects
Sixty-seven out of seventy-one elderly completed the study. Dropouts were patients
whofailed incompleting the study because of death (1subject), move (1subject) or
personal reasons (2 subjects). Data on dietary intake and body weight could be
obtained from all subjects. Besides, we also have results on appetite feelings and
79
Chapter6
depression from 42 subjects who were capable to understand and answer the different
questionnaires.
Table3:generalbaselinecharacteristicsoftheelderlysubjectswhocompletedthestudy
Control group 'flavor' group
Variable
n=31 n=36
Age [mean(SD) years] 83.0 (5.5) 84.6(6.1)
Gender (male/female) 6/25 7/29
Livingwith spouse 4 10
Dentures (%)
Complete 13 9
Partial 74 83
None 13 8
Smoking behavior(%)
No smoking 84 92
Smoking 16 8
Medicine uses (mean number/day) 2.1(1.6) 2.1(1.8)
Restrained physical mobility (% use)
Wheel chair 7 11
Walking frame 42 36
GDSscore [mean(SD)] 3.2(2.8)a 3.2(2.4)b
a)n= 18; b) n=24.GDS:geriatric depression scale (23)
Resident characteristics were similar for both groups at the start of the study (Table
3). Groups were comparable with respect to diseases and treatments and used
medications mainly prescribed for cardiovascular disorders, pain or digestive track
disorders. No differences in depression status were observed at baseline between the
two groups with a mean score of 3.2 indicating that participants were not depressed.
Compliance was high with on average 111 of 114 days (98%) of consumption of the
cooked meal.
Anthropometry
Table4:anthropometryanddailydietaryintakecharacteristicsandchangesafterthe
intervention[Mean (SD)]ofthenursinghomeelderlyresidents
Variable Control group (n=31) 'flavor' group (n=36)
Bodyweight (kg) at baseline 69.0(17.0) 72.0(17.5)
absolute changes6 -0.3(1.6) 1.1 (1.3)*§
Calculated Height (cm)a 160.0(10.6) 157.6(12.1)
BMI (kg/m2) 28.1(7.0) 28.4(7.1)
Daily energy intake at baseline (kJ) 5969(1641) 5821(1449)
absolute changes"0 -485(1245)* -208(1115)
Energy (kJ)/Weight (kg) at baselin 91(31) 86(30)
absolute changes -8(4)* -5(17)
BMI= body mass index, a: derived from Berkhout (22); b: absolute changes after 16 weeks
of intervention, c: control (n=29) and intervention (n=35). *: Significant difference in changes
within one group between start and end of the intervention period, (P<0.05); §: Significant
difference in changes between groups between start and end of the intervention period,
P<0.05.
80
Groups were comparable with respect of mean bodyweight, BMI and energy intake
beforethe study (Table4).Asshown inTable4, mean bodyweight increased during
the intervention in the experimental group (1.1±1.3 kg, P<0.001) period while it
remained stable in the control group (-0.4±1.6 kg, P=0.37). Changes between
groups differed significantly (P<0.001). Figure 1 shows the percentage of subjects
withstable bodyweight or losingorgainingweight overthe 16-weekperiod.
100n
90
80
1 70
« 60-
<D 50
•5 40
2? 30
20
10
0
>-0.5 [-0.5;0.5] >0.5
weight change (kg)
Figure 1: Distribution(%) ofnursinghome elderlyresidentslosingorgaining weightover

the
16-weekperiod(control(T\)n=31 and'flavor'(H) n=36).
Dailydietaryintake
Table 4 presents the daily dietary intake at baseline and absolute changes after 16
weeks of intervention. Inboth groups energy intake was low on average (5969±1641
kJforthecontrolgroup and 5821±1449 kJforthe 'flavor' group) and belowthe mean
Dutch requirement for elderly (7.8 MJ/Day). As expressed per unit of body weight,
dietary intake was also lower than the recommended intake (120 kJ/kg body weight)
with 91±31 kJ/kg body weight for the control group and 86±30 kJ/kg body weight for
the 'flavor' group. Percentage of energy provided by fat, carbohydrate and protein
were similar in both groups with 36 %, 46%, 17 % and 1%for fat, carbohydrate,
protein andalcohol, respectively.
After a 16-week intervention,energy intake ofthe controlgroup (-485±1245, P=0.03)
declined while it remained relatively stable (-208±1115kJ; P=0.28) in the 'flavor'
group. A similar trend was observed for intake expressed per unit of body weight.
Percentage of daily energy intake provided by fat (-2%, P<0.05) declined in the
control group while the energy provided by other macronutrients remained
unchanged. No changes occurred in the 'flavor' group regarding the contribution of
macronutrienttothedailyenergy intake.
When body weight variation after 16weeks was related with changes in daily dietary
intake, positive association were found for changes indaily energy [Pearson r=0.345,
81
Chapter6
P=0.04] and fat intake [Pearson r=0.407, P=0.01] in the 'flavor' group. Such
correlations were not found inthe control group.
Dietary intake at the cooked meal

Dietary intake at the cooked meal and absolute changes after 8 and 16 weeks of
intervention are shown in Table 5. Intake of energy, carbohydrate and fat increased
in the 'flavor' group after the 16-week intervention. Inthe control group little changes
were also observed for the carbohydrate intake. In both groups, intake of vitamins
and minerals remained stable (data not shown).
Table5: energy and macronutrient intake [mean±SD] of nursing home elderly residents at
the cooked meal, baseline values and absolute changes after 8 and 16 weeks of
interventionascomparedto valuesatthestartofthestudy.
Control (n=31) 'flavor' (n=36)
Variable Before Change Change Before Change Change
0-81a 0-162 0-8T 0-162
Energy (kJ) 1880±657 124±366 85±392 1907±560 87±380 133±367*
Protein(g) 25±8 2±6 1±6 27±8 0±6 0±5
Carbohydrate (g) 41±15 3±10 3±10* 43±13 0±11 3±9*
Fattotal(g) 20±9 1±6 0±7 19±7 2±6* 2±7*
Change 0-8 changes observed after 8 weeks as compared to baseline; Change 0-16
changes observed after 16weeks ascompared to baseline,a) N=30.*: Significant difference
inchanges withingroup as compared to agiventime (P<0.05).
Table 6: Mean score (SD) of the Appetite, hunger feelings and sensory perception
questionnaireand absolute changes after 16weeks of intervention in nursing homeelderly
residents.
Control(n=18) 'flavor (n=24)
Possible
Variable Absolute Absolute
Range Baseline Baseline
changes changes
Appetite 6-30 17.4(6.7) 1.2(3.1) 18.8 (5.3) 0.5(2.7)

Dailyfeelingof hunger 9-45 33.2(7.4) -0.3(5.8) 29.3 (7.5) 3.0(4.3)*§
Subjectivefeeling of present
taste perception 8-40 20.7(7.0) 0.8(3.5) 23.0(6.1) -0.0(3.0)
Subjectivefeeling of present
smell perception 3-15 11.2(1.6) 0.8(2.1) 9.8(2.4)§ 1.3(2.7)*
Present smell perception
comparedtothepast 3-15 8.2(1.6) 0.3(3.2) 7.5 (2.3) 0.8(2.7)
*:Significantdifferenceinchangeswithinonegroupbetweenstartandendoftheinterventionperiod,
P<0.05.§:Significantdifference betweengroups,P<0.05.
82
Appetite, hungerfeelingsandsensoryperception questionnaire (AHSP)

Table 6 presents mean scores and absolute changes of theAHSP questionnaire. No
differences wereobserved betweenscoresatbaseline.After 16weeksofintervention,
higher scores were observed in the 'flavor' group for daily feelings of hunger and
present smell perception. Changes in daily feelings of hunger in the 'flavor' group
differedfromthoseinthecontrolgroup.
Discussion
Resultsofthis interventionstudyshowedthreemajorfindingsafter 16week:
- Repeated consumption of a flavor enhanced cooked meal led to an increase in
dietary intakeatthismealandastabledailydietaryintake.
- Increased body weight was noticeable after consumption of a flavor enhanced
cookedmeal.
- Repeated consumption offlavor enhanced foods resulted in increased daily feelings
ofhunger.
These findings are in agreement with former studies (9;20;29;30) suggesting that
adding flavor enhancers might improve appetite and dietary intake in an elderly
population. Until now none of these studies could show an increase in actual food
intake. Our intervention lasted for 16weeks and the compliance was high. By using
such atime period weassume thatthe establishment of acceptance and preferences
for the foods with added flavor enhancers could be achieved and thereby could be
reflected bya rise inenergy intake atthe cooked meal.This might not have beenthe
caseinshorterstudies.
Daily energy intake was relatively stable in the 'flavor' group while a decrease of
about 0.4MJ occurred inthecontrolgroup.Atfirst sight,thisfinding is not in linewith
the observed increase in body weight in the experimental group and the relatively
stable weight in the control group. However from other studies it is well known that
the assessment of food intake gives lower intake values with repeated
measurements(31). Therefore we believe that the obtained values for the
measurement at the end of the experiment are underestimated. Besides, a
significant correlation betweenchanges indaily intakeand bodyweightvariationwas
observed and impliesthat the 'flavor' group hasactually increased rather than being
stable inenergy intake.We assumethat measured differences inmean bodyweight
give a better assessment of the overall changes incumulative differences in energy
intake over a 16-weekperiod.
Changes in dietary intake at the cooked meal are likely to be the result of an
increased enjoyment of food. This hypothesis was confirmed by an increase in
energy intake at the cooked meal in the 'flavor' group. Since intake data at the
83
Chapter 6
cooked meal are derived from repeated measures for 7 days we believe that we
haveagoodpictureof intakeatthecookedmeal.
The rather high BMI suggests a well nourished population(32). However,
interpretation of BMI should be more liberal than in younger adults. First, elderly
shrink so their measured body height is somewhat underestimated (33). Second,
BMI is not related tothe presence of diseases. Onthe contrary, the attention should
be paid to weight since it is one of the major risk factor for morbidity in this
population considered as accelerated agers(2;3). Our results indicate that
consumption of food with enhanced chemosensory properties in this population
could provide nutritional benefits and help to prevent weight loss. These positive
observations should also be further confirmed with data on body composition or
biochemical indices.A long-term assessment of body weight would be necessary to
verify that the gain in body weight remains stable and not start to decrease as soon
asthe interventionstopped.
The use of flavor enhancers has been suggested to compensate for diminished
chemosensory functioning contributing to impaired control of appetite in the elderly
or socalled anorexia of aging (34;35). Flavor amplification could restorethe hedonic
functions of food and hereby promote a partial re-establishment of the original
attitude/behavioral response ofthis population towardsfood intake(30;36).
Previous studies suggested that the consumption of flavor-enhanced foods would
stimulate the limbic system and the endogenous opioid activity (20;36). The positive
effect on body weight paired with increased daily feelings of hunger observed in our
study strengthen this possible path suggesting that the opioid activity arisingfrom the
consumption of more palatable foods may promote nutritional and physiological
benefits intheelderly.
In the present study we were in favor of stimulating both olfactory and gustative
functions. This could be realized by using flavor enhancers containing monosodium
glutamate.Wewere confronted to an arguable issue:the repeated exposure toflavor
enhancers containing about 30 % of monosodium glutamate. Since sodium intake
remainsasensible matter inthe elderlywith slower renalfunction,apossible increase
in daily sodium intake through the use of flavor enhancers rich in monosodium
glutamate could not seem to be advisable at first sight. Considering a mean daily
sodium intake of 9g in this population (37), an additional daily dose of about 30 to
45mg of sodium, i.e. 3.9g to 5.4 gfor a 16-week period,will most likely havevery little
influenceonsodium metabolism,renalexcretionandwaterretention.
Sensory studies inelderly subjects showed that the concentration of MSG needed to
influence preference was lower than the detection threshold in that food (29;38;39).
This finding suggests that the flavor enhancing effect of MSG occurs even if its
84
Flavor enhancement offoodfor elderly
concentration is too low to be detected by the elderly consumer. Furthermore the

sodiumcontentofMSGisonethirdofthetablesalt (NaCI)(40).Fromahealthpointof
view these findings together with our results suggest that the uses of flavors
enhancerscontaining MSGcouldallowtheelderlypopulationtodecreasetheirsodium
intake from table salt while maintaining palatability and thereby hedonic function of
foods.
Adding ready to useflavor enhancers to the cooked mealwas inthe present study a
simple buteffective way to improve daily feelings of hunger, actual dietary intake and
bodyweight inanursing home populationwithastable healthstatus.
Acknowledgment
We gratefully acknowledged IFF BV (Hilversum, The Netherlands) for their donation of
flavors. We also would like to thank Friesland Coberco Research and the Suikerstichting for
their sponsorship. Further, we are grateful to Alma van der Greft, Jill Idzinga, Marieke
Spaan and Marjolein Homs for their help during data collection. We would also like to thank
the participants, the nurses as well as the kitchen staff especially Mr Pinkster and Mr
Hardeman of the nursing home 'Rustenburg' for their cooperation in this study.
References
1. Lowik MR, van den Berg H., Schrijver J, Odink J, Wedel M, Van Houten P. Marginal nutritional
status among institutionalized elderly women as compared to those living more independently (Dutch
Nutrition Surveillance System). JAm Coll Nutr 1992;11:673-681.
2. van der Wielen RP,van Heereveld HA, de Groot CP,Van Staveren WA. Nutritional status of elderly
female nursing home residents; the effect of supplementation with a physiological dose of water-
soluble vitamins. Eur JClin Nutr 1995;49:665-674.
4. Blaum CS, Fries BE, Fiatarone MA. Factors associatedwith low body mass index andweight loss in
nursing home residents. J GerontolA BiolSci Med Sci 1995;50:M162-M168
5. Sullivan DH. Impact of nutritional status on health outcomes of nursing home residents. J Am
Geriatr.Soc 1995;43:195-196.
6. Morley JE, Silver AJ. Nutritional issues innursing home care. Ann.Intern Med 1995;123:850-859.
7. Marcus EL, Berry EM. Refusal toeat intheelderly. Nutr Rev1998;56:163-171.
8. Hetherington MM.Taste and appetite regulation inthe elderly. Proc.Nutr.Soc.1998;57:625-631.
9. Schiffman SS. Perception of taste and smell in elderly persons. Crit.Rev.Food Sci.Nutr.
1993;33:17-26.
10. Zandstra EH, Graaf Cd. Sensory perception and pleasantness of orange beverages from
childhood to oldage. Food Quality.and.Preference. 1998;35:7-12.
11. Griep Ml, Mets TF, Vercruysse A, et al. Food odor thresholds in relation to age, nutritional, and
health status. J.Gerontol.A.Biol.Sci.Med.Sci. 1995;50:B407-B414
85
Chapter6_
12. Griep Ml,Verleye G, Franck AH, Collys K, Mets TF, Massart DL. Variation in nutrient intake with
dental status,age and odour perception. Eur JClin Nutr 1996;50:816-825.
13. Griep Ml,Verleye G, Franck AH, Collys K, Mets TF, Massart DL. Variation in nutrient intake with
dental status, ageandodour perception. Eur.J.Clin.Nutr. 1996;50:816-825.
14.deJong N,deGraaf C,Van StaverenW. Effect ofsucrose inbreakfast itemson pleasantness and
food intake inthe elderly. Physiol.Behav. 1996;60:1453-1462.
15. Rolls BJ.Appetite and satiety intheelderly. Nutr Rev 1994;52:S9-10.
1999;66:193-197.
17. Drewnowski A.Taste preferences andfood intake. Annu.Rev.Nutr. 1997; 17:237-253.
18. Drewnowski A, Henderson SA, Driscoll A, Rolls BJ. The Dietary Variety Score: assessing diet
quality inhealthy young andolderadults. J.Am.Diet.Assoc.1997;97:266-271.
19. de Jong N, Mulder I, de Graaf C, Van Staveren WA. Impaired sensory functioning in elders: the
relation with its potential determinants and nutritional intake [see comments].
J.GerontolABiol.Sci.Med.Sci. 1999;54:B324-B331
20. Schiffman SS, Warwick ZS. Effect of flavor enhancement of foods for the elderly on nutritional
status: food intake, biochemical indices, and anthropometric measures. Physiol.Behav. 1993;53:395-
402.
home and atdischarge]. Ned.Tijdschr.Geneeskd. 1995;139:123-127.
22. Schiffman SS, Warwick ZS. Flavor enhancement of foods for the elderly can reverse anorexia.
Neurobiol.Aging 1988;9:24-26.
23. Yesavage JA. Geriatric Depression Scale. Psychopharmacol.Bull. 1988;24:709-711.
24. Berkhout AM, Cools HJ, Mulder JD. [Measurement or estimation of body length in older nursing
home patients]. Tijdschr.Gerontol.Geriatr. 1989;20:211-214.
25. Donders-Engelen M.R., Van Der Heijden L, Hulshof KFAM R. Maten, gewichten and
codenummers 1997 (food portion sizes and coding instructions 1997). Rapport TNO Zeist. 1997 ed.
The Netherlands: 1997.
1997. The Hague,The Netherlands: Voorlichtingsbureau voor devoeding, 1997.
28. SAS Institute Inc. SAS/Stat user's guideversion 6. fourth ed.Cary, USA: SAS Institute Inc., 1989.
29. Bellisle F, Monneuse MO , Chabert M, Larue AC, Lanteaume MT, Louis SJ. Monosodium
glutamateasa palatability enhancer inthe European diet. Physiol.Behav. 1991;49:869-873.
31. de Jong N, Chin a Paw J , de Groot LC, de Graaf C, Kok FJ, Van Staveren WA. Functional
biochemical and nutrient indices in frail elderly people are partly affected by dietary supplements but
not byexercise. J.Nutr. 1999;129:2028-2036.
86
Flavor enhancement of foodfor elderly
32. Berkhout AM. [Limitations in feeding behavior in nursing home patients] (in Dutch). Tijdschr
Gerontol Geriatr. 1996;27:62-66.

ofelderly Europeans. SENECA Investigators. Eur JClin Nutr 1996;50 Suppl2:S9-15.
34. Blundell JE. Understanding anorexia in the elderly: formulating biopsychological research
strategies. Neurobiol.Aging 1988;9:18-20.
35. Morley JE.Anorexia ofaging: physiologic and pathologic. Am J Clin Nutr 1997;66:760-773.
36. Schiffman SS, Warwick ZS. Use of flavor-amplified foods to improve nutritional status in elderly
patients. Ann.N.Y.Acad.Sci. 1989;561:267-276.
1998.The Hague,The Netherlands: Voorlichtingsbureau voor devoeding, 1998.
38. Schiffman SS, Sattely ME, Zimmerman IA, Graham BG, Erickson RP. Taste perception of
monosodium glutamate (MSG) in foods in young and elderly subjects. Physiol.Behav. 1994;56:265-
275.
39. Bellisle F. Effects of monosodium glutamate on human food palatability. Ann.N.Y.Acad.Sci.

1998;855:438-441.
40. Bellisle F. Glutamate and the UMAMI taste: sensory, metabolic, nutritional and behavioural
considerations. A review of the literature published in the last 10 years. Neurosci.Biobehav.Rev.
1999;23:423-438.
87
7
General Discussion
Chapter7
Aging is often accompanied by the phenomenon of anorexia of aging, defined as a

decline in appetite, followed by unexplained weight loss (1-8). The present thesis
described research on anorexia of aging and focused on social and physiological
determinants of appetite, food intake and/or body weight in observation as well as in
intervention studies.
First a cross-sectional observation study was conducted to assess possible
differences in appetite in elderly with a different health status (chapter 2). Then
several intervention studies were carried out to explore the age-related pathways
that may contribute to loss of appetite, as described in Figure 1. Social factors
influencing appetite were investigated on a short term in a laboratory setting {chapter
3) and on a long term in a natural environment {chapter 4). Physiological influence
on appetite was explored by assessing the effect of food-related factors: modifying
the properties of food enhancing the sensory properties of the cooked meal {chapter
6) and by looking at both short-term and long-term responses to energy challenges
(chapter 3,5).
Social and environmental Physiological andfood <

factors related factors
Chapter3-4 Chapter3,5-6
Physical Sensory properties Macronutrient and

Presence of energy content
others environment Palatabiiity
Chapter6 Chapter3,5
7
Hedonicfunction offoods
Appetite
Dietary intake:
Indicator ofshort-term regulation of appetite
Chapter3-6
Bodyweight:
Indicator of long-term regulation of appetite
Chapter4-6
Figure 1: Potential factors influencing appetite in the elderly that were investigated in the
interventionstudiesdescribedinthisthesis.
The first part of this chapter summarizes the main findings of all studies, followed by
a discussion of methodological aspects. Next, results of this thesis will be put into
perspective and implications for daily nutritional care in the elderly will be suggested.
Finally, conclusions and recommendations for future research will be proposed.
90
GeneralDiscussion
Mainfindings
The prevalence ofdecline ofappetite varied amongdiverse groups ofelderly people.
Nursing home elderly reported a lower appetite than the free-living frail elderly, and
this group reported on average a lower appetite than the healthy free living group.
Thisvariation probably resultsfromthe large differences inhealth status andthe rate
of biological aging. These feelings of hunger and appetite were found to be
indicators of body weight in an apparently health elderly group, while in the
population of frail or nursing home elderly with a dwindling health status this
relationshipwas notapparent (chapter2).
The results of the frail group (chapter 2, (9)) confirmed the effect of a decreasing
health status on the ability to counterbalance the age-related loss of appetite.
Further, it insinuates that nursing home elderly, who are in a relatively new social
andphysicalenvironment anddepend onothersforthe basic caressuchaswashing
andeating,present ahigher riskfor lossofappetite.Theexternal lacksof stimulation
of appetite together with a poor health condition make it rather difficult for them to
counterweigh the age-related decline in appetite (chapter 4-6). This presumably
explainsthedisruption inthe relationappetite-bodyweight inthis population(chapter
2).
As indicated in Table 1, an overview of the main findings of the interventions, all
long-term studies ledto a beneficial effect on body weight while the effect on dietary
intakevaried.
Table 1: overviewofthe main effectsobservedinthe interventionstudies.
Studies Levelof Periodof Effect ondietary Effect onbody

described in intervention assessment intake weight
Social+
Chapter3 Short-term 0 n.a*.
Physiological
Chapter4 Social Long-term 0 ++
Chapter5 Physiological Long-term + +
Chapter6 Physiological Long-term + ++
*n.a:notassessed;0:noeffect;+(++):(very)positiveeffect.
In chapter 2, it was observed that changes in meal environment of apparently

healthy, free-living elderly affected meal duration, an intermediate factor of social
facilitation but did not result in an increased dietary intake. Besides, short-term
changes insocial environment did not affect physiological influences on appetite and
energy intake in this elderly group. This suggests that dietary intake of apparently
healthy elderly would not be really sensitive to short-term variations in social factors
in a laboratory setting (Chapter3). On the contrary, long term effect of an improved
91
Chapter 7
social ambience offood consumption in nursing home ledto a stabilization of health

and nutritional status of the elderly residents (Chapter4). This indicates that elderly
respondedto longtermchanges ofsocialfactors inarather naturalsetting.
Enhancing the sensory properties ofthe cooked meal, i.e. compensating for the age-
related decline in taste and smell, was shown to improve daily feelings of hunger,
actual dietary intake and body weight in a nursing home population with a steady
healthstatus (chapter6).These resultsshowthat itispossibleto restorethe hedonic
qualities offoodandthereby stimulate appetite intheelderly.
Either a single preload intake or a regular intake of an evening energy-rich drink did
not lead to subsequent energy compensation. This lack of response to energetic
challenges presumed an impaired regulation of energy intake on short- and long-
term period in both free-living and nursing home elderly residents (chapter 3,5).
However, both long-term physiological interventions resulted in an improvement of
dietary intakeandbodyweight (table 1, chapter5-6).
Methodological considerations for research onappetite intheelderly

Population
The study population inthis thesis varied according to the mean age and the health
status. Part of the study described in chapter 2 and chapter 3 were conducted in
free-living elderly. Other studies (part of chapter2,chapter 4-6)were conducted in
elderly nursing homeresidents.
Mostofour studies (chapter2, 4-6)concentrated ona hitherto neglected segment of
society in many research studies, namely the oldest members of each community.
However, the definition of 'old-elderly' varies: in most of the communities studied so
far as well as in the present thesis, it refers to men and women of 80 years and
more.
Results ofthe SENECA baseline study showed that the participation was somewhat
selective with atendency towards a higher participation of apparently healthy elderly
(10). In the follow-up study, this tendency persisted. Consequently data of the final
SENECA study participants used inthe presentthesis areconsidered asa reference
of normaltosuccessfulaging(11).
Recruitment of the free living subjects for the study described in chapter 3 was
realized on avoluntary basis. Results showed that the participants were still socially
and physically active in daily life and could be classified as relatively healthy or
normal agers. These subjects were also relatively young compared to our other
studies.
Thefrailfree-living elderly population described inchapter2was found after specific
selection to have limited physical fitness and health. Health and nutritional
92
General Discussion
parameters ofthese participants described elsewhere (12;13)showedthattheywere

below average and that the downward process of frailty might already have started
to proceed inthisgroup ofelderly.
Nursing home residents sufferfromchronic disabilities,are oftenwheel-chair bondor
use awalking frame and may present decrease in cognitive functioning. They have
low energy needs and a relatively poor health condition. Consequently, they are at
high risk for nutritional deficiencies. Therefore participating to an intervention study
may represent a heavy charge for them because of the duration, the intervention
itself, the questionnaires they have to answer and the task, if any, they have to
complete. In addition, it is important to mention that none of the nursing home
residents who joined our studies had diagnosed of severe impaired cognitive
function or dementia. However we were confronted to the fact that some patients
were not capable to understand or answer questionnaires especially developed for
elderly.
In summary, the population described inthe present thesis was heterogeneous and
the interventions conducted were 'group-specific'.Therefore, results obtainedfor one
group ofsubjects might needto be adapted for another group ofelderly and can not
always begeneralizedtotheelderly population asawhole.
Dietaryassessmentmethods
To assess dietary intake is important for two reasons:first it may act as an absolute
measure of appetite and second it is an important determinant of the nutritional
status. Further in the absence of reliable tools and/or biochemical indicators to
assess of appetite in the elderly, dietary intake remains one of the most objective
andfeasible assessment appetite indifferent groupofelderly.
Twotypes of modified 3-day record method were usedto assess daily dietary intake
in the intervention studies (chapter 4-5) described in this thesis. These methods
were found to be an appropriate method to get information on the actual food
consumption in an accurate way (14). In both cases, consumption at the main meal
was assessed by weighing all items before and after. Consumption at the bread-
based meal was in Chapter4-5 assessed as the cooked meal and between meals
consumption was observed and registered in a diary. In chapter 6, subjects had to
record bread-based meals and snacks consumption in a diary checked by a
dietician.The use ofasimilar basic method inallstudies makes iteasier to compare
thedietary intakebetweenthegroups butalsotodetermine possible underreporting.
Table 2 shows a comparison between total measured daily energy intake and
calculated basal metabolic rate (BMR)(15) i.e. the minimum energy requirement to
maintain vital functions. It indicates that data obtained in chapter 4-5 are more
93
Chapter 7
accurate than those obtained in the latter study (chapter 6) since the measured
energy intake is more or less equal to the calculated BMR and suggests no energy
expenditure dueto physical activity which isquite impossible. Results intable 2also
indicatethatwomenaremore likelyto report lessaccuratelythan menare.
Table 2:differences between calculated estimatedbasalmetabolic rate andmeasured daily
dietaryenergyintake with differentmethods atbaseline ofthestudiesinnursinghomes.
body weight Measured daily Predicted Difference

Study N sex
(kg) dietary El (MJ)(a) BMR (MJ) (b) (a)-(b)
Chapter 3: Obs. 10 M 70.3(16.3) 7.3(1.2) 6.0 1.3
+ weighing 28 F 60.0(11.1) 5.8 (0.9) 5.1 0.7
Chapter 4: Obs. 11 M 68.4(12.5) 7.8(1.8) 5.9 1.9
+ weighing 23 F 66.0(16.0) 6.3(1.8) 5.4 0.9
Chapter 5: diary 13 M 70.7 (12.0) 7.2 (2.0) 6.0 1.2
+ weighing 54 F 70.5(14.4) 5.6(1.2) 5.6 -0.0
El: energy intake, BMR basal metabolic rate, obs +weighing: combination of observation
andweighing backmethod,diary +weighing :combination offooddiary andweighingback
method, body weight and measured daily El are given as mean (SD). BMR i.e. minimum
energy requirement based on the FAC7WHO/UNU equation predicting BMR from body
weightinolderadults(15).
Although the method used in chapter 4-5 seems to be more precise, the
disadvantage is that it is atime consuming method.An average 3-days observation
ofonesubject usually requires another halfadaybeforethedata areallstored inthe
computer andsuitableforstatisticalanlysis.
Assessment of dietary intake is a relatively simple way to provide information on
appetite but attention should be given to the precision of the method used since it
might influence oreven biastheresults.
Anthropometry
The biological basis for the selection of the anthropometric variables was related to
their expected associations with food habits, health and well being in the elderly.
Body weight was included because it reflects the recent and present balance
between energy intake and energy expenditure and because weight and changes in
weight intheelderly are related to risk of mortality. (4;6;16). Weight can be obtained
with minimal errors of measurement, although it represents a difficult and relatively
heavy measure to conduct in wheelchair bound or bedridden patients. However,
body weight has a large variation within a group of subjects with similar health
condition (table 2). Further, mean body weight was not low despite a relatively poor
healthcondition inthe nursing homegroups (table2).
94
General Discussion
Inyoung or middle-aged adults, stature is an important index of nutrition and health

during growth. For corresponding reasons, stature should be an important measure
in the elderly. However, stature decreases with aging to extents that differ among
individuals (17). These decreases may lead to bias in data analyses with an
overestimationofbody mass index,the body massdivided bythesquareofheight.
Therefore inthepresent studies bodyweight andchanges inbodyweight rather than
BMIwerechosenasendpoint oftheintervention.
The useofquestionnaires toassessappetiteinthe elderly

Two types of questionnaires were used in this thesis to assess appetite in elderly
subjects. In chapter 3 Visual Analogue Scales (VAS) were used. The VAS's give
indication on hunger, satiety and appetite feelings in young and elderly subjects.
These scales refer to an interval of time between two eating moments. It gives
indication on hunger and satiety on a very short term. However the use of such
scales are not recommended for elderly with reduced cognitive function or impaired
memory functioning.We also assumethat itwould bebiased insubjects who havea
fixed time meal service, i.e. eat because they are served but not because they are
hungry (chapter3, (18)).
In nursing home elderly, direct questions to the elderly on appetite are not always
possible, but from observation and reports from the nurse may help to indicate how
appetite ofthe residents is. However, this istime consuming and results are likely to
be biased. In chapter2 and 6,we used a relatively new tool, the Appetite, Hunger
and Sensory Perceptions (AHSP) questionnaire indifferent health groups of elderly.
Responses on the AHSP do not inform on the little day-to-day or meal-to-meal
variations ofappetite, itgives a reliable globalassessment ofappetite and relateson
the longterm,e.g.feelings ofappetite before andafter retirement, inallcategories of
elderly. It isalso sensitive enough todetect differences between health categories of
subjects. Besides, the AHSP was well accepted and understood by nursing home
elderly.
Determinants and regulation ofappetite intheelderly: asocial approach

Social influence on food intake has been defined as the effect of environmental
circumstances leading to a choice and consumption of foods, i.e., eating and/or
enjoying eating (19). Culture, family customs and traditions, and religious beliefs
influence appetite. In addition to lifelong eating habits, other factors that affect food
intakeare livingarrangements,food availability, mentalandemotional state, physical
limitations, and home-making skills. Food habits formed in youth and middle-aged
maybedifficult tomodify atoldandveryoldage.
95
Chapter 7
It has been suggested that elderly eating alone may be eating less regularly
scheduled meals and reducing the amounts and types of foods eaten (20;21). In
Chapter 3, we investigated the effect of social facilitation in healthy elderly during
lunch. Duringthe lunch consumption subjects wereeither inavery cozy environment
eatingwith 6to 8 people on onetable or ina non-cozy environment with a delimited
space and only one person per table. Results showed that eating together did
promote longer mealduration but nochange indietary intake. Inthis study in healthy
elderly, eating out in different environments, but in a laboratory setting did not
interfere with both choice and consumption of the bread based-meal, probably
because their lunch intake is the result of lifelong eating habits. Results might have
been different if a cooked meal was served: proportions and quantities consumed
are less settled and might respond more tothe actual appetite than the lunch did.In
all cultures, eating with friends is important and particularly so for lonely or isolated
old people since it maintains social contacts. In this study the elderly were rather
active and were members of several elderly clubs of dance, bridge or tourism
association. Social network, social support and social influences are important
aspects of the social environment that provide reserves enabling persons to cope
with stressful situations in daily life. These aspects may correspond to the
environment of free-living elderly but might often be underestimated in a nursing
home environment where subjects are dependent on others for the most basic
activities ofdaily life. Further, it is importantto recognize that social patterns may be
theconsequences andnotthecausesofillness.
The routine for serving the food to nursing home elderly residents include several
aspects that might not be suitable for an optimal dietary intake. First an inadapted
distribution of the food throughout the day with the main meals served at 3
occasions, delimited in time and within 8 hours might not be the best to stimulate
appetite anddietary intake insubjects who report eating and appetite problems. The
serving of an evening energy-rich supplement, enough time to consume the meals
aswell asthe continuous availability of drinks throughout the day were shown to be
important factors to stimulate dietary intake in a nursing home setting {Chapter4,5).
Second the attention and the whole ceremonial including decoration and service
accompanying meals is often neglected for no reason while both staff and residents
would benefit from a less busy meal time (chapter 4, (22)). It also underlines the
importance of the careful attention as well as the respect of the wish of the elderly
subjects atthediningtableto improvetheirfood intake.
Actually, we assume that attention together with an improved ambiance of food
consumption may serve as a buffer against the negative effects of poor appetite on
dietary intake (Chapter4).These results are in linewith observation of other studies
96
General Discussion
where meal environment and/or the immediate and patient support offamily, friends
andneighbors might helpto prevent malnutrition (23).
Lack of attention to social function of meals, with its consequent frustrations, might
be an important contributor to nutritional problems observed in nursing home elderly
residents (24). Ina public health perspective, it means that a simple re-thinking and
reorganization of the meal service and environment might contribute to the
prevention of weight loss at little costs. This reorganization should obviously be
realized according to the will and wishes of the concerned elderly group and the
nursing staff.
Determinants and regulation of appetite in the elderly: a physiological

approach
The roleofflavorenhancers andpalatabilityoffoods
Older adults have a decline in the function of taste and smell senses, which may
lead to a decrease in food palatability and a possible failure to develop sensory-
specific satiety (25). These declines in sensory perceptions offoods may lead to an
alteration ofthe hedonicqualities offoodswith increasingage.
Cooking is important infood processing since it helpsto producethe desiredtexture,
flavor and palatability of adish. Nursing home elderly residents, who do not cook for
themselves anymore but receive their mealsfrom a central kitchen, do not have the
possibilitytomodifytheirdietasfree livingdo(26).
The influence of the enhancement of the taste and odor of the main meal was
investigated in a real life situation during a 4-month intervention study in nursing
home elderly residents. Results showed that a repeated use (about 120 days) of 4
savory flavor enhancers had a positive effect onthe pleasantness and palatability of
foods as indicated by the increase in hunger feelings, daily dietary intake and body
weight (Chapter 6). Since the repeated intensification of the actual sensory
properties of foods did not lead to a reduced intake or complaints from the
participants that could indicate boredom,weassumedthatthe useofflavor amplified
foods waswell accepted.This confirms earlier findings observed on a short term,in
2-day and 3-week experiments, suggesting that flavor amplification of foods may
change food preferences and/or consumption (27;28). Through a compensation of
the age-related olfactory and gustatory deficits, the addition of flavors at optimal
concentration for the elderly might have restored the hedonic functions of foods,
thereby promoting an increase in dietary intake and reflected by changes in body
weight. The positive results obtained with flavor amplified food (chapter 6 (27-29))
shows that in older adults the palatability of food remains a determinant of appetite
anddietary intakeasitisinyounger adults(30).
97
Chapter 7
Further, the loss of smell has been suggested to happen first and to befollowed by
loss intastewith decreasing health condition, i.e. increased risk of anorexia of aging
(chapter 2, (31)). These findings show the importance of the senses of taste and
smell in the regulation of appetite in the elderly. Decline in taste and smell might
actually beoneofthestarting pointsforanorexia ofaging.
The observed positive long-term behavioral and physiological effects of flavor
enhancers in a population with a diminished odor perception and a poor but steady
health shows the general importance that palatability of foods have in the age-
related decline of appetite and food intake. From a public health point of view, it
implies that the easy intervention, which consists of adding flavor enhancers to the
main meal, might be a beneficial and effective way to improve food palatability and
dietary intakeand might helpto reverse, inafirststage,anorexia ofaging.
Theroleofenergyandmacronutrientoffoods
Macronutrient and energy contents are known to be determinants of appetite and
dietary intake in adults (30). Few studies (18;32-34) have until now explored if their
role remains in elderly people. Impaired physiological regulation of energy and
macronutrient intake has been shown to occur indifferent studies in elderly subjects
(33;34).
We explored the short-term regulation of food intake at first exposure with no prior
experience in apparently healthy, free-living elderly subjects (chapter 3). A food
preload varying in fat, carbohydrate and energy content was offered to the subjects
90 minutes before the consumption of a test lunch meal. Results showed an
imprecise and incomplete energy and macronutrient compensation in line with prior
studies (18;33), i.e., the lunch energy intake was reduced but not by an amount
equal to the energy content of the preload. Hunger ratings also slightly differed
between preloads. The slight differences found in the dimension of energy and
macronutrient compensation in our group compared to prior studies may be partly
explained by the greater size ofthe group and by the time interval between preload
andtest meal (33). Rolls(35;36) (1991,1994) suggested that the accuracy of energy
intake regulation decrease with increasing time delay.We deliberately selected a90
minutes interval since we assume that a shorter interval such as 20 minutes would
not assess the post-ingestive and post-absorptive physiological effects, but only the
volume andweight effects ofthepreloads.
Results of Chapter3 were limited to a single exposure to the preload. Precision of
energy compensation might improve with repeated exposures to the preload
consequent to the development of learned associations between satiety, energy
content andflavor of the food (37). The effect of repeated exposures to an evening
98
General Discussion
vitamin and energy rich snack on the long-term regulation of energy intake was
investigated inChapter5.
Results showed almost noenergy intake compensation and aslight increase inbody
weight was observed. The observed mean change in body weight of 0.8kg
represents an extra energy consumption over 27 days of about 30100kJ. The snack
consumed in the same period represented an increased energy intake of about
32400 kJ so it seems that the increase in body weight was the consequence of the
extra energy intake. This suggests a loss of ability to adjust energy intake
subsequently to energy challenges and it confirms the diminished capacity to
regulate energy intake onthe longterm inelderly subjects.
Energy balance depends on two factors: energy intake and energy expenditure. All
long-term studies showed aweight gain inthe intervention group,which had,apriori,
nolowbodyweight problems. First itsuggests thatconcern should rather begivento
changes in body weight than simple measure of body weight. It also indicates that
energy balance could not be maintained. Ifthere was a lack of regulation of energy
intake, impairment inthe regulation of energy expenditure has also been suggested
lately (34) (38) andshould befurther investigated.
The incapacity to adjust energy intake in the elderly seems to be a non-reversible
process (chapter 3,5-6). In daily practice, this lack of regulation suggests that the
consumption ofenergy and nutrient dense supplements between meal could helpto
increase the daily energy intake and thereby have a beneficial effect in the
prevention ofweight loss inolder adultswith reduced homeostaticcapacities.
Conclusions
Appetite and the extent to which food is enjoyed varied greatly between people. In
the elderly, these differences may be explained by differences in the health
characteristics ofthegroupsstudied.
Long-term influence of social and environmental factors such as a good meal
environment and careful attention during food consumption was found to remain an
important determinant of appetite, especially in elderly with an unstable or poor
health condition.Therefore it is important to take the local situation and feasibility of
the activity/intervention for the individual into account when studying appetite in the
elderly.
Elderly subjects did not express energy compensation or regulation after single and
repeated exposure to energetic challenges whilethe restoration ofthe optimal flavor
concentration of foods stimulated appetite and dietary intake. If the macronutrient
and energy contents offoods do not seem to have much influence onthe regulation
99
Chapter 7
of appetite in later age, palatability of foods remains an important physiological

determinant ofdietary intake inolderadults.
The properties of foods and the context in which the foods are consumed remain
important determinants of dietary intake in the elderly and are thereby major risk
factors for anorexia of aging and its subsequent weight loss. Intervention studies in
the present thesis suggest that weight loss might be a modifiable public health
problem when the adequate nutritional interventions are applied. In this regard,
weight loss prevention should not be considered as a Utopia. From a public health
perspective,thediminished or lack of regulation inappetite anddietary intake should
be taken into account in nutritional care and/or policy, and encourage nutritional
interventions inelderly atrisk.
Further, the most important fact to keep in mind is that one should first try to know
what the elderly people want. Whether or not their will is to undergo nutritional or
social intervention to adjust their dietary intake if necessary should be up to them.
However, attractive adapted nutritional guidance and intervention may help to
improvetheir quality oflife intheir lastyears.
Recommendations forfuture research

The present thesis is the first to describe research on both social and physiological
determinants of appetite (figure 1). Longitudinal and somewhat more extensive
assessment of appetite, dietary intake and changes in body weight are now
necessary to confirm thefindings ofthe present thesis. Further research on appetite
intheelderly shouldgiveattentiontothefollowing points:
• Vulnerability to hunger feeling: when does it start? Is it possible to define
biomarkers ofnutritional status andappetite inelderly subjects?
• Elderly's food choice, preference and liking: the underlying mechanisms for food
preferences and appetite are biological (physiological and genetic) and cultural.
Cultural transmission of food preferences commences in early life, at a time when
social contacts are limited to the family. With aging, chronic degenerative diseases
surface. These phenomena may be contributory to shift of food preference
mechanisms. Information, cultureandeducation also affectfood preferences. Yet no
methodsexisttoassessthesevariations offood preferences intheelderly.
• A descriptive study on eating environment and social facilitation in free-living
elderlyto answer questions suchas:Withwhom doyou eat?Who doyou seeduring
a day? How many people? The emphasis should be on finding out about 'eating or
drinkingtogether' and howthey like itratherthan ifthey 'eatout'or'eatin'
100
General Discussion
• The influence of social support including emotional, information, physical on

nutritional and health status should beconfirmed andfurther explored.This could be
realized inalong-term prospective studywith achanged environment.
• Study on the effect of isolation in the community: inequalities in social support
might exist among the socio-economic groups of free-living elderly. Some
differences in findings may be less age or nutrient related than social support
related. Some apparent nutrient effects could be related to socialfactors responsible
forthedietdifferences.
• Longitudinal assessment of appetite feelings and its relation with unexplained
unintentional weight loss: are the elderly who report a low appetite at higher risk for
weight loss?
• Depression and its relation to food choice and intake. Depression is known to be
one of the most important factors influencing weight loss in nursing home elderly
(39). It was suggested that depression has not been enough looked for and taken
into consideration in nutritional studies. Observed depression, loneliness and/or
bereavement are amongthewarning signals for potential loss ofappetite andweight
loss. The identification of relevant warning signals and risk factors followed by
appropriate grass rootsactioncould bebuilt intofuture protocols andquestionnaires.
References
1. Morley JE, Mooradian AD, Silver AJ, Heber D, Alfin SR. Nutrition in the elderly. Ann.Intern Med
1988;109:890-904.
2. Morley JE, Silver AJ. Anorexia inthe elderly. Neurobiol.Aging 1988;9:9-16.
3. Morley JE.Anorexia of aging: physiologic and pathologic. Am J Clin Nutr 1997;66:760-773.
factors in629 menandwomen aged over65. Eur J Clin Nutr 1996;50:6-11.
lower body weight] (in Dutch). NedTijdschr Geneeskd 1997;141:2184-2188.
7. Blaum CS, Fries BE, Fiatarone MA. Factors associated with low body mass index and weight loss
innursing home residents. JGerontol A Biol Sci Med Sci 1995;50:M162-M168
1990;90:1697-1706.
9. de Jong N, Chin a Paw J, de Groot LC, de Graaf C, Kok FJ, Van Staveren WA. Functional
biochemical and nutrient indices in frail elderly people are partly affected by dietary supplements but
not by exercise. J.Nutr. 1999;129:2028-2036.
investigators. Eur J Clin Nutr 1991;45 Suppl 3:23-29.
11. de Groot C, Van Staveren W, Dirren H, Hautvast JG. Summary and conclusions of the report on
the second data collection period and longitudinal analyses of the SENECA Study. Eur J Clin Nutr
1996;50Suppl 2:S123-S124
101
Chapter7_
1-191. 1999. PhD thesis, Department of Human Nutrition & Epidemiology, Wageningen University,
The Netherlands.
13. Chin a Paw, JMM. Aging in Balance, physical exercise and nutrient dense foods for the
vulnerable elderly. 1-176. 1999. PhD thesis, Division of Human Nutrition and Epidemiology;
Wageningen University, The Netherlands
15. Joint FAO/WHO/UNU Expert consultation, energy and protein requirements. 724. 1985. Geneva,
World Health Organization, technical reportseries.
16. Payette H, Coulombe C, Boutier V, Gray-Donald K. weight loss and mortality among frail living
elders:a prospective study. JGerontolA BiolSci Med Sci 1999;54:M440-M445
of elderly Europeans. SENECA Investigators. Eur J Clin Nutr 1996;50 Suppl2:S9-15.
18. Zandstra E, Mathey M-F, de Graaf C, Van Staveren W. Short-term regulation of food intake in
children, young adults andtheelderly. European Journal of Clinical Nutrition 2000;54:239-246.
19. Meiselman HR. The contextual basis for food acceptance, food choice and food intake: the food,
the situation and the individual. In: Meiselman HR, MacFie HJH, eds. Food choice, acceptance and
consumption. London: Blackie, 1996:239-263.
20. Hendy HM, Nelson GK, Greco ME. Social cognitive predictors of nutritional risk in rural elderly
adults. Int.J.Aging Hum.Dev. 1998;47:299-327.
21. Jensen GL, Kita K, Fish J, Heydt D, Frey C. Nutrition risk screening characteristics of rural older
persons: relationto functional limitations and health carecharges. Am.J.Clin.Nutr. 1997;66:819-828.
22. Inspectie voor de gezondheidzorg. Vocht-en voedselvoorziening in de nederlandse
verpleeghuizen: beleid en praktijk. 1999.The Hague,The Netherlands.
23. Walker D, Beauchene RE. The relationship of loneliness, social isolation, and physical health to
dietary adequacy of independently living elderly. J.Am.Diet.Assoc. 1991;91:300-304.
medicine. I. Effects ofachanged mealenvironment. Compr.GerontolA 1987;1:29-33.
1999;66:193-197.
26. Drewnowski A.Taste preferences andfood intake. Annu.Rev.Nutr. 1997; 17:237-253.
27. Griep Ml, Mets TF, Massart DL. Effects of flavour amplification of Quorn and yoghurt on food
preference and consumption in relation to age, BMI and odour perception. Br.J.Nutr. 2000;83:105-
113.
28. Schiffman SS, Warwick ZS. Effect of flavor enhancement of foods for the elderly on nutritional
status: food intake, biochemical indices, and anthropometric measures. Physiol.Behav. 1993;53:395-
402.
30. McCrory MA, Fuss PJ,Saltzman E, Roberts SB. Dietary determinants of energy intake and weight
regulation in healthy adults. J.Nutr. 2000;130:276S-279S.
31. Schiffman SS.Taste and smell losses innormal aging and disease. JAMA 1997;278:1357-1362.
32. Rolls BJ.Regulation offood andfluid intake inthe elderly. Ann.NYAcad Sci 1989;561:217-225.
33. Rolls BJ, Dimeo KA, Shide DJ. Age-related impairments in the regulation of food intake. Am J
Clin Nutr1995;62:923-931.
34. Roberts SB, Fuss P, Heyman MB, et al. Control of food intake in older men. JAMA
1994;272:1601-1606.
102
General Discussion
35. Rolls BJ, Kim S, McNelis AL, Fischman MW, Foltin RW, Moran TH. Time course of effects of
preloads high in fat or carbohydrate on food intake and hunger ratings in humans. Am J Physiol.
1991;260:R756-R763
36. Rolls BJ, Kim HS, Fischman MW, Foltin RW, Moran TH, Stoner SA. Satiety after preloads with
different amounts of fat and carbohydrate: implications for obesity [see comments]. Am J Clin Nutr
1994;60:476-487.
37. Louis SJ,Tournier A, Verger P, Chabert M, Delorme B, Hossenlopp J. Learned caloric adjustment
of human intake. Appetite. 1989;12:95-103.
38. Toth MJ, Poehlman ET. energetic adaptation to chronic disease in the elderly. Nutr Rev
2000;58:61-66.
1994;42:583-585.
103
Summary
Summary
Aging is often accompanied by anorexia of aging, described as a

decline inappetite, a lower dietary intake and followed by unexplained weight
loss. The present thesis described research on anorexia of aging. Focus was
givento social and physiological determinants ofappetite andthe relationship
with dietary intake and/or body weight was examined in observation(chapter
2)aswellasinintervention studies (chapter3-6).
First a cross-sectional observation study (Chapter 2) on taste, smell
appetite and hunger feelings inelderly subjects was conducted.The aimwas
to detect possible differences in appetite between different health categories
of elderly using the Appetite, Hunger feelings and Sensory Perception (AHSP)
questionnaire. Three health categories ofelderly subjects were selectedforthis
study:free-livingwithnohelp,free-livingwith helpandnursinghomeelderly.For
each group, collected data were general characteristics, anthropometry and
answers tothe AHSP. TheAHSP questionnaire includes 29-items focusing on
feelingsofhungerandappetiteaswellastasteandsmellperceptionaddressing
boththepresentsituationandtheperiodbeforeretirement.
Results indicated a decrease in feelings of appetite and hunger with a
deterioration of health conditions. It also shows that in relatively healthy
elderly subjects appetite remained an indicator of body weight while this
relationdisappeared innursing homeelderly residents.
The decline in average food intake in elderly people is attributed to
both physiological and social factors. These factors are usually studied in
isolation.The studydescribed in Chapter3concerned anexperiment inwhich
the effect of social setting on food intake was compared with the effect of
physiological challenges on food intake in 24 elderly subjects. Physiological
effects were assessed using apreload-test-meal design with a no load, and4
preload conditions.Thepreloads consisted of 300g of strawberry yogurt drink.
Thetest-meal,served 90 minutes after the preload consumption,was a lunch
ofwhich subjects could eat adlibitum.During the lunch consumption subjects
were either inavery cozy environment eating with 6to 8people on onetable
or ina non-cozy environment with a delimited space and only one person per
table.
Results showed that the presence of others did only promote longer meal
duration but there was no increase in dietary intake. In apparently healthy
elderly subjects, physiological parameters have a stronger effect than
changes insocial environment on appetite and energy intake. However, meal
duration, an intermediate factor of social facilitation, was strongly affected by
mealenvironment inthiselderlypopulation.
106
Summary
In Chapter 4, long-term effects of a changed environment and

atmosphere ofmealconsumptiononappetite andhealthwere explored during
aone-year intervention study inanursing home(n=38).Aneffort wasmadeto
improvethequality ofthe mealtime experience ofsubjects inthe experimental
group by changes on the attractiveness of the dining room,the food service,
and the organization of the mealtime assistance. For control subjects,
mealtime experience was unchanged from the situation before the onset of
the study. Both groups were served the same meals and meal patterns were
also the same for the two groups. Dietary intake, indicators of nutritional and
health status and quality of life (Sickness Impact Profile and Philadelphia
Geriatric Center Moral Scale) were assessed before and after one year of
intervention. Bodyweight informationwasobtained everyfour months.
Increased body weight, stable biochemical indicators of health status and
quality of life scores in the experimental group indicated a relatively stable
health condition while negative changes in the control group suggested a
decline inhealth status.These resultssuggestthat relatively minor changes in
mealtime circumstances ofelderly nursing home residents can have beneficial
effects on their food intake and nutritional status. In other words, social
support and physical environment could be reported as one of the
determinants for dietary intake in elderly nursing home residents. Further
these changes were simple to implement ands could be made without
additional costtothe nursinghome.
Inchapter5we investigated the effects ofthe introduction of anevening
vitamin- and energy-rich snack on average dietary intake and body weight of
nursing-home elderly. A parallel intervention study design was used. Every
evening for 30 days, the experimental group (n=26) consumed an energy and
vitamin-rich drink (200 ml; 1,2 MJ)around 7:30 p.m.whereas thecontrolgroup
(n=26) received usualcoffeeandtea.
A regular intake of an evening energy and vitamin -rich drink in the evening
had a positive influence on dietary intake and led to an increase in body
weight, thereby helping in the prevention of weight loss in nursing-home
elderly.
Taste and smell losses occurwith aging.These changes are supposed
to decrease the enjoyment of food, subsequently reduce food consumption
and negatively influence the nutritional status of elderly, especially the frail
ones. Flavor amplified foods have been proposed to be a feasible way to
compensate the age-related decreases. In chapter 6, we determined if the
addition of flavor enhancers to the cooked meal for nursing home elderly
107
Summary
promotes food consumption and provides nutritional benefits. During 16

weeks, the cooked meal of the intervention group (n=36) was sprinkled with
flavor enhancers, while the control group (n=31) received the regular cooked
meal. Feelings of appetite, dietary intake and anthropometry were assessed
beforeandattheendtheintervention.
Intake at the cooked meal, hunger feelings and bodyweight increased inthe
intervention group but not in the control group. Since the repeated intake of
flavor amplified foods did not lead to boredom, it suggests that the
intervention was well accepted. Adding ready to use flavor enhancers to the
cooked meal was in the present study a simple but effective way to improve
daily feelings of hunger, actual dietary intake and body weight in a nursing
home populationwithasteady healthstatus.
Finally in chapter 7, the main findings of the studies described in this
thesis arediscussed in relationto thefindings of other. First, appetite andthe
extent to which food is enjoyed varied greatly between people. Inthe elderly,
these differences may be explained by differences in the health
characteristics of the groups studied. Second, social and environmental
factors remained important determinants of appetite, more especially in
elderly with an unstable or poor health condition. Third, the incapacity to
adjust energy intake inthe elderly on both short and long-term seems to bea
non-reversible process. In daily practice, this lack of regulation suggests that
the consumption of energy and nutrient dense supplements between meal
could helpto prevent weight loss inolderadults.
The properties of foods and the context in which the foods are consumed
remain important determinants ofdietary intake inthe elderly and arethereby
major risk factors for anorexia of aging and its subsequent weight loss. From
a public health perspective, the lack of regulation in appetite and dietary
intake should encourage the use and consumption of nutritional interventions
inelderly atrisk.
108
Resume
Resume
Le vieillissement s'accompagne souvent d'une anorexie caracterisee par une

diminutiondeI'appetitetde laprisealimentaire,etquiestsuivied'une pertedepoids
inexpliquee. La presente these propose les resultats de differentes etudes sur
I'anorexie lieeauvieillissement. Cesetudesonteupour butdedeterminer I'influence
des facteurs physiologiques et sociaux sur I'appetit, et leur relation avec la prise
alimentaire et/ou la masse corporelle a fait I'objet d'observations (Chapitre 2) et
d'interventions (Chapitres3a6).
En premier lieu, une etude observationnelle croisee (Chapitre2) sur le gout,
I'odoratetde I'appetitet de lasensation defaim aete effectuee chez des personnes
agees. Lebutdecetteetudeaetedemettreenevidence lesdifferences eventuelles
d'appetit entre differents groupes de sante de personnes agees en utilisant le
questionnaire " Appetite, Hunger feelings and Sensory Perception" (AHSP). A cet
effet, trois categories de sante ont ete selectionnees : (1) vivant a domicile sans
aide, (2) vivant a domicile avec aide, (3) vivant en maison de retraite. Pour chaque
categorie, des donnees ont ete recueillies sur les points suivants : caracteristiques
generales, anthropometric, reponses fournies au questionnaire AHSP. Ce dernier
comporte 29 questions portant principalement sur la sensation de faim et d'appetit,
et aussi sur le gout et I'odorat, tant en ce qui concerne la situation actuelle des
personnesageesquesurlaperiode precedant leurretraite.
Les resultats ont montre une diminution de la sensation d'appetit et de faim lorsque
les conditions de sante se deteriorent. On observe egalement que I'appetit reste un
indicateur de la masse corporelle chez les sujets ages relativement en bonne sante
maisquecette relationdisparaTtchezceuxquisontenmaisonderetraite.
La diminution de la prise alimentaire moyenne chez les personnes agees est
attribuee a des causes a la fois physiologiques et sociales mais celles-ci sont, en
general, etudiees separement. L'etude presentee dans le Chapitre3 a ete realisee
sur un groupe de personnes agees (n=24) chez lesquelles on a compare I'influence
de I'environnement social et celle des conditions physiologiques sur la prise
alimentaire. Les effets physiologiques ont ete evalues par la methode du
"supplement suivi d'un repas-test" (preload-test-meal design). Cette methode
consiste a servir aux sujets un repas-test (une fois) non precede d'un supplement
(temoin) et 4 fois un supplement suivi d'un repas-test. Le supplement consistait en
300g de yaourt aux fraises a boire et le repas-test, servi 90 minutes apres la
consommation du supplement, etait un repas froid dont les composantes etaient
servies a volonte. Les sujets consommaient ce dernier soit dans les conditions de
confort d'un restaurant, par tables de 6 a 8 personnes, ou dans des conditions
d'absence deconfort, mangeant seulsdans unespacedelimite.
110
R6sum6
On observe que la presence de convives entrame une prolongation de laduree des

repas mais n'augmente pas la prise alimentaire. Chez les personnes agees
apparemment en bonne sante, I'influence des parametres physiologiques sur
I'appetitet laconsommationenergetiqueestplusforte quecelledesmodifications de
I'environnement social. En revanche la duree des repas, facteur intermediaire de
"socialfacilitation", depend etroitement du type d'environnement dans lequel ils sont
pris.
Dans le Chapitre 4, les effets a long terme de la modification de
I'environnement etde I'ambiancedes repas sur I'appetitet sur lasanteonteteI'objet
d'une etuded'une anneedans unemaisonde retraite (n =38).A cet effet, laqualite
du moment du repas des sujets du groupe experimental a ete amelioree en ce qui
concerne I'ambiance et ledecor de lasalleamanger, la presentation materielle etle
service des repas tandis que rien n'a ete modifie pour les repas du groupe temoin.
Les deux groupes ont consomme les memes aliments, prepares de la meme
maniere. La prise alimentaire, indicateur de I'etat nutritionnel, de la sante et de la
qualite de vie a ete mesuree avant et a la fin de I'intervention. De plus, une pesee
dessujetsaeteeffectuee tous lesquatre mois.
L'augmentation de la masse corporelle, la stabilite des indicateurs biochimiques de
sante et le niveau de la qualite de vie du groupe experimental sont I'indice d'une
relativestabilitedesconditionsdevietandis que lamodification negativedesmemes
parametresdanslegroupetemoin suggere unediminution du niveaudesantedece
groupe. De tels resultats permettent de penser que des modifications relativement
mineures des conditions dans lesquelles les personnes agees en maison de retraite
prennent leurs repas, peuvent avoir des effets benefiques sur leur prise alimentaire
et sur leur etat nutritionnel. En d'autres termes, le soutien et le cadre physique
peuvent etre consideres comme importantsdeterminants de la prise alimentaire des
residents de maison de retraite. En outre, les modifications utilisees pour cette
intervention sont simples a mettre en ceuvre et n'ont pas entrame de depenses
supplementairespour lamaisonderetraite.
Le Chapitre5traitedeseffets de I'introduction, lesoir, d'une collation enrichie
en vitamines et en energie sur la prise alimentaire moyenne et le poids corporel de
residents de maisons de retraite. L'etude d'intervention parallele suivante a ete
congueeteffectuee :chaque soir, pendant 30jours,ungroupeexperimental (n=26)
a consomme une boisson riche en energie et en vitamines (200 ml, 1,2 MJ) tandis
que legroupetemoin (n =26) ne recevait, comme d'habitude, qu'un cafe ou unthe.
La consommation reguliere d'une telle boisson, le soir, aentrame une augmentation
de la prise alimentaire et de la masse corporelle, ce qui en fait un facteur de
111
R6sum&
prevention de la perte de masse corporelle pour les personnes agees residant en

maisonderetraite.
La perte de gout et d'odorat accompagne le vieillissement. Ce changement
entraTne,en principe, unediminution du plaisir de manger, reduit en consequence la
consommation de nourriture et, de ce fait, influence negativement le statut
nutritionnel des personnes agees, en particulier des plus fragiles. L'utilisation
d'exhausteurs degout aete proposee comme I'undes moyens de compensation de
cette perte de gout et d'odorat. Le Chapitre 6 examine si I'adjonction de tels
exhausteurs aux aliments cuisines augmente la consommation de nourriture et se
revele benefique sur le plan nutritionnel. Pendant 16 semaines, un groupe
experimental (n = 36) a consomme des aliments cuisines et saupoudres
d'exhausteurs de gout tandis qu'un groupe temoin (n = 31) ne recevait que des
alimentscuisinessansaucuneadjonction.Lasensationd'appetit, laprisealimentaire
et I'anthropometrie ontetemesureesavantetalafinde I'intervention.
La prise alimentaire de repas cuisines, la sensation de faim et la masse corporelle
ont augmente dans le groupe experimental mais pas dans le groupe temoin. Lefait
que laconsommation repeteed'alimentsadditionnes d'exhausteurs degout n'aitpas
entraTne de lassitude chez les sujets du groupe experimental suggere que
I'intervention aetebienacceptee. L'adjonction d'exhausteurs degout pretsaI'emploi
aete,pour lapresente etude,uneprocedure simple maisefficace pouraugmenter la
sensation defaim,la prise alimentaire reelle et la masse corporelle d'une population
vivant enmaisonde retraiteet presentant unetatdesante regulier.
Le Chapitre7discute lesprincipaux resultats de la presente these en relation
avec ceux d'autres recherches. En premier lieu, I'appetit et le plaisir de manger
varient considerablement selon les individus. Chez les personnes agees, ces
variations peuvent etre dues a des differences dans les caracteristiques de sante
entre lesgroupes etudies. Ensecond lieu, lesfacteurs sociaux etenvironnementaux
sont des determinants importants de I'appetit, en particulier chez les personnes
agees qui presentent des conditions de sante mauvaises ou instables. En troisieme
lieu, I'incapacite des personnes agees a adapter leur apport energetique a long
termesembleetreunphenomeneirreversible.
Les proprietes des aliments et le contexte dans lequel ils sont consommes
demeurent des determinants importants de la prise alimentaire chez les personnes
agees et sont, de ce fait, des facteurs de risques majeurs pour I'anorexie liee au
vieillissement. Du point de vue de la sante publique, le manque de regulation de la
prise alimentaire devrait amener a recommander des interventions nutritionnelles
chez lespersonnes ageesarisquepour prevenir lapertedepoids.
112
Samenvatting
Samenvatting
Veroudering gaat dikwijls gepaard met "anorexia of aging". Dit kan

beschreven worden als een vermindering van de eetlust en een verlaging van de
voedselinname bij veroudering met een niet intentioneel gewichtsverlies als gevolg.
In dit proefschrift worden een observationele studie (hoofdstuk 2) en een aantal
interventiestudies (hoofdstuk3-6)beschreven op hetgebied van "anorexia ofaging".
Dezestudieswarengericht opdesociale enfysiologische determinantenvaneetlust
inrelatie metvoedselinname en/oflichaamsgewicht.
Allereerst werd een cross-sectionele observationele studie uitgevoerd naar
smaak,geur eneetlust op gevoelens van honger bij ouderen (hoofdstuk2). Hetdoel
was om te onderzoeken of er verschil bestaat in eetlust tussen ouderen die zich in
verschillende gezondheidscategorieen bevinden. Ouderen uit de volgende 3
gezondheidscategorieen werden geselecteerd: thuiswonend zonder hulp,
thuiswonend met hulp en wonend in een verpleeghuis. Voor elke groep werden
gegevens verzameld met betrekking tot algemene karakteristieken en
anthropometrie. Met behulp van de Hunger Feelings and Sensory Perception
(AHSP)-vragenlijst, werden gegevens verzameld op het gebied van eetlust,
hongergevoelens enperceptievansmaakengeur.
De resultaten lieten een relatie zien tussen de vermindering van gevoelens van
eetlust en honger met een verslechtering van de gezondheid. Verder bleek eetlust
als indicator voor gewicht te kunnen fungeren bij relatief gezonde ouderen. Deze
relatiewas nietaanwezigwas bijverpleeghuisouderen.
Een vermindering van de voedselinname bij ouderen wordt aan zowel
fysiologische als sociale factoren toegeschreven. Deze factoren worden gewoonlijk
apart bestudeerd. Inhoofdstuk3wordt eenexperiment beschreven waarin het effect
van de sociale setting van de voedselinname werd vergeleken met het effect van
een fysiologische preload in de regulatie van de voedselinname bij 24 ouderen.
Fysiologische effecten werden bepaald met een zo genaamd "preload-test maaltijd
design". Depreloads bestonden uit4 aardbeien-yoghurtdranken (300g)varierend in
energie-, vet- en koolhydraatgehalten. Een vijfde conditie -geen preload- werd
gebruikt om de basisvoedselinname bij de testmaaltijd te meten. Negentig minuten
na de consumptie van de preload werd deze testmaaltijd in de vorm van een ad
libitum lunch geserveerd. De effecten van sociale factoren op de voedselinname
werdengemeten ineensaaieeneengezellige omgeving. Indesaaieomgevingaten
de proefpersonen afgescheiden door tussenschotten. In de gezellige omgeving was
de eetzaal gedecoreerd en aten de proefpersonen samen aan tafel met 6 tot 8
personen.
Deaanwezigheid vananderen verlengde alleende maaltijdduur. Dit hadechter geen
verhoging van de voedselinname tot gevolg. Onder deze gezonde ouderen hadden
114
Samenvatting
defysiologische parameters een sterker effect op eetlust en energie-inname dan de

veranderingen in sociale omgevingsfactoren. Maaltijdduur als intermediaire factor
van sociale facilitatie werd echter sterk be'invloed door maaltijdomgeving in deze
populatie.
In hoofstuk 4 werden de lange termijn effecten van een verandering in
omgeving en sfeer tijdens de maaltijdconsumptie onderzocht op eetlust en
gezondheid. Deze interventiestudie vond plaats ineenverpleeghuis enduurde 1 jaar
(n=38). In de experimentele groep is getracht om de kwaliteit van de
maaltijdconsumptie te verhogen door de eetzaal gezelliger te maken en de
organisatie random de voedselservice en de beschikbaarheid van
voedingsassistenten te verbeteren. Voor de controlegroep bleef de situatie
ongewijzigd. Het maaltijdpatroon en de maaltijden die geserveerd werden, waren
voor beide groepen hetzelfde. De voedselinname, indicatoren voor voedings- en
gezondheidsstatus en kwaliteit van leven werden voor en 1jaar na de interventie
bepaald. Hetlichaamsgewichtwerdelke4 maandengemeten.
In de experimentele groep bleef de gezondheidstoestand stabiel; het
lichaamsgewicht namtoe en de indicatoren voor gezondheidsstatus en kwaliteit van
leven bleven stabiel. Negatieve veranderingen in deze parameters lieten in de
controlegroep echter een vermindering van de gezondheidstoestand zien. De
resultaten suggereren dat relatief kleineveranderingen inmaaltijdsituatiebijouderen
in verpleeghuizen, een gunstige invloed kunnen hebben op voedselinname en
voedingsstatus. Sociale ondersteuning en omgevingsfactoren zijn dus beschreven
belangrijke determinanten van voedselinname bij verpleeghuisouderen. Deze
veranderingen bleken eenvoudig te implementeren zonder dat dit extra kosten met
zichmeebracht.
In hoofdstuk 5 onderzochten we de effecten van een energie- en
vitamineverrijkte snack op de gemiddelde voedselinname en het lichaamsgewicht
van verpleeghuisouderen. Hiervoor werd een "parallel intervention study design"
gebruikt. Gedurende 30 dagen consumeerden ouderen uit de experimentele groep
(n=26) iedere avond een energie- en vitamineverrijkte drankje (200 ml; 1,2 MJ). De
controlegroep kreegdegebruikelijke koffie ofthee.
Het bleek dat de vaste consumptie van een energie- en vitamineverrijkte drankje in
de avond een positieve invloed had op de energie-inname. Dit leidde tot een
verhoging van het lichaamsgewicht en helpt hiermee dus gewichtsverlies in
verpleeghuisouderen voorkomen.
Met verouderen vindt smaak- en geurverlies plaats. Dit heeft een negatieve
invloed op de voedselinname en hiermee op devoedingsstatus van ouderen,vooral
bij fragiele ouderen. Het toevoegen van smaakversterkers aan voedingsmiddelen
115
Samenvatting
lijkt een uitvoerbare manier te zijn om leeftijdsgerelateerde verminderingen van

voedselconsumptietot staante brengen. Inhoofdstuk6 is bepaald ofde toevoeging
van smaakversterkers aan de warme maaltijd van verpleeghuisouderen inderdaad
de voedselinname verhoogd. Gedurende 16 weken werden er smaakversterkers
over de warme maaltijd van de interventiegroep (n=36) gestrooid. De controlegroep
(n=31) kreeg de gebruikelijke warme maaltijd.Aan het begin en aan het eind vande
interventieperiode werden eetlustgevoelens, voedselinname en anthropometrie
bepaald.
Devoedselinname tijdens de warme maaltijd, hongergevoelens en lichaamsgewicht
namen toe in de interventiegroep maar niet in de controlegroep. Omdat herhaalde
consumptie van smaakverrijkte voedingsmiddelen niet leidde tot verveling, wordt
aangenomen dat de interventie goed geaccepteerd werd. De toevoeging van
smaakversterkers aandewarme maaltijdwas inde huidige studieeen simpelemaar
effectieve manier om gevoelens van honger, de voedselinname en lichaamsgewicht
bijverpleeghuisouderen meteenstabiele gezondheidsstatus teverbeteren.
Tot slot werden in hoofstuk 7de belangrijkste bevindingen van de studies uit
dit proefschrift besproken in relatie met de bevindingen van andere onderzoekers.
De belangrijkste conclusies worden hieronder genoemd. Ten eerste, waargenomen
eetlust en aangenaamheid van een product varieert enorm tussen oudere mensen.
Bij ouderen kunnen deze verschillen ten dele verklaard worden door verschillen in
gezondheidskarakteristieken van de bestudeerde groepen. Ten tweede, sociale en
omgevingsfactoren blijken belangrijke determinanten voor eetlust te zijn, vooral bij
ouderen met een onstabiele en slechte gezondheidstoestand. Ten derde, het
onvermogen om de energie-inname aan te passen aan de behoefte, lijkt een non-
reversibel proces op langetermijn. De consumptie van energie- en vitamineverrijkte
supplemententussen de maaltijden is een middel om gewichtverlies te beperken, in
de dagelijkse praktijk en insufficiente inname van vitaminen en mineralen te
voorkomen. In het kader van dit proefschrift is aangetoond dat dit binnen bestaande
middelen ineenverpleeghuis mogelijkis.
116
Remerciements
Apres 4 ans d'aller et venues sur mon petit velo noir et par tous les
temps (!), d'etudes dans des maisons de retraite aux quatre coins de la
hollande,j'ai fini de tout analyser, corriger, rediger... Maintenant il est temps
de souffler un peu!? Auparavant il me faut remercier un certain nombre de
personne sansquicelivre n'aurait puvoir lejour.
Tout d'abord, je souhaiterais remercier mon directeur et mon co-
directeur de recherche, professeur Wija van Staveren et Docteur Kees de
Graaf.
Kees, bedankt voor de vrijheid die je me hebt gegeven in het doen van het
onderzoek, hetvertrouwen datjij inmij had ooktijdensje sabbatical indeUS,
maarookjouw inzet en hetvinden van diplomatiekeoplossingen inallefasen
vandestudies. Datwasweleen heleklusomaaneen Francaisete latenzien
dat 'social facilitation' niet vanzelfsprekend is maar wel een belangrijke
determinant vandevoedselinneming.
Wija, bedankt voor de prettige en inspirerende samenwerking. Ik heb veel
bewondering voor het enthousiasme, de energie en de inzet waarmeeje de
relatietussen voeding enveroudering bestudeert. Ikwilje bedanken voor alle
stimulerende discussies die we hadden en voor de kritische wijze waarop je
mijnprotocollen en manuscriptenhebtbekeken.
Lisette, sinds het begin als Franse erasmus student was je bij betrokken!
Bedankt voorjouw interesseenadviezen betreffende mijn onderzoek.
Ikhebveelvanjullie driegeleerden nietalleenopwerkgebied.
Je tiens aussi a remercier tout particulierement Professeur Jean
Bezard pour ses conseils, ses encouragements ainsi que ses critiques au
debut demathese.
Au groupe de dieteticiennes et assistantes: Els Siebelink, Saskia
Meyboom, Henny Rexwinkels en Marieke Spaan. Beste Els: hoewel we
tijdens de studies een aantal (onverwachte) hobbels hebben moeten nemen,
isallestochsuccesvol verlopen. Bedanktvoorjouw enthousiasme, flexibiliteit,
inzet en gezelligheid tijdens alle fasen van de studies in het verpleeghuis.
Saskia: bedankt voor je goede organisatie, tijdschema... maar ook je hulp
door zo prompt op een zondag alle metingen te gaan uitvoeren. Ik heb het
zeer gewaardeerd. Henny en Marieke: bedankt voorjullie inzet en de prettige
samenwerking!
Ben en Dirk: bedankt voor jullie snelle computer oplossingen,
woordenboek (wel belangrijk alsje in 3talenwil werken), beeldjes voor dia's,
dagboekje...
Hetvaststellen vandevoedingstoestand van verpleeghuisbewoners en
de daarop volgende interventie met een avonddrankje werd in het
zorgcentrumTilburg Zuid uitgevoerd. Deverandering inambiance heeft inhet
verpleeghuis Aeneas, Breda plaatsgevonden. De verstrekking van 'flavour
enhancers' heeft in de keuken het verzorgingstehuis Rustenburg,
Wageningen. Mijn dank gaan uit naar alle medewerkers inde verpleeghuizen
die, naast hun toch ai drukke baan, toch nog tijd en belangstelling hadden
voor het werven van deelnemers, het verstreken van supplementen en/of het
uitvoeren van interventie en het verzamelen van gegevens. De harde kern
werdgevormddoor Drs.SchimmeluitTilburg,Drs.Vanneste en Drs. Poulsuit
117
Breda en Mw. Pijlman, Dhr Pinkster et Dhr Hardeman uit Wageningen. Allen
hartelijk bedanktvoorjullie inzet!
I would like to express my gratitude to all organisations that made it
possible to carry out and report the studies described in this thesis. A list of
financiers is given at the beginning of this thesis and at the end of every
chapter.
Alledeelnemersaande verschillende studies:dank voor de bereidheid
ommeetedoen.
Een aantal mensen heeft als student Voeding en Gezondheid of
Voeding en Dietetiek vol enthousiasme aan mijn onderzoek gewerkt: Bas,
Marloes, Letitia, Monique, Antine, Bianca, Hugo, Brenda, en, last but not
least,AlmaenJill.Allen hartelijk dank voorjullie bijdrage en veelsucces inde
toekomst.
Dr Daniella Schlettwein, Bernard Decarli and Henri Dirren: thanks for
giving metheopportunity toconductthe SENECAstudy inYverdon.
Xavier, Marianne etVeronique :travailler en votre compagniefut un plaisir et
je le referais sans hesiter. Veronique, Clementine, Benjamin et Una: sans
vous, mon sejour en Suisse aurait certainement ete beaucoup moins
mouvemente. Etdonemoinsenrichissantet nettement moinssympa...
A Angelica, Liesbeth, Juliawati, Elvina, Siti, Marjanka, Natasja en
Robert: het was druk op 320 maar oh zo gezellig. Andere collega's, aio's en
PhDfellows: allemaal bedankt voor de gezellige tijd. Liesbeth en Nynke wilik
graag bedanken voor de reeksvoorbeeldboekjes entips. Margjeen Dorien,ik
vond hetgezellig om metjullie naar de Exhausted Nutritionist Like to Partyte
gaan
A Juul: leuk zeg dat mijn kamergenoot meer dan een collega is
geworden! Dat je bij alles zo betrokken bent vind ik wel knap. Bedankt voor
jouw gezelligheid, belangstelling, brandnetel thee (ja, heel gezond) en
abrikozen-dadel koek als we een goede dag hadden. En tja, dat weet ik, we
kletsen te veel... maar ik vind het zo leuk! Niek, ik vond het wel leuk dat je
altijd bereidwas omonste helpenonze kamerteverbouwen. Enstraks kanje
opwintersport bijons indebuurt komen.
A Judith et Luc: ik heb alles zeer gewaardeerd wat jullie in de
afgelopen tijd voor mij en ons hebben gedaan. Jullie zijn echte goede
vrienden. Straks kunnen we weer een lekker glaasje wijn met z'n alle gaan
drinken! Inzuid Frankrijk misschien?
Judith en Juul, ik vind het hartstikke leuk dat jullie tijdens de
verdediging naast me zullen staan. Wel even goed die Franse stellingen
oefenen he?
A Monsieur et Madame Renard et ma belle-famille: merci pour les
moments dedetentes et lesbonrepas!
A Liliane &Michel,Christine &Pierre, Stephanie, Nathalie, Marisol y Henry et
tous les autres de mafamille et de mes amis, qui malgre queje sois 'si loin',
m'onttoujourssoutenue.
A mes grands-parents qui, s'ils avaient pu etre presents, auraient
certainement etetresfiers. Papy, cettetheset'es dedieecartu as reussi avec
quelques paroles a convaincre et motiver une petite fille de huit ans.
Maintenantje sais que si Tona la chance d'aller loin, ilfaut lasaisir et foncer.
MERCI!
118
Jacinthe et Fabrice : pas evident d'avoir une grande soeur qui veut
toujours avoir ledernier mot mais bon vous vous en sortez bien !Fabrice, tu
as choisi une autre route que celle que tes soeurs t'avaient tracee, vas-y,
fonce et reussi! Jacinthe, c'est maintenant a ton tour. Ne t'en fais pas, il y a
des hauts et des bas mais tu vas y arriver j'en suis certaine, et je serai la
premiere a te feliciter! Pour les week-ends de detente, t'inquiete, on s'en
occupe avecGreg.
Papa: Cayest! Cecoup-cij'aifini mesetudes. Papa et Maman, merci
pour votre soutien, votre ecoute, vos idees et votre interet tout au long de
monsejour enpays batave.J'espere pouvoirfaire aussi bienquevous.
Paul,etoui,commed'habitude,jegarde lemeilleurpour lafin.Desfins
de semaine passees a peser la consommation alimentaire de personnes
agees en maison de retraite, a faire des graphiques, preparer des posters,
demenager un bureau etc. tu I'as toujours fait sans sourciller...Et de pres
commede loin(je preferedepres mais bon...) tuasreussiatoujours mefaire
savoirquetuetaisavec moi,jet'en suis reconnaissante.
Et puis on n'a pas fini de bouger ...Allez viens,je t'emmene au vent, je
t'emmeneau-dessus desgens,etje voudraisquetuterappellesnotreamour
esteterneletpasartificiel...Encorepleind'aventuresenvue.
Merciatous.
119
About the author
Marie-FrancoiseAnne Madeleine Matheywas born onthe 13thof May 1972 inDijon,

France. In 1990,she passed her secondary high school in mathematics and physics
(baccalaureat C) at the French school "Lycee Lafontaine" in Niamey, Niger. The
same year she started her study in physiology at the "Universite de Bourgogne",
France. In 1994 she passed her BSc degree with distinction in physiology with
majors in human physiology, biochemistry and nutrition and she started her MSc in
nutrition andfood sciences atthe"Ecole Nationale Superieure de Biologie Appliquee
a la Nutrition et a I'Alimentation", Dijon, France. During her MSc, she went for 6
months tothe department of Human Nutrition ofWageningen Agricultural University,
the Netherlands,toconducta research onthe impact ofthetype ofmeals onoxygen
consumption duringexercise inyoungandelderlywomen. In 1995,sheobtained her
MScdegreewithdistinction.
From 1996 to 2000 she was appointed as a PhD fellow at the Division of Human
Nutrition & Epidemiology of Wageningen University where she carried out the
research resulting inthisthesis. In 1999she collaborated to the SENECA final study
and conducted the fieldwork in Yverdon-les-bains, Switzerland. She joined the
education program of the Graduate School VLAG (advanced courses in food
technology, Agrobiotechnology, Nutrition and Health sciences). She participated in
international courses on 'Regulation of food intake and its implications for nutrition
andobesity' (1996, 1998), 'Nutrition, Lifestyle and Epidemiology' (1997), 'Proteinand
functional foods' (2000). From 1997 till 2000, she was an editor of the Newsletter
published bythe PhD students ofthe Division of Human Nutrition & Epidemiology of
Wageningen University. She was selected to participate to the 6th Seminar of the
European Nutrition Leadership Program,June2000,Luxembourg.
120

Aging & Apetite

Uploaded by

Copyright:

Available Formats

Aging & Apetite

Uploaded by

Document Information

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Aging & Apetite

Uploaded by

Copyright:

Available Formats

Aging &Appetite

Socialandphysiological approaches intheelderly

Co-promotor: Dr. CdeGraaf

Afdeling HumaneVoeding &Epidemiologie

0000 0807 0928

Aging &Appetite. Socialandphysiological approaches intheelderly.

Coverdesign:"etje doisencore penser atoutca?!", Marie-FrancoiseMathey

Printing: Grafisch Service CentrumVan Gils B.V., Wageningen,the Netherlands

Stellingen -Theorems -Theoremes

2. Ambience of food consumption is an important determinant of dietary intake in

3. Sufficient individualised care in elderly institutions is not a finance-related but a

4. Monitoring body weight is a reliable indicator of nutritional status in

5. Scientific truth, which I formerly thought of as fixed, as though it could be

10.Aubouquet onjugelevin ;aI'odeur,lafleur;aulangage, I'homme.

12.Werken meteenbuitenlandsetaal isaccepteren om niet honderd procentvanje

Stellingen behorende bijhet proefschrift

Aging&Appetite:Socialand physiological approaches intheelderly

Ph.D.-thesisbyMarie-FrangoiseA.M.Mathey,DivisionofHuman Nutrition &

Aging isoften accompanied by anorexia of aging,described asadecline inappetite,

Chapter 1 General introduction 11

Chapter 3 Socialandphysiologicalfactors affecting food intake in 33

Chapter 4 Healtheffect ofimproved mealambiance inaDutch 47

Chapter 5 Effect ofaneveningsupplement providedto nursing-home 63

Chapter 6 Flavorenhancement offood improves dietary intakeand 75

Tablel: Evolution oftheEuropean elderlypopulation inthelast10years

Different human beingsageatadifferent rate,implyingthat asochronological ageis

Conception Old age

With aging, this complex system may be affected, leading to an appetite

Appetite dysregulation:Anorexia ofaging

Anorexia ofaging makes itdifficult for anelderly personto respondto environmental

Involuntary weight loss

with a physiological manipulation consisting of preload-test meal design in order to

Marie-FrancoiseA.M. Mathey, Nynke de Jong, Lisette C.P.G.M. de Groot,

SENECA population:free-livingwithout help

Frailelderly population:free-livingwith help

Nursing homeelderly residents: nursing home

Appetite, HungerandSensoryPerceptionquestionnaire (AHSP)

Differences in general characteristics between the three groups were assessed

AHSP questionnaire results

Table2: Meanscores of theAppetite, Hunger and SensoryPerception questionnaire in the

Table3: Spearman correlation coefficients between body weight as indicator of nutritional

Inthe nursing home group no association could be established between feelings of

Marie-FrancoiseA. M. Mathey, Elizabeth H. Zandstra, Cees de Graaf and

*Food, Quality &Preferences 2000,11(5):397-403

Subjects and methods

Table 4:ExperimentalSchedule (VAS=VisualAnalogue Scale)

Inbetween preload andtest-meal,subjectsweretogether playinggames, reading but

Subjective appetite feelings

Table 5:Energycompensation atlunchafterdifferentpreloadcondition

noload LF/LC HF/LC LF/HC HF/HC

Reduced intake(kJ) 90 224 261 265

time after the p r e l o a d c o n s u m p t i o n ( m i n u t e )

Figure 1: Absolute 'appetite forameal'ratings afterdifferentpreload conditions: no load (+),

7. MorleyJE,SilverAJ. Anorexia intheelderly. Neurobiol.Aging 1988;9:9-16.

Marie-FrancoiseA. M. Mathey, Vincent G. G.Vanneste, Cees de Graaf,

Study population: 60 invitation letters

Inclusion criteria: Non-responders: 18

weight loss weight stable weight gain

Table3Characteristicsofthedropoutgroups: ageandbody weightatbaselineandabsolute

Biochemicalindicators ofhealth status