Norsk Epidemiologi 2002; 12 (1): 55-61 55

Differences between married and unmarried men

and women in the relationship between perceived
physical health and perceived mental health
John Palner and Maurice B. Mittelmark
Research Centre for Health Promotion (HEMIL-senteret), University of Bergen, Bergen, Norway
Correspondence to: John Palner, HEMIL-senteret, Christiesgate 13, NO-5015 Bergen, Norway
Telephone: +47 55 58 89 88 E-mail: [email protected]

ABSTRACT
Married and unmarried men and women were compared with regard to the relationship between perceived
physical health and perceived mental health (well being, anxiety and depressive symptoms). Perceived
physical health were assessed by self-reported number of medical diagnoses, pain in upper body and in
lower body, and physical symptoms such as digestive problems. The study sample was population-based,
selected from 13.662 community-dwelling participants in the 1995-96 national health survey conducted by
Statistics Norway. Selected were all 275 unmarried men and 271 unmarried women ages 35-67, and equal-
ly sized, randomly selected comparison groups in the same age range. A main effect of marital status was
observed for all measures of perceived physical health. Married people reported significantly better mental
health than unmarried people at all levels of perceived physical health (p < 0.000). Possible buffering
effects (stronger protective effect of marriage in the presence of poor perceived physical health) were
investigated, but not observed. These results are the first in a community sample in Norway to confirm
consistent protective effects (although in cross-sectional data) of marriage.
1
This paper is based on a master's thesis completed by the first author under the direction of the second author, at the Research Centre
for Health Promotion. Both authors contributed to the writing of this paper.

INTRODUCTION tirement were diagnosed with muscoloskeletal pro-

blems. In 1997, the figures had increased to 39 percent
This paper reports on comparisons between married for women and 25 percent for men, respectively (4).
and unmarried Norwegian men and women with At the same time that the disturbing trends cited
regard to the relationship between perceived physical above are occurring, important social resources that
health (number of diagnoses, upper and lower body help one cope with chronic health stress may be less
pain, and physical symptoms) and perceived mental abundant in modern society. In Norway as elsewhere,
health (well being, anxiety and depressive symptoms). the ageing of the population is accompanied by shrin-
It is a common finding that people who experience king social networks, a mobile workforce results in
multiple health complaints and pain have also poorer scattered families, and changing familial norms means
psychological health, often in the form of depressive fewer enduring relationships such as those found in
symptoms (1,2). The nature of the relationship be- many marriages (5). Indeed, data from Norway indi-
tween poor physical health and poor mental health is cate fewer people getting married, the proportion of
most likely one of reciprocal influence (3). single households increasing and a growing tendency
Besides the personal burden and distress associated for the Norwegian population to live alone, at least for
with many common health complaints, the costs to so- some parts of their adult lives (5).
ciety are well documented and substantial. In Norway, The significance of these trends to the population's
musculoskeletal problems in the general population are health may be substantial. There is a large literature
the most common cause of visiting a general practitio- showing that persons with chronic diseases experience
ner, the most common cause of absenteeism from the better physical and mental health outcomes if they
work place, and the trends are worsening (4,5). In have a good social network (7-11). The health protec-
1997, more than 50 percent of all cases of medium- tion effect of having a spouse is an especially robust
term absenteeism (three days to two weeks) from the finding (12-14). For social networks in general and for
work place were due to musculoskeletal problems (6). spouse support in particular, two types of effects have
Also musculoskeletal problems account for an increa- been observed. ‘Main’ effects refers to protective
sing proportion of early retirement. In 1980, 26 percent effects that are exerted regardless of level of health.
of all women and 18 percent men receiving early re- ‘Buffer’ effects refer to protective effects that are more
56 J PALNER AND M.B. MITTELMARK

evident when health is poor, and modest or not present The analysis variables were constructed as follows.
when health is good. With the assistance of a principal components factor
Most of the health research on the main and buffe- analysis with varimax rotation performed on the Hop-
ring effects of social network has focused on morbidity kins Symptoms Checklist variables, five scales were
and mortality associated with major diseases such as constructed measuring: 1) depressive symptoms (11
the cardiovascular diseases and cancers. The very few items, Cronbach's alpha = 0.89); 2) anxiety (six items,
studies that have focussed on common health problems Chronbach's alpha = 0.84), 3) upper body pain (six
have limitations that call for additional research. items, Cronbach's alpha = 0.81); 4) lower body pain,
Jackson (15), for example, observed a buffering effect two items, Cronbach's alpha = 0.65); 5) physical symp-
of support from spouse and friends on the relationship toms, six items, Cronbach's alpha = 0.73) (Table 1).
between self-reported physical health problems and
depressive symptoms. A critical limitation of the study
was its restriction of the sample to married people on- Table 1. Mean and standard deviation for health complaints
ly. Similarly, Barstad (16) observed a moderating ef- and measures of mental health for married men and women.
fect of having a confidant on the relationship between Married Unmarried
self reported health complaints and symptoms of de- Variables Mean S.D. n Mean S.D. n
pression and anxiety. In this study as well, the possible Health complaints
influence of marital status was not investigated. Diagnoses
Men 1.85 1.25 184 1.93 1.10 180
The rationale and design of the present study,
Women 1.90 1.12 185 2.27 1.33 193
which is a secondary analysis of data from the study Upper pain
"Helseundersøkelsen-95", was founded on the obser- Men 2.29 2.29 275 2.10 2.79 275
vations discussed above. Perceived physical health Women 2.94 3.31 271 3.57 3.64 271
was hypothesised to be associated negatively with Lower pain
Men 0.41 0.41 275 0.51 1.04 275
perceived mental health. Marriage was hypothesised to
Women 0.60 1.10 271 0.64 1.20 270
exert both main and buffer effects on the perceived Physical symptoms
physical/mental health relationship. That is, it was Men 0.98 1.59 275 1.20 2.18 275
expected that married men and women, compared to Women 1.32 1.87 273 1.91 2.39 271
unmarried people, would report lower levels of mental Mental health
health at all levels of physical health. It was expected Well being
Men 6.48 1.84 275 5.94 1.97 275
also that this effect would be more pronounced among
Women 6.25 1.81 271 5.82 2.10 271
those with poor perceived physical health. Further it Depressive symptoms
was expected that the protective effect of marriage Men 1.70 3.29 274 2.73 4.31 275
would be observed after controlling statistically for Women 2.36 4.09 268 3.16 4.48 269
other factors presumed to be correlated with both mari- Anxiety
Men 0.92 1.97 275 1.31 2.30 275
tal status and health status, especially socioeconomic
Women 1.00 1.96 271 1.42 2.23 268
status.
Moderator
Marital status
METHODS Men -- -- 275 -- -- 275
"Helseundersøkelsen-95" was a cross-sectional survey Women -- -- 271 -- -- 271

conducted in Norway from September to December, Control variables

Negative impact
1995, by Statistic Norway. The study sample was
Men 1.38 1.05 191 1.48 0.99 189
population-based and included a total of 13,662 parti- Women 1.58 0.99 189 1.69 1.03 196
cipants. Information on respondents' perceived health, Age
social network and family relations, occupational Men 49.58 8.77 275 45.77 8.64 275
information, self-reported utilisation of health care Women 47.83 9.03 271 46.45 8.94 271

services and information on demographic variables

was obtained by interview and by questionnaire. Infor-
mation was collected also on self-assessed well being, At the interview the respondents were asked whether
self-reported number of medical diagnoses, and a 35 they had one or more of 12 chronic diseases with
item modified version of the Hopkins Symptom duration longer than six months; nervous diseases and
Checklist (17) was administered. The survey included symptoms; neurological diseases; cardiovascular
several components such that not all respondents re- diseases; respiratory diseases; dermatological diseases;
ceived the same questionnaires and interviews. musculoskeletal diseases; injuries. A maximum of 10
Of the 2,814 participants ages 35-67, the 546 parti- diagnoses was recorded for each respondent. A sum
cipants who reported being unmarried (single, widow, score of all diagnoses was calculated.
separated, divorced) at the time of the survey were se- Three questions were summed to create a scale of
lected for the present analysis. A comparison group of perceived mental health (Cronbach's alpha = 0.69).
the same size was selected at random from among the These were "Have you been happy and satisfied the
married participants, matched on gender. past two weeks?", "How much of the time have you felt
PERCEIVED PHYSICAL AND MENTAL HEALTH IN MARRIED AND UNMARRIED MEN AND WOMEN 57

in a good mood and had good energy recently?", and dichotomous variable. Age, personal income and nega-
"How do you perceive your own health in general?". tive health impact were included in the analyses as
The interview included a question on self-rated covariates. Estimated marginal means was obtained for
negative impact of diseases, conditions and symptoms, all predictors. Possible buffering effects (a protective
worded as follows: "We would like to know how you effect of marriage in the presence of poor perceived
evaluate the various illnesses and functional limita- physical health) were investigated by examining sta-
tions you have reported. To what degree do these tistical interactions between marital status and the
effect your every day living (all kinds of effects, variables of perceived physical health in the analysis.
including pain, anxiety, sleeping problems, exhaustive- A main effect of the diagnoses variable was obser-
ness and limitations in what you can do)?" This was ved (F = 13.57, p < 0.000), with respondents in the
used in multivariate analyses to control for severity of high category of the diagnoses variable having signifi-
illness effects. cantly lower perceived mental health scores compared
Marital status and age were determined from natio- to respondents in the low category of the diagnoses
nal registry data that were added to the data set imme- variable. A main effect was observed also for marital
diately after the interviews were completed. For the status (F = 14.21; p < 0.000), with married respondents
present study, the categories unmarried, widows, di- having higher perceived mental health scores than
vorced and separated were grouped into the category unmarried respondents. A main effect of annual
'unmarried'. Because of the well documented relation- personal income was observed (F = 18.48; p < 0.000).
ship between socioeconomic status, a variable The adjusted R2 for this model was 0.32.
assessing annual personal income was included as a A main effect of upper body pain was observed
covariate in the analysis. (F = 24.27, p < 0.000), with those reporting more pain
also reporting lower levels of perceived mental health
compared to the low pain group. A main effect of ma-
RESULTS rital status was also observed (F = 14.21, p < 0.000),
Perceived mental health with married respondents reporting higher levels of
perceived mental health than unmarried respondents.
Four analyses of variance were conducted with percei- A interaction between upper body pain by marital
ved mental health as the dependent variable, and each status was observed, however not significant (F = 3.69,
of the four measures of perceived physical health trea- NS). A main effect of annual personal income was also
ted as fixed variables and dichotomised into high and observed (F = 17.25; p < 0.000). The adjusted R2 for
low groups. Marital status was also treated as a fixed, this model was 0.33.

Figure 1. Mean score for perceived mental health, married and unmarried men and women, with low and
high upper body pain.
58 J PALNER AND M.B. MITTELMARK

A main effect of the lower body pain variable was effect of personal income was observed (OR = 0.46,
observed (F = 17.46, p < 0.000). Respondents in the CI: 0.32–0.66, p < 0.000). Respondents in the high ca-
high category of the lower body pain variable repor- tegory of the personal income variable had a signifi-
ting significantly lower perceived mental health com- cantly lower risk of depressive symptoms, compared to
pared to respondents in the low category of lower body respondents in the low personal income category. No
pain variable. A main effect was observed also for main effect of marital status was observed, and neither
marital status (F = 16.33, p < 0.000), with married res- was a buffering effect.
pondents having significantly higher scores on percei-
ved mental health compared to unmarried respondents. Anxiety
The adjusted R2 for this model was 0.32.
A main effect of the upper body pain variable was
A main effect of physical symptoms was observed
observed (OR = 3.04, CI: 1.75–5.28, p < 0.000), with
(F = 60.06, p < 0.000), with respondents in the high
respondents in the high category of the upper body
category of the physical symptoms variable having
pain variable having significantly higher risk of
significantly lower score on the perceived mental
anxiety, compared to respondents in the low category
health variable, compared to respondents in the low
of upper body pain. Similarly, a main effect of the
category of the physical symptom variable. A main
physical symptoms variable was observed (OR = 4.05,
effect was also observed for marital status (F = 13.83,
CI: 2.20–7.21, p < 0.000). Respondents in the high
p < 0.000). Unmarried respondents had significantly
category of the physical symptom variable had a
lower score on the perceived mental health variable,
significantly higher risk of anxiety, compared to
compared to the married respondents. As in the pre-
respondents in the low category of the physical
vious analysis, a main effect of income was observed
symptom variable.
(F = 12.1, p < 0.001) The adjusted R2 for this model
was 0.35.
In the four analyses described above, no statisti- DISCUSSION
cally significant buffering effects were observed.
The present study has limitations that should be noted
The dichotomous depressive symptoms and an-
from the outset. Because of small sample sizes in the
xiety variables were analysed with logistic regression.
marital status categories separated, divorced,
The four dichotomous health stressor variables were
widowed, and never married, these were collapsed into
used as predictor variables in turn. Other predictor
one category of unmarried respondents. However,
variables included marital status, gender, personal in-
evidence suggests that mental distress is more
come (in two approximately equal categories), nega-
prevalent among separated and divorced individuals,
tive health impact (in two approximately equal catego-
compared to married individuals, and this pattern may
ries) and age (in two approximately equal categories).
extend to never married persons as well (18). That
Possible buffering effects were investigated by inclu-
these various unmarried subgroups were not studied in
ding the interactions of marital status by each health
their own right is potentially important, since some
stressor variables in turn.
researchers argue that combining the separated, divor-
ced, widowed and never married into one category
Depressive symptoms
may result in an underestimation of the level of mental
A main effect of upper body pain was observed (OR = distress among the separated/divorced persons and
2.91, CI: 1.60–5.28, p < 0.000) in an analysis in which overestimation among widowed persons (19).
depressive symptoms was the predicted variable. Res- In this study and some other cross-sectional
pondents in the high category of the upper body pain studies, evidence of a protective effect of marriage on
variable had a significantly higher risk of depressive health has been demonstrated. The seeming protective
symptoms, compared to respondents in the low cate- effect of marriage may however be spurious, in part or
gory of the upper body pain variable. A main effect of in whole. It is possible, for example, that good mental
personal income was observed (OR = 0.41, CI: 0.30– health may be more frequently observed among indi-
0.58, p < 0.000). Respondents in the high category of viduals living under more affluent social and material
the personal income variable had a significantly lower conditions, compared to those that live under relatively
risk of depressive symptoms, compared to respondents poorer conditions, regardless of marital status.
in the low category of the personal income variable. Marital status differentials of health have been ob-
No main effect of marital status was observed, and served for a number of socioeconomic indicators such
neither was a buffering effect. as income, housing tenure and employment. Waite et
A main effect of the physical symptoms variable al. (20) observed a higher per capita income, higher
was observed (OR = 6.52, CI: 3.3–12.1, p < 0.000), hourly wages and a lower risk for children of dropping
with respondents in the high category of the physical out of school among married people compared to
symptoms variable having a higher risk of depressive cohabiting and divorced, separated and widowed
symptoms, compared to respondents in the low cate- people. In the same study, married people had less
gory of the physical symptoms variable. Also a main alcohol related problems and were less likely to
PERCEIVED PHYSICAL AND MENTAL HEALTH IN MARRIED AND UNMARRIED MEN AND WOMEN 59

engage in negative health behaviour (20). tes, cancer, arthritis in knee and hip, or cardiac disea-
In Hope et al's investigation of marital transitions ses) and depressive symptoms. The sample consisted
and distress (21), housing tenure was observed as an of participants aged 55-85 years. The present study,
important moderator of the change in mental health in with an age range from 35-67 years, extends the find-
conjunction with marital transition. Poor mental health ings to a much younger age group.
persisted among married women who were downward Finally, a relationship between physical health
mobile in terms of housing tenure compared to those (such as functional disability and diagnoses) and men-
who were upward mobile. The difference was even tal health (depressive symptoms and anxiety) was also
more pronounced after divorce. observed by Barstad (16) in a study based on the same
In yet another study, a beneficial effect of marriage data as the present report. In Barstad's analysis, having
on psychosomatic symptoms was observed for women a confidant influenced the physical health/mental
who were unemployed. The differential persisted in health relationship, but Barstad (16) did not make the
each of two successive five-year follow-up intervals. marital status comparisons that were the focus of the
Unemployed women without an alternative source of present investigation.
financial and instrumental support also had the poorest In this study, married persons, compared to un-
health. Also unemployed women had a higher risk of married persons, consistently reported better mental
marital dissolvement compared to employed women health regardless of their physical health status. This
(22). observation is consistent with findings of Walen et al.
In the present study all exploratory analysis in- (27), who found that social exchange from ones'
cluded variables measuring annual personal income, partner was a significant predictor of well being (life
socioeconomic status based on the standard Statistics satisfaction, positive affect, and negative affect), and
Norway formulation (23), and educational level. Only also with Sherbourne et al. (25), who observed a main
annual personal income emerged as significantly asso- effect of marital status on mental health (depression
ciated with the perceived mental health variables. and anxiety), after controlling for perceived negative
Therefore, personal income was included as a co- health impact and age. The present results are con-
variate in all the final statistical models presented in sistent also with the findings of Penninx et al. (26),
this paper. Consistent with many other studies, in who observed a main effect of having a partner on the
almost all the models, a main effect of income was relationship between self-reported arthritis (in knee
observed. However, protective effects of marriage was and hip), and depressive symptoms.
evident even when income was included in the models. At the same time that main effects of martial status
Thus the marital status health differentials observed were observed consistently in this study, there was no
are likely due to factors other than societal/macro level evidence of buffering effects, and this bears some dis-
factors, a conclusion supported by other analysis of cussion. Researchers commonly differentiate between
Norwegian data (24). two possibilities regarding the health protection effects
In this study, a relationship between self-reported of social support: Effects than happen only when stress
common health complaints and mental health mea- is present (buffering effects) and effects that happen
sured in different ways was observed. The results are regardless of level of stress (main effects, sometimes
consistent with, and extend, findings from several referred to as marginal effects). Some authorities in the
other similar studies. A similar finding to that reported field hold such strong opinions on the nature of these
here, between self-reported health (physical functio- effects, stated almost as laws; for example, “close
ning, role limitations due to physical health problems, relationships have a buffering effect, but networks of
and satisfaction with ones' own physical abilities) and friends have a main effect—that is, they work whether
mental health (anxiety, loss of control, and positive stress is present or not” (28).
affect), was observed by Sherbourne et al. (25). How- In fact the situation is more complicated than that.
ever, Sherbourne et al.'s sample was restricted to chro- Vilhjalmsson (29), for example, has demonstrated that
nically ill patients. This paper generalises the finding, in community-based studies of life stress, social sup-
since a broad range of health status was represented in port and clinical depression, linear multivariate statis-
the population-based sample of the present study. tical analyses tend to reveal buffering effects, while
Jackson (15) also observed a relationship between non-linear analytic approaches tend to reveal main
self-reported physical health (perceived health status, effects in the same data. Variability in analytic out-
perceived physical ability, and perceived health satis- come may also be influenced by the nature, intensity
faction), and self-reported depressive symptoms, but and duration of the stressor, the investigator's concep-
the study included only married persons. In the present tualisation and measurement of support (enacted/
investigation, since both married and unmarried parti- received/perceived), the source(s) of support, the
cipants were included, important protective effects of appropriateness or inappropriateness of support
marriage could be investigated, and indeed, were attempts, the nature, duration and severity of the health
found to be of importance. outcome in question, and untold numbers of combi-
More recently, Penninx et al. (26) observed a rela- nations of these variables.
tionship between self-reported health (such as diabe- In the context of the present study, in which stres-
60 J PALNER AND M.B. MITTELMARK

sors were conceptualised as common health differences among men and women on the relationship
complaints and components of mental health was between perceived physical and mental health.
conceptualised as outcome, both linear multivariate As discussed above, the literature on the influence
analysis and non-linear analysis was used. Some of marital status on the relationship between chronic
evidence of a buffering effect of marital status was health stressors and mental health is not consistent.
observed but did not reach statistical significance. This One reason for inconsistent findings may be due to the
was in part due the fact that the level of significance categorising of key variables. As noted by several
chosen for this study was set to 0.005, implying a con- authors (34) categorising of variables results in loss of
servative estimate of probability. Thus ‘really’ signifi- information that may consequently limit the ability to
cant results may be discarded in this manner. detect interaction effects and mask real differences
Perceived negative health impact (severity) of within analysis subgroups. Another plausible explana-
health stressors was strongly associated with all out- tion for the inconsistency of results in this arena may
come measures in the present study. It was fortunate be that marriage per se is not a resource in times of
that this potential confounder was measured so that it stress. Intimate relationships may have a dark side
could be controlled for in the multivariate analyses, a (35). This argument is consistent with data from Phen-
step that was not taken in any of the other studies nix et al. (26), who observed that receiving instru-
discussed above. mental support was associated with more depressive
In the literature, some studies have controlled sta- symptoms. Some results from the present study seem
tistically for the effect of gender (26,16) or have not to support this possibility. As described earlier, mean
included the variable at all (25). Other studies have score on the perceived mental health variable among
examined gender as was done in the present investi- married men in the high category of lower body pain
gation, in which no differences were observed between approached the score for unmarried men. Lower body
men and women in the relationship between perceived pain usually implies functional disability, which in
physical and mental health. This is inconsistent with turn may restrain the possibility to participate in social
findings from several other studies indicating that men activities and imply a need for instrumental support.
in general benefit more from marriage than women do, Thus, being in need of considerable instrumental sup-
although such findings mostly have been observed in port from a partner or feeling socially isolated may
studies of mortality (30) and morbidity (31). cause feelings of helplessness, or dependency, both of
Consistent with the present findings, however, are which may lead to a decreased level of perceived men-
those of Jackson (15), who observed no differences be- tal health. However, this conjecture could not be tested
tween men and women in her study on the relationship in the present study due to limitations in the data.
between self-reported physical health and depressive Concluding with what may be the most obvious
symptoms. Women have been observed also to benefit point of discussion, because the study used data from a
more from social support sources outside marriage, population-based sample, very few people reported
such as having (non-spousal) confidant support (27), high levels of depressive symptoms, anxiety and poor
friendship (32), and material resources (18). perceived mental health. This resulted in highly
As mentioned in the introduction to this paper, the skewed distributions with the large majority of partici-
prevalence of chronic health stressors such as pants reporting no or very few problems. Under this
musculoskeletal conditions and mental illness is higher circumstance, it seems noteworthy that the protective
among women compared to men in Norway and else- effects of marriage were observed. Had the sample
where (33). Therefore, the explicit study of the role of included a more heterogeneous mix of symptoms, an-
gender in stress and health studies is needed. This stu- xiety and perceived mental health, effects might well
dy is among the first in Norway to explore the possible have been more pronounced.

REFERENCES

1. Blackburn-munro G, Blackburn-munro R. Chronic pain, Chronic stress and depression: Coincide or conse-
quence? J Neuroendocrinol 2001; 13: 1009-1023.
2. Grossi G, Soares J, Ängeslevä J, Perski A. Psychosocial correlates of long-term sick leave among patients
with musculoskelatal pain. Pain 1999; 80: 607-619.
3. Lazarus RS, Folkman S. Stress, Appraisal, and Coping. New York: Springer, 1984: 445.
4. Helsedepartementet. Folkehelserapporten 1996. Oslo: Statens insitutt for folkehelse, 1996.
5. Helsedepartementet. Folkehelserapporten 1999. Oslo, 1999.
6. Rikstrygdeverket. Trygdestatistisk årbok 1998. Oslo, 1998.
7. Berkman L. Assessing the physical health effects of social networks and social support. Annu Rev Public
Health 1984; 5: 413-32.
PERCEIVED PHYSICAL AND MENTAL HEALTH IN MARRIED AND UNMARRIED MEN AND WOMEN 61

8. Schwazer R, Leppin A. Social support, health, and health behavior. In: Hurrelmann K, Losel F, eds. Health
Hazards in Adolescence: Prevention and Intervention in Childhood and Adolescence. Berlin: Walter de
Gruyter, 1990: 528.
9. Cohen S, Kaplan J, Manuck S. Social support and coronary heart disease: Underlying psychological and bio-
logical mechanisms. In: Shumaker S, Czajkowski S, eds. Social Support and Cardiovascular Disease. New
York: Plenum Press, 1994.
10. Broadwell S, Light K. Family support and cardiovascular responses in married couples during conflict and
other interactions. Int J Behav Med 1999; 6: 40-63.
11. Seeman TE. Social ties and health: The benefits of social integration. Ann Epidemiol 1996; 6: 442-451.
12. Ross C, Mirowsky J, Goldsteen K. The impact of the family on health: A decade in review. J Marriage Fam
1990; 52: 1059-1078.
13. Umberson D, Chen M, House J, Hopkins K, Slaten E. The effect of social relationships on psychological
well-being: Are men and women really so different? Am Sociol Rev 1996; 61: 837-857.
14. Horwitz A, McLaughlin J, White H. How the negative and positive aspects of partner relationsships affects
the mental health of young married people. J Health Soc Behav 1997; 39: 124-136.
15. Jackson P. Specifying the buffering hypothesis: Support, strain and depression. Soc Psychol Q 1992; 55: 363-
378.
16. Barstad A. Fortrolighet forebygger? Samfunnspeilet 1997; 2: 22-31.
17. Derogatis L, Lipman R, Rickels K, Uhlenhut E, Covi L. The Hopkins Symptom Checklist (HSCL). A self-
report symptom inventory. Behav Sci 1974; 19: 1-15.
18. Joung I, Stronks k, Van de Mheen H, Van Poppel F, Van de Meer J, Mackenbach J. The contribution of inter-
mediary factors to marital status differences in self-reported health. J Marriage Fam 1997; 59: 476-490.
19. Cotten S. marital status and mental health revisited: Examining the importance of risk factors and resources.
Fam Relat 1999; 48: 225-233.
20. Waite L. Does marriage matter? Demography 1995; 32: 483-507.
21. Hope S, Rodgers B, Power C. Marital status transitions and psychological distress: longitudinal evidence from
a national population sample. Psychol Med 1999; 29: 381-389.
22. Waldron I, Hughes M, Brooks T. Marriage protection and marriage selection – Prospective evidence for reci-
procal effects of marital status and health. Soc Sci Med 1996; 43: 113-123.
23. Øien A. Standard for classification of socio economic status. Kongsvinger: Statistics Norway, 1984:32.
24. Elstad J. Inequalities in health related to women's marital, parental, and employment status – a comparison
between the early 70s and the late 80s, Norway. Soc Sci Med 1996; 42: 75-89.
25. Sherbourne C, Hays R. Marital Status, Social support and health transitions in chronic disease patients. J
Health Soc Behav 1990; 31: 328-343.
26. Phennix B, Tilburg VT, Boeke A, Deeg D, Kriegsman D, Eijk vJ. Effects of social support and personal
coping resources on depressive symptoms: Different for various chronic disease? Health Psychol 1998; 17:
551-558.
27. Walen H, Lachman M. Social support and strain from partner, family, and friends: Costs and benefits for men
and women in adulthood. J Pers Soc Relat 2000; 17: 5-30.
28. Argyle M. Interpersonal accounts – a social psychological perspective. Br J Soc Psychol 1992; 31: 393.
29. Vilhjalmsson R. Life stress, social support and clinical depression: A reanalysis of the literature. Soc Sci Med
1993; 37: 331-342.
30. Smith K, Waitzman N. Double jeopardy: Interaction effects of marital and poverty status on the risk of
mortality. Demography 1994; 31: 487-507.
31. Shumaker S, Hill D. Gender differences in social support and physical health. Health Psychol 1991; 10: 102-
111.
32. Antonucci T, Akiyama H. An examination of sex differences in social support among older men and women.
Sex Roles 1987; 17: 11-12.
33. Sandanger I, Moum T, Ingebrigtsen G, Sørensen T, Dalgard O. Prevalence, incidence and age at onset of
psyciatric disorders in Norway. In Press 1999.
34. Cohen S, Wills T. Stress, social support, and the buffering hypothesis. Psychol Bull 1985; 98: 310-357.
35. Spitzberg B, Cupach W. The Dark Side of Close Relationships. Mahwah: Lawrance Erlbaum Assoc., 1998.