Wellbeing N Health Questionnaire

View metadata, citation and similar papers at core.ac.
uk brought to you by CORE

provided by Memorial University Research Repository
The Simple Lifestyle Indicator Questionnaire and its association
with health-related quality of life and well-being.
By
© Jacqueline Fortier
A thesis submitted to the
School of Graduate Studies
In partial fulfillment of the requirements for the degree of
Masters of Science (Clinical Epidemiology)
Faculty of Medicine
Memorial University of Newfoundland
May 2015
St John’s, Newfoundland & Labrador

Abstract
A person’s lifestyle may be related to their quality of life and well-being. This cross-
sectional survey examined the association between the Simple Lifestyle Indicator
Questionnaire, a measure of lifestyle, and health-related quality of life and well-being
among a sample of 100 adults living in St John’s, Newfoundland & Labrador. Lifestyle
was significantly, positively correlated with well-being (r=0.47, p<0.01), self-perceived
health (r=0.59, p <0.01) and mental health-related quality of life (r=0.41, p<0.01), but not
physical health-related quality of life (r=0.13, p=0.19). This study benefitted from the use
of validated questionnaires but the generalizability of these results is limited by a sample
population that was younger, more educated and of higher household income than the
general population. Lifestyle appears to be related to well-being and quality of life, and
lifestyle factors may predict quality of life in populations similar to the population in this
study.
ii
Acknowledgements
This project would not have been possible without the assistance of the staff and
administrators at the Family Practice Unit, Shea Heights Community Health Centre, Ross
Family Medicine Centre, and the Avalon Mall. Thanks are also due to the participants,
who showed a great deal of interest in the project and kindly took time out of their busy
days to fill out questionnaires.
My supervisory committee, Shabnam Asghari and Cheri Bethune, provided invaluable

advice, input and suggestions throughout the course of my research, and both the project
and this manuscript were much improved by their suggestions. My colleagues at the
Primary Healthcare Research Unit have been fantastic on both professional and personal
levels, and my experience in St. John’s would not have been the same without them.
My parents were completely supportive when I moved to the other side of the country to
pursue my Masters and have been cheering me on ever since. Thank you to Emily, Matt
and Hannah for always being on my team. Hollis provided levelheaded advice (and the
occasional pep talk) at all stages of the project, and I couldn’t ask for a better friend.
There are too many others to mention, including good friends, fabulous roommates, and
my wonderful family. Thank you all for your love, kindness, and (endless!) patience over
the past two years.
Finally, I would like to thank my supervisor, Marshall Godwin. From the first proposal to
the final draft, he has provided support, advice, encouragement, constructive criticism,
and motivation in just the right amounts. Throughout the process he remained steadfast in
his belief that I could complete a study and write it up on time – he was right. Marshall
has been a truly wonderful teacher, supervisor and mentor, and for that I’ll say a
tremendous thank you.
iii
Contents
Abstract ........................................................................................................................................... ii
Acknowledgements ........................................................................................................................ iii
List of Tables ................................................................................................................................. vi
List of Figures ............................................................................................................................... vii
List of Abbreviations and Symbols .............................................................................................. viii
List of Appendices ......................................................................................................................... ix
1.0 Introduction and Overview ....................................................................................................... 1
1.1 Context .................................................................................................................................. 1
1.2 Health-related quality of life ................................................................................................. 2
1.3 Well-being ............................................................................................................................. 6
1.4 Lifestyle and the Simple Lifestyle Indicator Questionnaire.................................................. 8
1.5 Relationship between lifestyle and health-related quality of life ........................................ 11
2.0 Objectives and Research Questions ........................................................................................ 18
Primary question ....................................................................................................................... 18
Secondary questions .................................................................................................................. 18
3.0 Methods................................................................................................................................... 19
3.1 Study design ........................................................................................................................ 19
3.2 Pilot study............................................................................................................................ 19
3.3 Study population ................................................................................................................. 19
3.4 Sample size calculations...................................................................................................... 20
3.5 Sampling strategy and recruitment procedures ................................................................... 20
3.5.1 Recruitment at shopping mall ...................................................................................... 21
3.5.2 Recruitment in family physician waiting rooms .......................................................... 22
3.5.3 Recruitment at Faculty of Medicine ............................................................................ 22
3.5.4 Other ............................................................................................................................ 23
3.6 Instruments and variables .................................................................................................... 23
3.6.1 Measuring lifestyle....................................................................................................... 24
3.6.2 Measuring HRQoL....................................................................................................... 25
iv
3.6.3 Measuring well-being .................................................................................................. 25
3.7 Data entry and cleaning ....................................................................................................... 26
3.7.1 Missing data ................................................................................................................. 27
3.8 Data analysis ....................................................................................................................... 27
4.0 Results ..................................................................................................................................... 29
4.1 Response rate....................................................................................................................... 29
4.2 Descriptive statistics ............................................................................................................ 29
4.3 Correlation between overall SLIQ score and measures of HRQoL and well-being ........... 34
4.4 Correlation between the raw and category scores of the five SLIQ dimensions and
measures of HRQoL and well-being ......................................................................................... 36
4.5 Overall SLIQ score categories as distinct populations........................................................ 37
4.6 Multiple variable linear regression model of HRQoL and well-being outcomes ............... 41
5.0 Discussion ............................................................................................................................... 44
6.0 Conclusions ............................................................................................................................. 55
v
List of Tables
Table 1 Correlation between SLIQ scores and validated measures of lifestyle. Adapted
from Godwin et al. ..................................................................................................... 11
Table 2 Inclusion and exclusion criteria used to determine eligibility to participate in

research project. .......................................................................................................... 20
Table 3 List of variables and the questionnaires and components used to measure them. 26
Table 4 Description of the socio-demographic characteristics of the study population

(n=100). ...................................................................................................................... 30
Table 5 Measures of central tendency and variability for questionnaires.......................... 32
Table 6 Pearson correlation coefficient between lifestyle as measured by the SLIQ and
the outcome measures for HRQoL and well-being. ................................................... 35
Table 7 Pearson correlation coefficient for the relationship between lifestyle dimensions
of the SLIQ and the outcome measures for health-related quality of life and well-
being. .......................................................................................................................... 37
Table 8 Mean and standard deviation of the outcome measures of health-related quality of
life and well-being separated by overall SLIQ category. ........................................... 38
Table 9 Analysis of variance (ANOVA) of mean score on surveys for three categories of
SLIQ score .................................................................................................................. 39
Table 10 Post hoc tests of differences in mean variance for SLIQ categories using
Bonferroni correction.................................................................................................. 40
Table 11 Multiple variable linear regressions for the outcomes of HRQoL and well-being
using lifestyle and sociodemographic variables. ........................................................ 42
Table 12 Examples of the literature examining the association between lifestyle and
HRQoL and well-being evaluated in the literature review. ........................................ 66
vi
List of Figures
Figure 1 Venn diagram demonstrating domains unique to and shared by measures of

health-related quality of life and well-being................................................................. 7
Figure 2 Percent of the population between the ages of 20 and 65 by 5-year age
increments for study population, the St John’s metropolitan area and the province of
Newfoundland & Labrador. ........................................................................................ 31
Figure 3 Percentage of individuals achieving stated levels of education for study

population, the St John’s metropolitan area and the province of Newfoundland &
Labrador. ..................................................................................................................... 31
Figure 4 Percentage of individuals from household income levels for study population,
the St John's metropolitan area and the province of Newfoundland & Labrador....... 32
Figure 5 Histogram displaying the distribution of overall SLIQ scores. ........................... 33
Figure 6 Histogram displaying the distribution of (clockwise from top left) the SF-12
PCS, SF-12 MCS, PGWB-I and EQ5D VAS. ............................................................ 34
Figure 7 Correlation of overall SLIQ scores with outcomes (clockwise from top left) of
the SF-12 MCS, SF-12 PCS, EQ5D VAS, and PGWB-I. .......................................... 36
vii
List of Abbreviations and Symbols
ANOVA - Analysis of variance
BMI - Body mass index
CLAS – Computerized Lifestyle Assessment Scale
EQ5D – EuroQoL EQ-5D questionnaire
EQ5D VAS - EuroQoL EQ-5D visual analogue scale
HRQoL - Health-related quality of life
MCID – Minimal clinically important difference
NL – Newfoundland & Labrador, Canada
PGWB - Psychological General Well-being questionnaire
PGWB-I - Psychological General Well-being Index score
SD - Standard deviation
SF-12 - Short form 12 Questionnaire
SF-36 – Short form 36 Questionnaire
SF-12 MCS - Short form 12 – Mental component score
SF-12 PCS - Short form 12 – Physical component score
SLIQ - Simple Lifestyle Indicator Questionnaire
SPSS - Statistical Package for the Social Sciences
viii
List of Appendices
Appendix 1: Overview of studies evaluated during literature search
Appendix 2: Human Resarch Ethics Authority approval and study information sheet
Appendix 3: Survey package provided to participants
Appendix 4: Scoring template for the Simple Lifestyle Indicator Questionnaire
ix
1.0 Introduction and Overview
1.1 Context
Health is “a state of complete physical, mental and social well-being,”1 and there is
general consensus that a positive health state cannot be limited simply to the absence of
disease 1,2. A variety of factors contribute to an individual’s health and their risk of illness,
including environment, economic and social circumstances, and a person’s characteristics
and behaviours 2. In the latter category, behaviours and habits such as smoking, diet, and
physical activity3 contribute to a construct known as lifestyle, which can significantly
affect health. The increasing prevalence of chronic diseases has brought attention to the
role that lifestyle factors play in a person’s disease risk. Many lifestyle habits can
contribute to a person’s risk of developing a disease or illness, including tobacco use 4,
alcohol consumption 5, physical inactivity 6, an unhealthy diet 7, and psychological stress 8.
Health-related quality of life (HRQoL) was first described in the 1970s as a measure of
how an individual’s physical and mental health affects their day-to-day functioning 9, 10. In
contrast to many clinical measures of health and disease, HRQoL is a patient-reported
outcome with roots in the social sciences that focuses on health, well-being and the way a
person’s health affects their life from the patient’s perspective 11. For many patients
quality of life can be among the most important indicators of their own health 12. Studies
suggest that although HRQoL is subjective, and patients with the same illness may differ
in their HRQoL 13, the measure is generally sensitive to clinically-relevant changes in
health11.
1
There is evidence that lifestyle may also affect a person’s current quality of life. Patients
recovering from serious illnesses, such as cancer, or living with chronic conditions, such
as cardiovascular disease, have a higher quality of life if they have more healthful
lifestyle habits 14, 15, 16, 17. More recently there has been an interest in measuring
differences in HRQoL that may be attributed to lifestyle factors among an otherwise
healthy population, which shall be discussed in further detail.
1.2 Health-related quality of life
There is a significant body of literature on the value of HRQoL as a measured outcome in
a variety of clinical research settings9, 10, 11. HRQoL narrows the general concept of
quality of life, which may include components such as socioeconomic status and
environment, to the components specifically pertaining to health13. Including this measure
as an outcome in clinical study acknowledges two important realities: that physiologic
clinical indicators are not always the most important outcomes for patients; and that
patient-reported outcomes are of interest to clinicians, policymakers and researchers as
well11,13,18.
Measures of HRQoL are divided into categories of general instruments, which provide a
measure of HRQoL for a general population, and specific instruments, which measure
HRQoL in a disease-specific manner13. General instruments measure dimensions or
domains of quality of life, such as the ability to care for oneself, satisfaction with one’s
physical or emotional role, or feelings of anxiety or depression13. Specific instruments
incorporate additional domains relevant to the subpopulation of interest; for example,
there a number of instruments designed to measure HRQoL in people with diabetes which
2
include components of treatment satisfaction, stress related to blood glucose, and food-
related problems alongside the generic measures of mobility, well-being and social role
fulfillment 19.
There are a number of widely used, validated questionnaires that measure general
HRQoL. Some questionnaires, including the Short Form 36 (SF-36), the Nottingham
Health Profile and the Dartmouth COOP Charts, are based upon a health profile that
produces individual scores for a number of dimensions 20. The SF-36, for example,
measures HRQoL based on 36 likert scale questions which generate scores in the
domains: physical functioning, role-physical, bodily pain, general health, vitality, social
functioning, role-emotional, and mental health20, 21. The Nottingham Health Profile
measures energy level, emotional reactions, physical mobility, pain, social isolation, and
sleep based on participants’ responses of yes or no to 38 questions20. Alternatively,
HRQoL can be measured using questionnaires built upon preference-based measures that
generate single scores to represent an individual’s HRQoL, often based on scoring
algorithms that combine a number of dimensions20. Examples of preference-based
questionnaires include the EuroQoL EQ5D, which measures mobility, self-care, usual
activity, pain and discomfort and anxiety and depression, and the Quality of Well-Being
Scale, which measures of mobility, physical activity, social activity, and
symptoms/problems20.
When choosing a generic measure of HRQoL, researchers must consider carefully the
differences between validated instruments. For example, the Nottingham Health Profile is
a generic measure of HRQoL with reasonable validity but few reliability studies have
3
been carried out and it is a fairly lengthy questionnaire; in contrast, the EQ5D, with
comparable levels of validity and better reliability, is designed to be completed in
approximately five minutes to reduce participant burden20.
The SF-36 is a commonly used measure of HRQoL that underwent extensive
psychometric testing and has proven to be reliable and valid20. In addition to the
previously mentioned eight dimensional scores, the SF-36 can be scored to generate a
general physical component score (PCS) and mental component score (MCS) based upon
the four physical and mental domains of HRQoL to allow the testing of hypotheses
related to HRQoL with fewer outcome measures 22. The SF-36 questions were based on
questions in established questionnaires, and both the criterion validity and construct
validity of this instrument have been evaluated, as has the reliability and sensitivity to
change 23, 24. A comparison between a number of generic measures of HRQoL found that
the SF-36 was most efficient at distinguishing between patients with differences in their
illness severity20. A shorter version of the SF-36, known as the SF-12, was developed to
provide a generic measure of HRQoL that was comparable to the SF-36 but with fewer
questions to reduce the burden on participants. Completion of the SF-36 takes
approximately twelve minutes, while the SF-12 can be completed in under 2 minutes21.
The SF-12 provides scores based on the same eight domains as the SF-36 but reduces the
number of likert scale questions to 12. Scores on the SF-12 are comparable to, although
less precise than, the SF-36 25, but the benefits of the shorter instrument may outweigh the
drawbacks of reduced precision, particularly if participants must complete a number of
4
questionnaires13, 25. The SF-36, and later the SF-12, was adapted for use in a Canadian
population and psychometrically tested to ensure the preservation of its validity26,27.
The EQ5D was designed to be a simple, generic measure of HRQoL that was short
enough to be administered with other questionnaires20,28. As with the SF-36, the EQ5D
gathers information on a participant’s HRQoL using likert scale questions about five
dimensions of quality of life and a visual analogue scale (EQ5D VAS)20, 28. The likert
scale, which is used to generate five component scores and one summary score, can be
either a three-point (EQ5D 3L) or five-point (EQ5D 5L) scale, with the latter reducing the
observed ceiling effect of the former 29,30. The EQ5D VAS measures a participant’s self-
rated health by asking participants how they would rate their health on a scale of zero to
100, with zero being the worst health they can imagine and 100 being the best health they
can imagine28. The summary measure of HRQoL provided by the five questions on the
EQ5D and the VAS are comparable but not identical measures 31, and studies may use
either or both as outcomes20, 31. The EQ-5D was tested against the SF-12 to examine its
construct validity, and correlations between summary scores of the SF-12 and EQ-5D
correlated in the range of 0.41 to 0.55, although it was noted that the EQ-5D was less
sensitive than the SF-12 with respect to differentiating between patients of varying
severities of illness20, 25. The validity of the EQ5D has been studied in a Canadian
population 32, and population norms have been described27. The EQ5D is a popular
measure for studies that include a health economics component, as the preference-based
summary score can be more readily integrated into cost-utility analyses20, but it also has
wide use as a general measure of HRQoL31,27.
5
Shorter scales measuring HRQoL may improve response rates due to reduced participant
burden20,25. Shorter instruments also provide an opportunity to use more than one
instrument, which may provide valuable information on additional outcomes13.
Investigators have administered both EQ5D and SF-12 instruments to study participants25,
31, 27
, and investigators note that, “combining the EQ5D and SF-12 instruments provides a
broad coverage of health dimensions.”27
1.3 Well-being
Well-being is a broad concept that evaluates a person’s perception of how well their life
is going 33. Well-being and HRQoL are both holistic measures that incorporate physical
and mental components of health33, both are patient-reported outcome measures, and
instruments used to quantify them may include common dimensions such as vitality or
general health20,34. Well-being and HRQoL can be considered related and overlapping
constructs, and both serve as valid measures of a patient’s experience of their individual
health state. Commonalities between domains included in some validated questionnaires
that measure both HRQoL and well-being are demonstrated in Figure 1.
6
Figure 1 Venn diagram demonstrating domains unique to and shared by measures of
health-related quality of life and well-being.
As with HRQoL, a number of questionnaires have been developed to measure well-being.
Major population-based studies such as the US National Health Interview Survey have
used measures such as the Quality of Well-being scale33, which measures mobility, social
activity, and physical activity35, while the US National Health and Nutrition Examination
Survey uses the General Well-Being Schedule33, which includes dimensions such as
positive well-being, vitality, anxiety and self-control 36.
The Psychological General Well-Being Index (PGWB) is a validated tool used to quantify
well-being. Based upon the General Well-Being Schedule, the PGWB was been in
relatively wide use since the 1990s and is used to measure well-being in clinical
research 37. The PGWB consists of 22 likert scale questions, the answers to which are used
7
to calculate scores in the domains of anxiety, depression, positive well-being, general
health, vitality, self-control, as well as an overall index score (PGWB-I)37. Psychometric
testing, which included comparison with SF-36 scores to evaluate convergent validity and
evaluations of variation in PGWB scores with different health states to evaluate criterion
validity 38, and the PGWB has been shown to be a reliable measure of well-being that has
subsequently gained wider use in population-based studies34,37,39.
1.4 Lifestyle and the Simple Lifestyle Indicator Questionnaire
Lifestyle is a broad concept that includes behaviours such as diet, exercise, alcohol
consumption, tobacco use and psychosocial factors3-7,40. Researchers choose which
lifestyle risk factors to include in their analyses based upon the subpopulation they are
studying; for example, occupational exposure to chemical substances or environmental
exposure to pollution may be included alongside diet, tobacco use and alcohol
consumption in an analysis of the lifestyle risk factors for chronic obstructive pulmonary
disease 41, while physical activity, sedentary behaviours, and consumption of salty foods
or confectionaries may be of greater interest to researchers investigating adolescent
obesity42,43.
There are a large number of instruments and methods that can be used to quantify
lifestyle behaviours. For the category of diet alone there are food frequency
questionnaires 44, 24-hour dietary recall interviews 45, and self-reported questionnaires
such as the Mini Nutritional Assessment 46. As with measures of HRQoL, each tools has
benefits and limitations; 24-hour dietary recalls are commonly-used and validated, but
they require a skilled interviewer45 and may be prone to bias 47, while food frequency
8
questionnaires, also widely-used and validated, are long and can be intimidating to study
participants. Researchers must consider the participant burden – and potential decreased
response rate – when they used multiple questionnaires to measure multiple
lifestyle20,25,48,49.
An alternative is to use a generic questionnaire that includes multiple dimensions of
lifestyle. An Australian study generated an improvised lifestyle assessment tool based on
dimensions of diet, physical activity, alcohol consumption and BMI 50,51. Participants
were evaluated to see if lifestyle behaviours predicted mortality in healthy older men and
older men with vascular disease. Despite finding that the improvised lifestyle assessment
tool significantly predicted mortality in both healthy and unhealthy men,51 there is no
evidence that psychometric testing was pursued. The Computerized Lifestyle Assessment
Scale (CLAS) is another generic, multi-factorial tool intended to identify potential
lifestyle issues in order to discuss them further with their healthcare practitioner 52.
Completed electronically via computer, the CLAS was designed for used in family
practices and is a more clinically-oriented tool with a goal of stimulating discussion about
a patient’s risk52.
The Simple Lifestyle Indicator Questionnaire (SLIQ) was developed as a short, self-
administered questionnaire to evaluate the dimensions of diet, exercise, alcohol
consumption, tobacco use and psychosocial stress, and provide a single summary score.
Two family physicians and a nutritionist developed the SLIQ with the intention of
creating a reliable, valid summary measure of lifestyle that would allow researchers and,
eventually, clinicians to quantify lifestyle 53. The first iteration of the SLIQ consisted of
9
25 questions, including nine for the dimension of diet. Feedback from health
professionals with experience in lifestyle assessment, including family physicians,
nutritionists and nurses, was used in conjunction with factor analysis to reduce the
number of items in the SLIQ to 1253.
Initial psychometric testing on a group of family practice patients in a small city in
Ontario was undertaken to compare scores on the SLIQ to subjective lifestyle assessments
by a family physician, a nurse practitioner and a nutritionist, and to evaluate test-retest
reliability. The questionnaire was judged to have reasonable content validity, with strong
correlation (r = 0.77, p<0.001) between SLIQ scores and the blinded assessments by
health professionals and test-retest reliability that ranged from 0.63 to 0.97 for the
dimensions53. Further testing of the concurrent and convergent validity of the SLIQ was
carried out in a population of adults living in St. John’s, Newfoundland & Labrador (NL),
where scores on the SLIQ were compared with objective measures of lifestyle such as the
Diet History Questionnaire, the Social Readjustment Rating Scale, the SF-36, and
physical activity levels measured by pedometer 54. The SLIQ was found to correlate well
with these validated measures (Table 1), with the exception of the Stress scale which did
not correlate well with the Social Readjustment Rating Scale, and the authors suggest that
the SLIQ offers researchers a short, relatively simple method of assessing lifestyle in
study participants54. Further psychometric testing and the generation of population norms
are ongoing, but the SLIQ has been shown to be a reliable measurement of lifestyle
behaviours.
10
Table 1 Correlation between SLIQ scores and validated measures of lifestyle. Adapted
from Godwin et al. 54
Dimension Validated questionnaire R P
Diet Dietary Health Questionnaire 0.679 0.001

(vegetables, fruits, grains)
Physical Activity Pedometer (steps/day) 0.455 0.002
Alcohol Dietary Health Questionnaire 0.665 0.001

(alcohol)
Stress Social Readjustment Rating Scale -0.264 0.001
SLIQ lifestyle score Eight-question scale developed by 0.475 0.002

Spencer et al.
1.5 Relationship between lifestyle and health-related quality of life
The relationship between individual lifestyle risk factors and HRQoL has been described
fairly extensively in the literature. Physical activity and exercise are consistently linked to
improved HRQoL and well-being in a variety of populations 55,56,57. There are also a
number of studies supporting the link between a healthier diet and improved HRQoL 58,59.
For some lifestyle behaviours the relationship with HRQoL is less clear. Some large
studies have shown that people who smoke have a reduced quality of life 60, 61, but other
studies suggest that the relationship between smoking and lower HRQoL is rendered
nonsignificant when regression models control for potential confounders such as BMI and
depression 62. Alcohol consumption, a risk factor for some chronic diseases 63, 64, seems to
predominantly have a negative impact on HRQoL among heavy, rather than moderate,
consumers of alcohol 65. People with high levels of psychosocial stress also seem to have
reduced HRQoL 66,67. The majority of studies tend to look at one or two individual
11
lifestyle risk factors and how they affect HRQoL, but comparatively fewer examine
lifestyle comprehensively as a multi-dimensional variable that may affect HRQoL and
well-being 68.
A large cross-sectional study was carried out to evaluate lifestyle risk factors among
Chinese civil servants and the effects of those lifestyle factors on HRQoL 69. Of the
15,000 eligible participants employed in the civil service in five regions of China who
were at least 18 years of age, over 14,000 agreed to participate. The researchers measured
lifestyle factors including tobacco smoking, alcohol consumption, physical activity,
amount of sedentary time, and regular consumption of breakfast using an unvalidated
questionnaire and compared lifestyle with HRQoL as measured by the SF-36. Using
multivariate analysis, sleep duration (+3.743, p<0.01), consumption of breakfast (+2.491,
p<0.01), physical activity (+1.200, p<0.01), alcohol consumption (+0.691, p<0.01) and
smoking (-0.682, p=0.027) were significant coefficients for SF-36 PCS while
consumption of breakfast (+3.842, p<0.01), sleep duration (+3.565, p<0.01), and physical
activity (+1.271, p<0.01) were significant coefficients for SF-36 MCS. The large sample
size and the high response rate add strength to this cross-sectional study, although the use
of unvalidated instruments to measure lifestyle is a limitation. These results also may not
be generalizable to the general population, as the civil servants were overwhelmingly
married (82.06%) with at least a college education (92.12%); the degree to which the civil
service is representative of China’s population is unclear. It must also be noted that while
a number of lifestyle factors were statistically significant predictors of outcome scores in
the multivariate analysis, they may be of limited clinical significance. For example, the
12
difference in MCS scores between categories of sleep duration was just 3.565 points,
which is only a 3.5% difference.
A prospective cohort in eastern Finland evaluated the association between an unhealthy
lifestyle and reduced quality of life among a cohort of 560 adults68. The lifestyle risk
factors of nutrition, tobacco smoking, alcohol use and exercise were measured using a
series of structured questions, while HRQoL was measured using the 15D, a validated
questionnaire. The questions were based upon Finnish national guidelines for health, and
included questions of alcohol and tobacco consumption alongside items querying types of
cooking fats used; vegetable, berry and fruit intake; typical beverage consumed with
meals; and habit of adding salt to food. Lifestyle scores were calculated based on the
responses to each item (-1 for unhealthy choice, 0 for intermediate choice, +1 for healthy
choice), and the points were summed to generate an overall lifestyle score. Additional
factors such as BMI, waist circumferences and blood pressure were measured and
included as additional comparisons between lifestyle groups. The researchers separated
participants by lifestyle score into tertiles, which they categorized as healthy, neutral and
unhealthy and compared the differences in 15D scores using total scores and ANOVA.
Participants with a healthier lifestyle were significantly more likely to be female (p trend
= 0.001) and with a higher level of education (p trend <0.001), while those with an
unhealthier lifestyle were more likely to be living alone (p trend = 0.032). Certain
lifestyle factors clearly demonstrated trends with respect to the tertiles, but others failed to
show a clear trend; for example, only the healthy tertile had positive scores for all four
dietary items, which contrasts with the alcohol category where all three tertiles showed a
13
positive score. Overall there was a significant difference in HRQoL as measured by the
15D when comparing the highest and lowest lifestyle tertiles. Categorizing participant
lifestyle by tertile makes it more difficult to compare these results to other studies; the
unhealthiest tertile in this Finnish population may not be comparable to the unhealthiest
tertile in another city in Finland, let alone another country. The use of unvalidated set
questions to measure lifestyle also limits the strength of the study; without psychometric
testing it is impossible to judge the validity of the measures upon which these results rely.
A prospective cohort study examined the effect of lifestyle behaviours on healthy aging
over a 16-year period 70. Using data from an established British cohort study of over
10,000 civil servants, the researchers included participants who were over the age of 60 at
the time of follow-up with no history of serious diseases such as stroke, cancer or heart
attack, which resulted in a cohort of 5100 participants. Lifestyle behaviours were
measured through a series of questions focusing on the domains of tobacco smoking,
alcohol consumption, exercise, and diet, while healthy aging was defined as participants
with “no history of cancer, coronary artery disease, stroke or diabetes; good cognitive,
physical, respiratory and cardiovascular functioning, and the absence of disability; and
good mental health,” as assessed through clinical data, physiologic measurements and
validated questionnaires such as the SF-36. Some aspects of healthy aging are considered
measures of HRQoL, including the MCS from the SF-36 and the measures of the ability
to carry out activities of daily living. Participants were classified as healthy aging, normal
aging, or dead at the time of follow-up. Approximately one fifth of the participants fell
into the category of healthy aging, and these participants were younger and more likely to
14
be university-educated and married than their counterparts in the normal aging group.
Participants were significantly more likely to be in the healthy aging category if they
never smoked (OR 1.29), consumed alcohol in moderate quantities (OR 1.31), were
physically active (OR 1.45) and consumed daily servings of fruits and vegetables (OR
1.35) after adjusting for age, sex, level of education and marital status. The researchers
noted a positive, significant trend of healthy behaviours on the adjusted OR of good
mental health as judged by the SF-36 MCS (p trend < 0.001). Also of note was the
conclusions that the authors drew, that “although individual healthy behaviours are
moderately associated with successful aging, their combined impact is quite substantial.”
In contrast to many studies of lifestyle, the binary classification of lifestyle risk factors
precludes delving further into their potential effects on HRQoL; for example, physical
activity is divided into the categories of >2.5 hours of moderate or >1 hour vigorous
physical activity per week versus no physical activity, but it is probable that the range of
physical activity is more nuanced and detail is lost when these behaviours are recorded as
merely present or absent. As with the study of Chinese civil servants, this cohort of
British civil servants may not be representative of the wider British population, and likely
has a higher socioeconomic status, which may limit the generalizability of the results.
A cross-sectional survey examined potential clustering between healthy and unhealthy
lifestyle behaviours and how those clusters are associated with self-rated health and
quality of life 71. The data analyzed were part of the Survey of Lifestyle, Attitudes and
Nutrition 2007 conducted in Ireland. Lifestyle was measured through individual questions
about behaviours of interest, such as smoking, and questionnaires such as the
15
International Physical Activity Questionnaire, the Alcohol Use Disorders Identification
Test – Consumption, and a Food Frequency Questionnaire combined with the Dietary
Approach to Stop Hypertension guidelines. HRQoL was quantified using the WHO’s
Quality of Life Survey and mental health was assessed using two subscales of the SF-36.
A total of 7,350 study participants were included from a total sample of 10,364 eligible
adult participants; participants were excluded if they did not complete the Food
Frequency Questionnaire or if their International Physical Activity Questionnaire scores
were too extreme, the latter judgement made based on the questionnaire’s validated
scoring algorithm. Cluster analysis identified six common groupings of lifestyle
behaviours, including a healthy lifestyle cluster (physically active, never smokers,
nutritious diet), temperate cluster (moderately active, never smokers, moderate alcohol
consumption), and multiple risk factor cluster (moderate physical activity, current
smokers). Some clusters had significantly different levels of quality of life when
compared to the healthy lifestyle cluster, such as the multiple risk factor cluster and the
physically inactive cluster. The researchers included common confounders such as age,
sex and social class in their analysis, and their overall conclusion was that there are trends
in lifestyle behaviours that are related to lower HRQoL. Cluster analysis of this type
provides insights into patterns of lifestyle clustering and variability, but as with the
Finnish cohort it makes direct comparison with other studies more difficult. Strengths of
this study include the use of validated questionnaires to assess lifestyle, and a relatively
large, representative sample size.
16
These four examples are generally representative of the present consensus of the
relationship between lifestyle and HRQoL and well-being, and details of additional
studies 72, 73, 74 can be found in Appendix 1. A consistent limitation seen throughout this
area of the literature is the use of unvalidated measures of lifestyle. Validated
questionnaires are extensively tested and evaluated to ensure that they quantify values of
interest accurately75, but an additional benefit is comparability across different studies
and population groups; six studies that evaluate HRQoL using the SF-12 are easier to
compare than six studies using a variety of different measures. Additionally knowledge of
the relationship between lifestyle and HRQoL is based upon relatively few studies of
specific subpopulations, and would benefit from additional studies in more diverse
populations. Further investigation of the associations between a multi-dimensional
construct of lifestyle and HRQoL and well-being using validated questionnaires is
warranted as this time.
17
2.0 Objectives and Research Questions
The objective of this study is to examine the relationship between lifestyle behaviours and
an individual’s current health-related quality of life and psychological well-being.
Healthy and unhealthy lifestyles can have a positive or negative relationship, respectively,
with a person’s risk of developing a chronic disease, but it is less clear to what degree
lifestyle is associated with current HRQoL and well-being in a healthy population and
whether lifestyle is predictive of HRQoL. This study examines the relationship between
lifestyle and HRQoL and well-being.
Primary question
Is a healthy lifestyle associated with higher HRQoL and well-being? Specifically, is there
a significant, positive correlation between lifestyle, as measured by the SLIQ, and
HRQoL and well-being, as measured by the SF-12, the EQ5D and the PGWB, in adults
living in St. John’s, Newfoundland & Labrador?
Secondary questions
Is there a significant, positive correlation between the five dimensions of lifestyle
measured by the SLIQ and health status, health-related quality of life and well-being?
Do the people whose lifestyles are rated as healthy, intermediate, or unhealthy by the
SLIQ represent distinct populations when comparing their health-related quality of life
and well-being?
Does age, sex, household income or level of education affect the relationship between
scores on the SLIQ and scores on the EQ5D, SF-12 or PGWB questionnaires?
18
3.0 Methods
3.1 Study design
This study was designed as a cross-sectional survey to assess the relationship between
lifestyle, as measured by the SLIQ, and current health-related quality of life and well-
being. The study protocol and all survey instruments were reviewed and approved by the
Human Research Ethics Authority for Newfoundland & Labrador, protocol #13.140 (see
Appendix 2).
3.2 Pilot study
A pilot study was conducted prior to starting data collection to identify common
questions or concerns that participants may have and to determine the time commitment
for participants. Ten participants were recruited for the pilot study, and the time taken to
read all study documentation and complete the questionnaires was recorded. Pilot surveys
were examined for completeness, but were neither scored nor included in the final
sample.
Based on the pilot study, common questions that participants had were identified and
standard answers were generated. For example, some pilot study participants asked if they
could indicate between choices on the likert scale questions; participants recruited for the
study were informed that they must choose the one best answer.
3.3 Study population
The target population was adults between the ages of 18 and 65 living in St John’s, NL.
Inclusion and exclusion criteria (Table 2) were chosen to include as broad a sample as
possible while excluding participants likely to have experienced a significant change in
19
their lifestyle or quality of life in the past year, such as women who are pregnant or
people diagnosed with a serious disease.
Table 2 Inclusion and exclusion criteria used to determine eligibility to participate in

research project.
Criteria Rationale
Inclusion
Age 18-65 years Surveys validated for use in adults.
Living in St. John’s, NL Population of interest is adults living in
St. John’s, NL.
Able to understand study information sheet, Must be informed in order to participate,
questionnaires must be able to complete questionnaires.
Exclusion
Currently pregnant or pregnant within last Excluding participants who may have
year undergone significant lifestyle or HRQoL
Serious health condition or issue within past changes in the past year.
12 months (e.g. heart attack, stroke, cancer)
3.4 Sample size calculations
The study was powered to detect a correlation of at least 0.30 (low-moderate relationship
strength) 76, 77. Although 0.3 is a relatively low correlational strength, that value was
chosen to adequately power the study to detect correlations of 0.3 or higher. Using α=0.05
and β=0.20, and based on the sample size calculations for two-tailed tests 78, a sample size
of 84 was required. To account for incomplete responses and missing data, 100
participants were recruited.
3.5 Sampling strategy and recruitment procedures
Recruitment locations were chosen in an effort to enroll participants with diverse lifestyle
habits. Participants were recruited from a shopping mall, waiting rooms in family
medicine clinics, and through the Faculty of Medicine at Memorial University.
20
Potential participants were verbally given the criteria for inclusion in the study (Table 2)
and asked whether they met the criteria; those who stated that they met the criteria were
included in the study. The researcher did not ask for details on any medical conditions to
respect participants’ privacy. A consent form was not required, as the Human Research
Ethics Authority deemed completion of the surveys as implied consent; in its place, a
study information sheet was provided to all participants outlining the purpose of the study
alongside the benefits and risks of participating. Participants were provided with both a
study information sheet (Appendix 2) and a questionnaire package (Appendix 3).
Participants were encouraged to ask questions, and based on the pilot study, standard
answers to common questions were used by the researcher to ensure consistency of
information.
Policies at different locations meant that the recruitment process varied slightly, but all
participants were provided with the same information and questionnaires. Detailed
descriptions of the recruiting procedures that vary by location are described in greater
detail below.
3.5.1 Recruitment at shopping mall
Participants at the largest shopping mall in St. John’s, NL were recruited at a small table
set up in a foyer. Mall policy required that people must approach the table prior to being
greeted. Participants were informed of the purpose of the study and the inclusion criteria
and were provided with a survey package as outlined in Section 3.5. Participants had the
option to sit at the table to complete the questionnaires, or to take the study package along
with a postage-paid envelope to complete at a later time and return by mail. Participants
21
who took the study package to complete at a later time were given telephone and email
contact information in case they had questions; participants completing the survey in
person were encouraged to ask for clarification if needed.
3.5.2 Recruitment in family physician waiting rooms
Participants were recruited from the waiting rooms of family medicine clinics associated
with the Discipline of Family Medicine at Memorial University. In these clinics, the
researcher was allowed to politely approach patients and invite them to participate.
Participants were informed of the purpose of the study and the inclusion criteria and were
provided with a survey package as outlined in Section 3.5. Participants had the option to
complete the survey in the waiting room with a clipboard, or they could take the study
package along with a postage-paid envelope to complete at a later time and return by
mail. Participants who chose to take the study package to complete at a later time were
given both telephone and email contact information in case of further questions;
participants completing the survey in person were encouraged to ask for clarification if
they had any questions.
3.5.3 Recruitment at Faculty of Medicine
While obtaining permission from staff and clinicians to recruit from family medicine
clinics, these staff and clinicians were informed of the study via email. Some staff and
clinicians volunteered to participate. Participants were informed of the purpose of the
study and the inclusion criteria and were provided with a survey package as outlined in
Section 3.5. These participants returned the survey packages as scanned email
attachments or through interdepartmental mail.
22
3.5.4 Other
Participants were also recruited opportunistically through events such as a graduate
student seminar series. Participants were informed of the purpose of the study and the
inclusion criteria and were provided with a survey package as outlined in Section 3.5.
Participants had the option to complete the questionnaires right away, or to take the study
package along with a postage-paid envelope to complete at a later time and return by
mail. Participants who chose to take the study package to complete at a later time were
given both telephone and email contact information in case of further questions;
participants completing the survey in person were encouraged to ask for clarification if
they had any questions.
3.6 Instruments and variables
The study package included four surveys and a demographic information sheet (Appendix
3). The SLIQ was used to measure of lifestyle. HRQoL was measured with the SF-12
(version 2) and the EQ5D (version 5L). Well-being was measured using the PGWB. The
demographic information sheet asked participants to report their age, sex, total household
income and level of education. All responses were anonymous. A full list of variables is
shown in Table 3.
Demographic and socioeconomic information was collected using a demographic form
that has been previously used in the validation of the SLIQ. Participants indicate their age
in years; whether they are male or female; their household income (<$25,000; $25,001-
$35,000; $35,001-$50,000; $50,001-$75,000; $75,001-$100,000; $100,001-$150,000;
$150,001-$200,000; $200,001+; prefer not to say); and their level of education (did not
23
complete high school; completed high school; some college or university studies;
completed college diploma or university degree; some postgraduate or professional
training; completed postgraduate or professional training).
All questionnaires were administered in full, regardless of the outcomes being measured,
to preserve the validity of the instruments75.
3.6.1 Measuring lifestyle
For lifestyle, the independent or explanatory variables were the overall SLIQ score,
which is provided on a scale of one to ten; overall SLIQ category scores; and dimensional
scores, which measure diet, exercise, alcohol consumption, smoking status, and stress.
The SLIQ was scored according to the scoring template (Appendix 4) that has been used
in previous validation studies53, 54. When scoring the SLIQ, a raw score is calculated for
each lifestyle dimension; for example, in the alcohol category the raw score is the number
of units of alcohol consumed per week. For all dimensions except alcohol consumption, a
higher raw score indicates healthier behavior. The raw scores are converted into category
scores from zero to two based on the scoring guide, with 0 indicating a poor score in that
dimension, one indicating an intermediate score in that dimension and 2 indicating a
healthy score in that dimension. The category scores are summed to determine the overall
score on a scale of zero to ten, which is also classified into overall categories of unhealthy
(score 0-4), intermediate (score 5-7) and healthy (score 8-10).
24
3.6.2 Measuring HRQoL
For HRQoL, the PCS and MCS from the SF-12, and the VAS from the EQ5D were used.
The SF-12 and the EQ5D provide ten and six scores, respectively. The PCS and MCS
scores from the SF-12 were chosen for their broad coverage of physical and mental health
components. The EQ5D VAS provides a measure of the patient’s self-assessed health,
and has been used independently of the other five EQ5D scores in other studies20, 31. For
all three measures, the score is given as a continuous point on a scale of 0 to 100, with a
higher score indicating a higher HRQoL.
To calculate the PCS and the MCS for the SF-12, QualityMetric scoring software using
normalizing T-scores was used 79, with data adjusted for a mean of 50 (SD 10). Such
normalization is generally not desirable when comparing population means, but the goal
of this study is to evaluate the directionality and strength of the relation between lifestyle
and HRQoL, so the normalization did not affect the analysis.
3.6.3 Measuring well-being
Well-being was measured with the PGWB-I. The PGWB-I is the summary measure of the
PGWB, and was selected as a general measure of well-being that is generated based on
the scores of the six dimensions of the questionnaire (anxiety, depressed mood, positive
well-being, self-control, general health, and vitality). The PGWB was scored based on the
validated scoring scheme. The PGWB-I is a continuous score given on a scale of 0 to 100,
with a higher score indicating a higher level of well-being.
25
Table 3 List of variables and the questionnaires and components used to measure them.
Variable type Variable Measurement tool(s) Type

Age Continuous
Sex Dichotomous
Demographics
Education level Ordinal
Household income Ordinal
Independent
Simple Lifestyle Indicator
Questionnaire
Lifestyle Overall score Continuous
5 dimensions (diet, exercise, Continuous
alcohol, smoking, stress)
Overall score category Ordinal
EuroQol EQ5D
Visual analogue scale (VAS)
Health Related
Continuous
Quality of Life SF-12 v2
Physical component score (PCS)
Dependent Mental component score (MCS)
Psychological General Wellbeing Index

Well-being Continuous
Index score (PGWB-I)
3.7 Data entry and cleaning
When participants completed their surveys, the date was recorded on the study package.
For surveys taken and returned by mail, the date of receipt was noted. Surveys were
entered into a Microsoft Access database in batches of at least 10. Upon entry, the paper
copies were given an identification number and marked with the date of entry. All
surveys were stored in a locked filing cabinet.
26
3.7.1 Missing data
Eleven missing data points were identified out of 5800, a rate of 0.19%. Given the small
number of missing data points, sensitivity analysis was not conducted prior to imputation.
For the three participants who did not list an age, the median age of participants recruited
from the same location was used. A computer randomly generated either male or female
for two participants who did not indicate their sex. Missing values on the SF-12 were
generated using the overall sample median for that question. One participant indicated on
the SLIQ that they were not a current smoker, but did not indicate whether they had ever
smoked; as 75% of current non-smoking participants were never smokers, that is the
value that was imputed. One participant did not indicate their level of light exercise on the
SLIQ, so the median value for that item was used.
3.8 Data analysis
The data were transferred to IBM Statistical Package for the Social Sciences (SPSS,
version 20) for scoring and analysis.
Descriptive statistics were used to describe the sample population. To compare the
demographic characteristics of this sample population against the general population,
census data for both the city of St. John’s and the province of NL were obtained from
provincial and federal statistics agencies 80 and plotted against study demographic data.
All questionnaires were scored according to the validated scoring procedures and
established normalizing algorithms. Normality assumptions for parametric tests were
evaluated using histograms. Pearson correlation coefficients were calculated to determine
the direction, strength, and significance of the relationship between the explanatory and
27
outcome variables 81. The strength of each correlation was interpreted using two published
statistical references; although there is some variability in the literature, a value of less
than 0.3 is considered weakly or negligibly positive, values between 0.3 and 0.5 are
considered moderately positive, values between 0.5 and 0.8 are considered strongly
positive, an values above 0.8 are considered very strongly positive76, 77, with a
corresponding scale for negative correlations.
Analysis of variance (ANOVA) was used to determine whether the three categories of
SLIQ scores represent distinct populations with respect to the outcome measures, with
post hoc comparisons using Bonferroni correction.
To examine the effect of age, sex, education, income and lifestyle have on the outcomes
of interest, multiple variable linear regression was performed. The ordinal variables of
household income and education were coded as dummy variables for the regression with
the lowest categories were used as reference values; for household income the category of
< $25,000 was used and for level of education the category ‘finished high school’ was
used as there were no participants who had not finished high school. This study was not
powered for bivariate multiple variable linear regression, and all measured variables,
regardless of their significance, are shown.
28
4.0 Results
4.1 Response rate
A total of 177 people were asked to take part in the study, of whom 117(66.1%) agreed to
be part of the study. People who declined to participate were not asked the reason they
declined. One hundred of those participants met the eligibility criteria, for an overall
response rate of 56.5%. Eighty-one participants completed a survey at the time of
recruitment. Thirty-six participants took a survey to return by mail, of whom 19 (52.8%)
returned the completed survey package. Survey packages returned by mail did not include
location information, which precluded analysis by recruitment location.
4.2 Descriptive statistics
Of the 100 participants who completed the survey, 60 were female and the average age
was 37.5 years (SD 13.1 years) (Table 4). Fifty-three (53%) of the participants had
household incomes of $75,000 or greater. All participants had completed high school, and
93 (93%) had at least some post-secondary education, with 20 (20%) completing post-
graduate or professional training. When compared to the general population of the city of
St. John’s and the province of NL, this sample population was younger, with a higher
proportion of participants in the higher categories of household income and education
(Figures 2-4). Note that the census data had fewer categories for level of education, and
the study data were collapsed into the appropriate categories to facilitate comparisons.
The average score on the SLIQ was 7.29 (SD 1.5) out of ten. On the SF-12, the mean
scores for the MCS and PCS were 50.93 (SD 11.3) and 53.59 (SD 8.2), respectively, out
of 100. The EQ5D VAS had a mean score of 78.96 (SD 13.7) out of 100 and the mean
29
score on the PGWB-I was 73.15 (SD 16.5) out of 100 (Table 5). The primary explanatory
and outcome variables showed negative skew in their distributions (Figures 5 and 6).
Using the categories for overall SLIQ scores, there were three, 46, and 51 participants
with unhealthy, intermediate and healthy lifestyles, respectively.
Table 4 Description of the socio-demographic characteristics of the study population

(n=100).
Characteristic n
Age mean (SD) 37.5 (13.1)
Female 60
Income
$25,000 or less 19
$25,001-$35,000 5
$35,001-$50,000 9
$50,001-$75,000 14
$75,001-$100,000 18
$100,001-$150,000 15
$150,001-$200,000 10
$200,001+ 6
Prefer not to say 4
Education level
Completed highschool 7
Some college or university 16
Completed college or university 39
Some postgraduate or professional training 19
Completed postgraduate or professional training 19
Recruitment
Completed in person 81
Received by mail 19
Note: there were 100 participants recruited, thus the n is equal to the %, and only the
former is shown.
30
28
22
17 Study
Percent
St John's
NL
11
0
20-24 25-29 30-34 35-39 40-44 45-49 50-54 55-59 60-67
Figure 2 Percent of the population between the ages of 20 and 65 by 5-year age
increments for study population, the St John’s metropolitan area and the province of
Newfoundland & Labrador.
100
80
Percent
60
40 Study
St John's
20 NL
Figure 3 Percentage of individuals achieving stated levels of education for study

population, the St John’s metropolitan area and the province of Newfoundland &
Labrador.
31
50
40
Study
30
Percent
St John's
NL
20
10
Figure 4 Percentage of individuals from household income levels for study population,
the St John's metropolitan area and the province of Newfoundland & Labrador
Table 5 Measures of central tendency and variability for questionnaires
Observed Possible
Variable Mean (SD) Median
range range
SLIQ overall score 7.29 (1.5) 8.00 4-10 0-10
SLIQ Diet raw score 8.48 (3.5) 8.00 0-15 0-15
SLIQ Exercise raw score 11.85 (5.3) 13.00 2-24 0-24
SLIQ Alcohol raw score 3.13 (3.3) 2.00 0-12 0 - ∞*

SLIQ Smoking category
1.64 (0.6) 2.00 0-2 0-2
score
SLIQ Stress raw score 3.80 (1.3) 4.00 1-6 1-6
SF-12 PCS 53.59 (8.2) 55.94 23.50-69.70 0-100
SF 12 MCS 50.93 (11.3) 53.24 7.19-68.22 0-100
EQ5D VAS 78.96 (13.7) 80.00 20-100 0-100
PGWB-I 73.15 (16.5) 76.82 17-99 0-100

*Note: There is no upper limit for the alcohol raw score as participants report the number of
drinks consumed per week. SLIQ scoring documentation lists the top category as “14 or more.”
32
Figure 5 Histogram displaying the distribution of overall SLIQ scores.
33
Figure 6 Histogram displaying the distribution of (clockwise from top left) the SF-12
PCS, SF-12 MCS, PGWB-I and EQ5D VAS.
4.3 Correlation between overall SLIQ score and measures of HRQoL and well-being
Given the skew distribution of the outcome measures, the use of non-parametric
correlation techniques (e.g. Spearman rank correlation) was considered, as were
geometric transformations of the data to achieve a more normal distribution. In tests of
34
transformed data and when using non-parametric measures the results were similar in the
direction and strength of the relationship and the significance, so parametric tests were
used on untransformed data.
Figure 7 and table 6 show the strength and directionality of the Pearson correlation
between the overall score on the SLIQ and the outcome measures. The correlation
between the overall score on the SLIQ was statistically significant and positive for the
outcomes of SF-12 MCS (r=0.41, p<0.01), EQ5D VAS (r=0.59, p<0.01) and the PGWB-I
(r=0.47, p<0.01). There was no significant correlation between the overall SLIQ score
and the PCS (r=0.13, p=0.19).
Table 6 Pearson correlation coefficient between lifestyle as measured by the SLIQ and
the outcome measures for HRQoL and well-being.
Outcome Pearson’s R P value
SF-12 MCS + 0.41 < 0.01*
SF-12 PCS + 0.13 0.19
EQ5D VAS + 0.59 < 0.01*
PGWB-I + 0.47 < 0.01*

Note: values denoted with an asterisk (*) are statistically significant at p < 0.05
35
Figure 7 Correlation of overall SLIQ scores with outcomes (clockwise from top left) of
the SF-12 MCS, SF-12 PCS, EQ5D VAS, and PGWB-I.
4.4 Correlation between the raw and category scores of the five SLIQ dimensions and
measures of HRQoL and well-being
Pearson correlations were calculated between the SLIQ’s individual lifestyle dimensions
and the outcomes of interest (Table 7). For all SLIQ dimensions the raw score was used
except for smoking, for which there is only a category score. The SLIQ raw score for diet
correlated significantly and positively with the SF-12 MCS and EQ5D VAS outcome
measures. The exercise raw score and the smoking category score were positively
correlated with the SF-12 PCS and the EQ5D VAS. The alcohol category was not
36
significantly correlated with any of the outcome measures. The stress category correlated
significantly positively with the SF-12 MCS and PGWB-I and negatively with the SF-12
PCS.
Table 7 Pearson correlation coefficient for the relationship between lifestyle dimensions
of the SLIQ and the outcome measures for health-related quality of life and well-being.
SF-12-MCS SF-12 PCS EQ5D VAS PGWB-I
0.25* 0.043 0.45* 0.26*

Diet
(p=0.01) (p=0.67) (p<0.01) (p=0.01)
-0.02 0.32* 0.43* 0.09

Exercise
(p=0.85) (p=0.01) (p<0.01) (p=0.37)
-0.06 0.17 0.05 -0.02

Alcohol
(p=0.56) (p=0.10) (p=0.61) (p=0.85)
-0.06 0.33* 0.23* 0.05

Smoking
(p=0.53) (p=0.01) (p=0.02) (p=0.60)
0.64* -0.25* 0.15 0.62*

Stress
(p<0.01) (p=0.01) (p=0.14) (p<0.01)
4.5 Overall SLIQ score categories as distinct populations
Analysis was conducted to examine mean differences in scores on the HRQoL and well-
being questionnaires between the three categories of the overall SLIQ scores. There were
few participants with a SLIQ score in the unhealthy category (Table 8), which limited the
statistical power of the analysis of this group.
37
Mean scores for the three categories were calculated for each outcome measure (Table 8).
ANOVA was used to determine whether the healthy, intermediate and unhealthy
categories of SLIQ scores had significantly different mean scores on the HRQoL and
well-being questionnaires (Table 9). There was a significant difference between the
categories for the SF-12 PCS, the SF-12 MCS, the EQ5D VAS, and the PGWB-I. Using
Bonferroni post hoc testing, it was found that for the EQ5D VAS, the differences between
all three categories were significant, whereas the PGWB-I and the SF-12 PCS showed
significant differences between only some groups (Table 10). For the SF-12 MCS, the
Bonferroni post hoc testing, which is more conservative than some other post hoc tests,
failed to detect significant differences in group means.
Table 8 Mean and standard deviation of the outcome measures of health-related quality
of life and well-being separated by overall SLIQ category.
SLIQ N MCS PCS EQ5D PGWB

category
Mean (SD)
Unhealthy 3 41.46 (13.67) 42.64 (15.77) 49.19 (13.63) 47.59 (13.60)
Intermediate 46 48.39 (12.57) 53.01 (9.46) 51.76 (9.39) 51.30 (7.86)
Healthy 51 53.78 (9.01) 54.74 (5.84) 54.28 (7.35) 53.64 (6.57)
38
Table 9 Analysis of variance (ANOVA) of mean score on surveys for three categories of
SLIQ score
Outcome Sum of Degrees of

Mean Square F stat. P value
Squares freedom
SF-12 MCS
Between
980.64 2 490.32 4.12* 0.02
Groups
Within groups 11544.26 97 119.01
Total 12524.90 99
SF-12 PCS
Between
444.34 2 222.17 3.46* 0.04
Groups
Within groups 6229.49 97 64.22
Total 6673.83 99
EQ5D VAS
Between
5042.61 2 2521.31 18.02* <0.01
Groups
Within groups 13571.23 97 139.91
Total 18613.84 99
PGWB Index score

Between
3255.60 2 1627.80 6.64* <0.01
Groups
Within groups 23778.21 97 245.14
Total 27033.81 99
39
Table 10 Post hoc tests of differences in mean variance for SLIQ categories using
Bonferroni correction
Category A Category B Mean P value 95% confidence

Difference interval
SF-12 Unhealthy Intermediate -6.93 0.87 -22.77 8.90

MCS
Healthy -12.32 0.18 -28.11 3.47
Intermediate Unhealthy 6.93 0.87 -8.90 22.77
Healthy -5.39 0.05 -10.80 0.013
Healthy Unhealthy 12.32 0.18 -3.47 28.11
Intermediate 5.39 0.05 -0.02 10.80
SF-12 Unhealthy Intermediate -10.37 0.10 -22.00 1.26

PCS
Healthy -12.11* 0.04 -23.71 -0.52
Healthy -1.74 0.86 -5.71 2.23
Healthy Unhealthy 12.11* 0.04 0.52 23.71
Intermediate 1.74 0.86 -2.23 5.71
EQ5D Unhealthy Intermediate -24.57* <0.01 -41.74 -7.39

VAS
Healthy -34.63* <0.01 -51.75 -17.51
Intermediate Unhealthy 24.57* <0.01 7.39 41.74
Healthy -10.06* <0.01 -15.92 -4.20
Healthy Unhealthy 34.63* <0.01 17.51 51.75
Intermediate 10.06* <0.01 4.20 15.92
PGWB Unhealthy Intermediate -15.96 0.27 -38.69 6.77

Index
score Healthy -25.06* 0.03 -47.72 -2.40
Healthy -9.10* 0.02 -16.86 -1.34
Healthy Unhealthy 25.06* 0.03 2.40 47.72
Intermediate 9.10* .016 1.34 16.86
40
4.6 Multiple variable linear regression model of HRQoL and well-being outcomes
In order to determine the association between the explanatory variables of SLIQ score
and sociodemographic characteristics and the outcome variables of HRQoL and well-
being, multiple variable linear regression was performed (Table 11). Variables included
age, sex, household income, level of education, and overall SLIQ score. For every
outcome measure except the SF-12 PCS, the SLIQ score was a statistically significant
variable in the regression. For every outcome except the EQ5D VAS, age was a
significant variable. The only other sociodemographic characteristic that was significantly
predictive was level of education for the PGWB-I (completed high school versus
completed post-graduate or professional training) and SF-12 PCS (completed high school
versus completed college/university or completed post-graduate/professional). The
amount of variation in outcome measures explained by the regression variables (R2)
ranged from a low of 31% for the SF-12 PCS to a high of 44% for the EQ5D VAS (Table
11).
41
Table 11 Multiple variable linear regressions for the outcomes of HRQoL and well-being using
lifestyle and sociodemographic variables.
Outcome Variable B coefficient P value R2

SF-12 MCS Constant 23.03 * < 0.01
SLIQ score 2.93 * < 0.01 0.34
Age 0.27 * < 0.01
Sex 0.63 0.77
Education (reference: Completed high school)
Some college/university -5.69 0.25
Completed college / university -8.21 0.07
Some post-graduate / professional -1.60 0.74
Completed post-graduate / professional -9.66 0.05
Household income (reference: < $25,000)
$25,001-$35,000 5.70 0.26
$35,001-$50,000 -4.29 0.32
$50,000-$75,000 4.198 0.24
$75,01-$100,000 4.640 0.17
$100,001-$150,000 4.312 0.24
$150,001-$200,000 -3.132 0.44
$200,001+ -.882 0.85
SF-12 PCS Constant 50.33 * < 0.01
SLIQ score 0.72 0.20 0.31
Age -0.22 * < 0.01
Sex -1.01 0.53
Some college/university 4.00 0.27
Completed college / university 9.69 * < 0.01
Some post-graduate / professional 5.86 0.10
Completed post-graduate / professional 7.96 * 0.03
$25,001-$35,000 -7.07 0.06
$35,001-$50,000 5.324 0.10
$50,000-$75,000 2.267 0.39
$75,01-$100,000 -3.482 0.16
$100,001-$150,000 3.627 0.19
$150,001-$200,000 2.962 0.32
$200,001+ 4.895 0.17
42
Outcome Variable B coefficient P value R2
EQ5D VAS Constant 37.68 * < 0.01
0.44
SLIQ score 5.67 * < 0.01
Age .061 0.53
Sex -0.96 0.69
Some post-graduate / professional 0.61 0.91
Completed post-graduate / professional -5.70 0.30
$25,001-$35,000 5.02 0.37
$35,001-$50,000 1.84 0.70
$50,000-$75,000 5.05 0.21
$75,01-$100,000 -2.52 0.50
$100,001-$150,000 5.19 0.21
$150,001-$200,000 -4.59 0.31
$200,001+ 7.53 0.16
PGWB Constant 30.5 * 0.01
Index 0.35
SLIQ score 5.25 * < 0.01
Age 0.29 * 0.02
Sex 0.83 0.79
Some post-graduate / professional -4.61 0.51
Completed post-graduate / professional -16.91 * 0.02
$25,001-$35,000 4.17 0.57
$35,001-$50,000 -1.44 0.82
$50,000-$75,000 4.64 0.37
$75,01-$100,000 3.48 0.48
$100,001-$150,000 7.83 0.15
$150,001-$200,000 -3.43 0.56
$200,001+ -1.34 0.85
43
5.0 Discussion
The results indicate that there is a significant relationship between lifestyle and some
measures of HRQoL and well-being, and that lifestyle as measured by the SLIQ may be
associated with HRQoL and well-being. The positive correlation between lifestyle and
HRQoL, as measured by the EQ5D VAS, is considered moderate, with a value above 0.5.
The correlation between lifestyle and the SF-12 MCS and the PGWB-I were also positive,
moderate in strength and statistically significant. A significant correlation between
lifestyle and the SF-12 PCS was not detected. With respect to the dimensions of the
SLIQ, there were significant, positive associations between diet and the SF-12 MCS, the
EQ5D VAS and the PGWB-I; exercise and the SF-12 PCS and EQ5D VAS; smoking
status and the SF-12 PCS and the EQ5D VAS; and stress levels and the PGWB-I. There
was a significant, negative association between stress and scores on the SF-12 PCS. There
were too few participants with an unhealthy lifestyle to explore that category adequately,
but significant differences in the mean scores of the intermediate and healthy groups were
observed for EQ5D VAS and the PGWB-I. In multivariable linear regression models age
was a significant factor for the SF-12 MCS, SF-12 PCS and PGWB-I, and level of
education was a significant factor for PGWB-I and SF-12 PCS; all other
sociodemographic variables were not statistically significant. These results are generally
in line with the literature; despite variability in the definition of lifestyle and measures of
HRQOL the correlation between the two has been shown significant in diverse
populations68, 70, 74.
44
Lifestyle was particularly well correlated with HRQoL, specifically self-perceived health,
as illustrated by the EQ5D VAS. There is a moderate, significant correlation between the
overall lifestyle score, the dimensions of diet and exercise, and the EQ5D VAS, and
participants in the intermediate and healthy SLIQ categories had significantly different
mean VAS scores. This suggests that, in this sample population, lifestyle is predictive of a
person’s self-perceived health. There is general agreement with these findings in the
literature, including a study using Statistics Canada’s National Population Health Survey
data, with the finding that lifestyle habits are significant determinants of self-perceived
health 82. The National Population Health Survey is a representative survey of Canadians
living across the country, although people living in remote communities and on First
Nations reserves are not included. That study indicates that physical activity and smoking
status are among the strongest predictors of self-perceived health.
The relationship between lifestyle and the physical and mental components of the SF-12
are less clear. The SF-12 physical component was not significantly correlated with the
overall SLIQ score, although significant correlations were detected for the dimensions of
exercise (positive, weak/moderate relationship strength, statistically significant), smoking
(positive, weak/moderate relationship strength, statistically significant), and stress
(negative, weak/moderate relationship strength, statistically significant). The mental
component of the SF-12 was significantly correlated with the overall SLIQ score, as well
as the scores for diet (positive, weak/moderate relationship strength, statistically
significant) and stress (positive, moderate/strong relationship, statistically significant).
This indicates that individual lifestyle dimensions correlate differently with physical and
45
mental HRQoL, and the degree and strength of those relationships affects how an overall
lifestyle score correlates with the measures of HRQoL.
The PGWB-I correlates quite well with the stress dimension of the SLIQ, with a strong,
significant correlation observed in this sample. This suggests that a 6-point self-reported
scale of stress strongly correlates with a score generated through a 22-item questionnaire;
the latter is a simpler measure, and may be helpful in situations where researchers or
clinicians wish to generate a general measure of well-being relatively quickly. There was
a significant difference of just over 10 points in PGWB-I scores between individuals
categorized as intermediate and healthy by the SLIQ, indicating a significant difference in
well-being for these groups. Other studies have suggested an independent link between
psychosocial stressors and poor lifestyle, which is then further associated with reduced
quality of life73, and additional investigation of these three related factors might be useful.
The only SLIQ dimension that was not significantly associated with of any outcome
measures was alcohol consumption. Correlation coefficients were weak and not
statistically significant. Participant self-reports ranged from 0-12 drinks per week, with an
average of about two drinks per week and a strong positive skew in distribution. Concerns
about the validity of self-reported alcohol consumption have been raised in the
literature 83,84, and inaccurate reporting, which may have precluded the detection of a
relationship between alcohol consumption and the outcomes of interest, must be
considered. Mild to moderate alcohol consumption is relatively benign, but high
consumption of alcohol is associated with increased risk of developing some chronic
diseases5. With respect to HRQoL, one study found that among a cohort of men followed
46
from middle- to old-age, HRQoL was negatively affected in those who reported
consuming three or more drinks per day 85. Similar results were seen in a cross-sectional
study, which found a significant decrease in HRQoL among drinkers with DSM-IV-
classified alcohol dependence 86. Participants in this study reported consuming an average
of two drinks per week, so it is not surprising that no significant decrease in quality of life
related to alcohol consumption was found.
There was no significant correlation between overall SLIQ scores and the SF-12 PCS.
This is somewhat at odds with the literature, which has generally shown significant
positive relationships between physical activity, physical fitness and quality of life 87, 88.
The exercise category score on the SLIQ did correlated significantly with the SF-12 PCS
as well as the EQ5D VAS, which is more in line with the published literature in
suggesting an association between physical activity and dimensions of HRQoL56,89. One
possible explanation for this disparity is in the way the PCS is calculated. On the SF-12,
physical health is assessed through questions asking about how a participant’s health
affects their ability to carry out “moderate activities such as moving a table, pushing a
vacuum cleaner, bowling, or playing golf”, their ability to “climb several flights of
stairs”, whether they accomplished less or were limited in the kinds of activities they
could do, whether they experienced pain, and whether their physical health affected their
social activities. The sample population was generally healthy, community-dwelling
adults, and the strong central tendency of the PCS scores suggests that the SF-12 may not
have adequate sensitivity in this population. In contrast, the SLIQ measures mild,
moderate and intense physical activity, and scores in the study population showed a much
47
wider range of values. A more detailed evaluation of physical activity or physical fitness
than is used in the SF-12 and EQ5D may be needed to differentiate between the physical
function of healthy, community-dwelling adults.
Although a statistically significant association between lifestyle and HRQoL and well-
being was observed, the relevance to physicians, patients, and the general population is
less clear. The minimal clinically important difference (MCID) for measures of HRQoL
and well-being is around 10% 90, although there is some variance depending on the health
of the population and whether an improvement or decline is observed 91. It has been
suggested that defining an MCID is difficult because patients may not always understand
the context of their improvement or decline, and MCID will vary depending upon the
general health of the population being observed 92. For the EQ5D VAS studies have
suggested that the MCID, defined as the mean difference in scores associated with a
transition between the instrument’s health classification system, is 10 points 93, 94. For the
SF-12, a difference of 5 points on either the MCS or PCS is considered minimally
important93. There seems to be some consensus that a difference in 10 percent in an
instrument’s scoring range reaches the threshold of MCID, although this will vary by
population and health status93,95. The coefficients in the linear regression models indicate
that lifestyle may affect some measures of HRQoL and well-being in a clinically
important way. In the regression of both the PGWB-I and the EQ5D VAS, a difference in
SLIQ scores of two points would yield a difference of just over ten percent, while the
difference in SLIQ scores required to see a MCID on the SF-12 MCS is four points. The
weak, nonsignificant association between the overall SLIQ score and the SF-12 PCS does
48
not permit the determination of a threshold for MCID. These results indicate that a
change in lifestyle that resulted in two to four point difference in SLIQ scores would
result in a clinically meaningful change in HRQoL and well-being in this population. A
combination of lifestyle changes would yield that magnitude of change in SLIQ,
including quitting smoking (+ 1 point), reducing alcohol consumption from fourteen
drinks per week to fewer than seven (+ 2 points), adding vigorous exercise to one’s
lifestyle habits (+ 1 or 2 points, depending on previous exercise), and consuming two or
more servings of leafy greens, fruit and high-fibre carbohydrates to the diet (+ 1 or 2
points). Physicians counselling patients on making lifestyle changes and people
considering changing their lifestyle should be aware that it may take a number of
moderate changes or a one or two large changes to their lifestyle before they notice a
difference in their quality of life.
Sociodemographic characteristics varied in the significance of their association with
HRQoL and well-being. The results of this study suggest that some sociodemographic
characteristics, specifically age and education level, are significantly associated with
HRQoL and well-being. Age was a significant predictor of SF-12 MCS, PCS, and
PGWB-I, although the directionality of the coefficients varied; older participants had
lower values for PCS scores but higher values for MCS and PGWB-I; this is similar to a
large Canadian study of people with chronic diseases, which found that advancing age
was associated with lower PCS scores but higher MCS scores 96. Some studies have
suggested that overall HRQoL declines with age in certain populations 97, particularly for
physical domains in those with underlying health issues 98, 99. Higher levels of income and
49
education, which have been independently associated with increased quality of
life68,100, 101, may ameliorate the effects of age in certain studies as well; at least one study
reported lower HRQoL among older participants, but noted that participants in the oldest
age category who were more educated and with higher household income actually had a
higher HRQoL than younger participants in the lowest categories of education and
household income 102. Household income was not significantly associated with any of the
outcomes, and education was only associated with the PCS and PGWB-I in a few
categories. As a secondary outcome, this study was not statistically powered to evaluate
relationships between sociodemographic factors and the outcomes of interest, and a large
sample of this population may have yielded results more in line with the literature.
One significant limitation of this study relates to the study population and the ways in
which it differs from the general population in the city of St John’s and the province of
NL. This study population was younger and with a higher proportion of people who are
highly educated and with high household incomes than is representative of either the St
John’s metropolitan area or the province as a whole. Recruitment locations were selected
strategically to recruit people of different backgrounds in areas of varying socioeconomic
status. Despite these efforts, the study population differed from the general population
both in the city and the province, which reduces the generalizability of these results. To
some degree this may be expected, as exclusion criteria restricted the sample population
to people who had been free of serious disease or illness in the past year; the sample was
compared to the general population of the city of St. John’s and the province of NL, not
to healthy adults within those areas. Loosening those exclusion criteria still may not have
50
yielded a representative sample; researchers may distribute surveys to a random sample of
their target population, but the people who agree to participate and complete the survey
may not be representative of the population as a whole 103. Recruiting more diverse
participants, particularly having greater representation of individuals with lower levels of
education or household income, may have added diversity to the observed lifestyle scores,
as studies in the literature suggest that these characteristics are associated68, 100, 101. The
exclusion criteria, as well, may have prevented some people with unhealthy lifestyles
from participating, but the population of interest for this study was otherwise healthy
adults so excluding people who had experienced a major health crisis in the last year was
reasonable.
The strong central tendencies and skew of the distribution of lifestyle, HRQoL and well-
being variables may violate normality assumptions required for correlational and linear
regression analysis. This possibility was investigated by comparing parametric and non-
parametric tests, which produced correlation and regression coefficients of the same
direction, magnitude, and significance as the parametric tests. Statistical references
suggest using Pearson correlation and other parametric tests when the data are generally
normally distributed, stating that the effect on results is minimal as long as the data are
independent 104,105.
A number of variables that could have affected the results of this study were not
measured. A person’s lifestyle, HRQoL and well-being may be affected by medical
conditions such as arthritis or depression, medications they are currently taking, and
physiologic characteristics such as BMI. Social and personal events, such as a family
51
member’s illness, a divorce, or lost job may also affect lifestyle and HRQoL. Participants
who had experienced a serious health issue or pregnancy within the last year were
excluded, but attempting to screen for and measure or exclude all possible confounders
would have been impractical. The sociodemographic factors included in the study are
similar to those in the literature, but future studies in this area should consider including a
wider range of factors that may be confounders or effects modifiers in their analysis.
The necessary variability in recruitment techniques and study procedures may have
introduced some bias into the results. For example, any potential participant could be
directly approached in family medicine clinic waiting rooms, while participants at the
shopping mall had to approach a table and inquire about the study before they could be
recruited. Also, some participants completed their questionnaire packages at the time of
recruitment while others took the questionnaire packages to complete at a later time and
return by mail. This variability was largely unavoidable, as study procedures had to
comply with the regulations in the venues at which participants were recruited; the
shopping mall provided access to a diverse population of potential participants but had
firm rules that participants must approach the table. As well, the Human Research Ethics
Authority required that participants have the option to complete the questionnaires
privately and at their own pace, so the choice to take the surveys and return them by mail
was provided. The study procedures attempted to reduce this variability by developing
standard language for recruiting participants, and standard answers to common questions
were identified during the pilot study phase. All participants were recruited by the same
52
person (the candidate), which ensured consistency of information given and helped
reduce further variability.
The validated questionnaires used in this study used different time periods as frames of
reference for questions. The SLIQ, for example, asks participants to consider their
lifestyle habits over the last year, the EQ5D asks about a person’s HRQoL on the day it is
filled out, the SF-12 asks about quality of life in the past four weeks, and the PGWB does
not give a time period, instead asking how things have been going. It would be preferable
for all instruments to evaluate the same period of time, but questionnaires are validated
with specific wording and cannot be modified75. By querying lifestyle over the last 12
months, the SLIQ avoids some of the seasonal sensitivity and general variability that may
affect shorter time periods; people may be less physically active during the winter
months, for example, or may experience short periods of time during which their lifestyle
habits depart from normal, such as a vacation or holiday season. In contrast, the measures
of HRQoL and well-being seek the evaluate individuals on a much shorter timescale.
Although HRQoL and well-being are, no doubt, subject to variation throughout the year
and based on life events, the surveys are designed to give a snapshot of how an
individual’s quality of life is around the time they are surveyed.
As a cross-sectional survey, this study cannot provide evidence for causation or order of
events, but it can provide evidence for associations. This study cannot determine whether
people who have a good HRQoL and well-being are better able to lead a healthy, active
lifestyle, or if a healthy, active lifestyle improves a person’s quality of life; likely it is
some combination of both. This study is also unable to determine whether changes in
53
lifestyle consistently result in changes to HRQoL and well-being, as this sample
population was measured at just a single time point.
Studies in the literature use a number of measures to quantify lifestyle, and this variability
precludes making comparisons between studies. Just as quality of life research benefitted
greatly from the wider use of validated tools, so would the study of lifestyle risk factors
benefit from more consistent definitions and measurement tools. Validated instruments
such as the SLIQ provide consistent, comparable measurements of lifestyle, and
researchers should consider the use of such tools in place of the improvised assessment
measures that are often used at present.
Further understanding of the relationship between lifestyle and HRQoL and well-being
would benefit greatly from additional studies in more generalizable populations using
validated questionnaires.
54
6.0 Conclusions
In a sample of community-dwelling adults in St John’s, NL, there were significant
associations between lifestyle, as measured by the SLIQ, and some measures of HRQoL
and well-being. This study also found significant associations between certain dimensions
of lifestyle, especially diet, exercise and stress as measured by the SLIQ, and some
measures of quality of life. The three categories of overall SLIQ scores represent
populations with significantly different mean scores for the measures of HRQoL and
well-being. The SLIQ seems to have at least moderately associated with HRQoL and
well-being, which is a conclusion in line with previously-published literature on the
subject.
This study is strengthened by the use of validated questionnaires, which add reliability to
the results. The generalizability of these results may be limited by a sample population
that was younger, more educated and with higher household income than the general
population of adults in St. John’s, NL, and the province of NL as a whole. A paucity of
participants with a lifestyle categorized as unhealthy meant that some analyses were
underpowered.
These results indicate that there is a significant association between lifestyle and the
outcomes of HRQoL and well-being. Further study in this area is needed to assess
whether similar associations are present in more diverse populations, and to promote the
understanding of lifestyle as a source of disparity in both quality of life and well-being in
healthy, community-dwelling populations.
55
7.0 References
1
Huber M, Knottnerus JA, Green L, van der Horst H, Jadad AR, Kromhout D, Leonard
B, Lorig K, Loureiro MI, van der Meer JWM, Schnabel P, Smith R, van Weel C, Smid H.
How should we define health? BMJ. 2011; 26: 343.
2
World Health Organization. Health Impact Assessment: The determinants of health
[internet]. Geneva:World Health Organization. [cited 2015 Feb 20]. Available from:
http://www.who.int/hia/evidence/doh/en/
3
Mozaffarian D, Wilson PWF, Kannel WB. Beyond established and novel risk factors:
lifestyle risk factors for cardiovascular disease. Circulation. 2008l 117: 3031-3038.
4
Ezzati M, Henley SJ, Thun MJ, Lopez AD. Role of smoking in global and regional
cardiovascular mortality. Circulation. 2005; 112: 489-497
5
Corrao G, Bagnardi V, Zambon A, La Vecchia C. A meta-analysis of alcohol
consumption and the risk of 15 diseases. Preventative medicine. 2004; 38:613-619
6
Bassuk SS, Manson JE. Epidemiological evidence for the role of physical activity in
reducing risk of type 2 diabetes and cardiovascular disease. Journal of Applied
Psychology 2005; 99(3): 1193-1204.
7
Brunner E, Rees K, Ward K, Burke M, Thorogood M. Dietary advice for reducing
cardiovascular risk (Review). The Cochrane Library. 2009; 1.
8
Ohlin B, Nilsson PM, Nilsson JA, Berglund G. Chronic psychosocial stress predicts
long-term cardiovascular morbidity and mortality in middle-aged men. European Heart
Journal. 2004; 25(10): 867-873.
9
National Center for Chronic Disease Prevention and Health Promotion, Division of
Population Health. Health-Related Quality of Life [internet]. Atlanta (US): Centres for
Disease Control; 2011 Mar 15 [updated 2011 Mar 17, cited 2014 Aug 7]. Available from:
http://www.cdc.gov/hrqol/concept.htm
10
Armstrong D, Caldwell D. Origins of the concept of quality of life in health care: a
rhetorical solution to a political problem. Social Theory and Health. 2004; 2: 361-371.
11
Wilson IB, Cleary PD. Linking clinical variables with health-related quality of life.
JAMA. 1995; 273(1): 59-65
12
Covinsky KE, Wu AW, Landefeld CS, Connors AF, Phillips RS, Tsevat J, Dawson
NV, Lynn J, Fortinsky RH. Health status versus quality of life in older patients: does the
distinction matter? The American Journal of Medicine. 1999; 106: 435-440
13
Guyatt GH, Feeny DH, Patrick DL. Measuring health-related quality of life. Annals of
Internal Medicine. 1993; 118: 622-629.
56
14
Hatton DC, Haynes RB, Oparil S, Kris-Etherton P, Pi-Sunyer FX, Resnick LM, Stern
JS, Clark S, McMahon M, Morris C, Metz J, Ward A, Holcomb S, McCarron DA.
Improved quality of life in patients with generalized cardiovascular metabolic disease on
a prepared diet. Am J Clin Nutr. 1996; 64(6): 935-943.
15
Ravasco P, Monteiro-Grillo I, Camilo ME. Does nutrition influence quality of life in
cancer patients undergoing radiotherapy? Radiother Oncol. 2003; 67(2): 213-220.
16
Tung H-H, Tseng L-H, Wei J, Lin C-H, Wang T-J, Liang S-Y. Food pattern and quality
of life in metabolic syndrome patients who underwent coronary artery bypass grafting in
Taiwan. European Journal of Cardiovascular Nursing. 2011; 10: 205-215.
17
Alfonso-Rosa RM, Del Pozo-Cruz B, Del Pozo-Cruz J, Del Pozo-Cruz JT, Sanudo B.
The relationship between nutritional status, functional capacity, and health-related quality
of life in older adults with Type 2 diabetes: a pilot experiment. The Journal of Nutrition,
Health and Aging. 2013; 17(4): 315-321.
18
Padua L, Evoli A, Aprile I, Caliandro P, Mazza S, Padua R, Tonali P. Health-related
quality of life in patients with myasthenia gravis and the relationship between patient-
oriented assessments and conventional measurements. Neurol Sci. 2001; 22: 363-369.
19
El Achab Y, Nejjari C, Chikri M, Lyoussi B. Disease-specific health-related quality of
life instruments among adults diabetic: A systematic review. Diabetes Res Clin Pract.
2008; 80(2): Epub ahead of print.
20
Coons SJ, Rao S, Keininger DL, Hays RD. A comparative review of generic quality-of-
life instruments. Pharmacoeconomics. 2000; 17(1): 13-35.
21
Côté I, Grégoire JP, Moisan J, Chabot I. Quality of life in hypertension; the SF-12
compared to the SF-36. Canadian Journal of Clinical Pharmacology. 2004; 11(2): e232-
e238.
22
Ware JE, Kosinski M, Bayliss MS, McHorney CA, Rogers WH, Raczek A. Methods
for the scoring and statiscal analysis of SF036 health profile and summary measures:
summary of results from the Medical Outcomes Study. Med Care 1995; 33(4Suppl):
AS264-AS279.
23
Brazier JE, Harper R, Jones NMB, O’Cathain A, Thomas KJ, Usherwood T, Westlake
L. Validating the SF-36 health survey questionnaire: new outcome measure for primary
care. BMJ. 1992; 305(6846): 160-164.
24
Jenkinson C, Wright L, Coulter A. Criterion validity and reliability of the SF-36 in a
population sample. Quality of Life Research. 1994; 3(1): 7-12.
25
Johnson JA, Coons SJ. Comparison of the EQ-5D and SF-12 in an adult US sample.
Quality of Life Research. 1998; 7: 155-166.
26
Dauphinee SW, Gauthier L, Gandek B, Magnan L, Pierre U. Readying a US measure of
health status, the SF-35, for use in Canada. Clin Invest Med. 1997; 20(4): 224-238.
57
27
Johnson JA, Pickard AS. Comparison of the EQ-5D and SF-12 health surveys in a
general population survey in Alberta, Canada. Meidcal Care. 2000; 38(1): 115-121.
28
The EuroQol Group. EuroQol – a new facility for the measurement of health-related
quality of life. Health Policy. 1990; 16: 199-208.
29
Janssen MF, Pickard AS, Golicki D, Gudex C, Niewada M, Scalone L, Swinburn P,
Busschbach J. Measurement properties of the EQ-5D-5L compared to the EQ-5D-3L
across eight patient groups: a multi-country study. Quality of Life Research. 2013; 22(7):
1717-1727.
30
Herman M, Gudex C, Lloyd A, Janssen M, Kind P, Parkin D, Bonsel G, Badia X.
Development and preliminary testing of the new five-level version of EQ-5D (EQ-5D-
5L). Qual Life Res. 2011; 20(10): 1727-1736.
31
Sach TH, Barton GR, Doherty M, Muir KR, Jenkinson C, Avery AJ. The relationship
between body mass index and health-related quality of life: comparing the EQ-5D,
EuroQol VAS and SF-6D. International Journal of Obesity. 2007; 31: 189-196.
32
Bansback N, Tsuchiya A, Brazier J, Anis A. Canadian valuation of EQ-5D health
states: preliminary value set and considerations for future valuation studies. PLoS One.
2012; 7(2): e31115.
33
National Center for Chronic Disease Prevention and Health Promotion, Division of
Population Health. Well-being Concepts [internet]. Atlanta (US): Centres for Disease
Control; 2013 Mar 6 [updated 2013 Mar 6, cited 2014 Aug 7]. Available from:
http://www.cdc.gov/hrqol/wellbeing.htm
34
Grossi E, Groth N, Mosconi P, Cerutti R, Pace F, Compare A, Apolone G.
Development and validation of the short version of the Psychological General Well-Being
Index (PGWB-S). Health and Quality of Life Outcomes. 2006; 4:88-86
35
Kaplan RM, Ganiats TG, Sieber WJ, Anderson JP. The quality of well-being scale:
critical similarities and differences with SF-36. International Journal for Quality in Health
Care. 1998; 10(6): 509-520.
36
Fish J. General well-being schedule. Encyclopedia of Clinical Neuropsychology. 2011.
New York: Springer. P 1139-1140.
37
Lundgren-Nilsson A, Jonsdottir IH, Ahlborg G, Tennant A. Construct validity of the
psychological general well being index (PGWBI) in a sample of patients undergoing
treatment for stress-related exhaustion: a rasch analysis. Health and Quality of Life
Outcomes. 2013; 11:2
38
Rasmussen NA, Norholm V, Bech P. The internal and external validity of the
Psychological General Well-being Schedule (PGWB). Quality of Life Newsletter. 1999;
22: 7.
58
39
Carotenuto A, Fasanaro AM, Molino I, Sibilio F, Saturnino A, Traini E, Amenta F. The
psychological general well-being index (PGWBI) for assessing stress of seafarers on
board merchang shipts. Int Marit Health. 2013; 64(4): 215-220.
40
Sagner M, Katz D, Egger G, Lianov L, Schulz KH, Braman M, Behbod B, Phillips E,
Dysinger W, Ornish D. Lifestyle medicine potential for reversing a world of chronic
disease epidemics: from cell to community. International Journal of Clinical Practice.
2014; 68: 1289-1292
41
Kant S, Gupta B. Role of lifestyle in the development of chronic obstructive pulmonary
disease: a review. Lung India : Official Organ of Indian Chest Society 2008;25(2):95-101.
42
Plotnikoff RC, Karunamuni N, Spence JC, Storey K, Forbes L, Raine K, Wild TC,
McCargar L. Chronic disease-related lifestyle risk factors in a sample of Canadian
adolescents.
43
Gopinath B, Louie JCY, Flood VM, Burlutsky G, Hardy LL, Baur LA, Mitchell P.
Influence of obesogenic behaviors on health-related quality of life in adolescents. Asia
Pac J Clin Nutr. 2014; 23(1): 121-127.
44
Willett WC, Sampson L, Stamper MJ, Rosner B, Bain C, Witschi J, Hennekens CH,
Speizer FE. Reproducibility and validity of a semiquantitative food frequency
questionnaire. Am J Epidemiol. 1985; 122(1): 51-65.
45
Karvetti RL, Knuts LR. Validity of the 24-hour dietary recall. J Am Diet Assoc. 1985;
85(11): 1437-1442.
46
Cereda E. Mini Nutritional Assessment. Current Opinion in Clinical Nutrition &
Metabolic Care. 2012; 15(1): 29-41.
47
Kristal AR, Peters U, Potter JD. Is it time to abandon the food frequency questionnaire?
Cancer Epidemiol Biomarkers Prev. 2005; 14: 2826-2828.
48
Galea S, Tracy M. Participation rates in epidemiologic studies. Ann Epidemiol. 2007;
17: 643-653.
49
Rolstad S, Adler J, Ryden A. Response burden and questionnaire length: is shorter
better? A review and meta-analysis. Value in Health. 2011; 14(8): 1101-1108.
50
Spencer CA, Jamrozik K, Norman PE, Lawrence-Brown M. A simple lifestyle score
predicts survival in healthy elderly men. Preventive Medicine. 2005; 40(6): 712-717.
51
Spencer CA, Jamrozik K, Lawrence Brown M, Norman PE. Lifestyle still predict
mortality in older men with established vascular disease. Preventive Medicine. 2005;
41(2): 583-488.
52
Ahmad F, Hogg-Johnson S, Skinner HA. Assessing patient attitudes to computerized
screening in primary care: psychometric properties of the Computerized Lifestyle
Assessment Scale. J Med Internet Res. 2008; 10(2): e11.
59
53
Godwin M, Streight S, Dyachuk E, van den Hooven EC, Ploemacher J, Seguin R,
Cuthbertson S. Testing the Simple Lifestyle Indicator Questionnaire: initial psychometric
study. Canadian Family Physician. 2008; 54(1): 76-77
54
Godwin M, Pike A, Bethune C, Kirby A, Pike A. Concurrent and convergent validity of
the Simple Lifestyle Indicator Questionnaire. ISRN Family Medicine. 2013; article ID
529645: 6 pages.
55
Penedo FJ, Dahn JR. Exercise and well-being: a review of mental and physical health
benefits associated with physical activity. Curr Opin Psychiatry. 2005; 18: 189-193.
56
Olivares PR, Gusi N, Prieto J, Hernandez-Mocholi MA. Fitness and health-related
quality of life dimensions in community-dwelling middle aged and older adults. Health
and Quality of Life Outcomes. 2011; 9: 117-125.
57
Bize R, Johnson JA, Plotnikoff RC. Physical activity level and health-related quality of
life in the general adult population: a systematic review. Prev Med. 2007; 45(6): 401-415.
58
Wanden-Berghe C, Sanz-Valero J, Escriba-Aguir V, Castello-Botia I, Guardiola-
Wanden-Berghe R, Red de Malnutricion en Iberoamerica – Ciencia y Tecnologia para el
Desarrollo (Red MeI – CYTED). Evaluation of quality of life related to nutritional status
(systematic review). British Journal of Nutrition. 2009; 101: 950-960.
59
Bonaccio M, Di Castelnuovo A, Bonanni A, Costanzo S, De Lucia F, Pounis G, Zito F,
Donati MB, de Gaetano G, Iacoviello L. Adherence to a Mediterranean diet is associated
with a better health-related quality of life: a possible role of high dietary antioxidant
content. BMJ Open. 2013; 3(8): e003003.
60
Coste J, Quinquis L, D’Almeida S, Audureau E. Smoking and health-related quality of
life in the general population. Independent relationships and large differences according
to patterns and quantity of smoking and to gender. PLOS One. 2014; 9(3): e91562.
61
Wilson D, Parsons J, Wakefield M. The health-related quality of life of never smokers,
ex-smokers, and light, moderate, and heavy smokers. Preventive Medicine. 1999; 29(3):
139-144.
62
Vogl M, Wenig CM, Leidl R< Pokhrel S. Smoking and health-related quality of life in
English general population: implications for economic evaluations. BMC Public Health.
2012; 12: 203
63
Ronksley PE, Brien SE, Turner BJ, Mukamal KJ, Ghali WA. Association of alcohol
consumption with selected cardiovascular disease outcomes: a systematic review and
meta-analysis. British Medical Journal. 2011; 342: d671.
64
Patra J, taylor B, Irving H, Roerecke M, Baliunas D, Mohaptra S, Rehm J. Alcohol
consumption and the risk of morbidity and mortality for different stroke types – a
systematic review and meta analysis. BMC Public Health. 2010; 10:258.
60
65
Daeppen J-B, Faouzi M, Sanchez N, Rahhali N, Bineau S, Bertholet N. Quality of life
depends on the drinking pattern in alcohol-dependent patients. Alcohol and Alcoholism.
2014; 49(4): 457-465.
66
De Frias CM, Whyne E. Stress on health-related quality of life in older adults: the
protective nature of mindfulness. Aging & Mental Health. 2014; [epub ahead of print].
67
Reibel DK, Greeson JM, Brainard GC, Rosenzweig S. Mindfulness-based stress
reduction and health-related quality of life in a heterogeneous patient population. General
Hospital Psychiatry. 2001; 23(4): 183-192.
68
Savolainen J, Kautianen H, Miettola J, Niskanen L, Mantyselka P. Low quality of life
and depressive symptoms are connected with an unhealthy lifestyle. Scandinavian journal
of public health. 201; 42: 163-170.
69
Xu J, Qiu J, Chen J, Zou L,Feng L, Lu Y, Wei Q, Zhang J. Lifestyle and health-realted
quality of life: a cross-sectional study among civil servants in China. BMC Public Health.
2012; 12: 330-339.
70
Sabia S, Singh-Manoux A, Hagger-Johnson G, Cambois E, Brunner EJ, Kivimaki M.
Influence of individual and combined healthy behaviours on successful aging. CMAJ.
2012; 14(18): 1985-1991.
71
Conry MC, Morgan K, Curry P, McGee H, Harrington J, Ward M, Shelley E. The
clustering of health behaviours in Ireland and their relationship with mental health, self-
rated health and quality of life. BMC Public Health. 2011; 11: 692-702.
72
Myint PK, Smith RD, Luben RN, Surtees PG, Wainwright NWJ, Wareham NJ, Khaw
K-T. Lifestyle behaviours and quality-adjusted life years in middle and older age. Age
and Aging. 2011; 40: 589-595.
73
Seib C, Whiteside E, Lee K, Humphreys J, Tran THD, Chopin L, Anderson D. Stress,
lifestyle and quality of life in midlife and older Australian women: results from the stress
and the health of women study. Women’s Health Issures. 2014; 24(1): e43-d52.
74
Pisinger C, Ladelund S, Glumer C, Toft U, Aadahl M, Jorgensen T. Five years of
lifestyle intervention improved self-reported mental and physical health in a general
population: the Inter99 study. Preventive Medicine. 2009; 49: 424-428.
75
Juniper EF. Validated questionnaires should not be modified. Eur Respir J. 2009; 34:
1015-1017.
76
Mukaka MM. Statistics corner: a guide to appropriate use of correlation coefficient in
medical research. Malawi Medical Journal. 2012; 24(3): 69-71
77
Campbell MJ, Swinscow TDV. Statistics at Square One. 9th ed. London: BMJ Books;
c1997. Chapter 11, Correlation and Regression.
61
78
Algina J, Olejnik S. Sample size tables for correlation analysis with applications in
partial correlation and multiple regression analysis. Multivariate Behavioral Research.
2010; 38(3): 309-323.
79
Quality Metric (2011). QualityMetric Health OutcomesTM Scoring Software 4.5
Quality Metric, Lincoln RI. URL http://www.qualitymetric.com/
80
Newfoundland & Labrador Statistics Agency. 2011 Census and 2011 National
Household Survey (NHS). c2011 [cited 2014 Aug 08]. Available from:
http://www.stats.gov.nl.ca/Statistics/Census2011/
81
Zou KH, Tuncali K, Silverman SG. Statistical concepts series: correlation and simple
linear regression. Radiology. 2003; 227(3): 617-622.
82
Shields M, Shooshtari S. Determinants of self-perceived health. Health Reports. 2001;
13(1): 32-52.
83
Midanik LT, Validity of self-reported alcohol use: a literature review and assessment.
British Journal of Addiction. 1988; 83: 1019-1029.
84
Sawyer Sommers M, Dyehouse JM, Howe SR, Lemmink J, Volz T, Manharth M.
Validity of self-reported alcohol consumption in nondependent drinkers with
unintentional injuries. Alcoholism: Clinical and Experimental Research. 2000; 24(9):
1406-1413.
85
Strandberg AY, Strandberg TE, Salomaa VV, Pitkala K, Miettinen TA. Alcohol
consumption, 29-y total mortality and quality of life in men in old age. The American
Journal of Clinical Nutrition. 2004; 80(5): 1366-1371.
86
Volk RJ, Cantor SB, Steinbauer JR, Cass AR. Alcohol use disorders, consumption
patterns, and health-related quality of life of primary care patients. Alcoholism: Clinical
and Experimental Research. 1997; 21(5): 899-905.
87
Gill DL, Hammond CC, Reifsteck EJ, Jehu CM, Williams RA, Adams MM, Lange EH,
Becofsky K, Rodriguez E, Shang YT. Physical activity and quality of life. J Prev Med
Public Health. 2013;f 46 (Suppl 1): S28-S34.
88
Phillips SM, Wojcicki TR, McAuley E. Physical activity and quality of life in older
adults: an 18-month panel analysis. Qual Life Res. 2013; 22(7): 1647-1654.
89
Okano G, Miyake H, Mori M. Leisure time physical activity as a determinant of self-
perceived health and fitness in middle-aged male employees. J Occup Health. 2003;
45(5): 286-292.
90
Escobar A, Quintana JM, Bilbao A, Arostegui I, Lafuente I, Vidaurreta I.
Responsiveness and clinically important differences for the WOMAC and SF-36 after
total knee replacement. Osteoarthritis and Cartilage. 2007; 15(3): 273-280.
91
Angst F, Aeschlimann A, Stucki G. Smallest detectable and minimal clinically
important differences of rehabilitation intervention with their implications for required
62
sample sizes using WOMAC and SF-36 quality of life measurement instruments in
patients with osteoarthritis of the lower extremetitis. Arthritis Rheum. 2001; 45(4): 384-
391.
92
Cook CE. Clinimetrics corner: the minimal clinically important change score (MCID):
a necessary pretense. J Man Manip Ther. 2008; 16(4): E82-E83
93
Warkentin LM, Majumdar SR, Johnson JA, Agborsangaya CB, Rueda-Clausen CF,
Sharma AM, Klarenbach SW, Karmali S, Birch DW, Padwal RS. Weight loss required by
the severly obese to achieve clinically important differences in health-related quality of
life: two-year prospective cohort study. BMC Medicine. 2014; 12: 175.
94
Luo N, Johnson JA, Coons SJ. Using instrument-defined health state transitions to
estimate minimally important difference for four preference-based health-related quality
of life instruments. Meidcal Care. 2010; 48(4): 365-371.
95
Ringash J, O’Sullivan B, Bezjak A, Redelmeier DA. Interpreting clinically significant
changes in patient-reported outcomes. Cancer. 2007; 110(1): 196-202.
96
Hopman WM, Harrison MB, Coo H, Friedberg E, Buchanan M, VanDenKerkhof EG.
Associations between chronic disease, age and physical and mental health status. Chronic
Diseases in Canada. 2009; 29(2): 108-116.
97
Hazell ML, Morris JA, Linehan MF, Frank TL. Temporal change in health-realted
quality of life: a longitudinal study in general practice 1999-2004. Br J Gen Pract. 2009;
59(568): 839-843.
98
De Smedt D, Clays E, Annemans L, Doyle F, Kotseva K, Pajak A, Prugger C, Jennings
C, Wood D, De Bacquer D. Health related quality of life in coronary patients and its
association with their cardiovascular risk profile: Results from the EUROASPIRE III
survey. International Journal of Cardiology. 2013; 168(2): 898-903
99
Zabelina DL, Erickson AL, Kolotkin RL, Crosby RD. The effect of age on weight-
related quality of life in overweight and obese individuals. Obesity. 2009;17(7): 1410-
1413.
100
Mielck A, Vogelmann M, Leidl R. Health-related quality of life and socioeconomic
status: inequalities among adults with a chronic disease. Health and Quality of Life
Outcomes. 2014; 12: 58.
101
Ma X. McGhee SM. A cross-sectional study on socioeconomic status and health-
related quality of life among elderly Chinese. BMJ Open. 2012; 3(2): epub ahead of print.
102
Robert SA, Cherepanov D, Palta M, Cross Dunham N, Feeny D, Fryback DG.
Socioeconoic status and age variations in health-related quality of life: results from the
National Health Measurement Study. J Gerontol B Psychol Sci Soc Sci. 2009; 64(b): 378-
389
63
103
Coggon D, Rose G, Barker DJP, Epidemiology for the uninitiated, 4th ed. London:
BMJ Books; c1997. Chapter 4, Measurement Error and Bias.
104
Hubbard R. The probable consequences of violating the normality assumption in
parametric statistical analysis. The Royal Geographical Society. 1978; 10(5): 393-398.
105
Edgell SE, Noon SM. Effect of violation of normality on the t test of the correlation
coefficient. Quantitative Methods in Psychology: Psychology Bulletin. 1984; 94(3): 576-
583.
64
65
Appendix 1 – Overview of studies evaluated during literature search
Table 12 Examples of the literature examining the association between lifestyle and HRQoL and well-being evaluated in the
literature review.
Author, Purpose Study type Exposure variable Outcome Limitations Results & conclusions
year Population variable
Conry, To explore •Cross-sectional •Lifestyle - diet, •HRQoL. •Clustering •In this population there are
2011 clusters of survey exercise, alcohol •Self-rated analysis makes identifiable clusters of
lifestyle •Nationally- consumption, health comparisons lifestyle behaviours, both
behaviours representative tobacco use. difficult positive and negative.
and •Mental
sample of •Demographic health •Multi-factorial interventions
determine adults. Republic information – age, may be required to address
effect on of Ireland sex, social class clusters of poor lifestyle
health, (n=7350) behaviours in certain
mental populations.
health and
HRQoL
Savolainen, Examine the •Cross-sectional •Lifestyle – nutrition, •HRQoL. •Country-specific, •Participants in the healthiest
2014 association survey (baseline alcohol unvalidated lifestyle tertile were
between data from consumption, instrument for significantly more likely to
lifestyle and cohort) tobacco use, lifestyle have better HRQoL.
HRQoL •Community- exercise •Analysis based •"People who are expected to
dwelling adults •Demographic on tertiles, strive most to change their
in eastern information : age, making lifestyle have the lowest
Finland (n=560) sex, marital status, comparisons quality of life and
education difficult psychological welfare,
which should be taken into
account in both clinical
work and health promotion"
66
Xu, 2012 To evaluate •Cross-sectional •Lifestyle – tobacco •HRQoL • Sample may not •Smoking, alcohol
lifestyle and study smoking, alcohol represent wider consumption, decreased
HRQoL •Adults working consumption, population – high physical activity, short sleep
among civil in the Chinese physical activity, levels of duration, and high
servants in civil service sedentary time, university sedentariness were
China (n=14,021) sleep duration, education associated (p<0.05) with
breakfast habits. • No validated lower PCS and MCS scores
•Demographic instruments to •Lifestyle factors and
information – measure lifestyle behaviours affect HRQoL
martial status, level
of education
Pisinger, To •Interventional •Lifestyle •HRQoL •Lifestyle •Interventions to improve
2009 determine if study with 5- •Medical health measures at 3- lifestyle may also result in
an year follow-up assessment and 5-year changes to HRQoL.
intervention •Stratified followup not
to improve •Cardiovascular risk reported.
random sample assessment
lifestyle of adults aged •No validated
leads to an 30-60 years •Demographic instruments to
improvement from information - age, measure lifestyle
of HRQoL Copenhagen, sex, nationality,
Denmark employment status,
(n=7719) education,
•Disease history -
myocardial
infarction and
diabetes
Seib, 2014 To examine •Cross-sectional •Lifestyle – weight, •HRQoL •No validated • Women with history of
lifestyle, study physical activity, •Presence of instruments to personal trauma have poorer
stress and •Random sample alcohol and tobacco chronic measure lifestyle lifestyle. Poorer lifestyle is
quality of of women aged use, fruit and illness associated with poorer
life among 60-70 years vegetable outcomes such as HRQoL
older women living in consumption, sleep and chronic disease.
67
who have (or Australia quality •This suggests that lifestyle,
have not) (n=181) •Demographic life experiences and health
experienced information – age, are interrelated - life
life stressors. marital status, area stressors contribute to
of residence, unhealthy lifestyle
country of birth, behaviours, which in turn
education level, negatively affect health.
income,
employment
•History of stressful
life events
Myint, Evaluation •Cross-sectional •Lifestyle – tobacco •HRQoL •No validated •People with extremely poor
201166 of the study use, alcohol (converted instruments to lifestyles were 6.5 times
relationship • Healthy adults consumption, fruit to generate measure lifestyle more likely to die during
between recruited as part and vegetable QALYs) follow-up period (mean
lifestyle and of the Norfolk consumption, follow-up 11.4 yrs).
quality- site of the Eur. physical inactivity •Healthier lifestyle
adjusted life Prospective •Demogaphics – age, behaviours are associated
years Investigation BMI, education, with higher QALYs
into Cancer social class
(n=13,358)
Sabia, To evaluate •Cross-sectional •Lifestyle – history •HRQoL •Unrepresentative •Lifestyle behaviours have a
2012 64 the study of tobacco use, •Disability sample – high dose-response relationship
relationship participants in physical activity levels of with some aspects of
between the Whitehall II levels, consumption •Healthy university healthy aging.
lifestyle and cohort study of fruits and aging –no education
serious •Suggests that combination
healthy ≥60 years of age vegetables, •Binary measures of healthy lifestyle
aging at the time of consumption of illnesses,
good of lifestyle limits behaviours may confer
the study alcohol detailed analysis more benefit than individual
(n=5100) physical
•Age, sex, marital functioning, behaviours.
status, education
68
Appendix 2 – Human Research Ethics Authority approval and study information
sheet:
69
70
Primary Healthcare Research Unit
Janeway Hostel, 4th Floor
Health Sciences Centre
300 Prince Philip Dr
St. John’s NL A1B 3V6
Web: www.med.mun.ca/phru
Letter of Information
TITLE: The Simple Lifestyle Indicator Questionnaire and its Predictive Validity for
Health Status and Well-Being
INVESTIGATOR(S): Jacqueline Fortier (MSc. candidate) and Dr. Marshall Godwin

(supervisor)
You have been invited to take part in a research study. Taking part in this study is
voluntary. It is up to you to decide whether to be in the study or not. You can decide not
to take part in the study. If you decide to take part, you are free to leave at any time. This
will not affect your usual care.
Before you decide, you need to understand what the study is for, what risks you might
take and what benefits you might receive. This form explains the study.
Please read this carefully. Take as much time as you like. If you like, take it home to think
about for a while. Mark anything you do not understand, or want explained better. After
you have read it, please ask questions about anything that is not clear.
The researchers will:

• discuss the study with you
• answer your questions
• keep confidential any information which could identify you personally
• be available during the study to deal with problems and answer questions
1. Introduction/Background:
We are trying to learn more about lifestyle habits and a person’s health and well-
being. We know that some lifestyle habits increase your risk of getting a chronic
disease. We want to know whether those same lifestyle habits affect how you feel
right now. We are interested in how a person’s diet, exercise, tobacco use, stress, and
alcohol consumption affect their health.
71
This study will help us learn more about the link between lifestyle and health. We
hope it will help us understand how current habits can affect the way people feel.
2. Purpose of study:
The purpose of our study is to measure the lifestyle, health status and well-being of
adults. We will use these measurements to see whether lifestyle is connected to
health or well-being.
3. Description of the study procedures:
People who choose to participate in this study will provide some information about
themselves, including their age and sex. They will then complete four short surveys.
The researcher will help answer any questions that might come up.
4. Length of time:
Participation in this study will take between ten and twenty minutes, depending on
the time it takes to complete the questionnaires.
5. Possible risks and discomforts:
• Some participants may feel uncomfortable answering questions about their
emotional or physical health and well-being.
6. Benefits:
It is not known whether this study will benefit you.
7. Liability statement:
Completing our study questionnaires gives us your consent to be in this study. It tells
us that you understand the information about the research study. When you complete
the study questionnaires, you do not give up your legal rights. Researchers or
agencies involved in this research study still have their legal and professional
responsibilities.
8. What about my privacy and confidentiality?
Protecting your privacy is an important part of this study. Every effort to protect your
privacy will be made. However it cannot be guaranteed. For example we may be
required by law to allow access to research records.
When you complete the questionnaires you give us permission to

• Collect information from you
• Share information with the people conducting the study
• Share information with the people responsible for protecting your safety
Access to records
The members of the research team will see study records, but they will not identify
you by name.
Other people may need to look at the study record, but those records will not identify
you by name. This might include the research ethics board.
Use of your study information

The research team will collect and use only the information they need for this
research study.
72
This information will include your
• date of birth
• sex
• level of education
• household income
• your general health in the last year
• information from study interviews and questionnaires
Your name and contact information will not be collected. Your name will not appear
in any report or article published as a result of this study.
Information collected for this study will kept for five years.
If you decide to withdraw from the study, the information collected up to that time
will continue to be used by the research team. It may not be removed. This
information will only be used for the purposes of this study.
Information collected and used by the research team will be stored as encrypted files
on servers at the Primary Healthcare Research Unit. Jacqueline Fortier is the person
responsible for keeping it secure.
Your access to records

You may ask the researcher to see the information that has been collected about you.
9. Questions or problems:
If you have any questions about taking part in this study, you can meet with the
investigator who is in charge of the study at this institution. That person is:
Jacqueline Fortier
709-777-2942
[email protected]
Or you can talk to someone who is not involved with the study at all, but can
advise you on your rights as a participant in a research study. This person can be
reached through:
Ethics Office
Health Research Ethics Authority
709-777-6974 or by email at [email protected]
73
Appendix 3 – Study package provided to participants
PARTICIPANT DEMOGRAPHIC SHEET
GENDER: [ ] Male [ ] Female
AGE: _________
TOTAL HOUSEHOLD INCOME:
□ $25,000 or less
□ $25,001 to $35,000
□ $35,001 to $50,000
□ $50,001 to $75,000
□ $75,001 to $100,000
□ $100,001 to $150,000
□ $150,001 to $200,000
□ $200,001 plus
□ Prefer not to respond
Educational Level
□ Did not complete high school
□ Completed high school
□ Some college or university studies
□ Completed college diploma or university degree
□ Some postgraduate or professional training
□ Completed postgraduate or professional training.
74
Simple Lifestyle Indicator Questionnaire
Diet: To answer these questions, think about your eating habits in the past year.
Indicate how often you eat the following foods. Please include all meals, snacks and
eating out
1. Lettuce or green leafy salad, with or without other vegetables
[ ] less than 1/week [ ] 1/week [ ] 2-3x/week [ ] 4-6x/week [ ] once/day [ ] 2+/day
2. Fruit: include fresh, canned or frozen, but do not include juices
3. High fiber cereals or whole grain breads: this includes cereal such as Raisin bran,
Fruit and Fiber, cooked oatmeal, and breads which are whole wheat, multigrain,
rye or pumpernickel
Exercise: To answer the following questions please indicate how many times per
week you take part in the following activities for a duration of at least 30 minutes or
more at a time:
I. Light exercise, such as:
• light gardening and light housework (dusting, sweeping, vacuuming)
• leisurely walking (walking your dog)
• bowling, fishing, carpentry, playing a musical instrument
• volunteer work
[ ] 0/week [ ] 1-3x/week [ ] 4-7x/week [ ] 8 and more/week
II. Moderate exercise, for example:

• brisk walk
• bicycling, skating, swimming, curling
• gardening (raking, weeding, spading)
• dancing, Tai Chi or moderate exercise classes
75
III. Vigorous exercise, for example:
• running, bicycling, x-country skiing, lap swimming, aerobics
• heavy yard work
• weight training
• soccer, basketball or other league sports
Alcohol: Please indicate how many drinks of the following types of alcohol you
consume in an average week:
 Wine: _____ drinks (3-5 oz.)
 Beer: _____ drinks (10-12 oz or 1 bottle)
 Spirits: ____ drinks (1-1 ½ oz.)
Smoking: Please indicate your smoking habits below:
Are you a smoker? [ ] Yes [ ] No
If yes, how long have you been smoking? __________ years
If no, did you ever smoke? [ ] Yes [ ] No
If yes, how many years ago did you quit? ___________ years
Life Stress: To answer this question please circle the number which you feel best
corresponds to the level of stress in your everyday life
1 2 3 4 5 6
Not at all stressful Very stressful
76
EuroQoL EQ-5D-5L
Under each heading, please tick the ONE box that best describes your health TODAY
MOBILITY
I have no problems in walking about 
I have slight problems in walking about 
I have moderate problems in walking about 
I have severe problems in walking about 
I am unable to walk about 
SELF-CARE
I have no problems washing or dressing myself 
I have slight problems washing or dressing myself 
I have moderate problems washing or dressing myself 
I have severe problems washing or dressing myself 
I am unable to wash or dress myself 
USUAL ACTIVITIES (e.g. work, study, housework,
family or leisure activities)
I have no problems doing my usual activities 
I have slight problems doing my usual activities 
I have moderate problems doing my usual activities 
I have severe problems doing my usual activities 
I am unable to do my usual activities 
PAIN / DISCOMFORT
I have no pain or discomfort 
I have slight pain or discomfort 
I have moderate pain or discomfort 
I have severe pain or discomfort 
I have extreme pain or discomfort 
ANXIETY / DEPRESSION
I am not anxious or depressed 
I am slightly anxious or depressed 
I am moderately anxious or depressed 
I am severely anxious or depressed 
I am extremely anxious or depressed 
77
78
79
80
81
Psychological General Well-Being Index
READ: This section of the examination contains questions about how you feel and how
things have been going with you. For each question check [ ] the answer which best
applies to you.
1. How have you been feeling in general during the past month?
(Check one box)
In excellent spirits .............................................................................................. 5

In very good spirits ............................................................................................ 4
In good spirits mostly ........................................................................................ 3
I have been up and down in spirits a lot ............................................................ 2
In low spirits mostly .......................................................................................... 1
In very low spirits .............................................................................................. 0
2. How often were you bothered by any illness, bodily disorder, aches or
pains during the past month?
(Check one box)
Every day ........................................................................................................... 0

Almost every day ............................................................................................... 1
About half of the time ....................................................................................... 2
Now and then, but less than half the time ......................................................... 3
Rarely ................................................................................................................. 4
None of the time ................................................................................................ 5
3. Did you feel depressed during the past month?

(Check one box)
Yes - to the point that I felt like taking my life ................................................ 0

Yes - to the point that I did not care about anything ......................................... 1
Yes - very depressed almost every day ............................................................. 2
Yes - quite depressed several times .................................................................. 3
Yes - a little depressed now and then ................................................................ 4
No - never felt depressed at all .......................................................................... 5
82
4. Have you been in firm control of your behavior, thoughts, emotions
or feelings during the past month?
(Check one box)
Yes, definitely so ............................................................................................... 5

Yes, for the most part ........................................................................................ 4
Generally so ....................................................................................................... 3
Not too well ....................................................................................................... 2
No, and I am somewhat disturbed ..................................................................... 1
No, and I am very disturbed .............................................................................. 0
5. Have you been bothered by nervousness or your "nerves" during

the past month?
(Check one box)
Extremely so - to the point where I could not work or take care of things ...... 0
Very much so ..................................................................................................... 1
Quite a bit .......................................................................................................... 2
Some - enough to bother me ............................................................................. 3
A little ................................................................................................................ 4
Not at all ............................................................................................................. 5
6. How much energy, pep, or vitality did you have or feel during
the past month?
(Check one box)
Very full of energy - lots of pep ........................................................................ 5

Fairly energetic most of the time ...................................................................... 4
My energy level varied quite a bit .................................................................... 3
Generally low in energy or pep ......................................................................... 2
Very low in energy or pep most of the time ..................................................... 1
No energy or pep at all - I felt drained, sapped ................................................ 0
7. I felt downhearted and blue during the past month.

(Check one box)
None of the time ................................................................................................ 5

A little of the time .............................................................................................. 4
Some of the time ................................................................................................ 3
A good bit of the time ........................................................................................ 2
Most of the time ................................................................................................. 1
All of the time .................................................................................................... 0
83
8. Were you generally tense or did you feel any tension during the past month?
(Check one box)
Yes - extremely tense, most or all of the time .................................................. 0

Yes - very tense most of the time ...................................................................... 1
Not generally tense, but did feel fairly tense several times .............................. 2
I felt a little tense a few times ........................................................................... 3
My general tension level was quite low ............................................................ 4
I never felt tense or any tension at all ............................................................... 5
9. How happy, satisfied, or pleased have you been with your personal life
during the past month?
(Check one box)
Extremely happy - could not have been more satisfied or pleased .................. 5
Very happy most of the time ............................................................................. 4
Generally satisfied - pleased ............................................................................. 3
Sometimes fairly happy, sometimes fairly unhappy ........................................ 2
Generally dissatisfied or unhappy ..................................................................... 1
Very dissatisfied or unhappy most or all the time ............................................ 0
10. Did you feel healthy enough to carry out the things you like to do
or had to do during the past month?
(Check one box)
Yes - definitely so .............................................................................................. 5

For the most part ................................................................................................ 4
Health problems limited me in some important ways ...................................... 3
I was only healthy enough to take care of myself ............................................ 2
I needed some help in taking care of myself .................................................... 1
I needed someone to help me with most or all of the things I had to do ......... 0
11. Have you felt so sad, discouraged, hopeless, or had so many problems
that you wondered if anything was worthwhile during the past month?
(Check one box)
Extremely so - to the point that I have just about given up .............................. 0

Very much so ..................................................................................................... 1
Quite a bit .......................................................................................................... 2
A little bit ........................................................................................................... 4
Not at all ............................................................................................................. 5
84
12. I woke up feeling fresh and rested during the past month.
(Check one box)
None of the time ................................................................................................ 0

Some of the time ................................................................................................ 2
Most of the time ................................................................................................. 4
All of the time .................................................................................................... 5
13. Have you been concerned, worried, or had any fears about your health
(Check one box)
Extremely so ...................................................................................................... 0
Very much so ..................................................................................................... 1
Quite a bit .......................................................................................................... 2
Some, but not a lot ............................................................................................. 3
Practically never ................................................................................................ 4
Not at all ............................................................................................................. 5
14. Have you had any reason to wonder if you were losing your mind,
or losing control over the way you act, talk, think, feel or of your
memory during the past month?
(Check one box)
Not at all ............................................................................................................. 5

Only a little ........................................................................................................ 4
Some - but not enough to be concerned or worried about ............................... 3
Some and I have been a little concerned ........................................................... 2
Some and I am quite concerned ........................................................................ 1
Yes, very much so and I am very concerned .................................................... 0
15. My daily life was full of things that were interesting to me during
the past month.
Check one box)
None of the time ................................................................................................ 0

Some of the time ................................................................................................ 2
Most of the time ................................................................................................. 4
All of the time .................................................................................................... 5
85
16. Did you feel active, vigorous, or dull, sluggish during the past month?
(Check one box)
Very active, vigorous every day ....................................................................... 5

Mostly active, vigorous - never really dull, sluggish ....................................... 4
Fairly active, vigorous - seldom dull, sluggish ................................................. 3
Fairly dull, sluggish - seldom active, vigorous ................................................. 2
Mostly dull, sluggish - never really active, vigorous ....................................... 1
Very dull, sluggish every day ............................................................................ 0
17. Have you been anxious, worried, or upset during the past month?
(Check one box)
Extremely so - to the point of being sick or almost sick .................................. 0

Very much so ..................................................................................................... 1
Quite a bit .......................................................................................................... 2
A little bit ........................................................................................................... 4
Not at all ............................................................................................................. 5
18. I was emotionally stable and sure of myself during the past month.
(Check one box)
None of the time ................................................................................................ 0

Some of the time ................................................................................................ 2
Most of the time ................................................................................................. 4
All of the time .................................................................................................... 5
19. Did you feel relaxed, at ease or high strung, tight, or keyed-up
(Check one box)
Felt relaxed and at ease the whole month ......................................................... 5

Felt relaxed and at ease most of the time .......................................................... 4
Generally felt relaxed but at times felt fairly high strung ................................ 3
Generally felt high strung but at times felt fairly relaxed ................................ 2
Felt high strung, tight, or keyed-up most of the time ....................................... 1
Felt high strung, tight, or keyed-up the whole month ...................................... 0
86
20. I felt cheerful, lighthearted during the past month.
(Check one box)
None of the time ................................................................................................ 0

Some of the time ................................................................................................ 2
Most of the time ................................................................................................. 4
All of the time .................................................................................................... 5
21. I felt tired, worn out, used up, or exhausted during the past month.
(Check one box)
None of the time ................................................................................................ 5

Some of the time ................................................................................................ 3
Most of the time ................................................................................................. 1
All of the time .................................................................................................... 0
22. Have you been under or felt you were under any strain, stress, or
pressure during the past month?
(Check one box)
Yes - almost more than I could bear or stand ................................................... 0

Yes - quite a bit of pressure ............................................................................... 1
Yes, some - more than usual ............................................................................. 2
Yes, some - but about usual .............................................................................. 3
Yes - a little ........................................................................................................ 4
Not at all ............................................................................................................. 5
87
Appendix 4 – Scoring template for the Simple Lifestyle Indicator Questionnaire
88
89

Wellbeing N Health Questionnaire

Uploaded by

Copyright:

Available Formats

Wellbeing N Health Questionnaire

Uploaded by

Document Information

Original Title

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

Wellbeing N Health Questionnaire

Uploaded by

Copyright:

Available Formats

View metadata, citation and similar papers at core.ac.

uk brought to you by CORE

The Simple Lifestyle Indicator Questionnaire and its association

with health-related quality of life and well-being.

A thesis submitted to the

School of Graduate Studies

In partial fulfillment of the requirements for the degree of

Masters of Science (Clinical Epidemiology)

Memorial University of Newfoundland

St John’s, Newfoundland & Labrador

Questionnaire, a measure of lifestyle, and health-related quality of life and well-being

was significantly, positively correlated with well-being (r=0.47, p<0.01), self-perceived

of validated questionnaires but the generalizability of these results is limited by a sample

My supervisory committee, Shabnam Asghari and Cheri Bethune, provided invaluable

Table 2 Inclusion and exclusion criteria used to determine eligibility to participate in

Table 4 Description of the socio-demographic characteristics of the study population

Table 5 Measures of central tendency and variability for questionnaires.......................... 32

Figure 1 Venn diagram demonstrating domains unique to and shared by measures of

Figure 3 Percentage of individuals achieving stated levels of education for study

Figure 5 Histogram displaying the distribution of overall SLIQ scores. ........................... 33

ANOVA - Analysis of variance

BMI - Body mass index

CLAS – Computerized Lifestyle Assessment Scale

EQ5D – EuroQoL EQ-5D questionnaire

EQ5D VAS - EuroQoL EQ-5D visual analogue scale

HRQoL - Health-related quality of life

MCID – Minimal clinically important difference

NL – Newfoundland & Labrador, Canada

PGWB - Psychological General Well-being questionnaire

PGWB-I - Psychological General Well-being Index score

SF-12 - Short form 12 Questionnaire

SF-36 – Short form 36 Questionnaire

SF-12 MCS - Short form 12 – Mental component score

SF-12 PCS - Short form 12 – Physical component score

SLIQ - Simple Lifestyle Indicator Questionnaire

SPSS - Statistical Package for the Social Sciences

Appendix 1: Overview of studies evaluated during literature search

Appendix 3: Survey package provided to participants

Appendix 4: Scoring template for the Simple Lifestyle Indicator Questionnaire

including environment, economic and social circumstances, and a person’s characteristics

physical activity3 contribute to a construct known as lifestyle, which can significantly

contribute to a person’s risk of developing a disease or illness, including tobacco use 4,

alcohol consumption 5, physical inactivity 6, an unhealthy diet 7, and psychological stress 8.

contrast to many clinical measures of health and disease, HRQoL is a patient-reported

in their HRQoL 13, the measure is generally sensitive to clinically-relevant changes in

differences in HRQoL that may be attributed to lifestyle factors among an otherwise

healthy population, which shall be discussed in further detail.

1.2 Health-related quality of life

There is a significant body of literature on the value of HRQoL as a measured outcome in

environment, to the components specifically pertaining to health13. Including this measure

as an outcome in clinical study acknowledges two important realities: that physiologic

patient-reported outcomes are of interest to clinicians, policymakers and researchers as

HRQoL in a disease-specific manner13. General instruments measure dimensions or

physical or emotional role, or feelings of anxiety or depression13. Specific instruments

incorporate additional domains relevant to the subpopulation of interest; for example,

sleep based on participants’ responses of yes or no to 38 questions20. Alternatively,

generate single scores to represent an individual’s HRQoL, often based on scoring

algorithms that combine a number of dimensions20. Examples of preference-based

Scale, which measures of mobility, physical activity, social activity, and