Full text of DOI:10.1207/S15324796ABM2303_7

Self-Rated Quality of Life Measures: Effect of Change to a Low-Fat,
High-Fiber, Fruit and Vegetable Enriched Diet
Donald K. Corle, M.S.
National Cancer Institute
Carolyn Sharbaugh, M.S.
Westat
Donna J. Mateski, M.S.
Walter Reed Army Medical Center
Terry Coyne, Ph.D.
University of Pittsburgh
Electra D. Paskett, Ph.D.
Wake Forest University
Jack Cahill, M.A. and Cassandra Daston, M.A.
Westat
Elaine Lanza, Ph.D. and Arthur Schatzkin, M.D., Dr.P.H.
National Cancer Institute
The PPT Study Group
ABSTRACT
provides evidence that, given appropriate support, free-living
individuals can successfully alter their eating patterns in multi-
ple ways without a negative impact on quality of life.
The Polyp Prevention Trial (PPT) was a multicenter, ran-
domized clinical trial to determine the effect of a low-fat (20% of
energy from fat), high-fiber (18 g/1000 kcal/day),
high-fruit/vegetable (3.5 servings/1000 kcal/day) eating plan on
the recurrence of large bowel adenomatous polyps. The PPT
provided an opportunity to examine the impact of dietary
changes on quality of life. At baseline and annually for 4 years,
participants in the Quality of Life Substudy of PPT completed a
Quality of Life Factors (QF) Questionnaire, a modified
Block–National Cancer Institute Food Frequency Question-
naire, and a Health and Lifestyle Questionnaire. The 51-item
QF Questionnaire assessed changes in nine domains: taste, con-
venience, cost, self-care, social, health assessment, health be-
lief, health action, and life satisfaction. The analysis compared
annual changes in domain scores for intervention (n = 194) and
control (n = 200) participants. At Year 1, 363 (92%) completed
a questionnaire, and 325 (82%) participants completed a Year 4
questionnaire. There were no statistically significant differences
between treatment groups in the change from baseline to Year 1
for the convenience, cost, taste, health assessment, and life sat-
isfaction domains. At Year 1, intervention participants rated the
self-care (p < .001), health belief (p = .021), and health action
(Ann Behav Med
2001, 23(3):198–207)
INTRODUCTION
During the past 20 years, quality of life has emerged as an
important consideration in disease prevention and treatment
(1,2). A major goal in developing prevention and treatment
strategies is to help individuals live longer and preserve their
quality of life (3). It is important to understand the impact of the
prevention or treatment approach on daily life because strategies
with detrimental effects are not likely to be adopted and main-
tained. Although initial quality of life research centered on ther-
apies with profound side effects, more recent efforts have
looked at the impact of changes in everyday health habits, such
as diet, on quality of life (4–10).
Dietary factors are associated with 5 of the 10 leading
causes of death in the United States, including coronary heart
disease, some types of cancer, stroke, noninsulin-dependent dia-
betes mellitus, and atherosclerosis (11). Decreasing fat and in-
creasing fiber, fruits, and vegetables in the diets of Americans
has become a national priority (11,12). For recommendations
such as the Dietary Guidelines for Americans to be widely
adopted, it is important to understand their impact on overall
quality of life. Earlier studies on cholesterol- and fat-lowering
interventions found higher death rates from suicides, homicides,
and accidents (4,13–16), which may indicate unanticipated det-
rimental effects of the prevention strategy on various aspects of
life. However, observational studies have had more ambiguous
results (17–19), and recent randomized trials of the impact of
cholesterol-lowering diet interventions on women (5) or
(p < .001) domains significantly higher and the social domain
significantly lower (p < .001) than control participants. These
changes were consistent through Years 2, 3, and 4. This study
This work was supported by National Cancer Institute Contract No.
2
SC05221.
Reprint Address: D. K. Corle, National Cancer Institute, Executive
Plaza North, Suite 3131, 6130 Executive Blvd, MSC 7354, Bethesda,
MD 20892–7354.
198

Volume 23, Number 3, 2001
Self-Rated Quality of Life
199
mixed-gender (6,7) cohorts have found no impact of the diet on
mental health variables. Other studies suggest that healthier di-
ets improve well-being and that improved well-being and adher-
ence to a dietary regimen are related (8,9).
In assessing the true impact of dietary changes or dietary in-
tervention activities on quality of life, one must investigate
changes in physical and emotional well-being as well as look at
other aspects of daily life that are affected by a major change in
eating habits. These go beyond the “health” aspects of quality of
life and include whether the food tastes good, whether the eating
plan puts a crimp in one’s social life, whether the food is difficult
to prepare, and whether the eating plan costs too much.
The National Cancer Institute (NCI)-sponsored Polyp Pre-
vention Trial (PPT) provided an ideal opportunity to examine a
wide range of quality of life factors that may be influenced by
changes in eating habits and to determine how attitudes toward
these factors changed over time. In this randomized trial, partici-
pants in the intervention program were asked to make three con-
current dietary changes and to maintain these changes for 4
years, whereas the control participants were asked to continue
with their usual eating habits. The purpose of this article is to
evaluate the effects of a low-fat, high-fiber, and high-fruit and
vegetable eating plan on nutrition- and health-related quality of
life measures among PPT participants. The analysis is based on
quality of life questionnaire data collected from a subgroup of
PPT participants at baseline and Years 1 through 4; it focuses on
changes in measures such as self-perceived physical and emo-
tional well-being, satisfaction with the diet, and self-care.
study were asked to complete the Quality of Life Factors (QF)
Questionnaire at baseline and annually for 4 years, in addition to
other PPT baseline and annual measurements. Of these, 394
(97%) agreed to participate and completed the questionnaire at
baseline. Characteristics of participants in trials that accrue over
several years may change over the course of the study, and
substudies selected based on fixed time intervals of accrual may
not be representative of the trial as a whole. Although address-
ing this question of potential bias is not critical to the analyses
presented in this article, Table 1 shows the comparability of
QOL Substudy to participants in the remainder of the PPT trial.
Table 2 presents the demographic characteristics of the 200 con-
trol and 194 intervention participants completing a baseline QF
Questionnaire. In general, substudy participants were White,
married, well-educated nonsmokers with a mean age of 60.9
years at baseline. Thirty-seven percent of participants were
TABLE 1
Characteristics of Participants in the QOL Substudy and the
Remainder of PPT
Non-QOLa
QOLb
Intervention (%)
Age (M)
50.0 (1.2)
61.6 (0.2)
65.5 (1.2)
10.3 (0.7)
49.2 (2.5)
60.9 (0.5)
62.7 (2.4)
11.4 (1.6)
Men (%)
c
Minority (%)
More than high school
d
education (%)
64.3 (1.2)
13.9 (0.8)
27.6 (0.1)
95.5 (0.5)
66.8 (2.4)
11.7 (1.6)
27.5 (0.2)
96.2 (1.0)
Smoking at baseline (%)
Body mass index (M)
Alive at Year 4 (%)
METHOD
Overview of Study Design
Note. Numbers in parentheses are standard errors. None of the
differences between QOL and other PPT participants are statistically
significant. QOL = Quality of Life; PPT = Polyp Prevention Trial.
n = 1685. n = 394. Minorities include African Americans, Hispanics,
Indians/Native Americans, Asians/Pacific Islanders, and others. dOne
participant with missing education data in non-QOL.
The PPT study design, participant recruitment, randomiza-
tion, and nutrition intervention program have been described in
detail elsewhere (20,21). Briefly, the PPT was a multicenter,
randomized clinical trial to test the effect of a low-fat, high-fi-
ber, and high-fruit and vegetable eating plan on the recurrence of
adenomatous polyps of the large bowel. Participants were 2,079
men and women, 35 years of age or older, who had one or more
large bowel adenomatous polyps removed within 6 months be-
fore randomization. Participant recruitment and randomization
occurred between June 1991 and January 1994. During the 4
years of participation in the study, intervention participants re-
ceived ongoing nutrition and behavioral counseling to adopt an
eating plan providing daily 20% of energy from fat, 18 g of di-
etary fiber per 1,000 kcal in their usual diet, and 3.5 servings of
fruits and vegetables per 1,000 kcal (which ranged from five to
eight servings per day, depending on energy intake). Control
participants did not receive such counseling and were expected
to maintain their usual eating pattern. Self-reported dietary,
health and lifestyle data, and blood samples were collected at
study entry and on an annual basis.
a
b
c
TABLE 2
Baseline Characteristics of Control and Intervention
Participants in the QOL Substudy
Intervention
Group
b
Characteristic
Control Groupa
Age (M)
Men (%)
Minority (%)
61.6 (0.7)
61.0 (3.4)
9.0 (2.0)
60.2 (0.8)
63.4 (3.5)
13.9 (2.5)
c
More than high school
education (%)
Smoking (%)
65.0 (3.4)
12.5 (2.3)
27.4 (0.3)
83.4 (2.6)
68.6 (3.3)
10.8 (2.2)
27.6 (0.3)
77.3 (3.0)
Body mass index (M)
d
Married (%)
Note. Numbers in parentheses are standard errors. None of the
differences between the intervention and control groups are statistically
significant. QOL = Quality of Life.
Planning for the Quality of Life (QOL) Substudy described
in this article began early in 1993, more than 2 years after the
main PPT study began. All 407 participants accrued and ran-
domized into the trial between August 1993 and the end of the
a
b
c
n = 200. n = 194. Minorities include African Americans, Hispanics,
d
Indians/Native Americans, Asians/Pacific Islanders, and others. One
participant with missing marital status data in the control group.

2
00
Corle et al.
Annals of Behavioral Medicine
women, and 9.0% of control participants and 13.9% of interven-
tion participants were from minority populations.
consumption and eating behavior. The individual questions
within each domain were scored on Likert scales, where higher
scores represent greater satisfaction and higher quality of life.
All of the questions within the taste, convenience, cost,
self-care, and social domains are from the Southeast Cholesterol
Project QOL Questionnaire developed by Ammerman,
DeVellis, Keyserling, and Simpson (7). Twenty-two of the ques-
tions had been tested for reliability with participants in the phy-
sician-directed diet treatment program. An additional 11 ques-
tions (7 in the social domain and 1 each in the self-care, taste,
convenience, and cost domains) were under evaluation for valid-
ity and reliability at the time the QOL Substudy was being de-
veloped, and these were also included in the QF Questionnaire.
The 7-item health assessment domain includes four ques-
tions from the Medical Outcomes Study’s 36-item Short-Form
Health survey (22), two questions from the Health Locus of
Assessments
PPT QF Questionnaire. The QF Questionnaire was a
5
1-item, self-administered form completed by the participant at
baseline and again at annual follow-up visits for 4 years. The
purpose of the QF Questionnaire was to assess and monitor
changes in nine domains related to diet and health: taste, conve-
nience, cost, self-care, social, health assessment, health belief,
health action, and life satisfaction. Loading questions for each
domain (Table 3) provide a description of the main nutrition-
and health-related issues addressed in each domain. These do-
mains were selected because they focus on areas of everyday life
assumed to be affected by the adoption of major changes in food
TABLE 3
Internal Reliability and Description of Nine Domains Composing the QF Questionnaire
Domain
No. of Items
Cronbach’s α
Loading Questionsa
Taste
4
0.7
20. I enjoy most of the food I eat.
5. I am satisfied with the taste of the food I eat.
Responses: Disagree a lot, Disagree a little, Agree a little, Agree a lot)
18. Is it hard or easy to shop for the kinds of food you or your family eat?
9. Is it hard or easy to plan and prepare the kinds of meals you or your family eat?
Responses: Very hard, A little hard or don’t know, Pretty easy, Very easy)
36. The high cost of food is a problem for me.
7. Sometimes it’s hard for me to pay for the kind of food I eat.
Responses: Disagree a lot, Disagree a little, Agree a little, Agree a lot)
22. I feel like I’m doing something good for myself by eating the way I do.
9. Eating the way I do now makes me feel good about myself.
Responses: Disagree a lot, Disagree a little, Agree a little, Agree a lot)
8a. I enjoy the foods I eat at holiday meals and parties.
b. I enjoy getting together for meals with friends.
Responses: Disagree a lot, Disagree a little, Agree a little, Agree a lot)
2
(
Convenience
Cost
5
5
0.7
0.8
0.9
0.7
0.8
1
(
3
(
Self-care
Social
8
2
(
11
7
8
(
Health
13a. I have a lot of energy to accomplish what I need to.
assessment
(Responses: Disagree a lot, Disagree a little, Agree a little, Agree a lot)
1
4. During the past month, how have you been feeling generally?
(Responses: Poor spirits, Fairly good spirits, Good spirits, Very good spirits, Excellent
spirits)
Health belief
Health action
3
4
0.9
0.5
11. How likely is it that food choices you can make will:
1
1
1b. Improve your health?
1c. Help you feel better?
(Responses: Very unlikely, Somewhat unlikely, Somewhat likely, Very likely)
38. During the past month, how often have you received nutrition or health messages from
the following sources:
3
3
8a. Media (television, radio, newspapers, magazines)
8d. Food labels
(Responses: Never, Sometimes, Often)
Life
satisfaction
4
0.7
15. In general, how satisfying is your life?
(Responses: Very unsatisfying, Fairly unsatisfying, Fairly satisfying, Very satisfying)
b
1
6a. Overall, during the past month, how much satisfaction have you gotten from your
leisure time
Responses: Does not apply, No satisfaction, A little satisfaction, A moderate amount of
satisfaction, A good deal of satisfaction)
(
Note. QF = Quality of Life Factors.
a
b
Loading questions are the two questions in each domain that give maximal loading to the scale as determined by factor analysis. Question on job
satisfaction was not included as 40% of PPT population was retired.

Volume 23, Number 3, 2001
Self-Rated Quality of Life
201
Control Questionnaire (23), and one question from the Psycho-
logical General Well-Being Scale (24). The life satisfaction do-
main includes 4 items from the Bryant and Veroff QOL measure
tervention and control in the mean change for each domain score
was tested for significance using a two-sample t test. To deter-
mine whether the changes observed at the end of Year 1 were
maintained, the changes in each domain score from baseline to
Years 2, 3, and 4 were calculated for each participant. For those
domains that showed significant group differences at Year 1, re-
peated measures analysis of variance was used to test the consis-
tency of group differences over the 4 years of follow-up.
(25). The health belief domain includes questions on the per-
ceived benefits of healthy food choices, based on Becker’s
Health Belief Model (26). The four questions in the health ac-
tion domain, also based on the Health Belief Model, were
adapted from a published scale developed by Connell and
Crawford (27).
The reliability of each domain at baseline was assessed using
Cronbach’s alpha coefficient, which measures the internal consis-
tency of the questions in the domain (28). Reliability was 0.7 to
RESULTS
Domain Scores at Baseline
0
.9 at baseline (see Table 3) for all domains except health action,
As shown in Table 4, mean scores for each domain ranged
from 50.5 to 83.1 at baseline on a scale ranging from 0 (worst
quality of life) to 100 (best quality of life). The taste, convenience,
social, and health belief domains were all rated above 70. There
were no significant differences between control and intervention
participants for any domain except the taste domain. Control par-
ticipants scored the taste domain a statistically significant (p =
.005) 3.8 points higher on average than did intervention partici-
pants. We have no explanation for this difference, and as the ran-
domized trial progressed, taste domain scores for control partici-
pants steadily decreased from the high baseline levels.
which had a reliability of 0.5. Because the health action domain
includes separate queries on diverse sources of nutrition mes-
sages, we would not expect high intercorrelations among items.
Health and Lifestyle Questionnaire. This interviewer-ad-
ministered questionnaire was used to assess a variety of demo-
graphic, clinical, and behavioral characteristics, including ques-
tions on food shopping and meal preparation responsibilities in
the participant’s household. The questionnaire was completed at
baseline and annually for all PPT participants.
Relation of Baseline Demographic
Characteristics to Baseline Domain Scores
Food Frequency Questionnaire (FFQ). At baseline, and
annually thereafter, all participants completed a modified
Block–NCI FFQ (29). Modifications to the Block–NCI FFQ for
the PPT, discussed in more detail elsewhere (21), included the
addition of low-fat and nonfat food items and more high-fiber
foods, such as dried fruit, high-fiber cereals, and legumes. The
FFQ was self-administered at each annual visit, after which a
specially trained nutrition staff member reviewed the completed
questionnaire with the participant. Nutrient analysis of the com-
pleted FFQs was conducted using a modified Dietsys 2.0 pro-
gram (30).
Because there were minimal differences in domain scores
between treatment groups at the beginning of the QOL, baseline
scores for all participants were combined for the analysis of the
relation of selected baseline demographic and behavioral vari-
ables to each domain. As presented in Table 5, women scored
the convenience (p = .004) and social (p = .001) domains higher
and the cost (p = .011) domain lower than did men. Participants
older than 61 years of age rated themselves higher on the
self-care domain (p < .001) compared to younger participants.
Married participants more positively rated the cost (p = .009),
social (p = .008), health assessment (p = .019), and life satisfac-
tion (p < .001) domains than did their single counterparts. Simi-
larly, the higher the educational level, the more participants re-
Data Analysis
For each item in the QF Questionnaire, scores were coded
and standardized to a scale ranging from 0 (worst quality of life)
to 1 (best quality of life). Missing responses to individual items,
which were minimal (< 5%), were assigned a value of 0, the
lowest value for the question. The total score for each domain
was created by summing the responses to the individual ques-
tions in the domain, then dividing by the total number of domain
questions and multiplying this value by 100. Therefore, the total
score for each domain had a minimum value of 0 and a maxi-
mum value of 100, and each question had equal weight in creat-
ing the domain.
For each domain score, the mean and its standard error were
calculated at baseline. Simple regression analysis quantified the
relation of selected baseline demographic and behavioral vari-
ables to each domain score. The selected variables included gen-
der, race, age, marital status, body mass index, educational level,
and smoking status.
TABLE 4
Mean Baseline Domain Scores of Control and Intervention
Participants in the QOL Substudy
Domain
Control Groupa
Intervention Groupb
Tastec
Convenience
Cost
Self-care
Social
Health assessment
Health belief
Health action
Life satisfaction
83.1 (0.9)
72.7 (1.4)
66.1 (1.3)
61.0 (1.3)
74.5 (0.9)
64.0 (1.4)
76.3 (1.6)
51.1 (1.4)
62.4 (1.0)
79.3 (1.0)
72.1 (1.3)
65.7 (1.4)
59.0 (1.3)
73.1 (1.0)
62.8 (1.4)
78.8 (1.5)
50.5 (1.4)
63.7 (1.0)
Note. Numbers in parentheses are standard errors. QOL = Quality of
Life.
The change in each domain score from baseline to Year 1
was calculated for each participant. The difference between in-
a
b
c
n = 200. n = 194. Significant difference between groups, p = .005.

2
02
Corle et al.
Annals of Behavioral Medicine
TABLE 5
Differences in Mean Levels of Baseline Domain Scores for Dichotomous Demographic Variables
Domain
Health
Assessment
Health
Belief
Health
Action
Life
Satisfaction
Taste
Convenience
Cost
Self-Care
Social
Gender
Female = 149
Male = 245
Race
–2.38
(.092)
–5.77
(.004)
5.08
(.011)
–1.79
(.345)
–4.29
(.001)
3.64
(.072)
–4.80
(.036)
–4.20
(.041)
–0.21
(.888)
White = 349
Other = 45
1.32
(.541)
–4.28
(.164)
–4.97
(.103)
2.25
(.436)
–6.88
(<.001)
3.85
(.213)
–3.65
(.297)
1.26
(.688)
1.24
(.586)
Age
≤
>
61 years = 191
61 years = 203
1.95
(.155)
1.55
(.429)
–1.38
(.479)
7.22
(<.001)
–1.53
(.241)
–2.95
(.133)
–4.05
(.068)
1.73
(.387)
1.37
(.345)
Body mass index
≤
>
25 = 111
25 = 283
0.74
(.626)
–1.65
(.447)
–2.34
(.279)
–5.14
(.012)
2.16
(.136)
0.80
(.716)
–0.31
(.899)
0.21
(.926)
–2.45
(.129)
Education
≤
High school = 131
–2.57
(.078)
3.00
(.148)
13.93
(< .001)
–4.77
(.014)
3.47
(.012)
3.76
(.071)
0.04
(.986)
–0.91
(.669)
–0.62
(.687)
College = 263
Smoker
No = 348
Yes = 46
Marital status
–3.17
(.138)
–1.39
(.649)
–5.19
(.086)
–9.84
(< .001)
–1.86
(.360)
–4.90
(.109)
0.26
(.941)
–2.78
(.372)
–3.61
(.110)
a
Not married = 77
Married = 316
3.80
(.138)
2.61
(.290)
6.40
(.009)
2.18
(.346)
4.35
(.008)
5.77
(.019)
1.26
(.653)
2.60
(.299)
6.53
(<.001)
Note. Each factor was classified into two categories; the table entry is the difference between the means for the two categories. Values in parentheses are p
values for a t test that the difference is equal to zero.
a
One person had missing marital status at baseline.
TABLE 6
Self-Reported Daily Energy, Fiber, and Fruit and Vegetable Intake at Baseline, Year 1, and Year 4 in the Control (C) and
Intervention (I) Groups in the Quality of Life Substudy
Baseline
Year 1
Year 4
Ca
Ib
Cc
Id
Ce
If
%
energy from fat
35.3 (0.5)
9.4 (0.2)
35.2 (0.5)
9.6 (0.2)
33.5 (0.5)
10.3 (0.3)
24.2 (0.5)
17.3 (0.4)
33.2 (0.5)
10.2 (0.3)
23.5 (0.5)
17.7 (0.4)
Fiber (g/1000 kcal/day)
Servings of fruits and
vegetables (per day)
3.7 (0.1)
3.9 (0.1)
3.9 (0.1)
5.9 (0.2)
4.3 (0.1)
6.3 (0.2)
Note. Table entries are means with standard errors in parentheses. Seven and 15 participants in QOL provided Food Frequency Questionnaire data at Year
1
and Year 4, respectively, but failed to complete the Quality of Life Factors Questionnaire.
n = 200. n = 194. n = 190. n = 180. n = 171. n = 169.
a
b
c
d
e
f
ported enjoying eating in restaurants or with family and friends
p = .012) and less often reported cost as a barrier to the purchase
of food (p < .001).
Overweight participants (body mass index > 25) and partic-
ipants with at least some college education had lower ratings on
the self-care domain compared with lower weight participants
Reported Dietary Change
(
As shown in Table 6, intervention participants reported suc-
cess in adopting the PPT eating plan by the end of Year 1 of the
intervention and in maintaining the changes throughout the 4
years of the trial. Control participants did not report substantial
changes in their diet. Because energy intake remained fairly
constant, the Year 1 change in one of these dietary outcome
measures was highly correlated with each of the remaining two.
The three pairwise correlations among these factors ranged be-
tween .5 and .7, which were each significantly different from
zero (p < .001). These results are comparable to those for the en-
(p = .012) and individuals with less education (p = .014).
Smokers also reported lower scores on the self-care domain than
nonsmokers (p < .001). Although non-Whites scored lower than
Whites on the social domain (p < .001), there were no other sig-
nificant differences in domain scores between White and
non-White racial groups.

Volume 23, Number 3, 2001
Self-Rated Quality of Life
203
tire PPT cohort, which have been discussed extensively else-
where (31).
differences in the change from baseline to Year 1 between inter-
vention and control participants for the taste, convenience, cost,
health assessment, and life satisfaction domains. At Year 1, the
change of –2.4 points from the baseline taste domain score in the
control group was significant (p = .006). For intervention partic-
ipants, the change of –1.7 points in taste score was not statisti-
cally significant. These changes in the taste domain were not
statistically different between the treatment groups. The change
of –2.7 in the cost domain score for the intervention group was
statistically significant (p = .010). However, the changes in cost
domain scores were not statistically different between the
groups. There were significant differences between the treat-
ment groups in baseline to Year 1 changes in scores for the
self-care, social, health belief, and health action domains. In the
self-care domain, both intervention and control participants re-
ported significantly (p < .002) higher ratings at Year 1 than at
baseline. The increase in the self-care domain score of 18.2
points in the intervention group was significantly higher than the
Changes at Year 1 in Domain Scores
Figure 1 presents the average changes in domain scores
from baseline to Year 1 for 363 control and intervention partici-
pants that completed both surveys. There were no significant
3.2-point increase in the control group (p < .001). The social do-
main was the only domain the intervention group rated signifi-
cantly lower than the control group after Year 1 on the PPT eat-
ing plan. The social domain score for intervention participants
decreased by 3.8 points in Year 1, and this decrease was signifi-
cantly different (p < .001) from the negligible decrease in the
mean score for control participants.
Consistency of Changes in Domain Scores
Over Follow-Up Years
Figure 2 shows the control and intervention mean domain
scores over the 4-year study period for those four domains show-
ing significant group differences at Year 1. To test the consistency
of changes in each domain over the entire follow-up period, we
FIGURE 1 Mean change in domain scores from baseline to Year 1 for
the control and intervention groups in the Quality of Life Substudy (lim-
its are 1 SEM).
FIGURE 2 Mean baseline through Year 4 domain scores for intervention and control participants in the Quality of Life Substudy (limits are 1 SEM).

2
04
Corle et al.
Annals of Behavioral Medicine
TABLE 7
Baseline Characteristics of QOL Participants With Complete
personal health; and satisfaction with food taste, food cost, and
the convenience of buying and preparing food.
(Baseline and Year 4 QF Questionnaire)
and Incomplete Data Points
Positive Influences of Eating Pattern
Changes on QOL
Completea
Incompleteb
Intervention (%)
Age (M)
48.6 (2.8)
60.8 (0.6)
63.1 (2.7)
12.3 (1.8)
67.1 (2.6)
52.2 (6.0)
61.6 (1.3)
60.9 (5.9)
7.2 (3.1)
In this study, people encouraged to adopt a low-fat, high-fi-
ber, and fruit- and vegetable-enriched eating pattern were as sat-
isfied with the taste and enjoyment of food and the ease of gro-
cery shopping, meal planning, and meal preparation as on their
previous diet when compared with a control group. Intervention
participants reported significantly and substantially greater con-
fidence in their ability to care for their health compared with
control group participants. Those in the intervention program
reported more awareness of nutrition and health messages from
health professionals and food labels, and they expressed greater
belief that food choices would improve health.
Men (%)
Minority (%)
c
More than high school
education (%)
Smoking (%)
65.2 (5.7)
10.2 (1.7)
27.5 (0.2)
18.8 (4.7)
27.4 (0.5)
Body mass index (M)
Note. No differences are statistically significant between completers
and noncompleters. Numbers in parentheses are standard errors. QOL =
Quality of Life; QF = Quality of Life factors.
a
b
n = 325. Incomplete are those not completing the Year 4 questionnaire;
c
n = 69. Minorities include African Americans, Hispanics, Indians/Native
The overall positive impact of the PPT eating plan is consis-
tent with the results of several other studies. Hyman, Flora,
Reynolds, Johannsson, and Farquhar (32) examined the effects
of dietary changes to improve cholesterol levels among commu-
nity participants with elevated cholesterol levels. General
well-being (e.g., self-rating of health, level of anxiety or depres-
sion, energy level, and enjoyment of foods eaten) did not change
among those who made dietary changes to reduce cholesterol
levels. In a 4-month dietary treatment program to lower blood
cholesterol, Ammerman et al. (7) found that the intervention
group showed substantial and significant improvement on a
self-care index relative to controls. This intervention group also
had slight but significant increases in scores for the taste, cost,
and convenience indices, compared with the control group.
Bowen, Kestin, McTiernan, Carrell, and Green (5) reported that
women who lowered dietary fat intake in a breast cancer preven-
tion clinical trial experienced more positive psychological func-
tion than the control group; by the Year 1 follow-up, the inter-
vention participants reported less anxiety, less depression, more
vigor, and more positive perceptions of current health and health
outlook than the control participants.
These results contrast with previous reports identifying po-
tential negative outcomes of cholesterol- and fat-lowering inter-
ventions. Earlier randomized trials provided evidence that low-
ering serum cholesterol levels in men may result in increased
rates of death by violent means, including suicides, homicides,
and accidents (3,16,17). In our study, both men and women re-
ceived a dietary-only intervention to reduce fat and increase di-
etary fiber, fruit, and vegetable intake. The intervention did not
focus on lowering cholesterol. Although we did not administer a
full battery of mental health scales, our data suggest that partici-
pants did not experience a negative impact on general feelings of
overall well-being and general satisfaction with life, family rela-
tionships, and leisure activities. The overall death rate and the
rate of deaths resulting from accidents and injury were compara-
ble between the PPT control and intervention groups. Our data
suggest that diet-based, fat-lowering interventions can be rec-
ommended without fear of negative mental health sequelae.
Our results also contrast with reported barriers to adopting
and maintaining a low-fat eating plan. In prior studies, people
Americans, Asians/Pacific Islanders, and others.
computed the change from baseline as each participant’s ob-
served value of score change at each of the four annual follow-up
visits. Because some of the 394 participants in the QOL Substudy
failed to complete a questionnaire at all time points, we used a
mixed model repeated-measures analysis procedure that permits
some missing data. To adjust for potential biases induced by
missing observations, we adjusted for three factors. Table 7 shows
characteristics of those 69 missing a Year 4 QF Questionnaire
compared to the remaining 325 in the QOL Substudy. Of the 69
missing, 15 died prior to their Year 4 visit. We adjusted on gender,
minority status, and smoking status because some imbalances in
these three factors are suggested. We entered the three adjustment
and group effects as indicator variables and follow-up time as a
linear trend. We also tested for an interaction between group and
time. The repeated-measures analysis indicated differences in the
effect due to group assignment for the self-care (p < .001), social
(p < .001), health action (p < .001), and health belief (p = .051)
domain scores. We noted a significant (p = .028) negative time ef-
fect for the health belief domain. A hint of an interaction between
group and time (p = .044) was seen for the health action domain.
For each of the four domain scores, a model based contrast to test
the group difference at Year 4 indicated each significantly differ-
ent from zero (p < .002).
DISCUSSION
Our findings suggest that a low-fat, high-fiber, and fruit-
and vegetable-enriched eating plan can be adopted without neg-
ative impact on overall perception of quality of life. In fact,
many positive changes in quality of life perceptions were re-
ported by QOL Substudy participants. The ongoing challenge of
maintaining the eating plan in social situations was the only re-
ported negative perception. At baseline and prior to intervention
counseling, we found significant associations between selected
demographic and lifestyle characteristics and self-perceived
overall health; enjoyment of the social aspects of food; care of

Volume 23, Number 3, 2001
Self-Rated Quality of Life
205
adopting low-fat eating patterns to reduce cardiovascular dis-
ease risk have reported that a low-fat diet does not taste as good
Of the 11 items within the social domain, 2 in particular
were affected by adoption of the PPT eating pattern. After Year
1, intervention participants were significantly (p < .001) more
likely to state that the way they ate was different from the eating
pattern they observed in others. This may represent an accurate
comparison between the intervention eating pattern and the di-
etary choices made by the average American. Intervention par-
ticipants were also significantly more likely (p < .001) to report
difficulty in eating with others as a result of their modified eat-
ing patterns. This suggests that they did experience some degree
of challenge in maintaining their intervention eating pattern
when dining with other people. However, responses to questions
about the intervention participant’s enjoyment in eating meals
with family, with friends, or at restaurants were not different
from those of controls.
(8,33,34), costs more (9,33), and makes food shopping and meal
preparation difficult (33,35–37). Contrary to these common per-
ceptions, cost and convenience ratings did not change signifi-
cantly during 4 years of participation in the PPT intervention
program. Intervention participants did not rate the overall cost
or difficulty in shopping for and preparing foods as significantly
different from their previous diet, and their ratings on these fac-
tors were not significantly different from those of control partic-
ipants. After each year of participation in the intervention pro-
gram, participants rated their satisfaction with the taste and
enjoyment of food as similar to their level of satisfaction with
their previous diet. At baseline, control participants rated taste
3
.8 points higher than did intervention participants (p = .005).
By recognizing that eating away from home and eating with
others are major social aspects of life, the PPT intervention may
have helped participants maintain their enjoyment of eating with
friends and family. Strategies for meeting the challenge of main-
taining the PPT eating plan while eating away from home and
during holidays and celebrations were integrated throughout the
We have no explanation for this difference. However, comparing
the change in taste scores from baseline to Year 1 (see Figure 1)
showed no difference in the amount of change between the inter-
vention and control groups. Also, the multivariate analysis
across all years of the study showed no significant difference in
taste scores between intervention and control participants. We
conclude that participation in the PPT intervention program did
not result in an overall decline in this quality of life domain.
We speculate that the standardized PPT nutrition interven-
tion program, which included more than 60 hr of counseling on
behavior modification techniques and nutrition skills building in
Year 1 (21), helped participants overcome some of the commonly
perceived barriers (e.g., cost, convenience, taste of food) to adopt-
ing a healthier diet. The frequent face-to-face and telephone con-
tact with PPT nutritionists were intended to provide social sup-
port for adopting and maintaining the eating plan and help
participants with problem solving. The counseling stressed
self-monitoring, positive self-talk, practical meal preparation, and
recipe modification. The cost savings of increasing the complex
carbohydrate content of the diet while decreasing fat were dis-
cussed. In addition, the emphasis on increased fruit, vegetable,
and fiber intake was intended to provide a more positive message
compared to interventions focusing mainly on reducing fat. We
also believe that the increased availability of a wide range of
low-fat food products in the marketplace (38) may have made
adopting the PPT eating pattern more convenient for participants.
4-year intervention. Approximately 25% of the written materials
were devoted specifically to developing participant skills for so-
cial situations such as eliciting support from family and friends,
planning ahead for social occasions, making appropriate restau-
rant menu selections, making and refusing requests, and meeting
the challenge of vacations and travel. However, the limited selec-
tion of healthier menu items at most restaurants was a frequent
complaint of participants and also of study nutritionists.
Demographic and Lifestyle Differences
in QF Domains
Our data suggest that self-perceived physical and emotional
well-being and satisfaction with diet and self-care vary consider-
ably among demographic groups and lifestyles. At baseline, older
participants, those who were not overweight, nonsmokers, and
those with less education reported being better able to care for
their health, compared with younger participants, those who were
overweight, smokers, and those with more education. We were
surprised by the inverse relation of education level to the self-care
domain. We speculate that participants with higher levels of edu-
cation score themselves lower on the self-care domain at baseline
because they have a greater knowledge of healthful habits but
may not have adopted them. Compared to men, women found it
easier to shop for and prepare meals, reported more concerns
about the cost of food, and reported greater enjoyment of meals
with family and friends. Married participants had higher assess-
ments of their overall health, life satisfaction, the taste of food,
and enjoyment of the social aspects of eating and were less con-
cerned about food cost than their single counterparts.
Negative Influences of Eating Pattern
Changes on QF
Several studies (8,9,33–36) have identified difficulty eating
away from home or dining with family and friends as barriers to
adopting and maintaining a low-fat diet. After Year 1 on the PPT
eating plan, intervention participants indicated that eating in so-
cial situations was more difficult on the PPT plan than on their
previous diet, and their ratings on this domain were more nega-
tive than those of control participants. The social domain section
of the QF Questionnaire included questions about eating with
family and friends and eating at holidays, parties, and restau-
rants. There were also questions about eating differently than
others and difficulty eating with others.
Other studies have found significant associations between
demographic and lifestyle characteristics and the adoption of a
healthier eating pattern. In a diet and exercise intervention, poor
compliance was linked with smoking and younger age (38).
There was a significant decline in dietary satisfaction among
men, but not women, with kidney disease following the adop-

2
06
Corle et al.
Annals of Behavioral Medicine
tion of a low-protein diet (10). Male gender and fewer than 13
years of education are documented risk factors for low fruit and
vegetable consumption (39,40). Responsibility for food selec-
tion, purchase, and preparation were related to both short- and
long-term success in adopting a low-fat diet in the Women’s
Health Trial (41).
These studies support the importance of addressing the re-
lations between demographic and lifestyle factors and attitudes
and beliefs about food and health early in the intervention pro-
cess. Targeted strategies should be developed to help people
with specific demographic characteristics, lifestyle characteris-
tics, or both to overcome barriers to adopting a healthier eating
pattern. For example, unmarried people and those who smoke or
are overweight may require additional customized intervention
counseling to address negative attitudes and health beliefs.
(3) Guyatt GH, Cook DJ: Health status, quality of life, and the indi-
vidual. Journal of the American Medical Association. 1994,
272:630–631.
(
(
(
4) Cummings P, Psaty BM: The association between cholesterol
and death from injury. Annals of Internal Medicine. 1994,
120:848–855.
5) Bowen DJ, Kestin M, McTiernan A, Carrell D, Green P: Effects
of dietary fat intervention on mental health in women. Cancer,
Epidemiology, Biomarkers and Prevention. 1995, 4:555–559.
6) Hatton DC, Haynes RB, Oparil S, et al.: Improved quality of
life in patients with generalized cardiovascular metabolic dis-
ease on a prepared diet. American Journal of Clinical Nutrition.
1996, 64:935–943.
(
7) Ammerman AS, DeVellis RF, Keyserling TC, Simpson Jr. RJ:
Quality of life is not adversely affected by a dietary intervention
to reduce cholesterol. Circulation. 1993, 87:19.
(
8) Wright CA: Preliminary survey of attitudes to, and factors af-
fecting perceived success or failure among adults attending a
hospital lipid clinic. Journal of Human Nutrition and Dietetics.
SUMMARY
1994, 7:153–159.
In this study, the dietary intervention incorporated nutrition
skills building, behavior modification techniques, self-monitor-
ing, and extensive professional counseling and support to help
participants overcome barriers to adopting and maintaining ma-
jor changes in their eating habits. Before enrollment and ran-
domization in this study, an eligible individual was required to
commit to making a major change in his or her usual eating hab-
its. In addition, the individual agreed to attend more than 60 hr
of individual and group counseling, keep extensive food intake
records, and participate in regular telephone counseling be-
tween in-person sessions. This considerable personal commit-
ment resulted in the successful adoption of a low-fat, high-fiber,
and high-fruit and vegetable eating plan and did not negatively
affect overall perceptions of quality of life. In fact, the enhance-
ment of several quality of life factors was documented.
Participants who adopted and maintained the PPT eating
plan for 4 years reported greater confidence in their ability to care
for their health, greater belief that food choices would improve
health, and more awareness of health and nutrition messages.
Contrary to common perceptions of low-fat diets, participants did
not report any detrimental effects of the eating plan on taste, cost,
the convenience of shopping for and preparing foods, their overall
health assessment and general well-being, or satisfaction with
life. Difficulty in eating in social situations was the only reported
detrimental impact of the eating plan. This finding highlights the
need to develop additional strategies to facilitate the maintenance
of a modified dietary pattern while eating with others or dining
out. These data provide evidence that the adoption of a low-fat,
high-fiber, and high-fruit and vegetable eating plan can be recom-
mended without fear of negative quality of life sequelae.
(
9) Barnes MS, Terry RD: Adherence to the cardiac diet: Attitudes
of patients after myocardial infarction. Journal of the American
Dietetic Association. 1991, 91:1435–1437.
(
10) Coyne T, Olson M, Bradham K, et al.: Dietary satisfaction cor-
related with adherence in the Modification of Diet in Renal Dis-
ease Study. Journal of the American Dietetic Association.
1995, 95:1301–1306.
(11) U.S. Department of Health and Human Services: Healthy Peo-
ple 2000: National Health Promotion and Disease Prevention
Objectives (DHHS Publication No. (PHS) 91–50213). Wash-
ington, DC: U.S. Government Printing Office, 1990.
(
12) U.S. Department of Agriculture and U.S. Department of Health
and Human Services: Nutrition and Your Health: Dietary Guide-
lines for Americans (4th Ed.). Home and Garden Bulletin No.
232. Washington, DC: U.S. Government Printing Office, 1995.
(
13) Dayton S, Pearce ML, Hashimoto S, et al.: A controlled clinical
trial of a diet high in unsaturated fat in preventing complications
of atherosclerosis. Circulation. 1969, 39–40(Suppl. II):1–63.
(14) Frick MH, Elo O, Haapa K, et al.: Helsinki Heart Study: Pri-
mary-prevention trial with gemfibrozil in middle-aged men
with dyslipidemia. Safety of treatment, changes in risk factors,
and incidence of coronary heart disease. New England Journal
of Medicine. 1987, 317:1237–1245.
15) Frantz Jr. ID, Dawson EA, Ashman PL, et al.: Test of effect of
lipid lowering by diet on cardiovascular risk. The Minnesota
Coronary Survey. Arteriosclerosis. 1989, 9:129–135.
16) Muldoon MF, Manuck SB, Matthews KA: Lowering choles-
terol concentrations and mortality: A quantitative review of pri-
mary prevention trials. British Medical Journal. 1990,
301:309–314.
(17) Pekkanen J, Nissinen A, Punsar S, Karvonen MJ: Serum cho-
lesterol and risk of accidental or violent death in a 25-year fol-
low-up. Archives of Internal Medicine. 1989, 149:1589–1591.
(
(
(
18) Farchi G, Menotti A, Conti S: Coronary risk factors and sur-
vival probability from coronary and other causes of death.
American Journal of Epidemiology. 1987, 126:400–408.
REFERENCES
(
1) Sanders C, Egger M, Donovan J, Tallon D, Frankel S: Re-
porting on quality of life in randomised controlled trials: Bib-
liographic study. British Medical Journal. 1998,
(19) Chen Z, Peto R, Collins R, et al.: Serum cholesterol concentra-
tion and coronary heart disease in population with low choles-
terol concentrations. British Medical Journal. 1991,
303:276–282.
(20) Schatzkin A, Lanza E, Freedman LS, et al.: The Polyp Preven-
tion Trial. I: Rationale, design, recruitment, and baseline partic-
317:1191–1194.
(
2) Gill TM, Feinstein AR: A critical appraisal of the quality of
quality-of-life measurements. Journal of the American Medical
Association. 1994, 272:619–626.

Volume 23, Number 3, 2001
Self-Rated Quality of Life
207
ipant characteristics. Cancer Epidemiology, Biomarkers and
Prevention. 1996, 5:375–383.
21) Lanza E, Schatzkin A, Ballard-Barbash R, et al.: The Polyp Pre-
vention Trial. II: Dietary intervention program and participant
baseline dietary characteristics. Cancer Epidemiology,
Biomarkers and Prevention. 1996, 5:385–392.
22) Ware Jr. JE, Sherbourne CD: The MOS 36-item Short Form
Health Survey (SF–36). I. Conceptual framework and item se-
lection. Medical Care. 1992, 30:473–483.
23) Wallston KA, Wallston BS, DeVellis R: Development of the
Multidimensional Health Locus of Control (MHCL) Scales.
Health Education Monographs. 1978, 6:161–170.
24) Dupuy HJ: The Psychological General Well-Being (PGWB) In-
dex. In Wenger NK, Mattson ME, Furberg CD, Elinson J (eds),
Assessment of Quality of Life in Clinical Trials of Cardiovascu-
lar Therapies. New York: Le Jacq, 1984, 170–183.
25) Bryant FB, Veroff J: The structure of psychological well-being:
A sociohistorical analysis. Journal of Personality and Social
Psychology. 1982, 43:653–673.
26) Maiman LA, Becker MH, Kirscht JP, Haefner DP, Drachman
RH: Scales for measuring health belief model dimensions: A test
of predictive value, internal consistency, and relationships among
beliefs. Health Education Monographs. 1977, 5:214–231.
27) Connell CM, Crawford CO: How people obtain their health in-
formation: A survey of two Pennsylvania counties. Public
Health Reports. 1988, 103:189–195.
28) Bravo G, Potvin L: Estimating the reliability of continuous
measures with Cronbach’s alpha or the intraclass correlation
coefficient: Toward the integration of two traditions. Journal of
Clinical Epidemiology. 1991, 44:381–390.
(38) Naslund GK, Fredrikson M, Hellenius M, de Faire U: Determi-
nants of compliance in men enrolled in a diet and exercise inter-
vention trial: A randomized, controlled study. Patient Educa-
tion and Counseling. 1996, 29:247–256.
(39) Subar AS, Heimendinger J, Krebs-Smith SM, et al.: 5-A-Day
for Better Health. A Baseline Study of Americans’ Fruit and
Vegetable Consumption. Bethesda, MD: National Cancer Insti-
tute, National Institutes of Health, 1990.
(40) Laforge RG, Greene GW, Prochaska JO: Psychosocial factors
influencing low fruit and vegetable consumption. Journal of
Behavioral Medicine. 1994, 17:361–374.
(41) Bowen DJ, Henry H, Burrows E, Anderson G, Henderson MH:
Influences of eating patterns on change to a low-fat diet. Jour-
nal of the American Dietetic Association. 1993, 93:1309–1311.
(
(
(
(
APPENDIX
(
The PPT Study Group also includes L. S. Freedman, C.
Clifford, and J. Tangrea of the National Cancer Institute; M. R.
Cooper, S. Quandt, C. DeGraffinreid, K. Bradham, L. Kent, M.
Self, D. Boyles, D. West, L. Martin, N. Taylor, E. Dickenson, P.
Kuhn, J. Harmon, I. Richardson, H. Lee, and E. Marceau of the
Bowman Gray School of Medicine; M. P. Lance, J. R. Marshall
(currently at the University of Arizona), D. Hayes, J. Phillips, N.
Petrelli, S. Shelton, E. Randall, A. Blake, L. Wodarski, M.
Deinzer, and R. Melton of the State University of New York at
Buffalo; F. L. Iber, P. Murphy, E. C. Bote, L. Brandt-Whitting-
ton, N. Haroon, N. Kazi, M. A. Moore, S. B. Orloff, W. J.
Ottosen, M. Patel, R. L. Rothschild, M. Ryan, J. M. Sullivan,
and A. Verma of the Edward Hines Jr. Hospital, Veterans Ad-
ministration Medical Center; B. Caan, J. V. Selby, G. Friedman,
M. Lawson, G. Taff, D. Snow, M. Belfay, M. Schoenberger, K.
Sampel, T. Giboney, and M. Randel of the Kaiser Foundation
Research Institute; M. Shike, S. Winawer, A. Bloch, J. Mayer,
R. Morse, L. Latkany, D. D’Amato, A. Schaffer, and L. Cohen
of the Memorial Sloan-Kettering Cancer Center; J. Weissfeld,
R. Schoen, R. R. Schade, L. Kuller, B. Gahagan, A. Caggiula, C.
Lucas, S. Pappert, R. Robinson, V. Landis, S. Misko, and L.
Search of the University of Pittsburgh; R. W. Burt, M. Slattery,
N. Viscofsky, J. Benson, J. Neilson, R. McDivitt, M. Briley, K.
Heinrich, and W. Samowitz of the University of Utah; J. W.
Kikendall, R. Wong, E. Stoute, V. Jones-Miskovsky, A. Greaser,
S. Hancock, and S. Chandler of the Walter Reed Army Medical
Center; and M. Hasson, B. Brewer, T. Zimmerman, B. O’Brien,
L. Cranston, N. Odaka, K. Umbel, J. Pinsky, H. Price, and A.
Slonim of the Data and Nutrition Coordinating Center (Westat).
The central pathologists for the trial are K. Lewin (University of
California, Los Angeles) and H. Appelman (University of Mich-
igan). The PPT laboratory scientists are P. S. Bachorik and K.
Lovejoy (Johns Hopkins University) and A. Sowell (Centers for
Disease Control and Prevention). Members of the PPT Data and
Safety Monitoring Committee are E. R. Greenberg, chair
(
(
(
(
29) Block G, Woods M, Potosky A, Clifford C: Validation of a
self-administered diet history questionnaire using multiple diet
records. Journal of Clinical Epidemiology. 1990, 43:1327–1335.
30) National Cancer Institute: Health Habits and History Question-
naire: Diet History and Other Risk Factors Personal Computer
System. Bethesda, MD: Division of Cancer Prevention and
Control, National Cancer Institute, 1989.
(
(
31) Lanza E, Schatzkin A, Daston C, et al.: Implementation of a
4-year high fiber, high fruit/vegetable, low fat dietary interven-
tion: Results of dietary changes in the Polyp Prevention Trial.
American Journal of Clinical Nutrition (in press).
(
32) Hyman DJ, Flora JA, Reynolds KD, Johannsson MB, Farquhar
JW: The impact of public cholesterol screening on diet, general
well-being, and physician referral. American Journal of Pre-
ventive Medicine. 1994, 7:268–272.
33) Lloyd HM, Paisley CM, Mela DJ: Barriers to the adoption of re-
duced-fat diets in a UK population. Journal of the American Di-
etetic Association. 1995, 95:316–322.
34) Koikkalainen M, Lappalainen R, Mykkanen H: Why cardiac
patients do not follow the nutritionist’s advice: Barriers in nutri-
tional advice perceived in rehabilitation. Disability and Reha-
bilitation. 1996, 18:619–623.
35) Jeffery RW, French SA, Schmid TL: Attributions for dietary
failures: Problems reported by participants in the Hypertension
Prevention Trial. Health Psychology. 1990, 9:315–329.
36) Baird PC, Schutz HG: Life style correlates of dietary and bio-
chemical measures of nutrition. Journal of the American Di-
etetic Association. 1980, 76:228–235.
(
(
(
(
(
(
(
Dartmouth University); E. Feldman (Augusta, GA); C. Garza
Cornell University); R. Summers (University of Iowa); S.
Weiand (through June 1995, University of Minnesota); and D.
DeMets (beginning June 1995, University of Wisconsin).
37) Montgomery DA, Amos RJ: Nutrition information needs dur-
ing cardiac rehabilitation: Perceptions of the cardiac patient and
spouse. Journal of the American Dietetic Association. 1991,
91:1078–1083.