 

.1‘

 

44.“. $132.44!
5.. 5

lo]. #3..
x .

 

 

 

 

 

 

 

 

. #75:}. A .
J56: S v)
1.1.1. I (n ff...
EIQv 435:}: ft

.. lot: .2; f

...,. .

 

($.53

. [.1
:5: 3 I

-l?.¢3-§ . . . OJ, 1". 13"):ng til-A

52.55. ..F, . '4:

E§I$Itz

33.33:

‘15:...55

‘1'...

 

 

an. 4 . .3. . I, Eye,
Auxirl ‘ 354 :I (
..$ax..<ca).:.:.v.od).
v1; ...t:b>..$abf..!; v! ‘4
‘ .: tuthvs .29 a 1.3 $5.“): 515.05“
.191. IiiiQ 1...: .1 1: (vii an £0.31
.5133... . is 4 :.vlib\.Jx\rl_,.

(3’

{.2}:
.. £2.55. : a
O .
Egg; .
...S....... .
3.1.77
H :3 ,(rSi...
'I..¢IX\»
Six:

It‘ll-ii

.v) a... 1.}! \
aavrii... 3:1:

.:y.{>..:: 4:) .0.)
d...-.\.r...,.un.a... 3
s . w

v): I)... :15"... m

([3 '1 ; I

11111.9... .3:
((135. PK)

{#3
5.1.. . ,

’ «a . i:
liiikjlx.

53:3!) 4a.
.53\li;lx
it"! .11

3153/
1H.-4I

.54....11

 

 

 

 

 

//I!rta\.!14:

 

\ \ ‘rlvn

l THESiS

lllllﬁlllllﬁlllllll Till

3 1293 O 6 9976

 

 

 

 

 

 

 

 

 

 

 

 

 

 

\I

 

This is to certify that the

dissertation entitled

Predictors associated with dietary
modification after a worksite nutrition
program

presented by
Ya—Li Huang

has been accepted towards fulﬁllment
of the requirements for

DOCtOI‘ Jegree in Wition

 

Maj pr sor

Dme 8/22/97

 

MSU is an Afﬁrmative Action/Equal Opportunity Institution 0-12771

 

“LIBRARY

Mlchlgan State

Unlverslty

 

PLACE IN RETURN BOX
to remove this checkout from your record.
TO AVOID FINES return on or before date due.

 

DATE DUE

DATE DUE

DATE DUE

 

.I

L

 

 

 

 

 

 

 

 

 

 

 

 

 

 

1/98 cychlC/DateDuepGS-pJ 4

PREDICTORS ASSOCIATED WITH DIETARY MODIFICATION AFTER A
WORKSITE NUTRITION PROGRAM

By

Ya-Li Huang

A DISSERTATION
Submitted to
Michigan State University
in partial fulﬁlhnent of the requirements
for the degree of

DOCTOR OF PHILOSOPHY

Department of Food Science and Human Nutrition

1997

ABSTRACT

PREDICTORS ASSOCIATED WITH DIETARY MODIFICATION AFTER A
WORKSITE NUTRITION PROGRAM

By

Ya-Li Huang

Nutrition programs for employee health promotion have become common in the
past 10 years. A better understanding of individual determinants of dietary behavior and
behavioral changes is necessary for the development of effective nutrition programs. The
purpose of this study was 1) to identify predictors for changes in dietary fat intake at the
end and at 4-month follow-up of the worksite nutrition programs; 2) to compare those
who dropped out vs. completed the programs; and 3) to characterize people at each stage
of change for reducing fat intake and at each stage of change for increasing ﬁber intake
by dietary intake, blood lipids, body mass index (BMI), and psychosocial factors
surveyed at the time of enrollment. A total of 128 women was recruited from 10-week
worksite nutrition programs between 1994 and 1996. Dietary fat intake (% kcal) and
motivation at enrollment were signiﬁcant predictors for reduction of fat intake
immediately after the nutrition program whereas initial BM], previous weight loss
experiences and perceived beneﬁts of a healthy diet at enrollment as well as dietary fat
intake (% kcal) and motivation were the predictors at the 4-month follow-up. Dropouts
(n=40) were more likely to eat > 30% of kcal from fat (69% vs. 50%) and to be
overweight (67 % vs. 40%) than those who completed the programs. Blood lipids and

psychosocial factors did not diﬁer between the two groups. The majority of subjects were

already in Preparation and Action stages of change for fat and stages of change for ﬁber.
Compared to other stages of change, Maintenance stage had 1) lower fat and higher ﬁber
intake; 2) higher perceived beneﬁts of a healthy diet and motivation; 3) lower perceived
barriers to a healthy diet. Innovative strategies are also needed to retain those who drop
out from the programs as well as to advance the majority of the participants who enrolled
in the program from the Action stage to the Maintenance stage. Worksite nutrition
programs designed to increase motivation, the perceived beneﬁts of a healthy diet and the
prevention of relapse from previous weight loss experiences are needed to help

participants adopt more healthﬁJl diets.

ACKNOWLEDGMENTS

I would like to acknowledge my major advisor, Dr. Won Song, for bring my
potential to reality. In addition to her scholarship, I have beneﬁted immensely from her
professionalism, perseverance, and, boundless patience towards me. I also want to extend
a very special thank to my guidance committee members: Dr. Wanda Chenoweth, Dr.
Sharon Hoerr,and Dr. Larry Hembroff. They each have contributed to my present
research in a unique way. Thanks also go to my lab mates: Lydia Koemer, Prodromou
Prodromos, Janet Lawrence, and Saori Obayashi, for their friendship and support.

All my accomplishments, especially this dissertation, are the fruits of my parents
unconditional love and support. Finally, I want to thank my dear husband, Chengchang
Huang, whose love, friendship, and on-site, life-time PC consulting service are

always there whenever I need them.

iv

TABLE OF CONTENTS

List of Tables ..... vi
List of Figures ....viii
Chapter One Introduction ..... 1

Chapter Two Review of Literature

A. Worksite nutrition programs ...... 5
B. Relationship between various psychosocial factors and dietary
behaviors ..... 14
C. Application of stages of change model to dietary change ..... 19
D. High blood cholesterol and overweight ..... 23

Chapter Three Change in dietary fat intake of women after a worksite nutrition
programs predicted by blood lipids, BMI, dietary intake, stages of

change and various psychosocial factors ..... 26
A. Abstract ..... 26
B. Introduction ..... 27
C. Methods ..... 29
D. Results ..... 34
E. Discussion ..... 37

Chapter Four Dropouts from worksite nutrition programs differ from those who
completed in dietary intake and BMI but not in blood lipids and

various psychosocial factors. ..... 48
A. Abstract ..... 48
B. Introduction ..... 49
C. Methods ..... 50
D. Results ..... 55
E. Discussion ..... 56

Chapter Five Stages of change are associated with dietary intake, BMI and various
psychosocial factors but not with blood lipids among women in worksite

nutrition program ..... 63

A. Abstract ..... 63

B. Introduction ..... 64

C. Methods ..... 66

D. Results ..... 70

E. Discussion ..... 72

Chapter Six Conclusions and implications ..... 83
Chapter Seven Recommendations for future studies ..... 86
Appendices Appendix A Outline of the worksite nutrition program ..... 88
Appendix B UCRIHS Approval ..... 90

Appendix C Informed consent ..... 91

Appendix D Three-day diet record form ..... 92

Appendix E Psychosocial factors questionnaire ..... 98

Appendix F Stages of change fot fat: algorithm ....102

Appendix G Feedback on dietary intake and blood lipids ....103

Bibliography ....107

vi

LIST OF TABLES

Chapter Two
Table 1. Summary of worksite nutrition programs reported in the literature

Table 2. Previous studies reporting the association between various psychosocial
factors and dietary behaviors

Table 3. Previous studies of stages of change model on dietary behaviors

Chapter Three

Table 1. Dietary intake, blood lipids, BMI, and psychosocial factors of
subjects (n=65) at enrollment of the worksite nutrition program

Table 2. Changes in dietary intake, serum cholesterol levels, BMI of subjects at
the end and at 4-month follow-up of the worksite nutrition programs

Table 3. Backward multiple regression models for reducing fat intakes at the end
of the worksite nutrition programs and at the 4-month follow-up

Table 4. Dietary intake, blood lipids, BMI and psychosocial factors (collected at
the time of the enrollment) of beneﬁt and no-beneﬁt groups based on
their dietary changes at the end and at 4-month follow-up of the worksite
nutrition program

Chapter Four

Table 1. Dietary intake, blood lipids, BMI of subjects in dropout vs.
completion groups of worksite nutrition programs

Table 2. Psychosocial factors of subjects in dropout vs. completion groups of the
worksite nutrition programs

vii

..... 16

..... 22

..... 42

..... 43

..... 44

..... 45

..... 59

..... 6O

Chapter Five

Table 1.

Table 2.

Table 3.

Table 4.

Dietary intake, blood lipids, anthropometrics of subjects at various
stages of change for reducing fat intake

Psychosocial factors at various stages of change for reducing fat intake

Dietary intake, blood lipids, anthropometrics of subjects at various
stages of change for increasing ﬁber intake

Psychosocial factors at various stages of change for increasing ﬁber
intake

viii

..... 77

..... 79

..... 8O

..... 82

Chapter One

Introduction

Nutrition programs have been used to promote weight control, cholesterol
reduction and cancer prevention (Glanz and Seewald-Klein 1986; Kris-Etherton et al.
1988). Worksites are considered as an important channel to deliver these health
promotion programs because they provide access to over 60% of adults of varying ages
and health status (US. Department of Labor 1992). The percent of worksites that offer
nutrition programs increased sharply from 48% to 78% for large worksites (more than
750 employees), and from 9% to 22% for small worksites (50-100 employees) between
1985-1992 (USDHHS, 1992).

A growing number of studies have provided evidence that worksite nutrition
programs can be effective (Pelletier 1993; Haus et a1. 1994; Sorensen et a1. 1992; Briley
et a1. 1992; Hunt et a1. 1993; Gorbach et al. 1990; Masur-Levy et al. 1990). However,
most nutrition program are short, less than 6 months in duration and have not addressed
the challenge of how to support the maintenance of modiﬁed dietary behaviors.
Phenomena such as weight cycling demonstrate the questionable effectiveness of
intervention programs in achieving long-term modiﬁcations even though short-term

effects are encouraging. Health and medical researchers interested in reducing dietary

2
risks for chronic diseases have attempted to the identify fundamental factors which
inﬂuence modiﬁcation of dietary behaviors. Once those factors are identiﬁed, effective
program strategies can be developed to improve nutrition programs.

Another problem is that the majority of the ﬁndings on the eﬁcacy of worksite
nutrition are based solely on data of those who completed the programs in reference to
either baseline data of the same group or of a control group. High dropout rates (12-3 5%)
have been reported as one of the programmatic problems in worksite nutrition programs
(Bruno et al. 1983; Lovibond et a1. 1996; Carmody et al. 1980; Dishman 1988). Little is
known, however, about the characteristics of those who were apparently motivated to
enroll but discontinued participation in the program.

In order to improve the effectiveness of nutrition programs, many theories and models
have been used to explain an individual’s dietary behavior. The Stages of Change model has
been successﬁrlly used in helping people quit smoking and has only recently been tested in
relation to dietary change. The Stages of Change Model (Prochaska and DiClemente 1983)
proposes that change occurs through a series of stages: Precontemplation (unaware or not
thinking about changing), Contemplation (seriously thinking about making a change),
Preparation (making deﬁnite plans to change), Action (actively modifying an unhealthy
behavior) and Maintenance (maintaining the new behavior for some time). Currently, the
majority of nutrition interventions are “action-oriented” and provide skills and strategies for
people who are ready for action to change behavior. It appears, therefore, beneﬁcial to know
the stages of dietary change for people who participate in an nutrition program before an

intervention is introduced. Previous studies (Glanz et al.1994; Greene et a1. 1995) reported in

3

previous population-based studies that about 65% of people were in Action or Maintenance
stages of dietary change for fat. It is not clear about the distribution of the stages of dietary
change for people who were recruited into nutrition programs. In order to use the stages of
change model effectively in developing nutrition interventions, it is also important to
understand the characteristics of each stage of change.

Many studies have found that various psychosocial factors such as belief in diet-
disease connection, self-eﬂicacy, attitude, and nutrition knowledge are associated with
healthﬁrl diet behaviors (Patterson et a1. 1996; Shepherd and Stockley 1987; Smith et a1.
1995; Glanz et al. 1993; Brug et al. 1994; Kristal et a1. 1995; Laforge et al. 1994). Most of
these studies are cross-sectional in design and as such they cannot establish if psychosocial
factors are associated with dietary changes after intervention studies or further our
understanding of how these occur. Predictors of weight change after an intervention have
been reported (Haus et al. 1994; French et a1. 1994), but predictors of maintenance of weight
loss are not likely the same as those for dietary modiﬁcations. We know little about the
predictors of dietary modiﬁcations after the completion of nutrition intervention programs.
Additionally, high serum cholesterol and overweight are important risk factors for chronic
diseases. We wanted to know whether these risk factors are related to dietary changes in
worksite nutrition programs.

One approach to developing effective strategies for dietary changes is to identify the
predictive factors associated with dietary change. Once we can identify the factors that most
strongly inﬂuence people to select healthful diets, we can more rationally design worksite

nutrition programs. Therefore, the objectives of this study were:

4
To examine whether various psychosocial factors, the stages of change for reducing
fat intake, blood lipids (total cholesterol, LDL—C/HDL-C), dietary intake and BMI at
the time of enrollment can predict the dietary changes at the end and at a 4-month
follow-up of the worksite nutrition program. (Chapter Three)
To identify predictors which can differentiate participants who were more successful
at changing fat intake than those who were not at the end and at a 4-month follow-up
of the program. (Chapter Three)
To examine whether participants who dropped out from worksite nutrition programs
differ from those who completed the programs with respect to dietary intake, blood
lipid proﬁle, BMI, stages of change for fat and ﬁber, and various psychosocial
factors. (Chapter Four)
To identify the stages of change for reducing fat intake and the stages of change for
increasing ﬁber intake of those who participate in worksite nutrition programs.
(Chapter Five)
To examine the stages of change for reducing fat intake and the stages of change for
increasing ﬁber intake in relation to dietary intake, blood lipid proﬁles, BMI,

waist/hip ratio and various psychosocial characteristics. (Chapter Five)

Chapter Two

Review of Literature

A. Worksite nutrition Programs

Nutrition programs for employee health promotion have become common in the
past 10 years. Programs have primarily addressed weight control and cholesterol
reduction because of their link with obesity, hypertension, increased risk of
cardiovascular disease (Glanz et a1. 1986). Dietary intake, food selection, and dietary
behavior are paramount in the prevention of these risk factors and related diseases. (Kris-
Etherton et a1. 1988). Worksites are an important channel to deliver these health
promotion programs because they provide access to over 60% of adults of varying ages
and health status (U .S. Department of Labor 1992). The workplace provides many
opportunities for reinforcement and environmental support for health promotion
behaviors.

The success of worksite nutrition programs has been measured by the
improvement in blood lipid levels, dietary intakes, and weight control.
Previous studies indicated that the reduction of fat intake or an increase of ﬁber intake
can be implemented in a free-living population after the nutrition programs (Barratt et a1.

1994; Baer 1993; Hebert et a1. 1993; Briley et a1. 1992; Masur-Levy et a1. 1990). Weight

6

loss along with a decreased dietary fat intake were reported by most of the weight control
and cancer prevention programs (Fisher and ﬁsher 1995; Haus et a1. 1994; Johnston et a1.
1994; Harris et al.1994; Shannon et al. 1987). A decreased total serum cholesterol level
and LDL-C or improved ratio of total cholesterol to HDL-C were observed in the
participants with elevated total serum cholesterol level in most of the cholesterol
reduction programs (Perovich and Sandoval 1995; Byers et a1. 1995; Fisher and Fisher
1995; Angotti and Levine 1994; Johnston et a1. 1994; Baer 1993; Hebert et a1. 1993;
Briley et a1. 1992; Shannon et a1. 1987). However, most studies have focused on the
short-term evaluation of interventions (6 to 12 weeks) and their results may not be valid
or predictive for the long-term success of intervention. The health beneﬁts are likely to be
sustained only if positive dietary change is made and maintained for a long term. The few
studies which included follow-up surveys have reported limited success in maintenance
after interventions (Haus et a1. 1994; Harris et a1. 1994; Sorensen et a1 1996). In fact, a
high recidivism rate had been found in several weight maintenance studies (Foster et a1.
1988; Hovell et a1. 1988; Wooley et a1. 1991; Hans et a1. 1994).

Another problem is that average values for dietary fat, blood lipid levels and body
weight were reported to show the impact of the dietary intervention (Table 1). From such
aggregate data, we cannot tell whether everyone made the similar improvements on
dietary intake after participating nutrition interventions. If not, who are those individuals
with signiﬁcant dietary change after intervention and how are they different from those
who did not make signiﬁcant dietary changes. The third problem for worksite nutrition

programs is that the efﬁcacy of programs is based solely on data of those who completed

ouﬁqqoumosw >053on coon ”Ohm

 

 

m .VA 0
Be as new- ...52_ $38 -nesoeasoe
335 565 a com 030.3%: Seewoa 830m 38
Emmet,» boom 89835 6% @3086 $5088
930% ﬂop mean @005 .53 a 8388 ”m @580 Nm A #803,205
3 gm 28 #88835 838 ”$3302 628:me $6 88% 38 \B 53
:39 womanhoov baawomﬁmmm ”58-339805 £83 S 9sz 3388 ”4 @580 KEN: .3 Ho 8833
eases
E 3233 Son
.Qsasozﬁ an 350% Eon may: 0: ”Own - -28 3 8.58 83.52“
How 0.4% no Became—one 93:: .9305 a 3388 ”m 3on
820m :38 S omega 0 Z BR: @005 3.30:8 p.93: 3:8: .5an Wm A
”3.30:8 3 358w Son E “8:582 £8086 a com Boom: owmxoa 20: 359.865 883 33
38.8 «ﬂ @806 33883 sacrumombmowoum ”$003 w £3 a 8388 H< 965 :38 \B mwou: .E 6 ﬁwbem
0.43 532 ea: coca 82
use one Dina—$88865 “8&on SEEKS”: «$898 8304
Sow 33 vegan: 383833805 9.33 m 28 Swap 39950 m Zia EB gown/w
2.3..on 5:262 Shame—2.0
3:582 qouﬁﬁﬁﬁ 30>
338m 28 $609 mo ﬁmaoq 898285 madam 28 853‘

8388: on“ E wotoqoc 38%on cognac 0:38? we Saddam A 033.

 

.3088
NmV 8.3888 833
E8. 5.3 $890.58 8 8..

38mm. .58:

3.9886 .8 8.838.

Mm... 8,. 88888

 

:5 {5088 N.mA 8.8888 Ems... 38m .8832: 8.885
88% H8. .23 3:88....8 88: coo—m .838st 883230 33
8 8.8888 83.3. ”8.382 38.... Z on. :8... 362.3
E8. 38802. €825..me 58-608880... 8.83 v 8.502% on. 8 Baum 0mm": Ba 8308..
8032.3 .3.
88088680
53:85.86 88: 82m 88$. 88038.28
3988.8 8388 8388 @005 $82.88 8 maﬁa»...
.803 8.. 58% 8.8888 Emmy,» boom .8 Mm... on. 838. com. .8 8
838 R8. 8 woman“. 02 38.88.5885 8895 m 8 8223.5 vowQSOQdm 2N”: $04.82.:
8 case
.88 68082. 8 3.68..
ONES 88.8 mauam maﬁa on. .8 8330.. .5088
was 8.8888 833 E8. 8 88: .805 833 omega 8333 NmA 8.8.88.8
3886888. 8 8.20.. 0.83 ”8.382 08:88 8 88 083 83% 38:3 83
@8333 madam. E momawao 58-32388... 383 w mowemmoa Soﬁamﬁﬁm mvwn: .1. 8 588E
2.8..on 8.8.69. 8.88225
8.5982 8.89.8.3 80>
338m 2... ammmon ,8 583 8825.3 088.8 E... 883..

9.809 0.8.88: 08. a. 88.88.. magma... 82.52. 88.83 .8 553m A 033.

9

 

 

082809 8-30:8 .0088 88> 0888 N03
8806 080 0800—308 588% -2 03H 35—00: 800E: 008085
80888 80888 “m0ﬁmombm000i ”8.003 N 8080.8 098 0005 mm mun 080 08005
88885800 0380m0>
00 :03 00 0888 0808000 0808 090%.
88 < 8808.» 8000088 0888 8805
.0888 “0080002 883580800 @000
8098 8 008080.88 0: M0888 8880 83 88000800 .82
08 880% 8. 00m00000Q .m> 88803008 8808880 00085 ooomua .3 00 «800$
008080 880% 8%
8:00:00 8888: 0808
88% 088— 0005 N $888800 00888
8880 08 88 {em .0> 832 ”0080002 m 0080008 83 880000880
880000880 8 88 $00 808800 0003 8% 0080080 33
0 88 890% 88828008 .m> 88802008 8003 v 0803 0808883 noowﬂa .3 00 89am
080.80.:— 803262 88800—080
0080002 08803008 80>
88008 080 08009 .8 5808 88803008 088.8 000 8885.

90.803 085800: 08 8 “0008808 080808 808.88 08883 M0 @0885 ._ 050B

10

 

888388

 

Bad bawomﬂma
8888c 3303 $09 .3808
83 388838 888 80H EwEB buom N >88 wamomasoo
“ma 8m 8 ma m8 :8 2%: 255 3:22? .853
888028 688208 S85 8288 E85 897m 25 8388
”$3 8 $9 808 @8883 M8882 €88 EH 8038 N09
Gum Bob Rom Xv 8mm ~08on 388-3338on H8» H .2808 0 88m 08 mg mm“: .3 8 homum
.9.on 68:8 8-30:8
8 mowamao 0 Z .5880 28% @005 88mm 28 £8888
$8 888qu .88 309 28 5 30m 3on #880888
.5803 309 .mowtooﬁmg Ems? bucm $8 8383 88—8888
488838 @003 5850 how new 88:me
~88 8888a 88% 88% 8883 @805 58% 33838:
88 88833088 mo 0483 8880382 4826.588 88.88
88808 was 3.8838 28 #0880 53» 8.9%on 888:8
mote—8 mo 888 88889 .m> 888388 88% _ 828:8 o>a8n8maoo mm": 33 80mm
852ch 38:3 239$ 83 55860.. 3.88.2.0
8582 888285 H8>
838d 23 awwmom mo 8&qu 888283 oaawm 98 853w

9,888 8888: 05 E 88.8%: maimoa 8088:: 86MB? mo 583m A 2an

11

 

00.3008 80 0.80? 0380.8
080800. 5.3 080.8000

088 00 0-3-

 

000080800 800 0%.... 00.08. 00805
80880 5.3000 30.0000
08 0800 0.80.0088. 008 800000 .0.00m
00:50:00 80 0.803 0800.8 0088. 0.803
200 00 =0 00800 008 00303
n00.30.80 000 mu. N 8.3 880. 0.803 >00m. 00.3000.
0003 08 800080 $0082 0058 00-0 09:
8000.0.800 08 00 00.5 000 0088008880 .8808 0 80880 .8800 0.803 0Nn0 A0 80 0000.
003000..
8.88 000.800 00 00008008.
08 00038 008.80
0. 8080>808. 08008 m88. cm ”0.00 -
00.50% 008m 00.80880. 00.08. @0805 0.0038088
08 0008 .8800 0.803 000.0 8008.3. . .8800
08 8 0800800 0.8800 ”080080,. 0000800 3 08000000
000 0. 0.0 8 0800800 ”.8800 80880 .8800 30.
n8. Wm 003 80. 0.803 0003. .m> 00.80880. 8.003 0 880B 000 000.507. mhu0 .0 00 000000m
080.880 .8800 8.803
08882 0080880. 08>
8.08.". 000 088D 00 0800A 00.000880. 0.080m 000 008040

A0800. 800008.. 08 0. 08.0008 80880 000.800 8.8.003 00 @0885 .. 0.000.

12

 

00.00.00 .008 0. 0080000

008 088. gm. 8.080

8 080 030 .x. a $0500
808080800 030008003 00
0000008.. 0. 080000 8 080.8
0.0808. 003 080000 Em

8.080.

808 0320

0000. 808 000m
”0800000,. 000.
008.800 8880 08008 8. 80880 8.8080 000 00R. 0m .n0 ..0 00 0.80...

080880 .8800 2.8.03
080000.). 00.80280.
000 8.009 00 08004 00.80880. 0.080m 000 008000.

A0008. 88008.. 08 0. 080008 80880 00.0800 8.8.003 00 0.00885 .. 0.00H

13

the nutrition programs in reference to either baseline data of the same group or a control
group (Table 1). The forth problem is that high dropout rates (12-3 5%) have been
reported as one of the programmatic problem in worksite nutrition programs (Bruno et
a1. 1983; Livibond et a1. 1996; Dishman 1988; Carmody 1980). However, there is a lack
of research to characterize the people who joined these programs but dropped out during
the programs. These problems have made it difﬁcult to draw conclusions about the
efﬁcacy of worksite nutrition programs.

In order to understand more about the those problems such as predictors of dietary
change and characteristics of dropouts from nutrition program, we reviewed the previous
studies and summarized the variables which have been associated with dietary behaviors.
(Chapter two: B. Relationship between psychosocial factors and dietary behaviors; C.
Application of the stages of change model to dietary change; and D. High blood

cholesterol and overweight)

14

B. Relationship between psychosocial factors and dietary behaviors

Nutrition programs to improve health through promotion of desirable eating
patterns are more likely to be effective when based on an understanding of factors
inﬂuencing dietary behaviors. A number of theoretical model such as Social Learning
Theory (Bandura 1986), the Reasoned Action theory (Ajzen and Fishbein 1980), and the
Health Belief Model (Rosenstock 1983) have been used to investigate the association
between psychosocial factors and dietary behaviors. Previous research had addressed how
each model relates to healthful eating (Dittus et a1. 1995; Fenini et a1. 1994; Smith and
Owen 1992). Several studies have also tested combinations of variables based on more
than one model (Glanz et a1. 1993; Kristal et a1. 1995; Patterson et a1. 1995; Contento and
Murphy 1990). A summary of the association between various psychosocial factors and
dietary behaviors is presented in Table 2.

Key elements of these theoretical models are belief about diet and health,
perceived beneﬁts, perceived barriers, self-efﬁcacy, social support, nutrition knowledge,
and perceived norms (Table 2 ). Overall, some consistency is found in the ﬁndings of this
diverse research, in that belief in a strong relationship between diet and health and
knowledge of diet recommendation are associated with healthful dietary behaviors
(Patterson et a1. 1995; Contento and Murphy 1990; Smith and Owen 1992; Glanz et al.
1993) . There is less consistency about social support and perceived norms inﬂuencing
dietary habits (Glanz et a1. 1993; Patterson et a1. 1995; Shepherd and Stockley, 1987).

Most of these observations between psychosocial factors and dietary behaviors

are, however, based on cross-sectional data. Therefore, it is not entirely clear whether

15

psychosocial factors are associated with dietary change after nutrition interventions.
Attempts to identify indicators for predicting success in weight loss have been made
(Haus et a1. 1994; French et a1. 1994; Klesges et a1. 1992), but not for dietary
modiﬁcation. Demographic characteristics, dietary composition, social support, and
weight cycling history had been found as important predictors for weight control aﬁer
interventions (Haus et a1. 1994; French et a1. 1994; Klesges et a1. 1992). Predictors of
weight control, however, are not necessarily the same as those for dietary change. To our
knowledge, no previous studies have reported on the predictors of dietary modiﬁcation
after completion of a nutrition program. Since dietary habits pose a health risk factor to
many health conditions, identiﬁcation of predictors of successﬁil change in dietary habits

is important.

16

0000005000 50000000 000 m ”0mm

 

0800 00 000
8000080000 @805

000000 8 3003000005
00388 038%0> 000 00.0 8 80005

 

8030003 0088 038%0> 000 00.0 .80 80000m
00000000000 038%0> 000 00.0 805 0880080 “80080000 Eoomonoxmm 33
0003 0000800000 038%0> 000 030 8 80.5.3 08000 0000000080000 003n0 .00 00 0085
00800000000 88 w03008 00 00000 000000000 @800 m3 0%
8000 00.0000 00 0003 00.000 0 03 0300 8000000080 @805
00800 08 38000 030 0.000 .00 0000000000000: 0000 0%0000 8 000083
000003 w00000 00—88 88 E880 0880:0m
00000000000 08 000000 8 000.080 .00 880000 “80080000 0000000005 .002
0000000me 0 003 00—88 08 @800 088.3% 08000 300000000000 SmJH0 .00 00 mam
800% m 0,000,000 000 00>0
0%0000 880 300.300 0080000$0m
8000000080 5805
0000 Z
0%0000 080 0800000 0%003300M
003 08800000 0&0080 0003 00:00 00200000 m0=0m
00—88 080 00080000 0 003 08800000 80080000 Eoomonoamm 32
300005080 0003 %0005000 000 m0=0m 08000 0000000000000 «sq—H0 .00. 00 008803
8050a 00000002 00.0 08009 2ng 80> 000 8053‘

8030003 E880 00.0 8808 3000000me 000.08, 0003003 000800000 030 wan—00000 08030 000305 .N 0300.

17

 

00000000 0000 ©8000 000%3 000 00.0
©8000 0300 003 080000000 0003 000080090
000 00000 000 0000 0.00000 00300000 0002

0000 000 00.0 .00 0088
88000 003 08000000 03003 0003 0%0003000

000 800000 000000 .08000 800000 0000m
.0088 38000 003 080000000 08000 0003
0000.0 0.0.30— 000 8 0000008 000 0008080
000— 00803 003 00000080 .080 088.0%

0800 0:0 08
0008000000 00805

0000800900

00000 000 000 300m

000000000 0000000000000 ~02
0000000000000 vuwu0 0030 000 008m

088 000 000

080 508
0008000 000— 00803 003 0000000me
0000 w0%0000 00.0 >000€0u00m
0000 8800000 8 000088 80000002
800000 300m
0%000300M
0000 000.000 00.0 08000 00000000
0000 0000000 8 000.800 002000000
080 000000 .00 0000000 00300000

 

0000030

000000000 000003-080 8 .0000m $0306
00800.0 000000000000 00803 000 008 800
08000 00000000000000 w00003 Nm0u0 .00 00 N005
00000002 000 0800a 00080m 00> 000 80003

0.0008 8030000 @8000 000 00800.0 3000000000 00000> 0003000 000000000 000 8000000 08080 00030.5 .N 030 H

18

 

%0000

008000 0003 080000000 03003 003 0000000000m
.%0000 008000 00.0 88000 00800080 0008

000 0003 00000000 0800008000 8 00030000000
00000000 000 00000 800000 00300000

08000 00300000 0000 0008 0000.0 000-080
.00 0000800000 08000000 0008000 000000000
0000800

000.0 00.0 080 00 0000000000 0003 080000000
000 003 0000000800 000.0 .00 0%000300v0

00000 03.0 0000

000 00.0 %0000 008000 0800000000m
0008000000 00805

000800 8 0000000002

00000 000 0000 80000m

00000000..000m

0000000 0000000000

00030000000 00300000

800000 0000000000
”0000000000 000000000000
08000 0000000000000

8000 003000000
00802
0%0003000 000000070
88000 0000000000000
0380 0000000000000

0900000800 000-300 .0000

080 008::
00n0 000 808000

002 000800
Emma 80 00000000

 

00.0 800.0 00000 000 0000.0 00.00003 0038 00000002
0x. 00.0 000000> 000 .00 8.3 00800800 0000000800 0000000000 0000000000000 0000
00000 000 0000.0 000-080 00038 00000000.~ 0000000 00000000000000 2W0 .00 00 0000.08m
000>
0000030 00000002 000 0w000Q 00080m 000 000000<

60.00000 00000000 008000 000 00800.0 000000000000 00000> .00 0000000000 000 m0000000 000080 00030000 .N 030 .0.

19

C. Application of the stages of change model to dietary change

The stages of change model has been used successfully in predicting health-
related behaviors such as smoking cessation and exercise adherence (Prochaska and
DiClemente 1983; Marcus et a1. 1992); the model has only recently been tested in
relation to dietary change. The stages of change model proposes that behavior change
occurs through a series of ﬁve stages: Precontemplation (unaware/not thinking about
changing), Contemplation (seriously thinking about making a change), Preparation
(making deﬁnite plans to change), Action (actively modifying an unhealthy behavior) and
Maintenance (maintaining the new behavior for some time) (Prochaska and Diclemente
1983). While the stages are sequential, individuals do not necessarily progress through
the stages in a linear fashion, but rather, may relapse and repeat stages (Prochaska et a1.
1992)

Application of the stages of change model to dietary change poses methodological
challenges especially in classifying people according to the stages. Dietary behaviors are
changed rather than ceased as in the case of smoking. Furthermore, dietary references
such as 30% of energy ﬁ'om fat present complex concepts while dietary behaviors change
daily.

Currently, developing a method to identify an individual’s stage of change is an
area of much research. A summary of some published studies on developing the
algorithm for dietary change is presented in Table 3. Greene et a1 (1995) developed an
algorithm to assess stages of change for fat intake using the criterion for effective action

of fat intake 3 30% kcal and tested in a mail survey with adults and university staff and

20

graduate students (n=184). The algorithm was based on subjects’ behaviors related to the
avoidance of high-fat foods. Curry et a1. (1994) developed an algorithm based on
Prochaska and DiClemente’s series of mutually exclusive questions normally used for
smoking cessation and tested in a random-digit dial telephone survey (n=1083). Both
algorithms demonstrated that a liner decrease in % kcal as fat from Precontemplation to
Maintenance stages and signiﬁcant differences in fat intake were seen mainly between
those in pre-action stages of change and those in the Action or Maintenance stages. The
majority of people in these two studies were classiﬁed in the Action and Maintenance
stages (60-62%). Consistent with the ﬁndings of Cuny et a1. (1994) and Greene et a1.
(1994), Glanz et a1. (1994) tested an instrument designed to determine stage of change for
fat and ﬁber consumption in a large study in United States (n=17,121) and reported little
difference in % kcal from fat between the three pre-action stages. The largest difference
was found between those classiﬁed in the pre-action and those in the Maintenance stage.
Sixty-three percent of subjects were classiﬁed in Action or Maintenance stages for fat
while only 51% of subjects were classiﬁed in Action or Maintenance stages for ﬁber.
Little was known about the stages of change for reducing ﬁber intake (Loforge et a1.
1994; Glanz et a1. 1994). Loforge et a1 (1994), one of the few studies, tested an algorithm
relating to stages of change for ﬁ'uit and vegetables consumption (n=405). They found
that only 15% of subjects were classiﬁed in the Action or Maintenance stages, whereas
67% were classiﬁed in either Precontemplation or Contemplation and 19% were
classiﬁed in Preparation.

Age, education and gender were associated with stages of change in the previous

21

studies (Curry et a1. 1992; Spomy and Contento 1995; Glanz et a1. 1994). Those in
Maintenance stage for either fat or ﬁber behaviors are more likely to be female, older, and
highly educated. Spomy and Contento (1995) investigated psychosocial variables that
have been used in other theory-driven studies of everyday food selection in relation to
stages of dietary change. They found that reduction of perceived barriers, mostly in terms
of taste and perceived difficulty of performing needed behaviors, and increased overall
health concerns, social modeling, and self-efﬁcacy were associated with taking action and

maintaining the behavior change.

22

 

0000000 000000w0> 000 008.0 00.0 00000000 .00 00m000
000000000000 00 000000000 0003 $00 000 00w000
00000000000000 00 000000000000005 000000 00
000000000 0003 $00 0000003 .00w000 00000000002
00 00000< 00 0000000000 0003 0000.300 .00 $2

.000 00.0 0m0000 .00 00w000

000000000002 00 00000400 00 0003 0000.300 .00 $00
.0m000 00000000002

00 00.0 0000.0 0000 foam 8 00000008800005

0 00 820 080 $00 820 00000

.00w000 0000000000000 000 00000000080805
0003000 00.0 0000.0 0000 .X 00 000000.000 000000

.000 00.0 0m0000 .00 00w000
00000000002 00 00000< 00 0003 0000 .300 .00 XS

.00w000 000000000000 8 0w00000 08000080000005 00

00000 0003000 00.0 0000.0 0000 .x. 00 000000.000 0000.0

.0000 000 00w000 000000000000 00 00000<

00 0003 0000.300 .00 X: m 0000 0003 00.0 00.0 00w000
00000000002 00 00000400 00 0003 0000.300 .00 XS
.0m000 00000000002 00 00000 000 0w000 000000

-000 00 00000 0003000 003 A0000 00000000000 000w000
000 000 .00w000 000000000000 8 00000000080805 00
00000 0003000 00.0 0000.0 0000 0X. 00 000000.000 00005

=82 020 080-00 -8000 00020
000000
000000w0> 000 0000.0 00.0 0m0000 .00 0w00m

830. 0 00 00050 0588000
000000000 000 00.00.0000 E0005
000000 00.0 w0000000 00.0 0w0000 .00 0w00m

8000 .0 so 0050 088300
0000000000 000 00 000000 0000005
000000 00.0 w0000000 00.0 0m0000 .00 0w00m

00800 000 00.0 - 30000 .0320
0000 w000000000 00.0 0w0000 .00 0w00m
000000 000 m0000000 00.0 0m0000 .00 0w00m

000-8850
0000000 000000000.
m00000H0

00000000 002
00070

0000-050500
00003000 000000000.
0000.0":

00000000 0.02
000.2%

.003

.00 00 090.000

0000
.00 00 000000

0000
.00 00 D000

:2
.0 0o 0500

 

00000000

00000002

000000m 000% 000 00000400

00030000 00000000 00 000000 0&0000 .00 00w000 .00 0000000 00030000 .m 0000.0.

23

D. High blood cholesterol and overweight
High blood cholesterol

High blood cholesterol is deﬁned as a total cholesterol of 240 mg/dL or greater,
however, a blood cholesterol above 200 mg/dL has been shown to increase the risk of
developing heart and blood vessel disease (National Cholesterol Education Program
1993). Studies have documented an association between elevated level of blood
cholesterol, development of atherosclerosis and heart disease risk, and risk increases
directly as blood cholesterol level increases. Adults with high blood cholesterol are twice
as likely to have heart disease as those with normal blood cholesterol levels (below 200
mg/dL).

Effective dietary interventions have been demonstrated to result in decreased total
serum cholesterol level and LDL-C or improved ratio of total cholesterol to HDL-C in
participants with elevated total serum cholesterol level in most of the cholesterol
reduction programs (Perovich and Sandoval 1995; Byers et al. 1995; Fisher and Fisher
1995; Angotti and Levine 1994; Johnston et a1. 1994; Baer 1993; Hebert et a1. 1993;
Briley et al. 1992; Shannon et al. 1987) while a lack of changes in total serum cholesterol
levels was reported in some studies (Barratt et al. 1994; Masur-Levy et al. 1990). The
conﬂict in the literature may due to the difference in content of the programs or the
intervention protocol such as the length of intervention programs.

Perovich and Sandoval (1995) reported a signiﬁcant interaction between the risk
category and change in total serum cholesterol over time. People in high risk categories

of serum cholesterol decrease their serum cholesterol more that those in moderate or low

24

risk categories after nutrition programs. Thus, we wanted to know whether individuals
with high serum cholesterol at the time of enrollment in the program were more likely to

make dietary changes in worksite nutrition programs.

Overweight

Being overweight is a signiﬁcant contributor to risk for chronic disease (Berg
1992). The best current recommendation for reduction of chronic disease risk associated
with obesity is to prevent becoming overweight. However, some research suggests that
for those who are overweight even moderate weight loss (10% of body weight) does
reduce health risks of obesity (Glodstein 1992). Increased risk of cardiovascular disease is
generally recognized to be the primary relationship underlying the observed association
between obesity and increase mortality (Hubert et a1. 1983; Sorlie 1980). Obesity has also
been associated with an increased prevalence of other conditions, such as osteoarthritis
(Davis et al. 1988), gallbladder disease (Haffner et al. 1989), and some types of cancer
(Doll and Peto 1981).

The BMI reference standards used to categorize weight status were based on the
Second National Health And Nutrition Examination Survey 1976-1980 (NHANES II)
sex-speciﬁc BMI distributions for person 20-29 years of age (Nation Center for Health
Statistics 1987). The rationale for this choice was that people 20-29 years of age are
relatively lean, i.e., closest to a population with a “healthy” body weight, and that weight
gained after age 29 is predominantly fat (National Center for Health Statistics, 1983). The

deﬁnition of overweight for men was a BMI of 27.8 kg/m2 or greater; for women

25

overweight was deﬁned as a BMI of 27.3 kg/m2 or greater (greater or equal to the 85th
percentile).

Changing dietary habits are consistently emphasized for weight control program.
Weight loss along with a decreased dietary fat intake were reported by most of the weight
control and cancer prevention programs (Fisher and ﬁsher 1995; Haus et a1. 1994;
Johnston et al. 1994; Han'is et al.1994; Shannon et al. 1987). On the other hand, a high
recidivism rate had also been found in several weight maintenance studies (Foster, et a1
1988; Haus et al. 1994) Haus et a1. (1994) indicated people who regained the weight lost
after nutrition intervention had higher BMI at the time of enrollment than those who
maintained the weight loss. Thus, we wanted to know whether BMI at the time of

enrollment is related to dietary change after the worksite nutrition programs.

 

Chapter Three

Change in fat intake of women after a worksite nutrition programs predicted by blood
lipids, BMI, dietary intake, stages of change and various psychosocial factors

A. ABSTRACT

The objective of this study was to identify factors which predicted reduction in
dietary fat intake of women at the end and at 4-month follow-up of 10-week worksite
nutrition programs. Blood lipid, body mass index (BMI), dietary intake, stages of change
and psychosocial factors assessed in 65 women (44:4:9 yrs) at enrollment were used to
predict the change in dietary fat intake which was estimated from 3-day dietary records.
Compared to the fat intakes at the time of enrollment, % kcal from fat at the end and at
the follow-up of the program decreased on average 2.9213 .0% (meanisd; p<0.05) and
1.5i1.9% (n.s.), respectively. Dietary fat intake (% kcal) and motivation at the time of
enrollment predicted the reduction in fat intake after the nutrition program, explaining
37% of the variance. At the 4-month follow-up, dietary fat (% kcal), motivation, BMI,
previous weight loss experiences and perceived beneﬁts of a healthy diet at enrollment
were signiﬁcant predictors for reducing fat intake, explaining 50% of the variance. Blood
lipids and stages of change did not predict changes in dietary fat. Risk factors for chronic

diseases such as high fat intake or high BM] at the time of the enrollment were associated
with dietary changes after the nutrition program.

26

27

B. INTRODUCTION

Dietary modiﬁcation programs have been used to promote weight control, blood
cholesterol reduction and cancer prevention (Glanz and Seewald-Klein 1986; Kris-
Etherton et al. 1988) and a growing number of studies have provided evidence that such
programs can be effective (Pelletier 1993; Haus et al. 1994; Sorensen et al. 1992; Briley
et al. 1992; Hunt et al. 1993; Gorbach et a1. 1990; Masur-Levy et al. 1990). However,
most interventions are short, with less than 6 months duration, and do not address the
challenge of how to support the maintenance of modiﬁed dietary behaviors. Phenomena
such as weight cycling demonstrate the questionable effectiveness of intervention
programs in achieving long-term dietary modiﬁcation even though short-term eﬁ‘ects are
encouraging. Health and medical researchers interested in reducing dietary risks for
chronic diseases have attempted to identify fundamental factors which inﬂuence
modiﬁcation of dietary behaviors. Once those factors are identiﬁed, effective program
strategies can be developed to improve nutrition programs.

Many studies have found that various psychosocial factors such as belief in diet-
disease connection, self-efﬁcacy, attitude, and nutrition knowledge are associated with
healthful diet behaviors (Patterson et al. 1996; Shepherd and Stockley 1987; Smith et al.
1995; Glanz et a1. 1993; Brug et al. 1994; Kristal et a1. 1995; Laforge et al. 1994). Most
of these studies are cross-sectional in design and as such they cannot establish if
psychosocial factors are associated with dietary changes in interventions studies.
Previous studies have identiﬁed the predictive factors of weight change aﬁer the

intervention (Haus et al. 1994; French et al. 1994), but predictors of weight change are

28
not likely the same as those for dietary change. We know little about the predictive
factors of dietary change after completion of nutrition intervention programs designed to
reduce dietary fat intake. Since certain dietary behaviors such as high fat and low ﬁber
intakes pose signiﬁcant disease risk, it is important to identify the predictive variables to
successful dietary changes.
In order to improve the effectiveness of nutrition programs, many theories and

models have been used to explain an individual’s dietary behavior. The Stages of Change

Model addresses the readiness to make behavioral changes (Prochaska and DiClemente

1983). The model has been successﬁil in predicting smoking cessation, but only recently

 

has been applied to dietary behaviors, such as fat consumption. Researchers have found
an association between stages of change and dietary fat intake in cross-sectional studies
(Glanz et al. 1994; Greene et a1. 1994). However, this model has not yet been clearly
tested in relation to dietary changes over time.

Additionally, high serum cholesterol and overweight are important risk factors for

chronic diseases. We wanted to know whether or not these risk factors are related to

dietary changes after the worksite nutrition programs.

The objectives of this study were: 1) to examine whether various psychosocial
factors, stages of change for reducing fat intake, blood lipids (total cholesterol, LDL-
C/HDL-C), dietary intake (fat and ﬁber) and BMI at the time of enrollment can predict
the dietary changes at the end and at a 4-month follow-up of a worksite nutrition
programs; and 2) to identify predictors which can differentiate participants who were

more successful at changing fat intake than those who were not at the end and at a 4-

29

month follow-up of the program.

C. METHODS
Worksite nutrition programs

Ten-week nutrition programs have been offered since 1988 without charge as part
of the health promotion program at a large midwestem university. Faculty and staff are
informed of the nutrition program through brochures, ﬂyers, and university newspapers.
The program consisted of one-hour weekly meetings for ten weeks and the purpose was
to promote healthy eating habits by lowering fat intake and increasing dietary ﬁber
intake. The curriculum included recommendations for daily fat and dietary ﬁber intakes,
modiﬁcation of recipes, cooking techniques for low fat and high ﬁber foods, fat and ﬁber
contents of foods, interpretation of food labels, suggestions for dining out, principles of
weight control and exercise, and behavior modiﬁcation strategies. Both knowledge and
skills were provided for the participants to help them incorporate strategies for low fat
and high ﬁber dietary behaviors into their individual dietary preferences and lifestyle.
Two or three programs were offered each semester, averaging 12 participants in each
program. The nutrition programs were taught by three leaders: one nutrition specialist
with nutrition and exercise physiology degrees; and two well-trained senior dietetics

students.

Data collection

A total of 151 participants enrolled in the worksite nutrition interventions

 

30

conducted between 1994-1996. The small number of men were excluded from this study
(136 women; 15 men) to reduce the confounding effect of gender. Those who dropped
out during the ﬁrst four weeks of the program (n=40) or those who were unable to
complete one of the measurements (n=31) were excluded from ﬁirther analyses.
Comparison of those who dropped out vs. completed the nutrition program have been
presented elsewhere (Huang et al., 1995; Chapter Four). Data from the ﬁnal sample of 65

women were used for data analyses in this study.

This study was approved by the University Committee on Research Involving

 

Human Subjects (Michigan State University) and informed consent was obtained from all
subjects. During the ﬁrst week of the intervention program, the subjects were instructed
to keep a 3—day diet record and to complete a questionnaire on psychosocial factors and
stages of change. Blood lipids (total cholesterol, LDL-C, and HDL-C) and
anthropometric measurements (weight and height) were conducted at the Nutrition
Assessment Laboratory. Feedback on the dietary intake and blood lipids results were
provided to each subject in the subsequent meeting. The same procedures were repeated
at the end and at a 4-month follow-up of the program.

The number of participants who signed up for the nutrition programs did not
differ among seasons (fall semester vs. spring semester). No signiﬁcant differences were
observed in the demographic characteristics, BM] (kg/m2), or dietary intake at enrollment
among subjects in the eight programs conducted during these two years. All data,

therefore, were combined for subsequent analyses.

31

Variables and measurements

Wwew obtained from a 3-day food record. The
record required a detailed description of types and amounts of foods and beverages
consumed for two non-consecutive weekdays and one weekend day. A detailed
instruction for the food record was provided with an example of the types and amounts of
food and beverages. Nutrient composition of the diet was analyzed using MSU
NutriGuide diet analysis soﬁware (Song WO, Version 2.0, Michigan State University,

East Lansing, MI). The nutrient composition database of the software was originated

 

from Michigan State University’s main-frame database, supplemented with USDA’s
Revised Agriculture Handbook 8 and data from food manufacturers. The nutrient
database is 98% complete for fat and 99% for dietary ﬁber.

W included belief in diet-disease connection, perceived beneﬁts
of a healthy diet, perceived barriers to a healthy diet, social support, perceived norms for
healthy eating, motivation, self-efﬁcacy for change, and weight loss experience. The
instrument was a 24-item questionnaire which was developed and validated by Glanz et
al.(1993) for the "Working Well Study". The questions were written in a Likert scale
format with a ﬁve-category continuum from strongly negative=1 to strongly positive=5.

.g- o ...g- o -c .9 . ' .u .g- . ...g- o ' -. .3 u‘
intakewere determined based on combinations of ﬁve items from the psychosocial factor
questionnaire. Those ﬁve items were self-rated diet, how long a low-fat or high-ﬁber diet
had been followed, behavioral intentions to change diet, attempts to make dietary

changes, and the reported success of those change efforts. The ﬁve questions were

32

combined into an algorithm to classify the subject into one of ﬁve stages of changes for
reducing fat intake or increasing ﬁber intake: Precontemplation, Contemplation,
Preparation, Action, or Maintenance (Glanz et al. 1994).

WWW consisted of height and weight. The measurements
were done following standard procedures (Lohman 1988), using a stadiometer and a
calibrated balance beam scale (Holtain Portable stadiometers, Seritex Inc). Body mass
index was calculated based on the formula: weight (kg) / height (m)? Subjects with BMI

227 .3 kg/m2 were considered overweight (National Institutes of Health, Consensus

 

Development Panel 1995).

Bloodlipjds consisted of fasting plasma concentration of total cholesterol, HDL-
C, and triglyceride from a blood sample collected by a ﬁnger prick (approximately 0.3
ml). Plasma was separated from whole blood which was collected into a tube containing
lithium-heparin anticoagulant (Microcuvette LH CB 300, Sarstedt). Analytical procedures
were performed using Kodak Ektachem DT60 Analyzer, DTE module. Total cholesterol,
HDL-C, triglyceride were determined based on a colorimetric method. Plasma
concentration of LDL-C was calculated by the formula: LDL-C (mmol/L) = total
cholesterol- HDL—C-triglyceride/S. Cutoﬁ‘ points were used to classify subjects into

different risk levels of total serum cholesterol: moderate risk (>520 and _<_ 6.24 mmol/L);

high risk (>624 mmol/L) (National Cholesterol Education Program 1993).

Statistical analyses

Statistical Package for the Social Science (Windows version 6.1, 1996, SPSS Inc,

33

Chicago, IL) was used for statistical analyses. Multiple regression analyses were
performed using the changes in dietary fat at the end of intervention and at the 4-month
follow-up as the dependent variables. Changes in dietary fat were deﬁned as the
difference in dietary fat intakes (% kcal from fat) between the enrollment and at the end
of the program; and between the enrollment and the 4-month follow-up. Dietary fat (%
kcal ) and ﬁber intakes (g/ 1000 Kcal), BM] (5 27 .3 kg/mz; 27 .3-36 kg/mz; >36 kg/mz),
blood lipids (s 5.20 mmol/L; 5.21-6.24 mmol/L; >6.24 mmol/L), various psychosocial
factors (1-5 point scale), and stages of change (Precontemplation=l, Contemplation=2,
Preparation=3, Action=4, Maintenance=5) at enrollment of the intervention were used as
predictors (independent variables) for dietary changes. Initially, all variables described
above were included in the model. A backward elimination precedure was used with the
criterion for inclusion set at p<0. 10. Variables not reaching this criterion were eliminated
from the model.

Discriminate analyses were used to identify the differentiating predictors for
those who were successﬁil in changing fat intake (beneﬁt group) vs. those who were not
(no-beneﬁt group) at the end and at the follow-up of the program. Participants were
ranked and separated into tertiles based on the changes in fat intake (% kcal) at the end of
the program. Those who were at the top and bottom tertiles in changing fat intake
represented the beneﬁt group and no-beneﬁt group, respectively. Predictors (independent
variables) described earlier were considered in a stepwise discriminate model to
characterize the two groups. The same procedures were repeated to identify the predictors

for the dietary changes at 4-month follow-up. Discriminate analysis provided maximal

 

 

34

discrimination between the members of two groups with the coefﬁcients indicating the
estimates of the relative importance of the variables to discriminate between the beneﬁt

and no-beneﬁt groups.

D. RESULTS
Characteristics of subjects

A total of 65 women completed the measurements at enrollment and the end of

the program. They were predominantly Caucasian (97%) and staff (98%) on campus with

 

an average age of 44:1:9 years (mean is.d.): 27% between 25-3 9, 48% between 40-49,
22% between 50-59 and 3 % over 60 years of age. Descriptive statistics for dietary fat
and ﬁber intakes, blood lipids, BMI, and various psychosocial factors at enrollment are
summarized in Table 1. Mean % kcal from fat and ﬁber density were 30.8 :t 6.0 % (mean
i s.d.) and 9.5 i 3.6 g / 1000 kcal, respectively. Fifty-one and eighty-ﬁve percent of
subjects at enrollment did not meet the dietary recommendations for fat (5 30% kcal from
fat) and ﬁber (212.5 g/ 1000 kcal), respectively. About half of the subjects were at health
risk with total serum cholesterol >5.20 mmol/L and/or BMI > 27.3 kg/mz. According to
the stages of change algorithm, the majority of the subjects were in Action stages (55%),
followed by Preparation (23%) (Figure 1). The distributions of most of the psychosocial
factors were highly skewed toward strong agreement. Ninety percent of subjects had prior
weight loss experiences and 62% had previously lost the weight they wanted to lose but

either had gained back all or some of the weight.

35

Changes of dietary fat, body weight, and blood lipids

The mean (is.d.) dietary fat intake was 30.8i6.0, 27 .9d:6.8 and 29. 121:7.4% kcal
from fat at enrollment, at the end and at the 4-month follow-up of the program,
respectively (Table 2). Compared to fat intake at enrollment, % kcal from fat at the end
and at the follow-up of the programs decreased on average 2.9i 3.1% (p<0.05) and
1.5i1.9% (ns), respectively. Energy intakes had decreased by the end of program and at
the 4-month follow-up (p<0.05). Body weight, serum cholesterol and HDL-C had not

changed signiﬁcantly at the end or at the follow-up (Table 2).

Predictors of change in dietary fat

Table 3 shows the regression coefﬁcients and standard errors of the variables
which are signiﬁcantly associated with decreasing fat intakes at the end of program and at
the 4-month follow-up. Dietary fat intake and motivation (i.e., how important to you is
eating low-fat foods) at the time of enrollment predicted reduction in fat intake (p< 0.05)
at the end of the nutrition programs, explaining 37% of the variance. The reduction of %
kcal from fat was 4.7:l:6.0% (mean isd.) and 0.3 $1.5 % for those with >30% of kcal
from fat and those withs 30% of kcal from fat at enrollment, respectively. Subjects who
strongly agreed with the importance of eating low fat foods reduced their fat intakes more
than those who strongly disagreed. Dietary fat intake and motivation at the time of
enrollment also predicted the change at the 4—month follow-up. In addition, BMI, weight
loss experience, and perceived beneﬁts of a healthy diet at enrollment were signiﬁcant

predictors for reducing fat intake (p<0.05). These ﬁve predictors accounted for 50% of

 

 

36

the variance in reduction of fat intake at the follow-up. Subjects with higher BMI at
enrollment were less likely to reduce their dietary fat intake than those with lower BMI.
Those whose BMI s 27.3 kg/m2 reduced their fat intake from 31.1i5.4 % to 28.2i7.0%
whereas those with BMI > 27 .3 kg/m2 increased their fat intake from 295d: 5.9% to
31.1i7.0%. Blood lipids, stages of change, and some of the psychosocial factors (i.e.,
belief in diet-disease connection, social support) were not signiﬁcantly associated with
reduced fat intakes after the nutrition program or at the 4—month follow-up.

Differentiating predictors for benefit group vs. no—benefit group

 

At the end of the program, subjects categorized in the beneﬁt group (top—tertile;
=22) reduced their dietary fat intakes from 33.4:t4.5% to 23.5i5.2% (mean i s.d.), while

those who were in the no-beneﬁt group (bottom- tertile; n=22) increased their dietary fat
intakes from 27.4 i5.2% to 31.8 i5.4%. Those two groups could be discriminated by
their enrollment data of dietary fat, BMI and motivation (p<0.05) (Table 4). Overall,
subjects could be correctly classiﬁed into the beneﬁt group or no-beneﬁt group based on
those three differentiating predictors, 76% of the time (probability). At the 4-month
follow-up, subjects in the beneﬁt group (n=19) changed their fat intake from 32.714.4%
to 23.8 i6.1%, while those who were in the no-beneﬁt group (n=19) changed their fat
intake from 26.5:t4.4% to 35.4i6. 1% (Table 4). Among all of the variables, dietary fat
intake, EM, and self-efﬁcacy for change at enrollment were effective discriminators for
two groups at the 4-month follow-up (p<0.05). Based on these three diﬂ°erentiating
predictors, subjects could be correctly classiﬁed into the beneﬁt group and no-beneﬁt

group at the follow-up, 83% of the time (probability).

 

37

E. DISCUSSION

This study identiﬁed the predictors related to reduction of fat intake after the
worksite nutrition programs and but also at the 4-month follow-up. We found that dietary
fat intake and motivation at the time of enrollment predicted the changes in fat intakes at
the end and at the follow-up of the programs. In addition, BMI, perceived beneﬁt of a
healthy diet, and previous weight loss experience were also related to change in fat
intake at the 4-month follow-up, although these associations were not found at the end of
the program. Previous studies have attempted to identify psychosocial factors associated
with dietary change but not related to the intervention programs (Smith et al. 1995; Smith
et al. 1992; Patterson et al. 1996; Contento and Murphy 1990). Smith et al. ( 1995) found
that diet-related beliefs and nutrition knowledge were predictive of dietary change in 249
volunteers three-months after a 1-hour one-to-one nutrition education program. Patterson
et al. (1996) found in a population-based study that individuals who strongly believed in a
diet-cancer connection and those with knowledge of fat and ﬁber recommendations
decreased their percentage of energy from fat over 3 years. In this study, we examined
blood lipids, BMI, and stages of change as predictive factors as well as psychosocial
factors. Subjects who had a high risk for chronic diseases (i.e., BMI >27.3 kg/mz), were
less likely to reduce fat intake at the follow-up, although they needed most to modify and
maintain their dietary intake. Identiﬁcation of the underlying reasons for the diﬂiculty in
reducing fat intake for overweight people at follow-up is an important challenge for
health promotion programs.

Previous studies reported that maintenance of changes in dietary behaviors is

 

 

 

38

difﬁcult. We are not aware of any published research examining the predictors associated
with reduction of dietary fat intake at the follow-up of worksite nutrition programs.
However, several studies of weight management programs suggest that previous weight
loss attempts and history of participants in a formal weight loss programs were related to
increased body weight at 6-24 month follow-up (French et al. 1994; Haus et al. 1994). In
this study, previous weight loss experience predicted the change in fat intake at the
follow-up. These ﬁndings highlight the importance of prevention of relapse regarding
weight loss experiences (i.e., loss but part or all of the weight regained) in nutrition

interventions.

 

This is one of the few studies to show the association between stages of change
and dietary change over time. Consistent with the study of Patterson et al. (1996), we
found that the Stages of Change model was not a signiﬁcant predictor of change in fat
intake after a nutrition intervention program. One possible explanation is that 7 8% of the
subejcts in our study were in the Preparation and Action stages of change for reducing fat
intake, while a population-base study found only 55% of women to be in these two stages
of readiness to reducing dietary fat (Glanz et a1. 1994). The disproportionately smaller
percentage of our subjects in different stages of change may have reduced the likelihood
of ﬁnding associations with the dietary change even if such relations do exist. In this
study, the majority of subjects in maintenance stage already consumed s 30% kcal from
fat at the time of the enrollment and kept their fat intake 5 30% kcal from fat at the end or
at the 4-month follow-up of the program. Thus, there may have been a ceiling effect on

possible fat reduction for those who already had fat intakes less than 30% of kcal. A

38

diﬂicult. We are not aware of any published research examining the predictors associated
with reduction of dietary fat intake at the follow-up of worksite nutrition programs.
However, several studies of weight management programs suggest that previous weight
loss attempts and history of participants in a formal weight loss programs were related to
increased body weight at 6-24 month follow-up (French et a1. 1994; Haus et a1. 1994). In
this study, previous weight loss experience predicted the change in fat intake at the
follow-up. These ﬁndings highlight the importance of prevention of relapse regarding
weight loss experiences (i.e., loss but part or all of the weight regained) in nutrition

interventions.

 

This is one of the few studies to show the association between stages of change
and dietary change over time. Consistent with the study of Patterson et al. (1996), we
found that the Stages of Change model was not a signiﬁcant predictor of change in fat
intake after a nutrition intervention program. One possible explanation is that 7 8% of the
subejcts in our study were in the Preparation and Action stages of change for reducing fat
intake, while a population-base study found only 55% of women to be in these two stages
of readiness to reducing dietary fat (Glanz et al. 1994). The disproportionately smaller i
percentage of our subjects in different stages of change may have reduced the likelihood
of ﬁnding associations with the dietary change even if such relations do exist. In this
study, the majority of subjects in maintenance stage already consumed s 30% kcal from
fat at the time of the enrollment and kept their fat intake s 30% kcal from fat at the end or
at the 4-month follow—up of the program. Thus, there may have been a ceiling effect on

possible fat reduction for those who already had fat intakes less than 30% of kcal. A

 

39

study with EFNEP participants also suggested that participants with low dietary quality at
enrollment make more dietary changes than those with high dietary quality (Koemer and
Song 1997). In this study, we found that dietary fat consumption at enrollment in the
program was associated with the stages of change (Contemplation: 33% kcal from fat;
Preparation: 32% kcal from fat; Action: 31% kcal from fat; and Maintenance: 24% kcal
from fat) and the trend was similar with ﬁndings in previous studies (Glanz, et al. 1994;
Greenes, et al. 1994). This ﬁnding supports our conﬁdence in the use of the algorithm
developed by Glanz et a1 (1994) to classify participants into diﬂ‘erent stages of change for
fat at enrollment.

In this study, motivation and perceived beneﬁts of a healthy diet were important
predictors for reducing fat intake in women. This result is consistent with the ﬁndings
from a cross-sectional study by Kristal et al. (1995) in which similar psychosocial items
were examined in 16,287 working men and women, and perceived beneﬁts and
motivation were found to be strong predictors of dietary intake and intention to change
diet. In contrast to our ﬁndings, Patterson et al. (1996) and Smith and Owen (1992)
reported that beliefs in diet-disease connection was related to dietary changes. Probably
because belief in diet-disease connection was so skewed toward strong agreement (92%
strong or very strong agree) at the time of enrollment, it may have resulted in no
association with the reduction of dietary fat intake. Measurement of psychosocial factors
did not vary enough initially to explain the change in fat intake after the worksite
nutrition program.

Most of the nutrition intervention studies result in mixture of successful and

 

 

4o
unsuccessful participants (Sorensen et al. 1992; Briley et al. 1992; Hunt et al. 1993;
Gorbach et al. 1990; Masur-Levy et al. 1990). This study uniquely used both multiple
regression and discriminate analyses to explain the outcome which are changes in dietary
fat intake after intervention and at follow-up. Multiple regression models identiﬁed the
variables that are signiﬁcantly associated with the dietary changes. The discriminated
analyses provided an insight between the two groups (those who were successﬁil at
reducing fat intake vs those who were not) regardless of how much dietary change was

made. The predictors we found were similar for the two statistical approaches. We found

 

that dietary fat intake and BMI at enrollment were most important in distinguishing these
the two groups at the end and at the follow-up of the program. It appears that this
nutrition program may not be suitable for all the participants, especially for those with
high BMI and low fat intake at enrollment. Subjects who were less likely to reduce fat
intake could be identiﬁed in the beginning of the program based on the predictors we
found. Different strategies and program designs such as increased motivation, self-
efﬁcacy for dietary change and perceived beneﬁts of a healthy diet can be provided to
improve the effectiveness of nutrition programs such as the one described in this study.
Limitations of this study must be taken into account when interpreting the results.
Although subjects included in the present study were a self—selected convenience sample,
they may be representative of individuals likely to seek nutrition information in the
United States. Since the subjects in the study were women and predominately Caucasian,
results cannot be generalized to other populations. About 20% (17/88) of the participants

who did not complete the measurements at the end and follow-up were not signiﬁcantly

 

41

diﬂ‘erent in dietary intake, blood lipid, and BMI from those who were included in the

analysis, therefore, the results were quite robust.

 

42

Table 1. Dietary intakes, blood lipids, BMI, and psychosocial factors of subjects (n=65)
at enrollment of the worksite nutrition program

 

 

 

Variables Mean :1: s.d. % of participants at risk
Age 43.7:t8.9
Dietary intakes

% kcal from fat 30.8i6.0 51% ( >30% kcal from fat)

Fiber density 9.52136 85% (<12.5 g/ 1000 kcal)
Blood lipids

Total cholesterol(mmol/L)c 5.31:1:1.31 48% (>5 .20 mmol/L)

LDL-C/I-IDL-C 2.2i0.9
BMI (kg/m2) 28.1i6.2 45% (>27 .3 kg/mz)
Psychosocial factorsa

Belief in diet-disease connection 4.5i0.7

Perceived barriers to a healthy diet 2.521209

Perceived beneﬁts of a healthy diet 3.8i0.7

Perceived norms for healthy eating 2.9i0.9

Social support 2.9i1.0

Motivation 4. 1i0.9

Weight loss experienceb 3.6i1.0

 

a1-5 scale; 1: strongly negative, 5=strongly positive (Glanz et a1. 1993)

b1-6 scale; 1= overweight; never try to lose to weight, 2=did not lose any weight; still on
a diet now, 3= lost weight, but gained all of it back, 4: lost weight, but gained some of it
back, 5= lost part of the weight I wanted to lose; kept it off, 6=lost all the weight I want
to lose; kept it off (Glanz et al. 1993)

cTo convert mmol/L cholesterol to mg/dL , multiply mmol/L by 38.7. Cholesterol of 5.20
mmol/L=200 mg/dL.

Table 2. Change in dietary intake, serum cholesterol levels, BMI of subjects at the end
and at 4-month follow-up of the worksite nutrition programs

 

 

 

 

Variables Enrollment End Follow-up
(n=128) (n=65) (n=5 8)

Calories 1951i634a 1721i549* 1745:529“
Fat (g) 67 .1d:30.0 55.2:t25.9* 57.6dz28.9*
% kcal from fat 30.8160 27.9:l:6.8* 29.1:1:7 .4

> 30% kcalfromfat” 51% 41% 51%
Dietary ﬁber (g) 17.5i6.2 16.1i5 .9 16.9:t6.9
Fiber density(g/ 1 000 kcal) 9.6i3 .6 9.9:t3 .4 10.1i4.3

<12. 5g/1 000 kcal” 85 % 77% 78%
Serum cholesterol (mg/dL) 5.31i1.31 5.28:1:1.41 5.26:1:1.00

>5.20mmol/L” 48% 46% 4 7%
Body weight (1b) 174.3d:40.3 173.5:1:40.7 173.6i41.4
BMI (kg/m2) 28.8:62 28.7:1:6.1 28.7159
>27.3 kg/mz b 45% 44% 44%
a Mean :1: s.d.

b percentage of subjects

*p<0.05 by paired t-test compared to the enrolhnent

44

Table 3. Backward multiple regression models for reducing fat intakes at the end of the

worksite nutrition programs and at the 4-month follow-up

 

Predictors

B

SEofB

Beta

P

 

At the end of nutrition education program

 

Dietary fat intake(%kcal from fat) 0.85 0.18 0.64 0.000**
Motivation2 2.97 1.08 0.35 0005*
Weight loss experience 1.22 0.81 0.17 0.139
Dietary ﬁber density 0.34 0.27 0.17 0.220
Constant - 43.19 8.70 0.000**
R2 for the model 42%
R2 for dietary fat and motivation 37%
R2 for dietary fat alone 22%
At 4-month follow-up
Dietary fat intake (% kcal from fat) 0.78 0.20 0.52 0.000**
BMI (kg/m2) - 4.20 1.65 - 0.36 0.016*
Perceived beneﬁts of a health diet 4.82 2.03 0.34 0024*
Motivation 3.71 1.71 0.32 0.037 *
Weight loss experience 1.95 1.06 0.25 0.075
Constant -25.64 1 1.43 0.032
R2 for the model 50%
R2 for dietary fat and BMI 34%
R2 for dietary fat alone 24%
* p<0.05; **P< 0.001

1 Regression coefﬁcient

21-5 scale; 1= strongly negative, 5=strongly positive

5
4

 

 

 

 

mastic.“ 2: «n 5335820

 

Segue:— uo can 2: 3 5535320

mdﬁﬁm mix; adamd adawd noococomxo $2 BESS
odamé minor.“ *m.onmw.m mdﬁ .v coagaog
$43 03.3 233 34mm 8&5 .5.“ agomnuam
$43 on"? $43 3.1.: 5&3 38m
333 4.3.3 $3M 23; mass. been: 5 288 8383
3.03M vdﬂﬁm odawd mdhﬁm Home 3:8: a mo Emcee: coZooeom
Qoﬁxm adamd odﬁﬁm adaod Home 3:8: a 9 $293 33083
odaﬁv Yoamé odﬁqé demd 85288 omaommueomc ammozom
amuse-a _amuomoﬁzmm
Lehmdm véamdm *owaodm ﬁmamdm Em
maﬁam Yoﬁwg adﬂd odﬁQN can.“ AEAQA
£2“le wzacow 2.3mmw mvgﬂmw A§oﬁazeoumﬂono 3on
83: 25:.
«in: 343 Six: ”New 9862 823 93% “BE
*vdamém Davina»). *Nwavsm 5.33.9. E 88m :33 .x.
mow—SE b303—
33 $9 on? an“... 9?.
51¢ 51: ﬁne anus 2.2.525 2.. a
98..” E232... Z 9.9% 50qu 98am Egon—67H 98..” 52:5 8352: “51:056.:

 

gnawed 535:: 86x83 05 mo 3-26:0“ 5:084 um 98 c5 05 “a mowﬁﬁo FREE :2: so was «3:on “women

.0: 98 5059 we 9595360 2: mo 083 on“ “a vopoo=oov E83.“ 3008:0me ES .Em 62m: @003 .moxﬂﬁ @805 .v 2an

46

mombmsw oaﬁatommc no woman 8823a wﬁﬁuﬁbﬂﬁ *

A82 .3 s 536va 2 EE m32

8 an? H E“ “mo—Ho Ea z E3 ”32 8 @853 H Ems? 05 mo tag 52 Hm £03 a «o 088 Baum :5 Jams? 32 "v £93 x we =m
35mm Sn .EwEB 62 Hm .30: 86 a no 55 saws? b8 82 8: Evnm Jaws? 8 82 8 9 65a ”Emmozgo "L 638 “Eon 0-:
A82 .3 Ho 559 3:83 bwcobmum .0333: bmqobm NH 638 “Eon méo

A; H :82 a

83E 8.“ E owqaﬁ no waste“ Eaton 98 a8 05 Emmoaﬁ anew E2890: 2% gnaw “meson 2E. u

v 2an 8m ouoﬁoom

47

 

83E 5H wﬁoswm: Mom owsano b8 mowﬁm

8:28:35. co=o< cozﬁaaoi cozaanEoo cozmanEoooE

 

o

 

or

 

 

8
X.

 

 

cm

 

 

ow

 

 

 

om

 

 

hém cm

88on cough: 0:383 05
mo EoE=oEo 2: «a mwcwmmowtmm mo SEE E wﬁosva 8m owsmno mo mowﬁm mo :BSQEEQ
M 83?;

Chapter Four

Dropouts from worksite nutrition programs differ from those who completed in dietary
intake and BMI but not in blood lipids and psychosocial factors

A. ABSTRACT

The objective of this study was to compare the dietary intake, blood lipids, BMI, stages
of change for reducing fat and increasing ﬁber intakes, and various psychosocial factors
of participants who dropped out vs. those who completed worksite nutrition programs
offered at a major university in the United States. Dietary intake (3-day diet record),
blood lipids, BMI (kg/m2), stages of change for reducing fat and increasing ﬁber intakes,
and various psychosocial factors were surveyed during the ﬁrst week of a nutrition
program. A total of 128 women was recruited from 10—wk worksite nutrition programs
between 1994 and 1996. Those who dropped out during the ﬁrst four-weeks of the
program were classiﬁed into the dropout group (n=40) and the rest into the completion
group (n=88). T-tests and Chi-square analyses were used for continuous and categorical
variables, respectively. Compared to the completion group, the dropout group was more
likely to: 1) eat > 30% of kcal from fat (69% vs. 50%, p<0.05); 2) be overweight (67 %
vs. 40%, p<0.05); 3) have a lower percentage of subjects at the Maintenance stage of
change for reducing fat intake (6% vs. 18%, p=0.07) with no difference in other stages of

change; and 4) have a higher percentage of subjects at the Action stage of change for

48

49

increasing ﬁber intake (49% vs. 32%, p=0.07) and a lower percentage of subjects in the
Maintenance stage (8% vs. 21%, p=0.08). Innovative strategies are needed to retain those
who are overweight or consume > 30% of kcal from fat in worksite nutrition programs as
well as to advance the majority of the participants who enroll in such programs ﬁom the

Action to the Maintenance stage.

B. INTRODUCTION

Nutrition programs have been used at worksites to promote weight control,
cholesterol reduction and cancer prevention (Glanz and Seewald-Klein 1986; Kris-
Etherton et al. 1988). Worksites are an important channel to deliver health promotion
programs allowing access to over 60% of adults of varying ages and health status (U .S.
Department of Labor 1992). Between 1985-1992, the percentage of worksites that offer
nutrition programs increased sharply from 48% to 7 8% for large worksites (more than
750 employees), and from 9% to 22% for small worksites (SO-100 employees)(USDHHS
1992)

Several studies have reported the efﬁcacy of worksite nutrition programs
(Pelletier 1993; Haus et al. 1994; Sorensen et al. 1992; Briley et al. 1992; Hunt et al.
1993). These ﬁndings are based solely on data of those who completed the programs in
reference to either baseline data of the same group or of a control group. High dropout
rates (12-3 5%) have been reported as one of the programmatic problems in worksite
nutrition programs (Bruno et al. 1983; Lovibond et al. 1996; Carmody et al. 1980;

Dishman 1988). Little is known, however, about the characteristics of those who were

50

apparently motivated to enroll, but discontinued participation in the programs. Therefore
a better understanding of those individuals who drop out of worksite nutrition programs
can help develop eﬁ‘ective strategies, and also enhance efﬁcacy of the programs.

Dietary behaviors have previously been associated with socioeconomic and
psychological factors (Patterson et a1. 1996; Smith et al. 1995; Winkleby et al. 1994;
Glanz et al. 1993 ; Kristal et al. 1995; Brug et al. 1994; Laforge et al. 1994). The Stages of
Change Model addresses the readiness to make health behavioral changes (Prochaska and
Diclemente 1983). This model has been successfully used in predicting and modifying
behavioral changes across a wide range of health problems (i.e., smoking cessation,
exercise) and only recently has been applied to dietary behaviors. Whether the stage of
change at the enrollment is related with the adherence to the nutrition program has not
been tested clearly. Also it is important to evaluate whether or not the risk factors for
chronic diseases such as high cholesterol and overweight are associated with the
adherence to worksite nutrition programs.

The hypothesis of this study was that participants who dropped out from worksite
nutrition programs differed from those who completed the programs in dietary intake,
blood lipid proﬁle, BMI, stages of change for reducing fat and increase ﬁber intakes, and

various psychosocial factors.

C. METHODS
Worksite nutrition programs

Ten-week nutrition programs have been offered since 1988 without charge as part

51

of the health promotion program at a large midwestem university. Faculty and staﬁ‘ are
informed of the nutrition program through brochures, ﬂyers, and university newspapers.
The program consisted of one-hour weekly meetings for ten weeks. The purpose of the
program was to promote healthy eating habits by lowering fat intake and increasing
dietary ﬁber intake. The curriculum included recommendations for daily fat and dietary
ﬁber intakes, modiﬁcation of recipes, cooking techniques for low fat and high ﬁber foods,
fat and ﬁber contents of foods, interpretation of food labels, suggestions for dining out,
principles of weight control and exercise, and behavior modiﬁcation strategies. Both
knowledge and skills were provided for the participants to incorporate strategies for low
fat and high ﬁber dietary behaviors into their individual dietary preferences and lifestyle.
Two or three programs were offered each semester, averaging 12 participants in each
program. The nutrition programs were taught by three leaders: one nutrition specialist
with nutrition and exercise physiology degrees; and two well-trained senior dietetics

students.

Subjects and data collection

Between 1994-1996, 151 subjects enrolled in the worksite nutrition interventions.
A small number of men (n=15) was excluded from the study to reduce the confounding
effect of gender. Those who were unwilling or unable to complete any one of the
measurements (n=8) were also excluded. Subjects who discontinued the program during
the ﬁrst four-weeks of the program were classiﬁed as the dropout group and the rest of

subjects were classiﬁed as the completion group.

52

This study was approved by the University Committee on Research Involving
Human Subjects and informed consent was obtained from all subjects. During the ﬁrst
week of the nutrition program, subjects were instructed to keep a 3-day diet record and to _
complete a questionnaire on psychosocial factors and stages of change. Blood lipid
analyses (total serum cholesterol, LDL-C, and HDL-C) and anthropometric
measurements (weight and height) were conducted at the Nutrition Assessment
Laboratory by qualiﬁed technicians. Feedback on dietary intake and blood lipids data was
provided to each subject at the subsequent meeting.

No signiﬁcant differences were observed in demographic characteristics, BMI, or
dietary intake of the subjects among eight programs conducted over two years. All data,

therefore, were combined for subsequent analyses.

Variables and measurements

Dietamjntakulatajonfatmdﬁhﬁwere obtained from a 3-day food record. The
record required a detailed description of types and amounts of foods and beverages
consumed for two non-consecutive weekdays and one weekend day. A detailed
instruction for the food record was provided with an example of the types and amounts of
food and beverages. Nutrient composition of the diet was analyzed using MSU
NutriGuide diet analysis software (Song W0, Version 2.0, Michigan State University,
East Lansing, MI). The nutrient composition database of the software was originated
from Michigan State University’s main-frame database, supplemented with USDA’s

Revised Agriculture Handbook 8 and data from food manufacturers. The nutrient

53
database is 98% complete for fat and 99% for dietary ﬁber.

Bsxchosoﬂalfams included beliefs in diet-disease connection, perceived
beneﬁts of a healthy diet, perceived barriers to a healthy diet, social support, perceived
norms for healthy eating, motivation, self-efﬁcacy for change, and weight loss
experience. The instrument was a 24-item questionnaire which was developed and
validated by Glanz et al.(1993) for the "Working Well Study". The questions were
written in a Likert scale format with a ﬁve-category continuum from strongly negative=1
to strongly positive=5.

o ...g- . -o .3 . ' .n .1- o ...g- o ' -. .3 ”-
intakewere determined based on combinations of ﬁve items from the psychosocial factor
questionnaire. Those ﬁve items were self-rated diet, how long a low-fat or high-ﬁber diet
had been followed, behavioral intentions to change diet, attempts to make dietary
changes, and the reported success of those change efforts. The ﬁve questions were
combined into an algorithm to classify the subject into one of ﬁve stages of changes for
reducing fat intake or increasing ﬁber intake: Precontemplation, Contemplation,
Preparation, Action, or Maintenance (Glanz et al. 1994).

Anthropometricmeasnrements consisted of height and weight. The measurements
were done following standard procedures (Lohman 1988), using a stadiometer and a
calibrated balance beam scale (Holtain Portable stadiometers, Seritex Inc). Body mass
index was calculated based on the formula: weight (kg) / height (m)? Subjects with BMI

227 .3 kg/m2 were considered overweight (National Institutes of Health, Consensus

Development Panel 1995).

54

BlondJipids consisted of fasting plasma concentration of total cholesterol, HDL-
C, and triglyceride ﬁ'om a blood sample collected by a ﬁnger prick (approximately 0.3
ml). Plasma was separated from whole blood which was collected into a tube containing
lithium-heparin anticoagulant (Microcuvette LH CB 300, Sarstedt). Analytical procedures
were performed using Kodak Ektachem DT60 Analyzer, DTE module. The plasma
concentration of LDL-C was calculated by the formula: LDL-C (mg/dL)= total
cholesterol- HDL-C-triglyceride/S. A cutpoint was used to classiﬁed subjects into
different risk levels of total serum cholesterol: 5 200 mg/dL for low risk; > 200 mg/dL

for high risk (National Cholesterol Education Program, 1993).

Statistical analyses

The dropout group was compared with the completion group using t-tests for
continuous variables (i.e., dietary intake, blood lipids, and various psychosocial factors)
and chi-square analyses for categorical variables (i.e., stages of change for reducing fat
intake and percentage of overweight subjects). All statistical analyses were performed
using Statistical Package for the Social Science (Windows version 6.1, 1996, SPSS Inc,

Chicago, IL).

55

D. RESULTS

The majority of subjects who participated in this project (n=128) were Caucasian
women (97%) and staff on campus (98%). The mean age was 44:1:9 years (s.d.): 28%
between 24-39, 42% between 40-49, 28% between 50-59 and 3% over 60 years of age.
Of these, 33% of the women (n=40) discontinued the program during the ﬁrst four-weeks
of the program (dropout group) and the rest 67% of the women (n=88) completed the
program (completion group).

The dropout group had a higher average % kcal from fat (32.8i6.9%) in their diet
at the time of enrollment than the completion group (3 0.33:6.6%; p<0.05); and a higher
percentage of subjects not meeting dietary recommendation for fat i.e., > 30% of kcal
(69% vs. 50%; p<0.05; Table 1). Calories and ﬁber intake did not differ signiﬁcantly
between the two groups. The majority of the subjects in both groups did not meet the
dietary recommendation for ﬁber, i.e., 212.5 g/ 1000 kcal, (84% dropout group vs. 86%
completion group; n.s.; Table 1). Although the average BMI did not differ statistically
between the dropout group (30.9i7 .5 kg/mz) and the completion group (28.7i13.2
kg/mz), the dropout group had signiﬁcantly more overweight subjects (BMI 227.3 kg/mz)
than the completion group (68% vs. 47%; p<0.05; Table 1). About half of all subjects had
total serum cholesterol concentrations >200 mg/dL with no differences between the
dropout and completion groups (Table 1). Overall, the completion group seem to be
healthier than the dropout group.

Various psychosocial factors did not differ signiﬁcantly between the dropout and

completion groups (Table 2). Belief in diet-disease connection was skewed toward strong

56

agreement in both groups with the mean equal to 4.421: 0.1(s.d.) on a 1-5 scale. About 90%
of the subjects in both groups reported that they had tried to lose weight in the past.
Forty-four percent of the dropout group and 32% of completion group reported that they
had lost weight at least 10 pounds in the past, but gained back all of the weight(n.s.).

In terms of stages of change for reducing fat, the majority in both groups was in
Action stage (58% in dropout group vs. 54% in completion group; n.s.) and no one was
found in the Precontemplation stage (Figure 1). The distribution of subjects among
Contemplation, Preparation, and Action stages did not differ between the two groups. The
completion group had a higher percentage of subjects at the Maintenance stage (18%)
than the dropout group did (6%; p=0.07).

Nearly half of the subjects in both groups were classiﬁed into the early stages of
change for increasing ﬁber intake (Figure 2). Compared to the completion group, the
dropout group had a higher percentage of subjects in the Action stage for increasing ﬁber
intake (49% vs. 32%; p=0.07) and a lower percentage of subjects in the Maintenance
stage (8% vs. 21%; p=0.08; Figure 2). When the distributions between stages of change
for reducing fat and increasing ﬁber intakes were compared, a higher percentage of
subjects was found in the early stages of change for ﬁber than stages of change for fat.

About an equal percentage of subjects was in the Maintenance stage for fat and ﬁber.

E. DISCUSSION
We found that the dropout group had a higher percentage of subjects who were

overweight or whose diets contained > 30% kcal from fat. The dropout group,

57

unfortunately, needed the nutrition programs more than the completion group did to
modify their dietary intake.

Various psychosocial factors such as health beliefs, knowledge, motivation and
self-eﬁcacy have been related to dietary behaviors in other studies (Patterson et al. 1996;
Kristal et al. 1995; Smith et al. 1995; F errine et al. 1994; Contento and Murphy 1990).
However, dropout and completion groups in the present study did not differ in various
psychosocial factors; they did differ slightly with respect to the stages of change for
reducing fat intake or increasing ﬁber intake. The similar response to various
psychosocial factors by the two groups may be explained by 1) homogeneity of overall
subjects i.e., middle age, white, female staff; 2) strong agreement toward various
psychosocial factors questions; 3) volunteer-enrollment in the program because subjects
with similar response of psychosocial factors were likely to be recruited by the nutrition
programs; and 4) the small sample size in four subgroups of the chi-square test for stage
of change. Compared to the population-based studies (Glanz et al. 1994; Greene et al.
1994), a lower percentage of subjects in our study were found in the Precontemplation
and Contemplation stages of change for reducing fat intake and increasing ﬁber intake for
both the dropout and completion groups.

About half of the subjects in this study were overweight with BMI 227.3 kg/mz.
A campus-wide telephone survey conducted in 1993-1994 (Hembroff and Huang, 1995)
revealed that 32.1% of staff on campus were overweight. These ﬁndings suggest that
overweight individuals were likely to enroll in the worksite nutrition programs but also

tended to drop out of the programs. Compared to women in the Third National Health

58
and Nutrition Examination Survey (McDowell et al. , 1995), subjects in this study
consumed slightly less % kcal ﬁom fat (31% vs. 36%). This comparison suggests that
women consuming a high % kcal from fat were both less likely to be recruited into
worksite nutrition programs and more likely to drop out if they did enroll.

This study was conducted in a university setting with predominantly Caucasian
women. Thus the limitations of this study, such as participants’ education levels and work
environment, and limits of representation among diverse ethnic groups must be
considered when interpreting the results. Furthermore, additional testing of this

hypothesis is needed in different worksite settings and ethnic groups as well as in men.

If

 

59

Table 1. Dietary intake, blood lipids, and BMI of subjects in dropout vs. completion
groups of worksite nutrition programs

 

 

Variables Dropout group Completion group p2
n=40 n=88
Age 44.0:|:9.5 1 43.9dz8.5 ns
Dietary intakes(day)
Calories 1911i625 1913 i601 ns
Fat (g) 71.5:t 33.5 65.8 i301 ns
% kcal ﬁ'om fat 32.8:h6.9 30.3 d: 6.6 p<0.05
> 30 % kcal ﬁom fat 69% 3 50% p<0.05
Dietary ﬁber(g) 15.1 :t 5.1 16.9 :h 5.6 ns
Dietary ﬁber density 8.5 :1: 3.8 9.2 :t 3.2 ns
< 12.5 g/lOOO kcal 84% 3 86% ns

Anthropometric measurement

 

BMI (kg/m2) 30.9 3:7.5 28.7 :1: 13.2 ns
2 27.3 kg/m2 67% 3 47% p<0.05
Blood lipids (mg/dL)
Total serum cholesterol 208.8 :1: 43.6 202.2 :t 40.4 ns
> 200 mg/dL 50% 3 47% ns
HDL-C 54.7 d: 16.4 58.4 :1: 14.1 ns
LDL-C 119.9i43.6 115.4:1:33.8 ns
Triglyceride 165.2 :1: 124.5 1346 :t 70.5 ns
lMean i s.d.
2 t-tests and chi-square analyses were used for continuous and categorical variables,
respectively

3 percent of subjects with >3 0% kcal from fat, with ﬁber density <12.5 g/ 1000 kcal, BM]
227.3 kg/mz, or total serum cholesterol >200 mg/dL

gr

60

Table 2. Psychosocial factors of subjects in dropout vs completion groups of worksite
nutrition programs

 

 

 

Variables Dropout group Completion group p2
n=40 n=88

Belief in diet—disease connection 1 4.4:|:0.6 2 4.4i0.9 ns
Perceived barriers to a healthy diet 2.8i1.3 2.5i0.9 ns
Perceive beneﬁts of a healthy diet 3 .7i 0.6 3.7:}:09 ns
Perceived norms for healthy eating 2.8:tl .3 2.8i0.9 ns
Social support 2.9:i:1.3 2.9:I:0.9 ns
Self-eﬁicacy for change 3.9i1.3 39:1: 0.9 ns
Motivation 4.1i0.6 4.1 3:09 ns
lMean i s.d.

2 t-tests

3 1-5 Likert scale (1= strongly disagree, 5=strongly agree)

 
   
  

  

 

 

 

 

 

 

 

 

 

 

 

8.on .
MESS “ﬂ @5st com owcmno mo mowﬁm
coca—.3555. :o=o< cos—2.32m co=~_n_.:3:oo
II I F - c
. I .
S.
cm
3 _.~ Axe
m 2. mm on
3.
cm
drum so
mm
33.5 co=w_nEoo I @305 “30.3.5 I

 

mEahwea 5.5.5:: «28...?» 682988 .9» :5 nugget of» 085
u 8.35 an.“ wig—:5.— .Sm «wage .3 mow-3m no 55...:me
ﬂ «Suﬁ

62

8.ng a m Son 9 .

ESE Sam wﬁmmocoﬁ com owgﬁo mo mowﬁm

oocmcouﬁms. co=o< :osmnmnona co=m_QE3:oo

Ax.

 

 

 

 

 

 

 

 

 

3.0.5 :23.an0. 3.05 ”Sancho.

 

mﬁahwen 5.5.5:: axe—.83 698388 .m> :8 tonnage of» 32:
u 3.55 5.5 uni-3.55 .8.“ «wanna he 89% we E55535
N charm

Chapter Five

Stages of change are associated with dietary intake, BMI and various psychosocial factors
but not with blood lipids among women in worksite nutrition program

A. ABSTRACT

Stages of change for reducing fat intake and stages of change for increasing ﬁber
intake were assessed with dietary fat and ﬁber, body mass index (kg/m2), blood lipids and
various psychosocial factors among 128 women who enrolled in a worksite nutrition
program. According to the algorithm for readiness to change, subjects were classiﬁed into
ﬁve stages for reducing fat intake: Precontemplation (none), Contemplation (8%),
Preparation (20%), Action (56%), Maintenance (15%); and ﬁve stages of change for
increasing ﬁber intake: Precontemplation (none), Contemplation (16%), Preparation
(29%), Action (3 7%), Maintenance (18%). Compared to the other stages of change for
reducing fat intake, the Maintenance stage had 1) lower fat and higher ﬁber intake; 2)
lower BMI and waist/hip ratio; and 3) higher perceived beneﬁts of a healthy diet,
perceived norms for healthy eating, social support, self-efﬁcacy for change, and
motivation; 3) lower perceived barriers to a healthy diet. Compared to the other stages of
change for increasing ﬁber intake, the Maintenance stage had 1) lower fat and higher

ﬁber intake; 2) higher perceived beneﬁts of a healthy diet and motivation; and 3) lower

63

 

 

 

64

perceived barriers to a healthy diet. Blood lipids did not differ either among stages of
change for reducing fat intake or stages for increasing ﬁber intake. Strategies are
necessary to recruit those in early stages into worksite nutrition programs. It would be
important to evaluate further whether people in Preparation or Action Stages can be
moved into the Maintenance stage by inﬂuencing psychosocial variables differing among

the stages.

B. INTRODUCTION

 

Health professionals have put a lot of effort in promoting healthy eating, such as
decreasing fat and increasing ﬁber intakes, to reduce the dietary risk of chronic diseases.
In an effort to improve the effectiveness of nutrition programs, many theoretical models
have been used to explain an individual’s dietary behavior and help one make dietary
modiﬁcations. The Stages of Change model has been successﬁilly used in helping people
quit smoking and has only recently been tested in relation to dietary change. The Stages
of Change Model (Prochaska and DiClemente 1983) proposes that change occurs through
a series of stages: Precontemplation (unaware or not thinking about changing),
Contemplation (seriously thinking about making a change), Preparation (making deﬁnite
plans to change), Action (actively modifying an unhealthy behavior) and Maintenance
(maintaining the new behavior for some time).

Currently, the majority of nutrition interventions are “action-oriented” and
provide skills and strategies for people who are ready for action to change behavior. A

study by Campbell et al. (1994) demonstrated that nutrition messages tailored to a

65

person’s stage of change generated a signiﬁcantly greater reduction in dietary fat intake
than messages that were not so tailored. It appears, therefore, beneﬁcial to know the
stages of dietary change for people who participate in a nutrition program before an
intervention is introduced. A few published studies have attempted to categorize the
sample into stages of dietary change. Glanz et al. (1994) and Greene et al. (1995) reported
in population-based studies that about 65% of people were in Action or Maintenance
stages of dietary change for fat. Little is known about the stages of dietary change for
people who are recruited into various nutrition programs.

In order to use the stages of change model effectively in developing nutrition
interventions, it is important to understand the characteristics of each stage of change.
Psychosocial factors from several theories have been used to explain dietary behaviors.
Sporny and Contento (1995) reported that a reduction of perceived barriers and perceived
diﬂiculty and increased self-efﬁcacy were associated with Action and Maintenance stages
of change for reducing fat intake. Additionally, high serum cholesterol, overweight and
high waist/hip ratio are important risk factors for chronic diseases. We wanted to know
whether these risk factors are associated with stages of change for reducing fat and
increasing ﬁber intake.

The purpose of this study was: 1) to identify the stages of change for fat and ﬁber
of participants in a worksite nutrition program; and 2) to examine the stages of change for
reducing fat intake and stages of change for increasing ﬁber intake in relation to dietary
intake, blood lipid proﬁles, anthropometric measurements and psychosocial

characteristics.

 

 

 

66

C. METHODS
Worksite nutrition programs

Ten-week nutrition programs have been offered since 1988 without charge as part
of the health promotion program at a large midwestem university. Faculty and staff are
informed of the nutrition program through brochures, ﬂyers, and university newspapers.
The program consisted of one-hour weekly meetings for ten weeks and the purpose was
to promote healthy eating habits by lowering fat intake and increasing dietary ﬁber

intake. The curriculum included recommendations for daily fat and dietary ﬁber intakes,

 

modiﬁcation of recipes, cooking techniques for low fat and high ﬁber foods, fat and ﬁber
contents of foods, interpretation of food labels, suggestions for dining out, principles of
weight control and exercise, and behavior modiﬁcation strategies. Both knowledge and
skills were provided for the participants to help them incorporate strategies for low fat
and high ﬁber dietary behaviors into their individual dietary preferences and lifestyle.
Two or three programs were offered each semester, averaging 12 participants in each
program. The nutrition programs were taught by three leaders: one nutrition specialist
with nutrition and exercise physiology degrees; and two well-trained senior dietetics

students.

Subjects and data collection
Between 1994-1996, 151 subjects enrolled in the worksite nutrition interventions.

A small number of men (n=15) was excluded from the study to reduce the confounding

67

effect of gender. Those who were unwilling or unable to complete any one of the
measurements (n=8) were also excluded.

This study was approved by the University Committee on Research Involving
Human Subjects and informed consent was obtained from all subjects. During the ﬁrst
week of the nutrition program, subjects were instructed to keep a 3-day diet record and to
complete a questionnaire on psychosocial factors and stages of change. Blood lipids
analyses (total cholesterol, LDL-C, and HDL-C) and anthropometric measurements
(weight and height) were conducted at the Nutrition Assessment Laboratory by qualiﬁed
technicians. Feedback on the dietary intake and blood lipids data were provided to each
subject at the subsequent meeting.

No signiﬁcant differences were observed in demographics, BM], or dietary intake
of the subjects among eight programs conducted over two years. All data, therefore, were

combined for subsequent analyses.

Variables and measurements

WWwere obtained from a 3-day food record. The
record required a detailed description of types and amounts of foods and beverages
consumed for two non-consecutive weekdays and one weekend day. A detailed
instruction for the food record was provided with an example of the types and amounts of
food and beverages. Nutrient composition of the diet was analyzed using MSU
NutriGuide diet analysis software (Song WO, Version 2.0, Michigan State University,

East Lansing, MI). The nutrient composition database of the software was originated

68

from Michigan State University’s main-ﬂame database, supplemented with USDA’s
Revised Agriculture Handbook 8 and data from food manufacturers. The nutrient
database is 98% complete for fat and 99% for dietary ﬁber.

BsychgsociaLfactQLs included beliefs in diet-disease connection, perceived
beneﬁts of a healthy diet, perceived barriers to a healthy diet, social support, perceived
norms for healthy eating, motivation, self-efﬁcacy for change, and weight loss
experience. The instrument was a 24-item questionnaire which was developed and
validated by Glanz et al. (1993) for the "Working Well Study". The questions were
written in a Likert scale format with a ﬁve-category continuum from strongly negative=1
to strongly positive=5.

o ...3- o -c .'3 . ' .u .3- o ...3- o ' -. .3 "-
intakewere determined based on combinations of ﬁve items from the psychosocial factor
questionnaire. Those ﬁve items were self-rated diet, how long a low—fat or high-ﬁber diet
had been followed, behavioral intentions to change diet, attempts to make dietary
changes, and the reported success of those change efforts. The ﬁve questions were
combined into an algorithm to classify the subject into one of ﬁve stages of changes for
reducing fat intake and stages of change for increasing ﬁber intake, respectively:
Precontemplation, Contemplation, Preparation, Action, or Maintenance (Glanz et al.
1994)

Anthmpometdgmeasurements consisted of height, weight, waist/hip ratio and
percentage of body fat. The measurements of height and weight were done following

standard procedures (Lohman 1988), using a stadiometer and a calibrated balance beam

 

69

scale (Holtain Portable stadiometers, Seritex Inc) and ﬁberglass measuring tape. Body
mass index (BMI) was calculated based on the formula: weight (kg) / height (m)?
Subjects with BMI 227 .3 kg/m2 were considered overweight (National Institutes of
Health, Consensus Development Panel 1995). The percentage of body fat was determined
by bioelectrical impedance (BIA). BIA is based on the principle that resistance to a
standard electric current is proportional to the volume of a conductors or body, the square
of the length of a conductor path, the concentration of water and electrolytes, and the
compartmental distribution of the water and electrolytes (Lukaski 1987 ; Segal 1988).

mm consisted of fasting plasma concentration of total cholesterol, HDL,

 

and triglyceride from a blood sample collected by a ﬁnger prick (approximately 0.3 ml).
Plasma was separated from whole blood which was collected into a tube containing
lithium-heparin anticoagulant (Microcuvette LH CB 300, Sarstedt). Analytical procedures
were performed using Kodak Ektachem DT60 Analyzer, DTE module. The plasma
concentration of LDL-C was calculated by the formula: LDL-C (mg/dL)= total
cholesterol- HDL-C-triglyceride/S. A cutpoint was used to classiﬁed subjects into
different risk levels of total serum cholesterol: 5 200 mg/dL for low risk; > 200 mg/dL

for high risk (National Cholesterol Education Program, 1993).

Statistical analyses
Subjects in Precontemplation, Contemplation, Preparation, Action and
Maintenance stages of change were compared based on dietary intake, blood lipids, EM],

and various psychosocial factors. One-way analysis of variance (AN OVAs) were used to

7O

determine the signiﬁcance of group differences on these measures, and post-hoe contrasts
(least-signiﬁcant diﬁ‘erence) were used to determine signiﬁcant mean differences between
speciﬁc groups. Chi-square analyses were used for categorical variables (i.e., % subjects
with s 30% kcal from fat). All statistical analyses were performed using Statistical

Package for the Social Science (Windows version 6.1, 1996, SPSS Inc, Chicago, IL).

D. RESULTS

Characteristics of Subjects. The majority of subjects who participated in this project
(128 women) were Caucasian (97%) and staff (98%) on campus. The mean age was 442t9
(s.d.) years: 28% between 24-39, 42% between 40-49, 28% between 50-59 and 3% over
60 years of age.

Stages of change for reducing fat intake. The percentage of subjects in
Precontemplation, Contemplation, Preparation, Action and Maintenance stages for fat
were 0%, 8%, 20% 57%, and 15%, respectively. There was a linear trend in group means
for calories, % kcal from fat, and ﬁber density of four stages (Table 1). As indicated in
Table 1, average intake of calories and % kcal from fat decreased and ﬁber density
increased ﬁom Contemplation stage to Maintenance stage. The percentage of subjects
who met dietary recommendations for fat (s 30% kcal) and ﬁber (212.5 g/1000 kcal) also
increased with advanced stages. Subjects in the Maintenance stage had signiﬁcantly

lower % kcal from fat (24%) and higher ﬁber density (12.5 g/1000 kcal) than those in
other stages. Average BMI decreased with advanced stages of change for reducing fat

intake (p<0.05). Although percentage of body fat did not signiﬁcantly differ among

 

 

71

stages of changes for reducing fat intake (Table 1), it was found to be highly correlated

with BMI (r=0.80; p<0.001). This indicated that those with higher BMI did have higher
percentage of body fat. Blood lipids did not differ signiﬁcantly among stages of change

for reducing fat intake (Table 1).

The result of the ANOVAs of psychosocial factors are presented in Table 2.
Signiﬁcant F-ratios and linear trends were obtained for ﬁve out of seven psychosocial
factors: perceived barriers to a healthy diet, perceived beneﬁts of a healthy diet, social
support, self-efﬁcacy for change and motivation (Table 2). Those in the Maintenance
stage had signiﬁcantly higher scores than those in Action stages for perceived beneﬁts of
a healthy diet. Those in Action stage had signiﬁcantly higher scores than those in
Preparation stage for social support. Those in Preparation stage had signiﬁcant higher
scores than those in Contemplation stage for self-efﬁcacy for change and motivation
(Table 2).

Stages of change for increasing ﬁber intake. The percentage subjects in
Precontemplation, Contemplation, Preparation, Action and Maintenance stages for
increasing ﬁber intake were 0%, 16%,29%, 37%, and 18%, respectively (Table 3).
Subjects in Maintenance stage for ﬁber had higher ﬁber density and lower % kcal from
fat than other stages. BMI, blood lipids, and calories did not differ signiﬁcantly among
stages of change for increasing ﬁber intake (Table 3).

Signiﬁcant F-ratios and linear trends for group means were obtained for three out
of seven psychosocial factors: perceived barriers to a healthy diet, perceived beneﬁts of a

healthy diet and motivation. Subjects in the Maintenance stage had reported a lower

 

72

perceived barriers to a healthy diet than those in Action stage. Subjects in the Preparation
stage reported higher perceived beneﬁts of a healthy diet and higher motivation than
those in Contemplation stage. Subjects in Action stage did not differ signiﬁcantly from
Preparation stage on various psychosocial factors. About 55%, 62%, 55%, 62% of
subjects in Contemplation, Preparation, Action and Maintenance stages of change for
reducing fat intake, respectively, were also found in the same stages of change for

increasing ﬁber intake.

E. DISCUSSION

Most of the subjects in our study were in Preparation or Action stages of change
for reducing fat intake. A higher percentage of subjects in our study were in the
Preparation stage and a lower percentage of subjects in Precontemplation and
Contemplation stages for reducing fat intake compared to a population-based study where
the similar algorithm was used for 5,334 women (Glanz et al. 1994). Similar differences
were found in stages of change for increasing ﬁber intake between our study and the
population-based study of Glanz et a1. (1994). The differences between the two studies
could be predicted, since those who volunteer for programs aiming for particular
behavioral changes are more likely to be in advanced stages of change (Rossi 1996).

The distribution of stages for reducing fat intake and stages for increasing ﬁber
intake differed. Compared to the stages of change for reducing fat intake, stages of
change for increasing ﬁber intake had a lower percentage of subjects in Action stage

(3 0% vs. 56%) and a higher percentage of subjects in the Precontemplation and

73

Contemplation stages for ﬁber than for fat (16% vs. 8%). This may reﬂect the trends in
awareness of nutrition, speciﬁcally fat and ﬁber in the United States over the past decade
(Schucker et al. 1991). Concern about dietary fat has grown steadily since about 1980 and
dietary fat is now the leading nutrition concern (Schucker et al. 1991). Although there is
widespread general diet-disease awareness of dietary ﬁber, speciﬁc knowledge of
recommended intake and sources of ﬁber is still low (Fullmer et al. 1991; Ippolito and
Mathios 1991). About 60% of our subjects in each stage of change for fat were at the
same stages of change for ﬁber. These ﬁndings suggest that for the majority, but not all
subjects, stages of change for fat indicated the individual’s readiness to increase ﬁber
intake as well as readiness to reduce fat intake.

In this study, average % kcal from fat decreased with advanced stages of change
for reducing fat intake. The trend corroborates with the ﬁndings of Glanz et al. (1994) and
Greene et al. (1995). Subjects in Action and Maintenance stages who have tried and
believed that they had consumed a low-fat diet, indeed, had low % kcal from fat either
met or close to dietary recommendations (Action: 31% kcal from fat; Maintenance: 23%
kcal from fat). These ﬁndings support our conﬁdence in the use of such an algorithm to
classify participants into different stages of change for reducing fat intake. Consistent
with ﬁndings of another study (Sporny and Contento 1995), dietary ﬁber density of our
subjects also increased with advanced stages of change for reducing fat intake.

Although the dietary ﬁber density increased with advanced stages of change for
ﬁber, the average ﬁber density in Action and Maintenance stages for ﬁber was still only

8.8 g/1000 kcal and 10.8 g/1000 kcal, respectively. Most of the subjects (73%-88%) in

74

these two stages did not meet the ﬁber recommendation of 2 12.5 g / 1000 kcal. We do
not know if our ﬁnding is due to inadequate knowledge and skills necessary to rate ﬁber
intake of the subjects or truly inadequate ﬁber intake by the subjects. The algorithm used
in this study was based on the individual’s self-rated diet for classiﬁcation in either early
or later stages of change (i.e., how high in ﬁber is your overall diet? ). The algorithm thus
requires subjects to be knowledgeable on the ﬁber content of foods and ﬁber
recommendations to correctly classify themselves.

In this study, the largest reduction of % kcal from fat and increase in ﬁber density
were seen between Action and Maintenance stages. The most important psychosocial
factors for distinguishing those in Maintenance from Action stages were perceived
beneﬁts of a healthy diet in the case of stages of change for fat and perceived barriers to a
healthy diet in the case of stages of change for ﬁber. These ﬁndings suggest that different
strategies are needed for fat and ﬁber to advance those in Action stages to Maintenance.
Motivation was most important for both fat and ﬁber in distinguishing those in
Preparation from Contemplation stages. These ﬁndings support the suggestion of
Baranowski (1992) that motivations that lead to psychological state of readiness to take
action would be most important in the early stages of the dietary change process. For
both stages of change for fat and ﬁber, perceived barriers to a healthy diet were highest
for those in Contemplation stage and then linearly decreased with advanced stages;
perceived beneﬁts of a health diet were lowest in Contemplation and then linearly
increased with advanced stages. Thus, both a decrease in perceived barriers to a healthy

diet and increase in perceived beneﬁts of a healthy diet may help people move forward to

75

improve their diets relative to fat and ﬁber.

Previous studies found that older people were more likely to be in the Action or
Maintenance stages for reducing fat intake than younger people (Curry et al. 1992;
Spomy and Contento 1995; Glanz et al. 1994). In contrast, we found age was not
associated with stage of change for either reducing fat or increasing ﬁber intake. This
may be because the age of our sample (45% between 40-49 years) did not vary enough to
detect a diﬁ’erence.

We found that subjects in Contemplation and Preparation stages of change for fat
had higher BMI and waist/hip ratio than those in Action and Maintenance stages, but
these relationship was not found in stages of change for ﬁber. The cross-sectional nature
of this study does not permit us to see whether the lower BMl and low waist/hip ratio are
due to the their low fat diet or whether those with low BMI and low waist/hip ratio are
more likely to adopt low fat eating and maintain the change. However, the ﬁnding is
interesting and deserves ﬁirther study.

This study was conducted in a university setting with predominantly Caucasian
women. Thus the limitation of this study such as participants’ education levels and work
environment, and limits of representation of diverse ethnic groups must be considered
when interpreting the results. Furthermore, additional research is needed in different
worksite settings and ethnic groups as well as in men.

Most of subjects who signed up for the worksite nutrition program in this study
were in Preparation or Action stages. Such understanding can assist us in designing

nutrition programs that can more appropriately be targeted by stage of change. Strategies

76

are necessary to recruit those in early stages to the worksite nutrition program. It would
be important to evaluate ﬁirther if people in Preparation or Action stages can be moved to

Maintenance stage by inﬂuencing psychosocial variables differing among the stages.

F. IMPLICATIONS

The action-oriented activities of the worksite nutrition programs would appear to
be appropriate for the majority of subjects in Preparation and Action stages, but not for
those in Contemplation who had higher fat intakes, higher BMI and waist/hip ratio.
Prochaska and DiClemente (1983) and Campell et al. (1994) suggest different strategies
such as conscience raising, self-evaluation, and self-liberation should be developed to
move those in the Contemplation stage into the Preparation stage. The complete absence
of anyone in the Precontemplation stage in this worksite nutrition program reﬂects
readiness of people who enroll in worksite nutrition programs. Targeting those in the
Precontemplation stages with information messages about the health beneﬁts for adopting
healthy diet, along with the information about the health risk associated with high fat and
low ﬁber diet, may be more effective in recruiting those in early stage of change into the

nutrition programs.

77

 

 

modvm 8.on .433 among ..m.oﬂ._ .343 owe efﬁng
a. a. wsﬂ em 3.3% 23.2 2 3...? g... 3.8% 88803
.8 $8 :8 $6 $8 . .8? «RN
8.on 8va .9348 .933... .4358 .3 3.3m ANEEV Em
8muuo8eacuauu<
a. 3.. mm 343 Sﬂ .N 03 .N 4953
.2. $3 $9. $9. :2. meme 8%
a. a. 238 3.68 3.38 an: Swag 6.8.280 33
8...: 82m
.8.on xi $2 $3 .22 .182 82% mi N
80.on :5.on .wnnoﬁ .34; .343 .343 93 822588 spa
8.on modva 330.2 .332 .4962 .0392 953 can
.8.on $3 .3? $3 :8 we as... was <30
:5.on So.on .9308 .98: . z m 8.8%.? .333 E ace :3 .x.
80.on 80.on $2.39. .5338 38403 .3438 33$ 5
a. a. 83mm: £92.02 838$ 88.82.. assﬁoxv .285
3.3.: b.502—
2. a S 5.3. $32 :33. .33. :. cbswa.
$3.22 $3 $3 9.8 .28 a: .sw 3 .x.
.3585 mm 855382 :28... 8&3qu 59.388qu

 

”ESE Em $538 How owawao mo momma £593 «a $8.33 mo 85083055.. Ea 3a: @005 .8138 .0805 A 2an

78

385:8 093mg mam: .3.on Hm BENEc bpqwomawmm 8m Etoﬂomsm 08% 2: maﬁa 8: 9.802 3.
€85 2: $8 33 300 .33, mo 3083 m

momma Bow .«o 888 QSEW 08. a 255 Quad ..

33.28 0333.30.

owSEo mo momma macaw 839$ mo 3&3.» 33.89

.3 m. 9.02.

H 2an Ho.“ 808.com

79

.39338 oonéom maﬁa amodvﬁ S “:80wa bamomiwu 8w 238693 08% 2: maﬁa «on £802 2:

momma Sow go 8.88 9on 2: E gob Soﬁa ..

momma macaw count? mo $335 $3.25 m
dam.” :32 N

Gama bmaobmnm .oouwmmmu bwqobmuc 28m tow—ﬂ m4 _

 

 

80.on Sodva p.233 3&3”? so. 33 «0.32 nouguoz
8.on 8.on 333 $33 no. 3o. m 303m amuse é 588?:an
modva 3.on $33 Dwoﬁ .m 5.3% avoﬁa 5&3 38m
8 8 wig we“? 0. 33 0.33 3% 333 E 288 8283
8.on 8.on 9%.: .v 33$. m $.33 «0.33 3% 3:8: 3o £83 8383
:5.on modva $33 awoﬁa no. :3 «N. E: 5% 332 a 8 EBB 8383
a 8 no“: MS“? 013 No.3? ﬂ 88288 0985-56 338

‘35“an m 0285882 aouo< 833.305 comuwaaoﬁoo

 

3.33 “am memos? How omega mo momma £593 3 ﬁoﬁﬂ Bogonommm .N oEaH

 

 

8 8 8.3: Nona; We“: no“: 28 85883
8 8 33 .3 :83 38.3 8.362 E 88 8 88885
8 8% $8 :8 8% m N83 28 N
8 8 38.8 :38 8.3th 333 @888 Em
8508325:
8 8 38.8 38.3 383 0.33 15.35
8 8% $8 $3 $8 “8&8 8?
8 8 omﬂmm :82 288.6. 888 3283 388888 885
88: 32m
8 88 8S x: :2 “=88 coo:m mi N
0 8an _odvm 8.3:: 8.33 “383 8.32 93 08:3 £88 .5an
8 8 8 .332 38.2 8.38.2 33.: $23 sen
8.on Sam $8 $8 £8 “E 808 88 $88
8 8.38 328.8 ~38. 8 m 8.35. H m “.288. _ m E :88 _88 x
8 8 338.8 3.826 02E .3 >88 .8 98E 5
8 8 2382 8:802 2382 0:888 9828“: 8580
8:35 Ego:—
8 8 $8.8 $8 .3 3i .8 888.? A5 93
236%: 9i .3: $3 $8 a: $2 3 .x.
83.585 E 855332 aouo< 8098835 comps—@8880

 

cum—SE bow mammwouoﬁ Sm owqano mo momma 8553 “a 38.38 mo 85083038.» 28 Eva: woo—p .3qu 385 .m 2an

81

$3838 02:39 mam: a3.on 8 880% bummomammm 08 3.53095 08% 2: maﬁa 8: 882 a:
«C85 05 $8 33 38.3% mo 808$ n

momma Sow mo 388 95% 2: S 255 30:5 v

£338 8553-39

omega mo momma macaw 855?? $9328 33-89

ﬁn um 53?

m 2an 8m Beacon

82

.mwmabaoo 023mg mam: £0.on as “:80wa bugomﬁwa 8w Etoﬂoqsm 08$ 2: maﬁa Ho: made: 2..

momma 50m .«o :88 8:on 0% E 305 Soﬁa— V

mowﬁm macaw oo§E> mo maxim.» ~33.qu m
6.3" 932 N

385 bwaobmnm .oouwmmmu Emaobmuﬁv 03% his m; _

 

 

50.on :5.on 333 333 a 3“? be“: Eggs:
8 mm ”on: .v wdﬁoe 0. _ﬂ .v 036$ owSBQ How 5880.5an

8 m: N. $3 We“? oi“: 0.?an toga 38m

2 m: ”owe. m M33. m O13 0.0a: wage 3:8: 8m 288 83088
8.3a 8.on 333 335 m anew”: "we“? 5% 3:8: So 383 8385
80.on 8.on E33 “wind "03% boﬁwa $6 333 a 9 eggs 8588
a a 25+ moi; 33v “33¢ “8:858 88%-66 £23m

5585 E 855382 acacia 535335 :ouwaaopaoo

 

333 Sam mﬁmwohé Mom amaze mo momma msotg Hm Eoﬁﬂ EQOmoaowmm .v 2an

Chapter Six

CONCLUSIONS AND IMPLICATIONS

Conclusions

1.

Dietary fat intake and motivation at the time of the enrollment were strong
predictors for reduction in fat intake at the end of the nutrition program and at the
4-month follow-up. In addition, BMI, previous weight loss experience and
perceived beneﬁts of a health diet at the enrollment were signiﬁcant predictors for
reducing fat intake at the 4-month follow-up, but these associations were not
found at the end of the program.

Dropouts were more likely to be overweight or to have diets containing more than
30% kcal from fat. The dropouts needed the nutrition programs to modify dietary
intake more than those who completed the programs.

Most of the subjects who participated in the worksite nutrition programs were in
Preparation or Action stages of change for reducing fat intake and/ or reducing
ﬁber intake. Those who were in Maintenance stage consumed lower fat plus
higher ﬁber than those in other stages. The most signiﬁcant psychosocial factors
in distinguishing those in the Maintenance from Action stages were perceived

beneﬁts and barriers to a healthy diet. Motivation was most important

83

84

psychosocial factor in distinguishing those in Preparation from those in the

Contemplation stages

Implications

l.

The action-oriented activities of the worksite nutrition program would appear to
be appropriate for the majority of subjects in Preparation and Action stages, but
not for those in Contemplation who had higher fat intakes, higher BMI and
waist/hip ratio. Prochaska and DiClemente (1983) and Campell et al. (1994)
suggest different strategies such as conscious raising, self-evaluation, and self-
liberation should be developed to move those in the Contemplation stage into the
Preparation stage.

The complete absence of anyone in the Precontemplation stage in this worksite
nutrition program, potentially 12-20 % of the population, means that a large
number of people are neither ready nor willing to consider dietary change in fat
intake. Targeting those in the Precontemplation stages with information messages
about the health beneﬁts for adopting healthy diet, along with the information
about the health risk associated with high fat and low ﬁber diet, may be more
effective at recruiting those in early stage of change into the nutrition programs.
Worksite nutrition programs designed to increase motivation, the perceived
beneﬁt of a healthy diet and designed to prevent relapse from previous weight loss
experiences are needed to help participants adopt more healthful diets.

Prior to enrollment, two-thirds of our subjects previously lost the weight they

85

wanted to lose but had gained back all or some of the weight. In order to prevent
the inﬁnite diet—relapse cycle, the prevention of relapse and coping with the
negative psychological impact from relapse prior to the enrollment need to be
taken into the consideration in the beginning as well as throughout the program.
Innovative strategies to retain those who dropout from the programs are needed

for successful worksite nutrition programs.

Chapter Seven

RECOMMENDATIONS FOR FUTURE STUDIES

Based on the ﬁndings of this study the following recommendation are made for ﬁlture

studies:

1. Since this study was conducted in a university setting with predominantly
Caucasian women, additional research is needed in different worksite settings and
ethnic groups as well as in men.

2. It would be important to evaluate further if the predictors we found in this study
can help people to change their diet successﬁilly and improve the eﬁ‘ectiveness of
the worksite nutrition program.

3. It is necessary to identify the underlying reasons why those were overweight or
consumed > 30% kcal from fat at the enrollment were more likely to dropout from
the worksite nutrition programs.

4. We found that subjects in Contemplation and Preparation stages of change for
reducing fat intake had higher BMI and waist/hip ratio than those in Action and
Maintenance stages, but this relationship was not found in the stages of change for
increasing ﬁber intake. The cross-sectional nature of this study does not permit us

to evaluate whether the lower BMI and low waist/hip ratio are due to their low fat

86

87

diet or whether those with low BMI and low waist/hip ratio are more likely to
adopt low fat eating and maintain the change. However, the ﬁndings are

interesting and deserves further study.

 

APPENDICES

APPENDIX A

OUTLINE OF THE WORKSITE NUTRITION PROGRAM

88

Diet/Weight Management Program

Spring, 1995

1-16 Session 1 Eztablnhinzmsszlm

0 course overview
0 assessment of current health status

# announce health fair (pretest)

1-23 Session 2 With

0 risk factors (diet, anthropometric,
blood lipid) for chronic disease
0 interpretate the result of lipid
profile and body composition

# feedback from lipid proﬁle and body
composition
# collect a 3-day dietary record from
participants
1-30 Session 3 WM
0 sources of dietary fat
0 source of dietary fiber

0 interpretation of a 3-day dietary records
# feedback of the-3-day dietary record
2-6 Session 4 Cmﬁnuuatinuuanﬂiatlsﬂghmuau
0 being a smart consumers

0 alternative eating plans for lifestyles

# feedback from 3 day dietary record

2-13

2-20

2-27

3-6 - 3-10

3-13

3-20

3-27

Session 5

Session 6

Session 7

Session 8

Session 9

Session 10

89

Behavior change- The key to success

0 behavior chain
0 weight cycling
0 food label

0 easily prepared and low fat foods
0 foods prepared for different lifestyle

0 deve10p[ network of support
0 high risk situations(2 examples)
0 continue "behavioral change" (session 5)

spring break

0 importance of physical activity for health
0 various physical activities

0 develop physical activity program for lifestyle
0 strength and weakness of different of exercise
program

# announce health fair (posttest)

WWW

# announce health fair(follow-up) and
follow-up plan
# collect 3-day dietary record

 

 

APPENDIX B

UCRIHS APPROVAL

 

  

OFHCEOF
RESEARCH
AND
GRADUATE
STUDIES

University Committee on
Research Involving
HumanSuMcds
(UCRIHS)

thngHmUmnmw
225 Administration Building
En! Lifting. Mldﬁgan
4882‘4045

517/355-2180
FAX: 517/432-1171

 

90

MICHIGAN STATE

U PI I V’ E R S I 1' Y

December 16, 1994

To: Ya-Li Huan

208 Food Sgience Bldg.

RE: IRBI: 94—577
TITLE: HEALTH PROMOTION (HEALTHY-U) PROJECT-DIET/HEIGHT
MANAGEMENT PROGRAM (FOR STAFF AND FACULTY)
REVISION REQUESTED: N/A
: -A
APPROVAL DATE: 12/14/94

The University Committee on Research Involving Human Subjectsg(UCR§HS)
t at t e

review of this project is complete. I am pleased to adVise

heretore, the UCRIHS approved this project including any revision
listed above.

RENEWAL: UCRIHS approval is valid for one calendar year, beginning with
the approval date shown above. Investigators plannin 0
continue a project beyond one year must use the green renewal
form (enclosed with t e original approval letter or when a
project is renewed) to seek u dated certification. There is a
maXimum of four such expedite renewals ossible. Investigators
wishing to continue a project beyond the time need to submit it
again or complete reView.

REVISIONS: UCRIHS must review any changes in rocedures involving human

sub ects, rior to initiation of t e change. I: this is done at

.the time o renewal, lease use the green renewal form.

revise an approved protocol at an other time during the year,

send your written request to the CRIHS Chair, requesting revised

approval and referencing the project's IRB # and title. Include

in our request a description of the change and any revised

ins ruments, consent forms or advertisements that are applicable.
PROBLEMS]
CHANGES: Should either of the following arise during the course of the
work, investigators must noti UCRIHS promptly: (l) roblems
(unexpected side effects comp aints, etc.) involVing uman
subjects or (2) changes in the research environment or new
information indicating greater risk to the human sub'ects than
existed when the protocol was previously reviewed an approved.

If we can be of any tuture'help, lease do not hesitate to contact us
at (517)355-2180 or FAX (Sl7)3 6- 171.

avid E. Wright, Ph.
UCRIHS Chair

DEw:pjm

cc: Won 0. Song

APPENDIX C

INFORMED CONSENT

91

Healthy U Diet/Weight Management Program

Informed Consent

I enrolled voluntarily in this DIET/WEIGHT MANAGEMENT PROGRAM which is sponsored

by the Healthy U at MSU. The purpose of the program is to promote healthy eating habits by
reducing fat intake and increasing dietary ﬁber intake.

My full paru'cipation in the program will involve a) attendance of all 10 weekly classes and
follow-up meetings offered at designated sites; b) completing questionnaires on dietary and
health habits; c) body measurements such as height, weight, percentage of body fat, waist/hip
ratio; and (1) measurements for blood total cholesterol, HDL, LDL, triglyceride and hemoglobin
with a ﬁnger prick sample.

All of the above measurements will be done by trained Healthy-U Program staff and faculty
following the standard procedure. I understand that blood lipid measurements are considered
screening only. I am aware of no medieal conditions which would limit my participation in the
program. Should this change during the course of the program, I will immediately notify my
program leader.

The Healthy-U Program will keep all individual data strictly conﬁdential. I consent that my
result may be used in research projects which are to improve the efﬁeacy and quality of this
Program. All data will be reported in aggregate form with no individuals identiﬁed. My
consent to the research project is completely voluntary without affecting my enrollment in the
Program. I ean withdraw at any time from the research projects or the Program. I ean request
the Healthy-U Program any information regarding the projects or the Program at any time.

Name (Print) :

 

(Signature) : Date:

Address:

 

 

Phone: circle (Home or Work)

MSU is an Afﬁrmative Action/Equal 0pp0.".'.:".:ty Inslllullon

 

APPENDIX D

THREE-DAY DIET RECORD FORM

92

THREE - DAY DIET RECORDS

Accurate assessment of current dietary habits is important to
identify any changes to make. Please record all the food and drink
you consume on two typical weekdays and one weekend day. This
record should not reflect any modification of your current dietary
habits.

Be sure to include:
‘ 1) beverages: water, milk, soft drinks, juice, tea, coffee,
alcoholic beverages, etc.
2) condiments: butter, margarine, mayonnaise, catsup, mustard,
pickle relish, cream jelly, sauce, etc.
3) method of preparation: fried, baked, boiled, broiled , etc.
4) anything added during preparation :oils, milk, wine etc.
5) for combination foods, list all ingredients as accurately
as possible.

BE SURE TO ESTIMATE QUANTITY OF FOOD AS CLOSE AS POSSIBLE.
Describe portion sizes by ounces, cups, tablespoons, etc. rather
than a glass of milk.

1 cup = 8 ounces (fluid)
1 Tablespoon - 3 teaspoons
1/4 pound =4 ounces (weight)

Please r' 't e te 3-d d'et records to our second ass.
You will receive your diet analysis results with recommendations in-
the following class.

93

NAME: W2

 

Healthy-U ID # 620 Z

 

 

 

 

 

 

 

 

 

 

 

 

ITEM AND DESCRIPTION PORTION
MeaH (bud. C036“— 3W“ I 0055““ “‘19
0m?- galley, WW4. {/9- 0*?
W4. 5/4 Cup
“Wm-w. / at.“ (a 0L)
Joni: , obi» 0W / Alia.
MW, and, 1 +59.
0666“» c
Snack(s) u/ 0mm 6: W ifféffeaif
3% Maxi,
Meal 2 FBW, who!» W <3 ”(AW
Dmpaokzd w (.3an a oz.
Ma mam / +fp.
L WW4" W, (1:74:24 I ‘hp. mcﬂc
PW.MW I comma 02.)
714W Wﬁ (Dania-4.) a% 0i. (smuzl 5%)
Snack(s) W / W
o , _ .. . . . . a .
Mea13 ckéqux Moo/outdo”. I W
Ra’s, sme A Cu!
WCA UK? M V3- CW?
' (std) .2 +sp.
W(JNM£¢~.W¢MW) [.cup
' ( 3%, W) I TBSP.
#01:" J 35AM LEM.)
Snack(s)
goﬁan’f’ S'Muw (pl/WM) 8 02°
Vitamin/Mineral Supplement(s) 421$,
Is this a complete one-day intake? _1_ Yes No

WasthisaTYPlCAL day? __sk_ Yes __ N0

 

 

 

94

The following information are needed for the analysis of dietary intakes

 

Age : Gender :
Height : feet inches
Weight : lb

Type of Physical Activity

Time (Hour) Type of Physical Activity
Sleeping

Light activity
ex. studying, typing, cooking, driving. swimming,
walking

Moderate activity
ex. sailing, dancing, chid we, golf, swimming,
walking fast (3 mph)

Heavy activity

ex. basketball, aerobics, running, siding,
rowing, cycling, tennis

** The total of physical activity should be 24 hours in one day

95

(DAY 1)

NAME: Healthy-U ID #

 

 

ITEM AND DESCRIPTION PORTlON

 

Meal 1

 

Snack(s)

 

Meal 2

 

 

 

 

 

 

 

 

Snack(s)

 

 

 

Vitamin/Mineral Supplement(s)

 

ls this a complete one-day intake? Yes No

Was this a TYPlCAL day? Yes No

 

 

96

(DAY 2)
NAME:

 

Healthy—U ID #

 

 

'l‘ lTEM AND DESCRIPTION PORTlON

 

 

Snack(s)

 

Meal 2

 

Snack(s)

 

Meal 3

 

Snack(s)

 

 

 

 

 

Vitamin/Mineral Supplement(s)

 

ls this a complete one-day intake? Yes No

Was this a TYPICAL day? Yes No

 

 

 

97

(DAY 3
NAME:) Healthy—U ID #

 

ITEM AND DESCRIPTION PORTION

 

Meal 1

 

Snack(s)

 

 

Meal 2

 

Snack(s)

 

Meal 3

 

Snack(s)

 

 

 

Vitamin/Mineral Supplement(s)

 

Yes No
Was this a TYPlCAL day? Yes No

ls this a complete one-day intake?

 

 

APPENDIX E

PSYCHOSOCIAL FACTORS QUESTIONNAIRE

98

Healthy U Diet/Weight Management Program

 

 

Name: Healthy U ID:
strongly strongly
agree disagree

1. Eating a lot of fruits and vegetables decreases my chances
of getting serious diseases like heart disease or eancer. 1 2 3 4 5

2. Eating a lot of whole grain, bread, and cereals decrease
my chances of getting serious disease like heart disease 1 2 3 4 5
or cancer.

3. Eating a lot of fried foods increases my chances of
deve10ping serious illnesses like heart disease or cancer. 1 2 . 3 4 5

4. It is hard for me to get fruits and vegetables when I’m at
work.

 

5. There is so much advice about health ways to eat. I

don’t know what is good or bad 1 2 3 4 5
6. What I eat is one of the most important things for my 1 2 3 4 5
health.
7. Low fat foods taste good. 1 2 3 4 5'
8. There is a lot of information on health eating where I 1 2 3 4 5
work.
9. At my workplace, it’s easy to eat a healthy diet. 1 2 3 4 5
very
much none
10. How much encouragement for eating low-fat foods do
you get from your co-workers? 1 2 3 4 5
11. How much encouragement for eating low-fat foods do
you get from friends and‘family? 1 2 3 4 5
very
important none

12. How important to you is eating low-fat foods? 1 2 3 4 5

 

99

13. How conﬁdent are you that you will decrease the
amount of fat in your diet during the next 6 months?

14. How confident are you that you will eat more fruits and
vegetables in your diet during the next 6 months?

15. How conﬁdent are you that you will eat more whole
grain, bread, and cereals in your diet during the next 6
months?

16. How high in fat is your overall diet?

* If low or very low, please answer this one
How long have you followed a diet that is low fat?
(circle one)
s 6 months 1-6 month 6-12 month _>_ 12 months

17. How high in ﬁber is your overall diet?

* If high or very high, please answer this one
How long have you followed a diet that is high in
ﬁber?

(circle one)
_<_6 months 1-6 month 6-12 month _>_12 months

18. Over the next 6 months, do you plan to cut down on
fats?

19. Over the next 6 months, do you plan to eat more fruits
and vegetables?

20. Over the next 6 months, do you plan to eat more whole
grain, bread, and cereals?

very
conﬁdent
I 2
1 2
1 2
very
high
1 2
very
high
1 2
deﬁnitely
yes
1 2
1 2
1 2

not

conﬁdent

3 4 5

3 4 5

3 4 5
very
lo“!

3 4 5
very
low

3 4 5

deﬁnitely
no

3 4 5

3 4 5

3 4 5

 

 

100

21. Have you tried to make any changes to lower the fat in
your diet in the past 6 months?

Yes N o
extremely not

WOCCSS success
“ If yes, please answer this one 1 2. 3 4 5

How successful were you in making those changes?

22. Have you tried to make any changes to increase the
fiber in your diet in the past 6 months?

Yes No
extremely not
SUOCCSS success
' If yes. please answer this one
How successful were you in making those changes? 1 2 3 4 5
23. Have you ever tried to lose 10 pound or more? Yes No

If yes, please answer this one

Think about your most recent effort to lose weight.
How would you describe the results?

Lost all I want to and kept it off

Lost part of the weight I wanted to and kept is off
Lost weight, but gained some of it back

Lost weight, but gained all of it back

Didn’t lose any weight

Still on a diet now

Other:

9999939!"

 

24. In the past 2 years (except pregnancy), how many times has you weight gone up and
down by a1 least 8 to 10 pounds?

times

101

25. If you were trying to choose more low-fat foods, which food in each of the following
pairs would you select because it was lower in fat?

 

Pair I Pair II Pair III

a. saltines/soda crackers a.margarine a. potato chips
b. Ritz crackers b. butter b. pretzels

c. neither one c. neither one c. neither one
d. don’t know (1. don't know d. don’t know

26. If you were trying to choose more high-ﬁber foods, which food in each of the following
pairs would you select because it was higher in ﬁber?

 

Pair I Pair 11 Pair III

a. chile w/beans a. bran mufﬁn . a. canned pears
b. spag. w/meat balls b. bran cereal b. stewed prunes
c. neither one 0. neither one c. neither one

d. don’t know d. don‘t know d. don’t know

APPENDIX F

STAGES OF CHANGE FOR FAT: ALGORITHM

102

STAGES OF CHANGE FOR FAT: ALGORITHM

 

Stages Deﬁnition Items Used

Maintenance Healthy diet for > 6 months Self-rated diet

Action Healthy diet for < 6 months or Self-rated diet
tried to changes with some Reported changes

success in the last 6 months

Preparation Tried to make healthy diet Self-rated diet
changes in last 6 months but Reported changes
not successful, or

Deﬁnitely plan to change

Contemplation Maybe/probably plan to change Self-rated diet.
diet in the next 6 months; Behavior intentions to change diet
and no attempts to change in Reported changes

the last 6 months

Precontemplation No plan to change diet in the next Self-rated diet
6 months;and no attempts to Behavior intentions to change diet
in the last 6 months Reported changes

 

‘ Assignment to stage is done sequentially, beginning with maintenance. Once an individual is
assigned to a stage, the remaining response are not processed.

' Healthy dieF low/very low fat, or high/very high ﬁber

APPENDIX G

FEEDBACK ON DIETARY INTAKE AND BLOOD LIPIDS

103

MSU NutriGuide

NUTRITIONAL ANALYSIS FOR“ ON 00/03/99

ID (STUDENT) NUMBER:
SEX: F

0000000

This analysis is based on 42 years of age, 5 feet 6

and 116 pounds.

ENERGY EXPENDITURE

RESRING ENERGY EXPENDITURE *

TIME
(hour)

inches,

KCAL
(needed)

1284

SLEEPING ..................................

\l
O

O

LIGHT ACTIVITY .........
ex. studying, typing, cooking,driving,
watch TV, walking

422

H
m
0

H
O

MODERATE ACTIVITY ................. ........
ex. sailing, dancing, child care, golf,
swimming, walking fast (3 mph)

79

HEAVY ACTIVITY ....................
ex. basketball, aerobics, running,
skiing, rowing, cycling, tennis

00......

O
O
.— c— .— e— o— .- u— e— .- o— a— .— ._ ._ .— .-

c— o— a- a- e— o— o— 0-

 

DAILY TOTAL ENERGY EXPENDITURE ** 1963

* Resting Energy Expenditure refers to the energy needed to sustain

life at rest. ** Daily total energy expenditure includes
thermogenesis.

LIST OF FOODS CONSUMED

 

FOOD DESCRIPTION

 

MEAL AMOUNT
B MEDIUM BANANA 1 each
B CHEERIOS 1 cup
B SKIM MILK NONFAT .5 cup
3 DIET COKE W/ASPARTAME 12 fl oz
L BLUEBERRY MUFFIN .6 each
L CHICKEN SALAD .5 cup
L NECTARINE, MEDIUM 1 each
5 DIET COKE W/ASPARTAME 12 fl oz
3 MILK CHOCOLATE BAR 1 each
D ICEBERG LETTUCE-RAW,SHRED 1 cup
D MEDIUM RED TOMATOES, RAW 3 slice
D ITALIAN DRESSING(LOW CAL) .66 Tbsp
D LOCAL VEAL PARMIGIANA+VEG 1 serv

 

DIET ANALYSIS

104

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

NUTRIENT AMOUNT RDA
NAME CONSUMED % 0 50 100 150

TOT-PROT 50 gm 100 1

TOT-FAT 56 gm 76 *1

POLY-FAT 17 gm 70 *1

SAT-FAT 19 gm 78 *t

CHOLESTROL 13 8 mg 4 6 * ! ==============
ALCOHOL 0 gm --- i

VIT—A 1003 RE 125 I

VIT‘D 3 IU 2 1

VIT-E 13 mg 163 1

THIAMIN 0.9 mg 82 !

RIBOFLAVIN 1.6 mg 123 I

NIACIN 17 mg 113 I

PYRIDOXINE 2.0 mg 125 I

VIT-BlZ 3.2 ug 160 I

VIT-C 65 mg 108 1

IRON 10 mg 67 l

CALCIUM 592 mg 74 !

PHOSPHORUS 987 mg 123 1

POTASSIUM 2296 mg —-- l

MAGNESIUM 213 mg 76 1

Z I N C 7 mg 5 8 I =================
CAFFEINE 111 mg --— 1

 

l I I
v I

The Recommended Dietary Allowance (RDA) is a daily nutrient intake

believed to be appropriate for practically all healthy Americans of your
age and sex group.

other than the RDA.

Nutrients flagged with “*" are based on recommendations

105

NUTRITIONAL ANALYSIS SUMMARY

ENERGY INTAKE vs ENERGY OUTPUT

Based on your Restingl Energy Expenditure (calorie expended at rest),
energy for activity and thermogenesis, your body expends 1963 kcal/day.

Your diet analysis reveals that you consumed 1177 kcal on 00/03/99
From this, you can see that you have consumed 786 fewer calories
than your body requires to maintain current weight.

DESIRABLE BODY WEIGHT RANGE

Your desirable weight range is 117 - 143 lb.

ENERGY BALANCE

If you continue to consume fewer calories than your body needs,
weight loss will occur.

You are currently less than 90% desirable body weight and have consumed
fewer calories than your body needs. This is considered to be unhealthy.

ENERGY SOURCES

40% of the calories in your diet (excluding alcohol) came from
carbohydrate, 17% from protein, and 43% from fat. Alcohol
provided 0 kcal.

Current recommendations suggest that your diet consist of 55% or more
Of the total calories from carbohydrates, 15% from protein, and 30% or

less from fat. It is suggested that no more than 150 calories from
alcohol be consumed per day.

106

Healthy U Diet/Weight Management Program
Mini Health Fair, Spring Semester
January, 1995

Name:

 

Biochemical Measurement

 

 

 

0 Total cholesterol : mg/dl

0 HDL - cholesterol : mg/dl

0 LDL - cholesterol : - mg/dl

0 Triglyceride : _ mg/dl
Blood Pressure : mmHg

 

Anthropometric Measurement

0 Body Weight : lb
0 Percentage of body fat : %

0 Waist/Hip ratio :

HealthyUII)#:

 

<200 (desirable)
200-239 (borderline high)
2340 (high risk)

_>_60 (negative risk factor)
.535 (high risk)

< 130 (desirable)
130-159 (borderline high)
2160 (high risk)

<200 (desirable)
200-399 (borderline high)
2400 (high risk)

_>_ 140/90 mmHg (high risk)

Female > 0.8 (high risk)
Male > 1.0 (high risk)

 

BIBLIOGRAPHY

Bibliography

Ajzen I, Fishbein M. Understanding attitudes and predicting social behavior. Englewood
Cliffs, NJ :Prentic-Hall. 1980.

Alaimo K, McDowell MA, Briefel R, et al. Dietary intake of vitamins, minerals, and
ﬁber of persons ages 2 month and over in the united states: Third National Health and
Nutrition Examination Survey, Phase 1, 1988-1991.Advance data from vital and health
statistics; No.255. Hyattsville, Maryland: National Center for Health Statistics. 1994.

Angotti C, Levine MS. Review of 5 years Of a combined dietary and physical ﬁtness
intervention for control Of serum Cholesterol. J Am Diet Assoc 1994; 94:634-638.

Baer J T. Improved plasma cholesterol levels in men after a nutrition education program at
the worksite. J Am Diet Assoc 1993; 93:658-663.

Bandura A. Social foundations of though and action: A social cognitive theory.
Englewood cliffs, NJ: Prentice-Hall. 1986.

Baranowski T. Beliefs as motivational inﬂuences at stages in behavior change.
International Quarterly of Community Health Education 1992; 13 :3-29.

Barrett A, Reznik R, Irwig L, Simpson JM, Oldenburg B, Horvath J, Sullivan D. Work-
site Cholesterol screening and dietary intervention: The Staff Healthy Heart Project. Am J
Public Health 1994;84:779-782.

Berg FM. 1993 special report: Health risks Of Obesity. Hettinger, ND: Obesity & Health.
Borrows ER, Henry H], Bowen DJ, Henderson MM. Nutritional applications Of a
clinical, low fat dietary intervention to public health change. J Nutr Educ 1993; 25:167-
175.

Briley ME, Montgomery DH, Blewett J. Worksite nutrition education can lower total

cholesterol levels and promote weight loss among police department employees. J Am
Diet Assoc 1992; 92: 13 82-1384.

107

 

 

108

Brownell KD, Merlatt GA, Lichenstein E, Wilson GT. Understanding and preventing
relapse. Am Psychol 1992; 41 :765-782.

Brug J, Assema PV, Kok G, Lenderink T, Glanz K. Self-rated dietary fat intake:
association with objective assessment of fat, psychosocial factors, and intentions to
change. J Nutr Educ 1994; 26:218—223.

Bruno R, Arnold C, Jacobson L, Winick M, Wynder E, Randomized controlled trial of a
non-pharmacologic cholesterol reduction program at the worksite. Prev Med 1983;
12:523-532.

Byers T, Mullis R, Anderson J. The costs and effects of a nutritional education program
following work-site cholesterol screening. Am J Public Health 1995; 85:650-655.

Campbell RS, Devellis BM, Stretcher VJ, Ammerman AS, Sandler RS. The impact of
message tailoring on dietary behavior change for disease prevention in primary care
settings. Am J Public Health 1994; 84:783—787.

Carmody T, Senner J, Manilow M, Matarazzo J. Physical exercise rehabilitation: long-
term dropout rate in cardiac patients. J Behav Med 1980; 12237-267.

Contento 1, Murphy BM. Psycho-social factors differentiating people who reported
making desirable changes in their diets from those who did not. J Nutr Educ 1990; 22:6-
14.

Curry SJ, Kristal AR, Bowen D. An application of the stage model of behavior change to
dietary fat reduction. Health Educ Res 1992; 7297-105.

Davis MA, Ettinger WH, Neuhaus JM, Hauck WW. Sex differences in osteoarthritis of
the knee: The role of obesity. Am J Epi 1988; 127:1019-1033.

Dishman R. Exercise adherence (overview). Champaign, Illinois: Human Kinetics Books.
1988: 1-9.

Dittus KL, Hillers VN, Beerman KA. Beneﬁts and Barriers to fruit and vegetable intake:
relationship between attitudes and consumption. J Nutr Educ 1995; 27: 120-126.

Doll R, Peto R. The causes of cancer: quantitative estimates of avoidable risks of cancer
in the United State today. J Cancer Insti 1981; 66: 1 191-1308.

Ferrini R, Edelstein S, Barrett-Connor E. The association between Health Beliefs and
Health Behavior change in older adults. Prev Med 1994; 23: 1-5.

109

Fisher SP, Fisher MM. Development, implementation, and evaluation of a health
promotion program in a college setting. Am J Coll Health 1995; 44:81-83.

Foster JL, Jeffery RW, Snell MK. One year follow-up study to a worksite weight control
program. Prev Med 1988; 17:129-133

French SA, Jeffery RW, Forster JL, Mc Govern PG, Kelder SH, Baxter JE. Predictors of

weight change over two years among a population of working adults: the Healthy Worker
Project. Int J Obes 1994; 18:145-154.

Fullmer S, Ceiger CJ, Parent CR. Consumers’ knowledge, understanding, and attitudes
toward health claims on food labels, J Am Diet Assoc 1992; 91 :166-171.

Garner DM, Wooley SC. Confronting the failure of behavioral and dietary treatments for
obesity. Clin Psychol Rev 1991; 11:729-780.

Glanz K, Kristal AR, Sorensen G, and PaLombo R, Heimendinger J, Probart C.
Development and validation of measures of psychosocial factors inﬂuencing fat-and
ﬁber-related dietary behavior. Prev Med 1993; 22:373-387.

Glanz K, Patterson RE, Kristal AR, Diclemente CC, Heimendinger J. Linnan L, and
McLerran DE. Stages of change in adopting healthy diets: fat, ﬁber and correlates of
nutrient intake. Health Educ Res 1994; 21 :499-519.

Glanz K, Seewald-Klein T. Nutrition at the worksite: an overview. J Nutr Educ 1986;
18:81-812.

Goldstein DJ. Beneﬁcial health effects of modest weight loss. Int J Obes 1992; 16:397-
415.

Gorbach SL, Marl-LaBrode A, Woods MN, Dwyer JT, Selles WD, Henderson M, Insull
W, Golderman S, Thompson D, Clifford C, Sheppard L. Changes in food patterns during
a low-fat dietary intervention in women. J Am Diet Assoc 1990; 90:802-809.

Greene WS, Rossi SR, Reed GR,Willey CW, and Prochaska JP. Stages of change for
reducing dietary fat to 30% of energy or less. J Am Diet Assoc 1994; 94:1105-1110.

Greene GW, Strychar 1. Participation in a worksite cholesterol education program in a
university setting. J Am Diet Assoc 1992; 92:1376-1381.

Haffner SM, Diegl AK, Stern MP, Hazuda HP. Central adiposity and gallbladder disease
in Mexican Americans. Am J Epi 1989; 129:587-595.

Harris J K, French SA, Jeffery RW, McGovern PG, Wing RR. Dietary and physical

110

activity correlated of long-term weight loss. Obes Res 1994; 2:307-313.

Hartman TJ, McCarthy PR, Himes JH. Use of eating-pattem messages to evaluate change
in eating behaviors in a worksite cholesterol education program. J Am Diet Assoc 1993;
93:1119-1123.

Haus G, Hoerr SL, Marvis B, Robison J. Key modiﬁable factors in weight maintenance:
fat intake, exercise and weight cycling. J Am Diet Assoc 1994; 94:409-413.

Hebert JR, Harris DR, Sorensen G, Stoddard Am, Hunt MD, Morris DH. A work-site
nutrition intervention: its effects on the consumption of cancer-related nutrients. Am J
Public Health 1993; 83 :391-394

Hembroff L, Huang YL. 1993-1994 Healthy U Survey: summary of ﬁndings. Institute for
Public Policy and Social Research, Michigan State University 1995; January: 5

Hovell MF, Koch A, Hofstetter CR, Sipan C, Faucher P, Dellinger A, Borok G, Forsythe
A, Felitti VJ. Long-term weight loss maintenance: assessment of a behavioral and
supplemented fasting regimen. Am J Public Health 1988; 78:663-666.

Hubert HH, Feinlieb M, McNamara PM, Castelli WP. Obesity as an independent risk
factor for cardiovascular disease: a 26-year follow-up of participants in the Framingham
Heart Study. Circulation 1983; 67:968-977.

Huang YL, Song W0, Ball M. Characteristics of participants who do not complete
worksite health promotion program. FASEB Abstract 1996; 9:A255.

Hunt MK, Hebert JR, Harris DR, Heisen E, Morris DH, Stoddard AM. Impact of a

worksite cancer prevention program on eating pattern of workers. J Nutr Educ 1993;
25:236-244.

Ippolito PM, Mathios AD. The regulation of science-based claims in advertising, in
Mayer RN (ed.): Enhancing consumer choice; proceedings of the second international
conference on research in the consumer interest. Columbia, MO: American Council on
consumer interests, 1991 .

Johnston JM, Jansen R, Anderson J, Kendall P. Comparison of group diet instruction to a
self-directed education program for cholesterol reduction. J Nutr Educ 1994; 26:140-145.

Klesges RC, Klesges LM, Hadock CK, Eck LH. A longitudinal analysis of the impact of

dietary intake and physical activity on weight change in adults. Am J Clin Nutr 1992;
55:818-822.

Koemer L, Song W0. Dietary quality and dietary changes of EFNEP participants:

 

111

Michigan statewide EFNEP study. (In preparation)

Kristal AR, SeLett AC, Henry AJ, Fowler AS. Rapid assessment of dietary intake of fat,
ﬁber and saturated fat: validity of an instrument suitable for community intervention
research and nutritional surveillance. Am J Health Promot 1990; 4;288-295.

Kristal AR, Patterson RE, Glanz K, Heimendinger J, Herbert JR, Feng Z, Probart C.
Psychosocial correlates of healthful diets: baseline results from the working well study.
Prev Med 1995; 24:221-228.

Kristal AR, Stattuch A. Patterns of dietary behavior associated with selecting diets low in
fat: reliability and validity of a behavioral approach to dietary assessment. J Am Diet
Assoc 1990; 90:214-220.

Kris-Etherton PM, Krummel D, Russell ME, Dreon D, Mackey S, Borchers J, Wood PD.
The effect of diet on plasma lipids, lipoproteins and coronary health disease. J Am Diet
Assoc 1988; 88:1373-1400.

Laforge RG, Greene GW, Prochaska J. Psychosocial factors inﬂuencing low fruit and
vegetable consumption. J Behav Med 1994; 17:361-374.

Lovibond SH, Birrell PC, Langeliddecke P. Changing coronary heart disease risk factor
status: the effects of three behavioral programs. J Behav Med 1986; 9:415-43 6.

Lohman T, Roche A, Martorell R. Anthropometric standardized reference manual.
Champaign, III: Human Kinetics Books, 1988.

Lukaski HC. Methods for measurement of human body composition, traditional and
new. Am J Clin Nutr 1987; 46:537-556.

Marcus BH, Banspach SW, Lefebvre RC, Rossi J S, Carleton RA, Abrams DB. Using the
stages of change model to increase the adoption of physical activity among community
participants. Am J Health Promot 1992;6z424-429.

McDowell MA, Briefel RR, AlaimoK, et al. Energy and macronutrient intakes of
persons ages 2 month and over in the united states: Third National Health and Nutrition
Examination Survey, Phase 1, 1988-1991. Advance data from vital and health

statistics; No.255. Hyattsville, Maryland: National Center for Health Statistics. 1994

Masur-Levy P, Tavris DR, Elsey-Pica L. Cardiovascular risk changes in a work-site
health promotion program. J Am Diet Assoc 1990; 90:1427-1428.

National Center for Health Statistics, Naj jar MF, Rowland M. Anthropometric reference
data and prevalence o overweight, United States, 1976-1990. Vital and Health Statistics.

112

Series 11, no. 238 Washington, DC: Public health service, 1987 (DHHS Publication No.
(PHS) 87-1688).

National Center for Health Statistics, Abraham S, CarrollMD, Najjar MF, Fulwood R.
Obesity and overweight adults in the United States, 1976-1990. Vital and Health
Statistics. Series 11, no. 230 Washington, DC: Public health service, 1983 (DHHS
Publication No. (PHS) 83-1680).

National Cholesterol Education Program. Summary of the second report of the NCEP
expert panel on detection evaluation and treatment of high blood cholesterol in adults
(ATP 11). JAMA 1993; 269:3015-3023.

National Institutes of Health Consensus Development Panel. Health implications of
obesity: National Institutes of Health Consensus Development Conference Statement.
Annals of Internal Medicine 1985; 103:1073-1077.

Patterson RE, Kristal AR, White B. Do beliefs, knowledge and perceived norms about
diet and cancer predict dietary change? Am J Public Health 1996; 86: 1 394-1400.

Patterson RE, Kristal AR, Lynch J C, White B. Diet-caner related beliefs, knowledge,
norms and their relationship to healthful diets. J Nutr Educ 1994; 27:86-92.

Pelletier KR. A review and analysis of the health and coast-effective outcome studies of
comprehensive health promotion and disease prevention programs at the work-site: 1991-
1993; update. Am J Public Health 1993; 8:50-62.

Perovich SJ, Sandoval WM. Outcomes of a worksite cholesterol education program over
a 5-year period. J Am Diet Assoc 1995; 95:589-590.

Prochaska J, DiClemente C. Stage and Processes of self-change on smoking: toward an
integrative model of change. J Consult Clin Psychol 1983; 51:390-395

Prochaska J, DiClememente C. Transtheoretical therapy: toward a more integrative model
of change. Am Psychol 1992; 47:1102-1114.

Rosenstock IM. The Health Belief model: Explaining health behavior through
expectancies. In: Glanz K, Lewis FD, Rimer BK. Health Behavior and health education
theory, research and practice. San Francisco: Jossey-Bass, 39-60, 1990

Schucker B, Wittes JT, Santanello NC, Weber SJ, McGoldricj D, Donate k, levy A,
Riﬂdnd BM. Change in cholesterol awareness and action: results from national physician

and public surveys. Arch intern Med 1991; 151:666-672.

Shannon B, Hendricks M, Rollins P, Schwartz RM. A comprehensive evaluation of a

113

worksite nutrition and weight-control program. J Nutr Educ 1987;91:109-116.

Shepherd R, Stockley L. Nutrition knowledge, attitudes, and fat consumption. J Am Diet
Assoc 1987; 87:615-619.

Smith AM, Katrine B, Owen N. Socioeconomic status and personal characteristics as
predictors of dietary change. J Nut Educ 1995; 27;173-181

Smith A, Owen N. Associations of social status and health-related beliefs with dietary fat
and ﬁber density. Prev Med 1992; 21 :73 5-745.

Sorensen G, Thompson B, Glanz K, Feng Z, Kinne S, DiClemente C, Emmons K,
Heimendinger J, Probart C, Lichtenstein E. Work site-based cancer prevention: primary
results form the working well trial. Am J Public Health 1996; 86:939-947.

Sorensen G, Morris DM, Hunt MK, Hebert JR, Harris DR, Stoddard A, Ockene JK.

Work-site nutrition intervention and employees’ dietary habits: The Treatwell Program.
Am J Public Health 1992; 6:877-880.

Spomy LA, Contento IR. Stages of change in dietary fat reduction: social psychological
correlates. J Nutr Educ 1995; 27:191-199.

Staﬂeu A, Graaf C, Staveren WA. Attitudes towards high-fat foods and their lowfat
alternatives: vreliability and relationship with fat intake. Appetite 1994; 22: 1 83-196.

US Dept. Of Agriculture. Composition of Food: Raw, Processed, Prepared Agric,
Handbook No.8. AH 8-1 Dairy and Egg Products, 1976; AH 8-2, Spices and Herbs, 1977;
AH 8-3 Baby Foods, 1978; AH 8-4, Fat and oils, 1979; AH 8-5, Poultry, 1979; AH 8-6,
Soup, Sauces and Gravies, 1980; AH 8-9, Fruits and Fruit Juices, 1982; AH 8-10,
Breakfast Cereal, 1988; AH 8-11, Vegetables and Vegetable Products, 1984; AH 8-12,
Nut and Seed Products, 1984;AH 8-13, Beef Products, 1986; AH 8-14, Beverages, 1986;
AH 8-15, Fish and Shellﬁsh Products, 1989; AH 8-16, Legumes and Legume Products,
1986; AH 8-17, Lamb, Veal and Game Products, 1989; AH 8-20 Cereal, Grains and
Pasta, 1989; AH 8-21 Fast Foods, 1989; 1989 supplement, 1990. Washington, DC.

US Department of Health and Human Services. National survey of worksite health
promotion activities, summary report, 1992

US Department of Labor. “Employment and Earnings”. Bureau of Labor Statistics, 1992

Winkleby MA, Flora JA, Kraemer HC. A community-based heart disease intervention:
predictors of change. Am J Public Health 1994; 84:767—772.

Willett W. Short-term dietary recall and recording methods. In: Nutritional

114

Epidemiology. NY; Oxford University Press Inc, 1990.

Wooley SC, Garner DM. Obesity treatment: The high cost of false hope. J Am Diet
Assoc 1991;91:1248-1251.

 

 

HHHHHHHH LIBRRRIES

111111111111l1116111111111

 

312930170 9976