HEALTH EDUCATION RESEARCH Vol.20 no.

4 2005
Theory & Practice Pages 439–447
Advance Access publication 14 December 2004

Innovative techniques to address retention in

a behavioral weight-loss trial

Jennifer H. Goldberg1,2 and Michaela Kiernan1

Abstract Introduction

Given that retention rates for weight-loss trials Based on the 1999–2000 National Health and
have not significantly improved in the past 20 Nutrition Examination Survey, 64.5% of US adults
years, identifying effective techniques to en- are overweight or obese (Flegal et al., 2002). To
hance retention is critical. This paper describes validly test the efficacy of long-term obesity treat-
a conceptual and practical advance that may ments, randomized controlled trials must have
have improved retention in a behavioral weight- minimal participant dropout (Hansen et al., 1985;
loss trial—the novel application of motivational Ribisl et al., 1996; Ware, 2003). However, reten-
interviewing techniques to diffuse ambivalence tion over time is challenging (Wilson and Brownell,
during interactive group-based orientation ses- 1980; Brownell and Wadden, 1992). Across be-
sions prior to randomization. These orientation havioral weight-loss treatment studies, 32% of
sessions addressed ambivalence about making participants drop out (Davis and Addis, 1999).
eating and exercise behavior changes, ambiva- Given that retention rates for behavioral weight-loss
lence about joining a randomized controlled trials have not significantly improved in the past
trial, and unrealistic weight-loss expectations. 20 years (Wilson and Brownell, 1980; Brownell
During these sessions, overweight and obese and Wadden, 1992), identifying novel techniques
men and women learned about the health that improve participant retention is a critical pri-
benefits of modest weight loss as well as trial ority (Jeffery et al., 2000).
design, the importance of a control condition, Ambivalence, defined as ‘simultaneous and
random assignment and the impact of dropouts. contradictory attitudes or feelings (as attraction
Participants were then divided into groups of and repulsion) toward an object, person, or action’
three or four, and asked to generate two pros (Mish, 1990), is thought to undermine behavior
and two cons of being assigned to a control change. Motivational interviewing is ‘a directive
condition and an active condition. Participants client-centered counseling style for eliciting behav-
shared their pros and cons with the larger ior change by helping clients to explore and resolve
group, while the investigator asked open-ended ambivalence’ (Rollnick and Miller, 1995). One
questions, engaged in reflective listening and motivational interviewing technique is to build
avoided taking a ‘pro-change’ position. Reten- upon a decisional balance exercise (Janis and
tion was high, with 96% of the participants (N 5 Mann, 1977; Prochaska and DiClemente, 1983;
162) completing 18-month clinic visits. Miller and Rollnick, 1991; Prochaska et al., 1992,
1994; Miller and Rollnick, 2002) by making any
1 existing ambivalence explicit, and normalizing it
Stanford Prevention Research Center, Stanford University
School of Medicine, Stanford, CA 94305-5705, USA
using open-ended questions and reflective listening
2
Correspondence to: J. H. Goldberg; to acknowledge that the pros and cons exist
E-mail: jennifer.goldberg@stanford.edu simultaneously and may be contradictory (Miller

Health Education Research Vol.20 no.4, Ó Oxford University Press 2004; All rights reserved doi:10.1093/her/cyg139
Downloaded from https://academic.oup.com/her/article-abstract/20/4/439/632665
by Sandy Lemmon user
on 09 November 2017
 J. H. Goldberg and M. Kiernan

and Rollnick, 1991). This is especially effective high retention rates in a recently completed random-
when the counselor avoids taking or defending the ized behavioral weight-loss trial—the novel appli-
‘pro-change’ position (e.g. reinforcing pros and cation of motivational interviewing techniques to
problem-solving cons with participants) and, thus, diffuse ambivalence during orientation sessions
avoids provoking participants to take on the ‘status prior to randomization. In interactive small group
quo’ position [(Miller and Rollnick, 1991), p. 47]. orientation sessions, we addressed ambivalence
Recently, motivational interviewing techniques about joining a randomized controlled trial, ambiva-
have been used to diffuse ambivalence and increase lence about making eating and activity behavior
motivation during the intervention phase of weight- changes, and unrealistic weight-loss expectations.
loss, diet and physical activity trials (Smith et al., We explain how the orientation sessions were con-
1997; Resnicow et al., 2002). ducted, present a content analysis of participants’
In this paper, we speculated that these techniques responses during the sessions and report the trial
could be applied to diffuse ambivalence prior to trial retention rates. Given the current obesity epidemic,
randomization to improve retention. Participants the crucial importance of retention in weight-loss
may be ambivalent about being in a randomized research and the lack of even descriptive research on
weight-loss trial (Burke et al., 2003). Participants retention techniques in the weight-loss field, we
may see the benefits of joining a trial, such as free believe that this descriptive paper will be of value for
treatment and additional support, while at the same guiding practice and provoking further research on
time perceive limitations, such as being randomly effective retention strategies.
assigned to a control condition. Participants may
even plan to drop out of the trial if they are assigned
to the control condition. In one recent behavioral Method
weight-loss trial, a higher percentage of the control
condition (41%) did not complete follow-up clinic Overall design of original randomized
visits compared to the two active conditions (14 and trial
23%) (Ciliska, 1998). In addition, participants may The Stanford Healthy Weight Project is a random-
be ambivalent about whether they really want to ized weight-loss efficacy trial that recruited over-
change their eating and activity, resent or resist being weight and obese adults (ages 25–80 years; BMI
told to make these behavioral changes, and drop out 27–37 kg/m2) in a major metropolitan area who
(Windhauser et al., 1999; Sparks et al., 2001; wanted to lose a modest amount of weight (10–15
Shepherd, 2002). For example, participants in a pounds). Eligibility characteristics were similar to
controlled feeding study reported that ‘not having to other behavioral weight-loss studies (Davis and
shop for and cook food’ aided their compliance Addis, 1999) and are summarized in Table I (Beck
while at the same time the ‘lack of freedom to choose et al., 1961; Block et al., 1986; Pate et al., 1995;
what/when to eat’ challenged their compliance Stice et al., 2000).
(Windhauser et al., 1999). Finally, participants may Participants were randomly assigned to one of
experience ambivalence because of a contradiction three study conditions: a control condition or one of
between their initial (and unrealistic) expectations two active behavioral weight-loss conditions. The
that they will lose a lot of weight and their dissat- control condition was allowed to enroll in any
isfaction with the amount of weight they are actually behavioral treatment programs (e.g. Weight Watch-
losing (Bennett, 1986; Foster et al., 1997; Jeffrey ers) available in the community that did not include
et al., 1998; King et al., 2002; Wadden et al., 2003), medication or very-low calorie diets. Both active
prompting participants to drop out of active condi- conditions attended 14 weight-loss classes for the
tion classes or not return for follow-up clinic visits. first 6 months of the trial without continued contact
This paper describes a conceptual and practical after classes ended. All three conditions were asked
advance that we speculate may have contributed to to attend four clinic visits, one every 6 months for

440
Downloaded from https://academic.oup.com/her/article-abstract/20/4/439/632665
by Sandy Lemmon user
on 09 November 2017
 Innovative techniques to address retention

18 months. Participants were recruited in three Participants

cohorts via local radio stations, and articles in local Of the 72 potential participants who attended an
community, regional and university-based news- orientation session in Cohort 1, 51 participants
papers (Kiernan et al., 2001). (71%) were randomized to the trial. Of the 21
participants not randomized, 14 did not schedule or
Current descriptive study
attend a baseline clinic visit and seven attended the
The current descriptive study is based on orientation baseline clinic visit, but were not randomized due to
session data from Cohort 1 of the trial; however, schedule conflicts and/or uncontrolled high blood
trial protocols were similar for all three cohorts. pressure. There were no clinically significant differ-
To examine replicability, retention rates across all ences in initial characteristics between participants
three cohorts are presented. who were and were not randomized (Table II).

Orientation sessions
Table I. Eligibility criteria at screening timepoints Individuals who were eligible to participate after
Telephone screening eligibility criteria a phone and mail screening were invited to attend
Age >25 an interactive group-based orientation session prior
BMI 27–37 to the baseline clinic visit and randomization.
Non-diabetic
Willing to be randomly assigned to any of the three groups
Led by the trial principal investigator (M. K.),
Available for active condition class meetings these 1-hour sessions combined motivational inter-
Planning to remain in the area for the next 2 years viewing techniques (Miller and Rollnick, 1991,
Not pregnant or planning to become pregnant in the next 2 years 2002) with active learning principles (Meyers and
Not following a special diet (e.g. Pritkin) Jones, 1993) to explicitly address the demands of
Not participating in another research trial
Not participating in another weight-loss program
joining a randomized controlled trial, making eating
and activity changes, and weight-loss expectations.
Mail screening eligibility criteria
Completed and returned questionnaire packet
Total calorie intake 500–5000 kcal (Block Food Frequency
Questionnaire) (Block et al., 1986)
Total calories from fat >30% (Block Food Frequency Table II. Means, SDs and percentages for demographic
Questionnaire) (Block et al., 1986) characteristics of participants
Physically inactive (American College of Sports Medicine)
(Pate et al., 1995) Characteristic Attended orientation session
Able to participate in physical activity
Not randomized Randomized
Free of heart disease
Stable on blood pressure, cholesterol and hormone medications N 21 51
for >3 months Gender
Not taking weight-loss medication female 14 67% 33 65%
Not dysphoric (Beck Depression Inventory <18) male 7 33% 18 35%
(Beck et al., 1961) Ethnic group
Not binge eating or bulimic (Eating Disorder Diagnostic Scale) white 17 81% 45 88%
(Stice et al., 2000) non-white 4 19% 6 12%
Baseline clinic visit eligibility criteria Marital status
Completed questionnaire packet single 2 9% 7 14%
Refrained from eating, drinking besides water, exercising, and married/living 17 81% 36 71%
smoking for 2 hours prior to clinic visit with partner
No uncontrolled hypertension as measured at clinic visit separated/divorced/ 2 9% 8 16%
Written permission to participate from a physician if two or widowed
more cardiovascular risk factors were present Age (years) 49.3 10.9 53.1 10.4
Committed to attending subsequent clinic visits Education (years) 15.6 1.8 16.5 2.3
Committed to attending active condition classes if applicable Initial BMI (kg/m2) 30.1 2.2 30.0 2.4

441
Downloaded from https://academic.oup.com/her/article-abstract/20/4/439/632665
by Sandy Lemmon user
on 09 November 2017
 J. H. Goldberg and M. Kiernan

In the didactic portion of the session which in- asked open-ended questions and engaged in reflec-
cluded a handout of the session’s key points, poten- tive listening. The investigator gave equal weight
tial participants first heard an explanation about and consideration to all responses, avoided the ‘pro-
cardiovascular health benefits of losing a change’ position (Miller and Rollnick, 1991, 2002),
modest amount of weight (10–15 pounds) at a slow ensured that each small group shared at least one
rate of loss (Goldstein, 1992; National Institues of response, and wrote all responses in a 2 3 2 grid on
Health/National Heart Lung and Blood Institute, a white board. The pros and cons discussion began
1998; Tate et al., 2001; Knowler et al., 2002) and with a focus on the cons of the control condition (i.e.
were explicitly told that this trial would not be a good the most salient reasons not to participate). The pros
match for people seeking to quickly lose a lot of of the control condition were discussed next, fol-
weight. Participants then heard about the importance lowed by the pros of the active conditions. The dis-
of this trial and the specific commitments required cussion purposely ended with the cons of the active
(e.g. study conditions, clinic visits, classes). Partici- conditions. This section concluded with the state-
pants were told that if they were assigned to the active ment that the investigator was attempting the
condition classes they would be asked to complete ‘opposite of a hard sell’. The investigator encour-
homework, and make eating and activity changes. aged participants to consider all pros and cons and
Participants were given a schedule with the dates for to recognize that they would be making two
clinic visits and active condition classes, asked not to commitments—one to themselves (i.e. time, behav-
enroll if they knew in advance they would miss two or ior change) and one to ensure the trial’s scientific
more classes, and asked to commit to attending a quality (i.e. returning to all clinic visits).
makeup class for any missed classes.
To promote commitment to the scientific portion Additional retention techniques
of the trial, participants then learned about the Additional retention techniques advocated by other
importance of a control condition, random assig- epidemiological research studies and clinical trials
nment and attrition bias. For instance, participants were integrated throughout the trial’s recruitment and
reviewed graphs illustrating how trial results would retention phases (Table III) (Bindman et al., 1993;
be biased toward success if unsuccessful partici- Ribisl et al., 1996; Senturia et al., 1998; Kiernan et al.,
pants did not return to subsequent clinic visits. 2000; Janson et al., 2001; Prinz et al., 2001).
Participants were asked to think of reasons why the
next 18 months might not be a good time for trial
participation (e.g. planning a daughter’s wedding) Results
and whether they would return to clinic visits if
they ‘gained 15 pounds’. Orientation session results
Participants in each orientation session were then The 72 potential participants attended one of seven
divided into small groups of three or four, and asked orientation sessions before randomization. Because
to generate two pros and two cons of being assigned participants generated responses in small groups
to the control condition and to the active conditions. and shared them with the larger group, we analyzed
The principal investigator left the room during the the 130 responses by orientation session rather than
small group discussions, and then reassembled by individual. A similar number of responses (M =
the small groups to share their pros and cons with 18.6) was generated across all orientation sessions
the whole group. In this discussion, the investigator [v2 (6, N = 130) = 1.49; P = 0.96)]. The number of
did not follow a typical health education approach responses was equally distributed across the 2 3 2
(i.e. emphasizing the pros and encouraging partici- grid (pros/cons and active conditions/control con-
pants to problem solve the cons). Rather, consistent dition) [v2 (3, N = 130) = 1.26; P = 0.74)].
with motivational interviewing principles (Miller In a content analysis (Patton, 1980), two raters
and Rollnick, 1991, 2002), the principal investigator independently sorted the 130 pro/con responses

442
Downloaded from https://academic.oup.com/her/article-abstract/20/4/439/632665
by Sandy Lemmon user
on 09 November 2017
 Innovative techniques to address retention

Table III. Additional retention enhancement techniques pants generated the same response both as a pro
Create ‘project identity’ that participants can recognize by using and a con for the same type of study condition. For
similar colors and fonts on trial materials instance, illustrating participants’ ambivalence
Track eligibility status of potential participants on a computer about joining a randomized trial, the same response
database was generated both as a con (‘structured/inflexible’)
Write protocols to systematically address common participant
questions
and a pro (‘structure/discipline’) of being assigned
Adhere to trial protocols and procedures to the active conditions. Second, as would be
Provide support to all participants expected in a pro/con activity, participants also
Offer flexible scheduling generated opposite responses for different types of
Attempt to be on time for clinic appointments study conditions. For instance, regarding their
Make multiple attempts to contact participants for complete data
by phone and mail
ambivalence about making behavior changes, the
Encourage participants who move from the area to continue response (‘have to make behavioral changes/do
completing questionnaires, and have clinic data (e.g. weight, uncomfortable things/hard to change’) was gener-
blood pressure) collected and verified by another health ated as a con of being assigned to the active
professional condition, whereas the opposite response (‘eat or
Send birthday cards to all participants
Determine two secondary contacts by asking participants to sign
do what you want’) was generated as a pro of being
letters notifying contacts of trial participation and giving assigned to the control condition. Thus, participants
permission to provide forwarding information (letters also may have a preference for one condition even
served as an implicit behavioral commitment to complete though they would have to agree to random
the trial) assignment and the possibility of being assigned
to their non-preferred condition. Third, participants
also generated the same response for different types
into thematic categories. There was high inter-rater of study conditions. Illustrating participants’ re-
agreement (interclass correlation a = 0.99). Table alization that weight-loss expectations may not be
IV presents the number and percentage of orienta- fulfilled (and in fact may not lose any weight in this
tion sessions in which categories were generated. trial), the response (‘may not lose weight/may not
The thematic categories are presented in a 2 3 2 be successful’) was a frequently mentioned con not
grid, i.e. by pros/cons and by type of study only of being assigned to the control condition, but
condition (active conditions/control condition). also of being assigned to the active conditions.
Many thematic categories focused on losing
weight, making behavioral changes, structure/ Retention results
discipline, time commitment, social support and Of the 51 participants randomized to the trial in
learning. Some reasons for joining a randomized Cohort 1, 50 (98%) completed the 6-month clinic
controlled trial discussed in previous studies, e.g. visit, 48 (94%) completed the 12-month clinic visit
helping the trial or science (Mattson et al., 1985), and 48 (94%) completed the 18-month clinic visit.
were mentioned infrequently. Participants’ respon- Overall, of the 162 participants randomized across all
ses fell into distinct (and many) thematic categories three cohorts, 159 (98%) completed the 6-month
in each of the seven orientation sessions rather than clinic visit, 157 (97%) completed the 12-month clinic
into the same (and few) thematic categories in each visit and 156 (96%) completed the 18-month clinic
of the seven sessions. visit, with no differential dropout by study condition.
To determine whether participants expressed
ambivalence about joining a randomized trial,
ambivalence about making behavior changes and Discussion
unrealistic weight-loss expectations, we examined
three sets of comparisons. First, in perhaps the most Although orientation sessions are often used to
interesting demonstration of ambivalence, partici- recruit participants into randomized trials, this

443
Downloaded from https://academic.oup.com/her/article-abstract/20/4/439/632665
by Sandy Lemmon user
on 09 November 2017
 444

J. H. Goldberg and M. Kiernan

Table IV. Number and percent of orientation sessions in which responses were generated by pros/cons and type of study condition (active condition/control condition)a

No. and percent of No. and percent of

orientation sessions orientation sessions

N % N %

PROS of being assigned to an active PROS of being assigned to the control

condition condition
Support/encouragement 6 86 Less time commitment/no class time 7 100
Learning/education 5 71 Eat or do what you want/can continue 4 57
Structure/discipline 4 57 what already doing/status quo
Motivation/incentive 4 57 Fewer demands/less work/less effort 3 43
May lose weight/may be successful 4 57 Self-structured/flexible 2 29
may lose and maintain Eventually get class information 2 29
Change exercise/eating habits/behavior 3 43 Get feedback reports 2 29
Health improvement/benefits (not 3 43 Help the study/help science 2 29
explicitly eating or exercise) Can still do other programs 1 14
Structured monitoring 1 14 Have clinic visits 1 14
May feel better 1 14 Self-accomplishment 1 14
Sense of purpose 1 14 May lose weight/may be successful 1 14
Involvement 1 14
CONS of being assigned to the control
CONS of being assigned to an active condition
condition Unstructured/no direction/no control 6 86
Time commitment/class time 6 86 May not lose weight/may not be 6 86
May not lose weight/may gain/may not 4 57 successful/no benefits
be successful (may not work) No support system/alone 4 57
Have to make behavioral changes/do 3 43 No commitment/uninvolved 3 43
uncomfortable things/hard to change Less information/less learning 3 43
Structured/inflexible 2 29 Less motivating 2 29
May not like it/may not be appropriate 2 29 Eighteen-month delay 2 29
Demands/work/homework 2 29 May get worse 1 14
May not like groups 2 29 May lose interest 1 14
Peer pressure 1 14
Feel dependent on classes 1 14
May already know information 1 14
Family may not like it 1 14
Things happen that make class difficult 1 14
Instructor characteristics 1 14
a
There were seven orientation sessions.

Downloaded from https://academic.oup.com/her/article-abstract/20/4/439/632665

by Sandy Lemmon user
on 09 November 2017
 Innovative techniques to address retention

study describes a conceptual and practical advance definitively assess the impact of these orientation
that we speculate may enhance retention in a be- sessions, this retention technique itself would need
havioral weight-loss trial—the novel application to be tested using a randomized design. Future
of motivational interviewing techniques to diffuse research is also needed to systematically determine
ambivalence during interactive group-based orien- the impact of these sessions on participant trust
tation sessions prior to randomization. and satisfaction as well as the replicability across
Retention in this trial was very high—96% at participant subgroups (i.e. by education, gender,
18 months. In recent similar behavioral weight-loss age or ethnicity), session facilitators and types of
trials, between 13 and 41% of participants dropped behavioral interventions.
out by post-treatment follow-up (Perri et al., 1997; Whereas low retention rates can undermine the
Jeffery and French, 1999; Sbrocco et al., 1999; Tate findings of randomized trials by threatening internal
et al., 2001, 2003; Foster et al., 2003; Heshka et al., validity, there are limitations to using motivational
2003). However, direct comparisons are difficult interviewing techniques to improve retention. Like
given that eligibility criteria, number of classes, multiple eligibility criteria, rigorous screening pro-
follow-up length and other requirements vary cedures and burdensome trial requirements, these
widely across trials. For instance, although a few techniques may discourage unmotivated partici-
recent trials have also had little dropout (3–9%), the pants from entering a trial and compromise gener-
trials had weekly contact with participants through- alizability (Wilson and Brownell, 1980). However,
out the trial and/or the trials were shorter in total bias in participant selection may be an acceptable
duration (6–12 months total) (Blumenthal et al., cost for efficacy trials seeking to maximize retention.
2000; Stevens et al., 2001; Irwin et al., 2003; Jakicic Past research on retention in behavioral weight-
et al., 2003). In contrast, our 1-year follow-up after loss studies has primarily focused on participant
the 6-month classes ended did not include continued characteristics between dropouts and completers
contact and the trial was longer in total duration. (Hjoerdis and Gunnar, 1989; Clark et al., 1996;
Using motivational interviewing techniques in Davis and Addis, 1999). Research on techniques to
an open-ended and reflective manner to involve actually improve retention is rare. We hope this
participants in a discussion of the study design, descriptive paper will provoke further experimental
scientific rationale and trial’s pros/cons may have research about optimal retention strategies.
potentially increased motivation, decreased poten-
tial for disappointment, empowered participants to
make changes, fostered participant ‘buy-in’ and,
thus, encouraged high retention at follow-up (Miller Acknowledgements
and Rollnick, 1991, 2002). Consistent with active
learning principles, participants heard a variety of We gratefully acknowledge the individuals who
pros and cons, which may have carried more weight participated in the Stanford Healthy Weight Project.
because responses were generated by fellow par- We also thank Laurie Ausserer, Katie DaBell,
ticipants rather than by the investigator (Meyers and Lauren Durkin, Susan Kirkpatrick, Rebecca Lee,
Jones, 1993). Gabe Meeker, Suzanne Olson and Peggy Raymond
These descriptive results are only suggestive, in for their valued contributions to this study. This
part because the application of motivational inter- research was supported by an NIH/NHLBI FIRST
viewing principles in interactive group-based ori- Award (R29 HL60154) awarded to M. K., an
entation sessions was supplemented with other unrestricted research gift from Nutrilite Health In-
recommended retention techniques (King et al., stitute and an NIH/NHLBI Training Grant (5T32
1991; Bindman et al., 1993; Ribisl et al., 1996; HL07034). Portions of this paper were presented at
Senturia et al., 1998; Kiernan et al., 2000; Janson the Society of Behavioral Medicine’s 24th Annual
et al., 2001; Prinz et al., 2001). Most critically, to Meeting and Scientific Sessions, Salt Lake City, UT.

445
Downloaded from https://academic.oup.com/her/article-abstract/20/4/439/632665
by Sandy Lemmon user
on 09 November 2017
 J. H. Goldberg and M. Kiernan

References Pi-Sunyer, F.X. (2003) Weight loss with self-help compared

with a structured commercial program: a randomized
trial. Journal of the American Medical Association, 289,
Beck, A.T., Ward, C.H., Mendelson, M., Mock, J. and Erbaugh, 1792–1798.
J. (1961) An inventory for measuring depression. Archives of Hjoerdis, B. and Gunnar, E. (1989) Characteristics of drop-
General Psychiatry, 4, 561–571. outs from a long-term behavioral treatment program for
Bennett, G.A. (1986) Expectations in the treatment of obesity. obesity. International Journal of Eating Disorders, 8,
British Journal of Clinical Psychology, 25, 311–312. 363–368.
Bindman, A.B., Grumbach, K., Keane, D. and Lurie, N. (1993) Irwin, M.L., Yasui, Y., Ulrich, C.M., Bowen, D., Rudolph, R.E.,
Collecting data to evaluate the effect of health policies on Schwartz, R.S., Yukawa, M., Aiello, E., Potter, J.D. and
vulnerable populations. Family Medicine, 25, 114–119. McTiernan, A. (2003) Effect of exercise on total and intra-
Block, G., Hartman, A.M., Dresser, C.M., Carroll, M.D., abdominal body fat in postmenopausal women: a randomized
Gannon, J. and Gardner, L. (1986) A data-based approach to controlled trial. Journal of the American Medical Association,
diet questionnaire design and testing. American Journal of 289, 323–330.
Epidemiology, 124, 453–469. Jakicic, J.M., Marcus, B.H., Gallagher, K.I., Napolitano, M. and
Blumenthal, J.A., Sherwood, A., Gullette, E.C., Babyak, M., Lang, W. (2003) Effect of exercise duration and intensity on
Waugh, R., Georgiades, A., Craighead, L.W., Tweedy, D., weight loss in overweight, sedentary women: a randomized
Feinglos, M., Appelbaum, M., Hayano, J. and Hinderliter, A. trial. Journal of the American Medical Association, 290,
(2000) Exercise and weight loss reduce blood pressure in men 1323–1330.
and women with mild hypertension: effects on cardiovascular, Janis, I.L. and Mann, L. (1977) Decision Making: A Psychologic-
metabolic and hemodynamic functioning. Archives of Internal al Analysis of Conflict, Choice and Commitment. Free Press,
Medicine, 160, 1947–1958. New York, NY.
Brownell, K.D. and Wadden, T.A. (1992) Etiology and treatment Janson, S.L., Alioto, M.E. and Boushey, H.A. (2001) Attrition
of obesity: understanding a serious, prevalent and refractory and retention of ethnically diverse subjects in a multicenter
disorder. Journal of Consulting and Clinical Psychology, 60, randomized controlled research trial. Controlled Clinical
505–517. Trials, 22(6 Suppl.), 236S–243S.
Burke, L.E., Cartwright, M., Music, E., Kim, Y., Tschirpke, D., Jeffery, R.W., Epstein, L.H., Wilson, G.T., Drewnowski, A.,
Polakoski, T. and Klem, M.L. (2003) Weight loss programs: is Stunkard, A.J. and Wing, R.R. (2000) Long-term maintenance
there a mismatch between treatment preferred and treatment of weight loss: current status. Health Psychology, 19(Suppl.
sought? Annals of Behavioral Medicine, 25(Suppl.), S021. 1), 5–16.
Ciliska, D. (1998) Evaluation of two nondieting interventions Jeffery, R.W. and French, S.A. (1999) Preventing weight gain
for obese women. Western Journal of Nursing Research, 20, in adults: the Pound of Prevention study. American Journal of
119–135. Public Health, 89, 747–751.
Clark, M.M., Niaura, R., King, T.K. and Pera, V. (1996) Jeffrey, R.W., Wing, R.R. and Mayer, R.R. (1998) Are smaller
Depression, smoking, activity level and health status: Pre- weight losses or more achievable weight loss goals better in
treatment predictors of attrition in obesity treatment. Addictive the long term for obese patients? Journal of Consulting and
Behavior, 21, 509–513. Clinical Psychology, 66, 641–645.
Davis, M.J. and Addis, M.E. (1999) Predictors of attrition from Kiernan, M., Phillips, K., Fair, J.M. and King, A.C. (2000) Using
behavioral medicine treatments. Annals of Behavioral Medi- direct mail to recruit Hispanic adults into a dietary interven-
cine, 21, 339–349. tion: an experimental study. Annals of Behavioral Medicine,
Flegal, K.M., Carroll, M.D., Ogden, C.L. and Johnson, C.L. 22, 89–93.
(2002) Prevalence and trends in obesity among US adults, Kiernan, M., Raymond, M., Goldberg, J.H., DaBell, K., Meeker,
1999–2000. Journal of the American Medical Association, G. and Kirpkatrick, S. (2001) A physician-hosted radio talk
288, 1723–1727. show versus traditional recruitment methods. Annals of
Foster, G.D., Wadden, T.A., Vogt, R.A. and Brewer, G. (1997) Behavioral Medicine, 23(Suppl.), S020.
What is a reasonable weight loss? Patients’ expectations King, A.C., Haskell, W.L., Taylor, C.B., Kraemer, H.C. and
and evaluations of obesity treatment outcomes. Journal of DeBusk, R.F. (1991) Group- vs home-based exercise training
Consulting and Clinical Psychology, 65, 79–85. in healthy older men and women. A community-based clinical
Foster, G.D., Wyatt, H.R., Hill, J.O., McGuckin, B.G., Brill, C., trial. Journal of the American Medical Association, 266,
Mohammed, B.S., Szapary, P.O., Rader, D.J., Edman, J.S. and 1535–1542.
Klein, S. (2003) A randomized trial of a low-carbohydrate King, C.M., Rothman, A.J. and Jeffery, R.W. (2002) The
diet for obesity. New England Journal of Medicine, 348, Challenge study: theory-based interventions for smoking and
2082–2090. weight loss. Health Education Research, 17, 522–530.
Goldstein, D.J. (1992) Beneficial health effects of modest weight Knowler, W.C., Barrett-Connor, E., Fowler, S.E., Hamman,
loss. International Journal of Obesity and Related Metabolic R.F., Lachin, J.M., Walker, E.A. and Nathan, D.M. (2002)
Disorders, 16, 397–415. Reduction in the incidence of type 2 diabetes with lifestyle
Hansen, W.B., Collins, L.M., Malotte, C.K., Johnson, C.A. and intervention or metformin. New England Journal of Medicine,
Fielding, J.E. (1985) Attrition in prevention research. Journal 346, 393–403.
of Behavioral Medicine, 8, 261–275. Mattson, M.E., Curb, J.D. and McArdle, R. (1985) Participation
Heshka, S., Anderson, J.W., Atkinson, R.L., Greenway, F.L., Hill, in a clinical trial: the patients’ point of view. Controlled
J.O., Phinney, S.D., Kolotkin, R.L., Miller-Kovach, K. and Clinical Trials, 6, 156–167.

446
Downloaded from https://academic.oup.com/her/article-abstract/20/4/439/632665
by Sandy Lemmon user
on 09 November 2017
 Innovative techniques to address retention

Meyers, C. and Jones, T.B. (1993) Promoting Active Learning: decision-based treatment for obesity. Journal of Consulting
Strategies for the College Classroom. Jossey-Bass, San and Clinical Psychology, 67, 260–266.
Francisco, CA. Senturia, Y.D., McNiff Mortimer, K., Baker, D., Gergen, P.,
Miller, W.R. and Rollnick, S. (1991) Motivational Interviewing: Mitchell, H., Joseph, C. and Wedner, H.J. (1998) Successful
Preparing People to Change Addictive Behavior. Guilford techniques for retention of study participants in an inner-city
Press, New York. population. Controlled Clinical Trials, 19, 544–554.
Miller, W.R. and Rollnick, S. (2002) Motivational Interviewing: Shepherd, R. (2002) Resistance to changes in diet. Proceedings
Preparing People for Change. Guilford Press, New York. of the Nutrition Society, 61, 267–272.
Mish, F.C. (ed.) (1990) Webster’s Ninth New Collegiate Slack, M.K. and Draugalis, J.R. (2001) Establishing the internal
Dictionary. Merriam-Webster, Springfield, MA. and external validity of experimental studies. American
National Institutes of Health/National Heart Lung and Blood Journal of Health-System Pharmacy, 58, 2173–2181; quiz
Institute (1998) Clinical guidelines on the identification, evalu- 2182–2173.
ation and treatment of overweight and obesity in adults—the Smith, D.E., Heckemeyer, C.M., Kratt, P.P. and Mason, D.A.
evidence report. Obesity Research, 6(Suppl. 2), 51S–209S. (1997) Motivational interviewing to improve adherence to
Pate, R.R., Pratt, M., Blair, S.N., Haskell, W.L., Macera, C.A., a behavioral weight-control program for older obese women
Bouchard, C., Buchner, D., Ettinger, W., Heath, G.W., King, with NIDDM. A pilot study. Diabetes Care, 20, 52–54.
A.C., Kriska, A., Leon, A.S., Marcus, B.H., Morris, J., Sparks, P., Conner, M., James, R., Shepherd, R. and Povey, R.
Paffenbarger, R.S., Patrick, K., Pollock, M.L., Rippe, J.M., (2001) Ambivalence about health-related behaviours: an
Sallis, J. and Wilmore, J.H. (1995) Physical activity and public exploration in the domain of food choice. British Journal of
health. A recommendation from the Centers for Disease Health Psychology, 6, 53–68.
Control and Prevention and the American College of Sports Stevens, V.J., Obarzanek, E., Cook, N.R., Lee, I.M., Appel, L.J.,
Medicine. Journal of the American Medical Association, 273, Smith West, D., Milas, N.C., Mattfeldt-Beman, M., Belden,
402–407. L., Bragg, C., Millstone, M., Raczynski, J., Brewer, A., Singh,
Patton, M.Q. (1980) Qualitative Evaluation Methods. Sage, B. and Cohen, J. (2001) Long-term weight loss and changes
Beverly Hills, CA. in blood pressure: results of the Trials of Hypertension
Perri, M.G., Martin, A.D., Leermakers, E.A., Sears, S.F. and Prevention, phase II. Annals of Internal Medicine, 134, 1–11.
Notelovitz, M. (1997) Effects of group- versus home-based Stice, E., Telch, C.F. and Rizvi, S.L. (2000) Development and
exercise in the treatment of obesity, Journal of Consulting and validation of the Eating Disorder Diagnostic Scale: a brief self-
Clinical Psychology, 65, 278–285. report measure of anorexia, bulimia and binge-eating disorder.
Prinz, R.J., Smith, E.P., Dumas, J.E., Laughlin, J.E., White, Psychological Assessment, 12, 123–131.
D.W. and Barron, R. (2001) Recruitment and retention of Tate, D.F., Wing, R.R. and Winett, R.A. (2001) Using Internet
participants in prevention trials involving family-based technology to deliver a behavioral weight loss program. Journal
interventions. American Journal of Preventative Medicine, of the American Medical Association, 285, 1172–1177.
20(Suppl. 1), 31–37. Tate, D.F., Jackvony, E.H. and Wing, R.R. (2003) Effects of
Prochaska, J.O. and DiClemente, C.C. (1983) Stages and Internet behavioral counseling on weight loss in adults at
processes of self-change of smoking: toward an integrative risk for type 2 diabetes: a randomized trial. Journal of the
model of change. Journal of Consulting and Clinical American Medical Association, 289, 1833–1836.
Psychology, 51, 390–395. Wadden, T.A., Brownell, K.D. and Foster, G.D. (2002) Obesity:
Prochaska, J.O., DiClemente, C.C. and Norcross, J.C. (1992) responding to the global epidemic. Journal of Consulting and
In search of how people change. Applications to addictive Clinical Psychology, 70, 510–525.
behaviors. American Psychologist, 47, 1102–1114. Wadden, T.A., Womble, L.G., Sarwer, D.B., Berkowitz, R.I.,
Prochaska, J.O., Velicier, W.F., Rossi, J.S. and Goldstein, M.G. Clark, V.L. and Foster, G.D. (2003) Great expectations: ‘I’m
(1994) Stages of change and decisional balance for 12 losing 25% of my weight no matter what you say’. Journal of
problem behaviors. Health Psychology, 13, 39–46. Consulting and Clinical Psychology, 71, 1084–1089.
Resnicow, K., DiIorio, C., Soet, J.E., Ernst, D., Borrelli, B. Ware, J.H. (2003) Interpreting incomplete data in studies of diet
and Hecht, J. (2002) Motivational interviewing in health and weight loss. New England Journal of Medicine, 348,
promotion: it sounds like something is changing. Health 2136–2137.
Psychology, 21, 444–451. Wilson, G.T. and Brownell, K.D. (1980) Behavior therapy for
Ribisl, K.M., Walton, M.A., Mowbray, C.T., Luke, D.A. obesity: an evaluation of treatment outcome. Advances in
and BootsMiller, B.J. (1996) Minimizing participant attrition Behavior Research and Therapy, 3, 49–86.
in panel studies through the use of effective retention and Windhauser, M.M., Ernst, D.B., Karanja, N.M., Crawford, S.W.,
tracking strategies: review and recommendations. Evaluation Redican, S.E., Swain, J.F., Karimbakas, J.M., Champagne,
and Program Planning, 19, 1–25. C.M., Hoben, K.P. and Evans, M.A. (1999) Translating the
Rollnick, S. and Miller, W.R. (1995) What is motivational Dietary Approaches to Stop Hypertension diet from research
interviewing? Behavioural and Cognitive Psychotherapy, 23, to practice: dietary and behavior change techniques. DASH
325–334. Collaborative Research Group. Journal of the American
Sbrocco, T., Nedegaard, R.C., Stone, J.M. and Lewis, E.L. Dietetic Association, 99(8 Suppl.), S90–S95.
(1999) Behavioral choice treatment promotes continuing
weight loss: preliminary results of a cognitive-behavioral Received on February 3, 2004; accepted on October 18, 2004

447
Downloaded from https://academic.oup.com/her/article-abstract/20/4/439/632665
by Sandy Lemmon user
on 09 November 2017