|Obesity; Lifestyle modification; Portion-controlled
foods; Primary care
|Obesity and its related medical co-morbidities continue to be
among the most common problems encountered by primary care
physicians. One study, using national data from the U.S., conservatively
estimated that 30% of clinical items addressed in a visit are related to
excess weight . This same study reported that 8% of total primary
care physician time can be attributed to overweight and obesity .
Recent trials have begun to address obesity more routinely in clinical
settings [2-7] and to disseminate effective interventions to primary care
and community settings [8,9].
|Despite these efforts to integrate treatment of obesity with primary
care, which direction the field will take remains unclear. One of the
main questions relates to the cost of the interventions. Trials have taken
quite different approaches to this question. For example, one recent
study offered participants multiple interventions for treating obesity,
with relative disregard for cost. Specifically, participants were offered a
medically supervised low-calorie diet consisting of meal replacements,
plus intensive group behavioral treatment, plus the option to use
weight loss medication . In contrast, three recent trials (funded
under a cooperative UO1 mechanism from the U.S. National Institutes
of Health) tested the use of lower cost counseling personnel that did
not have professional degrees in nutrition or psychology. One trial
employed telephone coaches who worked with patients remotely .
The second trial used medical assistants who worked in the primary
care practices . The third trial employed a website and an interactive
voice response telephonic system .
|Closely related to the issue of cost is the issue of which interventions
should be reimbursed for primary care patients with obesity. In the
U.S., health care payers (employers and insurance plans) generally
do not reimburse for weight loss medications or for the use of meal
replacements or other foods formulated for weight loss . This is unfortunate, given that these are two of the most effective non-surgical
interventions for treating obesity. However, the U.S. national epidemics
of diabetes and pre-diabetes have led to increased attention on excess
weight . This attention, in turn, has led to efforts to disseminate the
highly successful intervention developed by the Diabetes Prevention
Program (DPP) . Some health care payers and health care
institutions are now offering the DPP at low or zero cost to individuals
at high risk for diabetes.
|This paper has three aims: 1) to describe the design and start-up
of a clinical trial targeting primary care patients with obesity and a
related co-morbidity; 2) to quantify outcomes between baseline and
6 months; and 3) to examine baseline predictors of weight loss. The
primary outcome at 6 months was weight change. Secondary outcomes
included changes in cardiovascular risk factors and health-related
quality of life.
|The overall study was designed to test the hypothesis that inperson
visits, provided during a period of weight maintenance, would
be associated with greater weight loss at the end of 18 months. This paper describes the outcomes of the first 6 months of treatment (i.e.,
the initial weight loss phase). All participants were offered a series of
12 in-person counseling visits during the first 6 months. As described
below, all participants also were offered discounted access to portioncontrolled
weight loss foods throughout the trial.
|We recruited individuals from two primary care internal medicine
practices at the University of Colorado. Individuals aged 18-79, with
a body mass index (BMI) 30-49.9 kg/m2 and an ICD-9 (International
Classification for Diseases) code for either type 2 diabetes, hypertension,
hyperlipidemia, or obstructive sleep apnea, were sent letters inviting
them to participate in a research trial for weight loss. Exclusion criteria
included a weight change of ≥ 5% of starting weight within the previous
six months, use of chronic glucocorticoids or second generation antipsychotics,
previous bariatric surgery, or if the individual was unable
or unwilling to travel for in-person visits and to use portion-controlled
foods for weight loss.
|Potential participants were screened by telephone to ensure
eligibility. Eligible individuals were then asked to attend two screening
visits. At the first screening visit, they provided informed consent and
completed baseline questionnaires. At the second screening visit, they
met with the study’s principal investigator for a clinical interview.
The purpose of the clinical interview was to assess readiness for
participation in an 18-month clinical trial. The flow chart (see Results
section) describes the flow of study participants.
|During the first 6 months, all participants were offered 12 in-person
weight loss counseling visits. The written materials used for these visits
were modified from the Diabetes Prevention Program (DPP). The
DPP showed that a weight loss of 7% of starting weight reduced the
incidence of type 2 diabetes by 58% after 4 years of follow-up, despite
partial weight regain . Follow-up of DPP participants showed that
the cumulative incidence of diabetes remained lower after 10 years .
We slightly modified the materials from the DPP and then reviewed
these with study participants at each visit. The modifications to the
materials reflected updates in the behavioral and dietary treatment of
obesity. For example, we modified the list of snacks from the original
DPP materials to focus more on whole grain carbohydrate foods with
lower glycemic index. Visits in the current study lasted 20-30 minutes
each time. Participants were assessed at baseline and at 6 months.
|The primary counselor was a professional research assistant at the
University of Colorado. She had no formal training in nutrition or
weight management but holds a Master’s degree in psychology. She had
collaborated with the Principal Investigator on one previous weight loss
|The use of portion-controlled weight loss foods has been
demonstrated in randomized trials to increase weight loss beyond the
provision of a calorie prescription with the same target for energy intake
[16,17]. Participants were offered the Nutrisystem® weight loss program
at a deeply discounted rate of $95 per month (the typical retail cost
of the program is $250-$300 per month). The Nutrisystem® program
includes three portion-controlled entrees and two portion-controlled
snack items per day. Participants are required to purchase their own
fruits, vegetables, and dairy items while using the Nutrisystem®
program. The Nutrisystem® meal plan (including provided items and
recommended grocery additions) provides approximately 1200-1500 calories per day, with approximately 50%, 25%, and 25% of energy
from carbohydrate, fat, and protein, respectively. The Nutrisystem®
meal plan is consistent with the Dietary Guidelines for Americans-2010
targets for fiber, saturated fat, and sodium.The program has been shown
to be effective in randomized controlled trials [18-20].
|Outcomes and assessments
|The primary outcome of the study was weight change. We
weighed participants in light clothing on a digital scale (model #740;
Tanita Corporation; Tokyo; Japan). Secondary outcomes included
changes in body mass index (BMI), waist circumference, glycemic
control, blood pressure, lipids, mood, and health-related quality of
life. Waist circumference was measured by trained staff at the Center
for Translational and Clinical Research (CTRC) at the University of
Colorado Hospital. Blood pressure was measured by CTRC staff, taking
three measurements and resting 1 minute between each measurement.
Blood for fasting lipids, highly sensitive C-reactive protein (hsCRP),
and hemoglobin A1c were drawn after an overnight fast and analyzed
on a Beckman-Olympus AU400e chemistry analyzer (lipids, glucose
and hsCRP), a Siemens DCA Vantage analyzer (hemoglobin A1c), or
a Perkin-Elmer Wizard 1470 Gamma Counter (insulin). Mood was
measured with the Patient Health Questionniare (PHQ-9). Dietary
outcomes were assessed with the Diet History Questionnaire (DHQ).
Health-related quality of life was measured with the SF-12 (Medical
Outcomes Study, Short Form 12) and with the feeling thermometer of
the EuroQol-5D (EQ5D).
|To analyze weight change between baseline and month 6, we used a
conservative assumption that participants who dropped out prior to 6
months regained weight. Specifically, we assumed that individuals who
dropped out gained 0.3 kg per month. We assumed that weight regain
occurred either until they reached their baseline weight or until they
reached 6 months (whichever event occurred first). This assumption of
weight regain has been empirically validated . Use of this method
facilitated an intention-to-treat (ITT) analysis in which we were able
to estimate weight change for all 106 individuals who entered the trial.
|Weight loss predictors
|Univariable regression analysis was conducted to explore the
association of baseline characteristics with the outcome of weight loss
at 6 months. Continuous variables (e.g., age, number of medications
taken) were collapsed into categories for ease of interpretation from
regression analyses, and t tests were used for dichotomous variables
(e.g., gender). Variables with a p value of < 0.2 in univariable analysis
were included in the multivariable model. Variables were added
sequentially to the multivariable model, ultimately keeping those only
with a p value < 0.05.
|Sample size and power
|We did not compute power for the outcome of weight loss at 6
months, as all participants received the same treatment during this
|Funding and role of sponsors
|The trial was funded through a scientist development grant from
the American Heart Association to the principal investigator (award
# 10SDG2610292). The funder did not have a role in the design of the
trial or in data analysis or writing. Nutrisystem subsidized the use of its
food products for study participants but did not pay funds to the study team. The study also received support from the CTRC at the University
of Colorado for lab assays. The study was approved by the Colorado
Multiple Institutional Review Board (protocol #10-0719).
|Table 1 shows the baseline characteristics of study participants.
Participants had an average age of nearly 55, most were women, and
more than half were college graduates. The Figure shows the flow of
participants during the trial. A total of n=240 individuals were screened
by telephone for participation. Of these, n=139 signed a consent form,
of whom n=106 completed all enrollment procedures and entered
the trial. A total of n=84 completed the first 6 months. Thus, attrition
during the first 6 months was 22/106 (20.7%) (Figure 1).
|Using intention-to-treat analysis with the assumption of weight
regain, average weight loss from baseline to 6-month follow-up was 7.0
± 7.0 kg, equal to 6.5 ± 6.1% of initial weight.Average weight loss among
completers (n=84) was 8.6 ± 7.0 kg, equal to 8.0 ± 5.9% of initial weight.
|Changes in secondary outcomes are summarized in Table 4. Highly significant improvements were noted for most outcomes,
including three out of four measures of health-related quality of life.
These were the SF-12 Physical Component Summary, the Impact of
Weight on Quality of Life (IWQoL), and the feeling thermometer of
the EQ5D. No significant change was observed in the SF-12 Mental
Component Summary Score. Significant improvements were noted in
the PHQ-9 (depression) score and in the two nutrition parameters of
most interest (total calorie intake and grams per day of fat). Significant
improvements also were noted in cardiovascular disease risk factors.
These risk factors included systolic and diastolic blood pressure, waist
circumference, and glycemic control, as measured by hemoglobin A1c
(the average participant went from a “pre-diabetes” value of 5.88% to
a normal A1c value of 5.63%). Improvements were also noted in ratio
of total to HDL cholesterol, in LDL cholesterol, and in triglycerides. Of the improvements in lipids, only the change in triglycerides reached
statistical significance (Tables 2 and 3).
|Univariable regression analysis
|Several baseline characteristics had an association with weight
change at 6 months. However, most associations did not reach statistical
significance in univariable testing (p<0.05). Factors associated with
greater weight loss at six months included: greater medication use; older
age; male gender; lower PHQ-9 (depression) score, greater number of
hours per week of television; and fewer stressors listed in the Weight
and Lifestyle Inventory  (Table 4).
|Multivariable regression analysis
|In multivariable analysis, only older age and greater number of
hour/week of television watched were associated with weight change
|In this summary of the results from the non-randomized phase
of a clinical trial, clinically significant weight loss was achieved,
even with conservative analysis assuming weight regain. Substantial
improvements also were observed in cardiovascular disease risk
factors, mood, and health-related quality of life. The weight loss of 8.6
kg among individuals completing 6 months of treatment was nearly
identical to the mean weight loss of 8.5 kg reported in the Weight Loss
Maintenance (WLM) trial among individuals completing the first 6
months of treatment . The current trial is one of the first to use
a trained layperson to treat individuals recruited from a primary care
medical setting. The WLM Trial, for example, used trained behaviorists
to lead their intervention.The current study used a layperson who
received training from the principal investigator.
|The six month weight losses in this study compare favorably with
other trials of participants recruited from primary care settings. For
example, Tsai et al. found a control-subtracted weight loss of 3.5 kg
after 6 months among individuals recruited from two primary care
clinics . In a larger trial, Wadden et al.  reported a controlsubtracted
weight loss of 1.5 kg after 6 months among participants
who received monthly counseling from medical assistants, based in
primary care clinics . These two trials offered moderate-intensity
lifestyle interventions (approximately monthly visits). In two trials
that used high-intensity intervention, Ma et al. reported a controlsubtracted
weight loss of 5.9 kg in their in-person coaching group ,
and Weinstock et al. reported a weight loss of 4.3 kg for their in-person
intervention (4.5 kg for a conference call intervention that used the
same curriculum) .
|In this trial, very few baseline characteristics predicted weight
loss. The only substantial predictor of weight loss was older age. This result is consistent with results of previous studies, including the
multi-center Look AHEAD Trial, in which older age was a consistent
predictor of greater weight loss . Older individuals may have more
predictable routines and fewer responsibilities (e.g., child care), giving them more time to exercise and to plan meals that facilitate weight
loss. Interestingly, in the current study, greater television watching at
baseline also was associated with better weight loss. This result suggests
that individuals who lost weight may have reduced their television
watching time during the first 6 months. [We did not inquire at the
6-month time point about television time].
|This study has at least three important limitations. First, this was a
relatively high socioeconomic status population, and thus, participants
likely had more time and resources to help control their weight, as
compared to lower income populations. Second, although individuals
were recruited from primary care settings for the trial, they were
provided high-intensity treatment in a University-based weight loss
center, with counseling visits lasting longer than they would in a typical
primary care office setting. Third, participants were offered subsidized
use of portion-controlled foods for weight loss, which is not typical of
how health systems in the U.S. reimburse treatment for obesity.
|In summary, this analysis shows excellent weight loss after the
first 6 months of treatment, along with substantial improvements
in secondary outcomes. Future publications will describe changes in
medication utilization as well as weight losses at 18 months (12 months
after randomization to a weight maintenance condition).
|This study was funded by a grant from the American Heart Association (award
# 10SDG2610292) to Adam Gilden Tsai. We also gratefully acknowledge in-kind
support for the study from Nutrisystem, who provided our participants with a
subsidized rate for their foods, as well as the Center for Translational and Clinical
Research at the University of Colorado, which provided free lab assays for our
participants.We also acknowledge the use of REDCap (Research Electronic Data
Capture) at the University of Colorado, funded by NIH/NCRR Colorado CTSI Grant
Number UL1 RR025780.
- Tsai AG, Abbo ED, Ogden LG (2011) The time burden of overweight and obesity in primary care. BMC Health Serv Res 11: 191.
- Appel LJ, Clark JM, Yeh HC, Wang NY, Coughlin JW, et al. (2011) Comparative effectiveness of weight-loss interventions in clinical practice. N Engl J Med 365: 1959-1968.
- Bennett GG, Warner ET, Glasgow RE, Askew S, Goldman J, et al. (2012) Obesity treatment for socioeconomically disadvantaged patients in primary care practice. Arch Intern Med 172: 565-574.
- Kumanyika SK, Fassbender JE, Sarwer DB, Phipps E, Allison KC, et al. (2012) One-year results of the Think Health! study of weight management in primary care practices. Obesity (Silver Spring) 20: 1249-1257.
- Ma J, Yank V, Xiao L, Lavori PW, Wilson SR, et al. (2013) Translating the Diabetes Prevention Program lifestyle intervention for weight loss into primary care: a randomized trial. JAMA Intern Med 173: 113-121.
- Wadden TA, Volger S, Sarwer DB, Vetter ML, Tsai AG, et al. (2011) A two-year randomized trial of obesity treatment in primary care practice. N Engl J Med 365: 1969-1979.
- Weinstock RS, Trief PM, Cibula D, Morin PC, Delahanty LM (2013) Weight loss success in metabolic syndrome by telephone interventions: results from the SHINE Study. J Gen Intern Med 28: 1620-1628.
- Ali MK, Echouffo-Tcheugui J, Williamson DF (2012) How effective were lifestyle interventions in real-world settings that were modeled on the Diabetes Prevention Program? Health Aff (Millwood) 31: 67-75.
- Ackermann RT, Finch EA, Brizendine E, Zhou H, Marrero DG (2008) Translating the Diabetes Prevention Program into the community. The DEPLOY Pilot Study. Am J Prev Med 35: 357-363.
- Ryan DH, Johnson WD, Myers VH, Prather TL, McGlone MM, et al. (2010) Nonsurgical weight loss for extreme obesity in primary care settings: results of the Louisiana Obese Subjects Study. Arch Intern Med 170: 146-154.
- Tsai AG, Asch DA, Wadden TA (2006) Insurance coverage for obesity treatment. J Am Diet Assoc 106: 1651-1655.
- Centers for Disease Control and Prevention (2011) National Diabetes Fact Sheet
- Knowler WC, Barrett-Connor E, Fowler SE, Hamman RF, Lachin JM, et al. (2002) Reduction in the incidence of type 2 diabetes with lifestyle intervention or metformin. N Engl J Med 346: 393-403.
- Diabetes Prevention Program Research Group, Knowler WC, Fowler SE, Hamman RF, Christophi CA, et al. (2009) 10-year follow-up of diabetes incidence and weight loss in the Diabetes Prevention Program Outcomes Study. Lancet 374: 1677-1686.
- Tsai AG, Felton S, Hill JO, Atherly AJ (2012) A randomized pilot trial of a full subsidy vs. a partial subsidy for obesity treatment. Obesity (Silver Spring) 20: 1838-1843.
- Casazza K, Fontaine KR, Astrup A, Birch LL, Brown AW, et al. (2013) Myths, presumptions, and facts about obesity. N Engl J Med 368: 446-454.
- Heymsfield SB, van Mierlo CA, van der Knaap HC, Heo M, Frier HI (2003) Weight management using a meal replacement strategy: meta and pooling analysis from six studies. Int J ObesRelatMetabDisord 27: 537-549.
- Figueroa A, Vicil F, Sanchez-Gonzalez MA, Wong A, Ormsbee MJ, et al. (2013) Effects of diet and/or low-intensity resistance exercise training on arterial stiffness, adiposity, and lean mass in obese postmenopausal women. Am J Hypertens 26: 416-423.
- Foster GD, Wadden TA, Lagrotte CA, Vander Veur SS, Hesson LA, et al. (2013) A randomized comparison of a commercially available portion-controlled weight-loss intervention with a diabetes self-management education program. Nutrition & Diabetes 3: e63
- Foster GD, Borradaile KE, Vander Veur SS, LehShantz K, Dilks RJ, et al. (2009) The effects of a commercially available weight loss program among obese patients with type 2 diabetes: a randomized study. Postgrad Med 121: 113-118.
- Wadden TA, Berkowitz RI, Sarwer DB, Prus-Wisniewski R, Steinberg C (2001) Benefits of lifestyle modification in the pharmacologic treatment of obesity: a randomized trial. Arch Intern Med 161: 218-227.
- Wadden TA, Sarwer DB, Fabricatore AN, Jones L, Stack R, et al. (2007) Psychosocial and behavioral status of patients undergoing bariatric surgery: what to expect before and after surgery. Med Clin North Am 91: 451-469, xi-xii.
- Svetkey LP, Stevens VJ, Brantley PJ, Appel LJ, Hollis JF, et al. (2008) Comparison of strategies for sustaining weight loss: the weight loss maintenance randomized controlled trial. JAMA 299: 1139-1148.
- Tsai AG, Wadden TA, Rogers MA, Day SC, Moore RH, et al. (2010) A primary care intervention for weight loss: results of a randomized controlled pilot study. Obesity (Silver Spring) 18: 1614-1618.
- Wadden TA, West DS, Neiberg RH, Wing RR, Ryan DH, et al. (2009) One-year weight losses in the Look AHEAD study: factors associated with success. Obesity (Silver Spring) 17: 713-722.