Effect of Behavioral Therapy With In-Clinic or Telephone Group Visits vs In-Clinic Individual Visits on Weight Loss Among Patients With Obesity in Rural Clinical Practice: A Randomized Clinical Trial

Abstract

Importance: Rural populations have a higher prevalence of obesity and poor access to weight loss programs. Effective models for treating obesity in rural clinical practice are needed.

Objective: To compare the Medicare Intensive Behavioral Therapy for Obesity fee-for-service model with 2 alternatives: in-clinic group visits based on a patient-centered medical home model and telephone-based group visits based on a disease management model.

Design, setting, and participants: Cluster randomized trial conducted in 36 primary care practices in the rural Midwestern US. Inclusion criteria included age 20 to 75 years and body mass index of 30 to 45. Participants were enrolled from February 2016 to October 2017. Final follow-up occurred in December 2019.

Interventions: All participants received a lifestyle intervention focused on diet, physical activity, and behavior change strategies. In the fee-for-service intervention (n = 473), practice-employed clinicians provided 15-minute in-clinic individual visits at a frequency similar to that reimbursed by Medicare (weekly for 1 month, biweekly for 5 months, and monthly thereafter). In the in-clinic group intervention (n = 468), practice-employed clinicians delivered group visits that were weekly for 3 months, biweekly for 3 months, and monthly thereafter. In the telephone group intervention (n = 466), patients received the same intervention as the in-clinic group intervention, but sessions were delivered remotely via conference calls by centralized staff.

Main outcomes and measures: The primary outcome was weight change at 24 months. A minimum clinically important difference was defined as 2.75 kg.

Results: Among 1407 participants (mean age, 54.7 [SD, 11.8] years; baseline body mass index, 36.7 [SD, 4.0]; 1081 [77%] women), 1220 (87%) completed the trial. Mean weight loss at 24 months was -4.4 kg (95% CI, -5.5 to -3.4 kg) in the in-clinic group intervention, -3.9 kg (95% CI, -5.0 to -2.9 kg) in the telephone group intervention, and -2.6 kg (95% CI, -3.6 to -1.5 kg) in the in-clinic individual intervention. Compared with the in-clinic individual intervention, the mean difference in weight change was -1.9 kg (97.5% CI, -3.5 to -0.2 kg; P = .01) for the in-clinic group intervention and -1.4 kg (97.5% CI, -3.0 to 0.3 kg; P = .06) for the telephone group intervention.

Conclusions and relevance: Among patients with obesity in rural primary care clinics, in-clinic group visits but not telephone-based group visits, compared with in-clinic individual visits, resulted in statistically significantly greater weight loss at 24 months. However, the differences were small in magnitude and of uncertain clinical importance.

Trial registration: ClinicalTrials.gov Identifier: NCT02456636.

Figure 1.. Participant Flow Through the Rural Engagement in Primary Care for Optimizing Weight Reduction Trial

^aOne practice in the in-clinic group intervention randomized in cohort 1 declined prior to enrolling patients, and the replacement practice was randomized in cohort 2. One practice in the in-clinic group intervention and 1 practice in the in-clinic individual intervention randomized in cohort 3 declined, after randomization but prior to enrolling patients, and were subsequently replaced with the next recruited practices. ^bMedian number of patients screened per clinic: for in-clinic group visits, 54 (interquartile range [IQR], 49-58); for telephone group visits, 55 (IQR, 51-59); and for in-clinic individual visits, 51 (IQR, 47-57). ^cMedian number of patients enrolled per clinic: for in-clinic group visits, 40 (IQR, 39-42); for telephone group visits, 40 (IQR, 38-42); and for in-clinic individual visits, 40 (IQR, 39-41).

Figure 2.. Observed Weight Change by Treatment Group Including Primary Outcome at 24 Months

In the box plots, the middle lines represent median observed change in weight (calculated as baseline weight subtracted from observed follow-up weight), open squares represent mean observed change, box tops and bottoms represent interquartile range, whiskers extend to the most extreme observed values with 1.5 times the interquartile range of the nearer quartile, and dots represent observed values outside that range. More negative values indicate greater weight loss. Mean weights at baseline were 102.9 (SD, 15.5) kg for the in-clinic group intervention, 102.7 (SD, 15.6) kg for the telephone group intervention, and 103.1 (SD, 15.4) kg for the in-clinic individual intervention. Analyses used linear mixed-effects multilevel models, which included random cluster (clinic) effects, and adjusted for randomized strata (affiliated academic medical center) to examine the primary outcome of group comparison of absolute change in weight at 24 months. An unstructured covariance matrix was used, and missing weights were treated as missing at random and addressed using maximum likelihood methods. The primary outcome, the difference in mean weight change at 24 months compared with in-clinic individual visits was –1.9 kg (97.5% CI, –3.5 to –0.2 kg; P = .01) for in-clinic group visits and –1.4 kg (97.5% CI, –3.0 to 0.3 kg; P = .06) for telephone group visits. There were no significant differences between in-clinic group visits and telephone group visits.