Outcomes Research in Review

What PCP-Related Factors Contribute to Successful Weight Loss Among Positive Deviant Low-Income African-American Women?

Banerjee ES, Herring SJ, Hurley KE, et al. Overcoming obesity: a mixed methods study of the impact of primary care physician counseling on low-income African-American women who successfully lost weight. Am J Health Promot 2017;94:81–90.



Study Overview

Objective. To evaluate factors related to interactions with primary care physicians (PCPs) that may contribute to successful weight loss and maintenance among low-income, African-American women.

Design. Mixed methods, positive deviance framework.

Setting and participants . Participants were African-American women aged 18–64 years from an urban university-based family medicine practice who received Medicaid, resided in Philadelphia, and had a body mass index (BMI) of ≥ 30kg/m 2. From among these, “positive deviant” cases were identified as patients with EMR-confirmed weight loss of at least 10% of patient’s maximum weight between 2007–2012 and maintenance of this loss for at least 6 months. Controls were defined as patients who had not lost a significant amount of weight during this time period. Patients were excluded if they were an amputee or wheelchair-bound; had bariatric surgery, severe illness during weight loss, EMR-documented unintended weight loss, pregnancy at time of weight loss, a psychiatric disorder or were taking antipsychotic medication; had an intellectual disability; or could not give consent to participate.

Main outcomes measures. PCP- and patient-reported weight variables were collected through the EMR (documentation of dietary counseling by PCP, documentation of a weight-related problem, diagnosis of overweight, obesity, or morbid obesity on the problem list), surveys (additional predicters of positive deviant membership, including patient-reported weight-related diagnosis or discussion of weight with PCP or health professional), and interviews. Logistic regression was used to determine whether a priori-identified EMR and survey variables could predict positive deviant group membership, adjusting for demographic variables significantly associated with the outcome of interest or hypothesized to be confounders of the associations between predictors and outcomes (results were adjusted for age in the EMR analysis and for employment status and education level in the survey analysis). Once thematic saturation was reached, interviews were analyzed by a 4-member coding panel using a modified approach to grounded theory to identify themes.

Main results. For the EMR analysis, data from 161 positive deviant cases and 602 controls were analyzed. For the survey analysis, data from 35 positive deviant cases and 36 controls matched for age and maximum BMI were analyzed. For in-depth interviews, thematic saturation was reached after collecting data from 20 positive deviant participants. In the EMR analyses, documentation of dietary counseling and a weight-related diagnosis were significant predictors of positive deviant membership after adjusting for age ( P < 0.001 and P = 0.011, respectively). However, documentation of obesity on the problem list was predictive of control group membership ( P = 0.032). In the survey analysis, neither patient-reported weight-related diagnosis nor discussion of weight with a medical provider were predictors of positive deviant membership ( P = 0.890 and P = 0.373, respectively). In the qualitative analysis of interviews with positive deviant participants, 5 themes emerged: (1) framing the problem of obesity in the context of other health problems provided motivation; (2) having a full discussion around weight management was important; (3) an ongoing conversation and relationship was valuable; (4) celebrating small successes was beneficial for ongoing motivation; and (5) advice was helpful but self-motivation was required in order to make a change.

Conclusions. PCP counseling may be an important factor in promoting weight loss in low-income, African-American women, a population at high risk for obesity. Patients may benefit from their PCPs drawing connections between obesity and weight-related medical conditions and enhancing intrinsic motivation for weight loss.


The increasing prevalence and clinical consequences of having obesity are well-documented, with low-income minorities disproportionately burdened by this condition [1,2]. The United States Preventive Services Task Force (USPTF) recommends that all patients be screened for obesity and offered intensive lifestyle counseling [3], yet evidence-based guidelines for best approaches to incorporate this into practice are few and unclear, and even fewer are specific to high-risk patient populations [4–9].

This study adds to the literature by using a positive deviance approach to identify PCP-related factors that predict successful weight loss among low-income African-American women. This approach has rarely been used in the obesity literature. In a few childhood obesity studies, this approach was used to identify motivations used by child “positive outliers” to improve their BMI [10], characterize variations of feeding and activity practices by parents of healthy children normally at high risk for obesity [11], and explore successful health and BMI reduction strategies used among positive outlier families [12]. Positive deviance has also been used to characterize and change nutritional behavior and understand successful weight-control practices among adults [13–15]. One study has suggested that studying “positive deviant” physicians that regularly provide weight counseling may help to provide practice methods to increase these practices in the primary care settings [16].

Thus, the study approach in using a positive deviance framework is an important and unique strength. Addi-tionally, the authors used a mixed-methods approach, analyzing EMR, survey, and interview data to assess PCP- and patient-reported weight-related factors that predict successful weight loss. As the authors describe, their results confirm findings from previous studies looking at counseling preferences among ethnic minority women and PCP attitudes and practices related to weight management.

They acknowledge important limitations of their study design, primarily the generalizability of findings only to urban, low-income, African-American women, the small sample size in the survey analysis, and the use of EMR data to collect data on PCP counseling (as opposed to interviews, for example). It important to also acknowledge that this study was conducted at a family medicine practice, and physician behavior and practices likely do not generalize to other PCPs and specialists. Additionally, while their intention was to use a positive deviance framework, conducting interviews among a subset of their control cases may have provided useful information regarding negative or ineffective PCP interactions regarding weight loss and management.

Applications for Clinical Practice

As the authors emphasize, the outcomes of this study are especially relevant for PCPs and other health practitioners, as the identified themes can help guide weight counseling that incorporates patient preferences and promotes successful weight loss. Importantly, these findings underscore that the role of the physician is important in promoting weight loss, yet it does not require in-depth knowledge and training in evidence-based weight loss strategies. While dietary counseling is still helpful, patients with successful weight loss value the supportive relationship with their physician, their physician drawing connections between obesity and weight-related medical conditions, and their physician enhancing intrinsic motivations for weight loss.

Next Article: