Implementing texting programs in the P.O.W.E.R. (preventing obesity with eating right) medical group visit for weight loss

Abstract Background The effect of incorporating mobile technology to support participants’ lifestyle change and weight loss in medical group visits has not been well studied in a safety‐net setting. Rationale and Design Thus, the rationale of the current study was to examine the effect of text messaging in a medical group visit, and test the effect of two texting programs (12 weeks and 20 weeks), compared to those who did not receive text‐messaging in the Preventing Obesity With Eating Right (POWER) group visit program. The primary outcome was weight loss. Results We found that those enrolled in the 20‐week and 12‐week texting programs attended more group visit sessions than those enrolled in the POWER group only (p < 0.001). Both POWER and POWER + 20‐week texting groups had a significant reduction in weight at their final group visit compared to their baseline (POWER, 114 ± 27 kg vs. 112 ± 26 kg, p < 0.001; POWER + 20‐week texting, 111 ± 28 kg vs. 109 ± 28 kg, p < 0.01), but not the 12‐week texting group (114 ± 29 kg vs. 113 ± 29 kg, p = 0.22), with no differences between the groups. The number of group visits was correlated with a decrease in weight (rs = 0.12, p < 0.05). Conclusion In conclusion, text messaging programs led to more attendance in the medical group visits, but not greater weight loss or reduction in HbA1c than the POWER group obesity program alone. Further studies are needed to maximize the beneficial effects of texting programs in medical group visits in underserved minority populations.


| INTRODUCTION
Obesity (a body mass index (BMI) ≥ 30 kg/m 2 ) has become a major public health concern worldwide in the last 40 years. In the United States, obesity contributes to 100,000-400,000 excess deaths per year and is attributed to $117 billion dollars in medical expenditure. 1,2 Obesity poses a significant health disparity, largely affecting socioeconomically disadvantaged individuals which are disproportionately members of racial and ethnic minority groups. 3 The burden of obesity in socioeconomically disadvantaged communities calls for innovative efforts that are both cost-effective and sustainable. A study assessing weight loss maintenance over seven years found that weight-loss maintainers used more behavioral strategies to control fat intake and more strenuous physical activity than weight-regainers. 4 These strategies include self-monitoring, goal setting, eating habit modifications, and health behavior reinforcement. 5 Two approaches that have been shown to significantly improve behavior change in chronic conditions are the medical group visit model and the use of mobile technology. 6 Medical group visits, also known as "cooperative health care clinics," "shared medical appointments," or "group medical visits," deliver care to patients with similar conditions in a group setting  individuals). 7,8 The model uses a multi-disciplinary team that takes an educational approach to teach patients effective self-management strategies in addition to providing individual medical attention. 9 Medical group visit models benefit from both group therapy and the physician-patient relationship. Studies demonstrate that medical group visits improve efficiency in healthcare delivery, patient satisfaction, and use of preventative services, while decreasing emergency clinical visits. [10][11][12] Further, a study with low-income women managing chronic conditions, largely of Hispanic descent, found that group medical visits increased personalized attention (77%), self-care education (69%), and access to medication refills (69%), and significantly decreased urgent care visits during the 9-month intervention compared to the 9 months prior to the intervention (p < 0.05). 13 In the treatment of obesity, medical group visits have shown to lead to significant weight loss in developmentally delayed adults. 14 Another study in a pediatric population who attended a medical group visit at least twice in a 3-year period revealed improvement in BMI, stress, and healthy behaviors such as exercise and sleep, while decreasing unhealthy habits including high sugar beverage consumption, fast food intake, and television viewing time. 14,15 Despite the clear benefits of medical group visits, this care model poses potential limitations including logistical barriers such as transportation, missing appointments, family obligations, and difficulty relating to or supporting other group members from different racial/ethnic backgrounds. 13 Innovative approaches such as incorporating mobile technology in medical group visit settings can address some of these logistical barriers, while continuing the behavior interventions outside of the visits. Mobile phones are used in a variety of domains including the improvement of medication adherence, attendance of medical appointments, and disease self-management. 16,17 More specifically, using mobile phones to deliver short message service (SMS), or text messages, can impact behavior modifications. Text messaging allows for the delivery of individualized health communication and reinforcement. 18 A systematic review of health promotion and behavior interventions found that personalized periodic reminders for modifying diet, activity, and weight are effective to encourage and reinforce healthy behaviors. 19 Studies evaluating the effectiveness of using text messaging for the promotion of healthy weight loss behaviors and improvement of diabetes control in racial and ethnically diverse communities show promise as demonstrated in the feasibility of text messages. [20][21][22] Wadden at al. 23 commented that digitally delivered obesity programs expand treatment reach and lower costs. Although a recent systematic literature review on mHealth technology in historically underserved and minority populations in the United States failed to find an article on mHealth and obesity in this population, 24 we found several references of weight loss in this population. A randomized control trial with 124 African American adults who were overweight or obese found that the group who participated in a 6-month text messaging program in addition to the standard of care had an added weight loss of 3.5 kg, on average, than standard-care control at 6 months. 25 Additionally, a 12-month effectiveness randomized control trial with socioeconomically disadvantaged patients with obesity and elevated cardiac risk found that a mobile app behavior change intervention plus physician counseling rendered a larger weight loss relative to usual primary care in a community health care system. 26 A randomized control trial in Belgium comparing a conventional faceto-face weight loss program, a weight loss mobile app program, a combination of both, and a control found that while all interventions achieved weight loss from baseline, there was a trend of a greater number of participants in the combination group losing at least 5% of baseline weight compared to the mobile app group alone, suggesting that the combination of in-person interaction and mobile technology might be more effective to reaching weight control. 27 Thus, we sought to test the hypothesis that text-messaging programs would be a useful and effective strategy to help socioeconomically disadvantaged adults in a safety-net setting improve weight outcomes, hemoglobin A1c (HbA1c), and lipid levels compared to those in the medical group visit alone.
for English-speaking participants and the 2 nd and 4 th Mondays were for Spanish-speaking participants. Each session began with a thirtyminute Zumba exercise, followed by a live discussion on various topics led by an Endocrinologist (Theodore C. Friedman). Topics included obesity complications and prevention, with an emphasis on lifestyle changes. Additionally, Theodore C. Friedman encouraged open discussions to review participant challenges and personal experiences. Guest speakers were also invited to give brief talks to help address participants' main concerns and provide additional information. A dietitian also gave a lecture for about 45 min based on the Diabetes Prevention Program (DPP). 28 The study was approved by the Charles R. Drew University IRB (CDU IRB # 13-08-2409) and all participants signed an informed consent to participate in the study.  Figure 1 shows the number in each group and dropouts.    Table 1). Aside from the significantly higher HDL levels in the POWER group visit plus texting program groups (49 ± 14 mg/dl vs. 46 ± 12 mg/dl, p < 0.05) group and the significantly higher non-HDL levels in the POWER only group (141 ± 58 mg/dl vs. 129 ± 39 mg/dl, p < 0.05), there were no other significant differences in baseline clinical values between both groups.

| Medical group visit attendance
Those enrolled in the 20-week and 12-week texting programs attended more group visit sessions than those enrolled in the PO-WER group only (p < 0.0001). There was no significant difference in Note: p-values were calculated using or the quantitative data, the Wilcoxon rank-sum test was used; for the qualitative data, the chi-square test for homogeneity.
Abbreviations: BMI, body mass index; HbA1C, hemoglobin A1c; HDL, high-density lipoprotein; LDL, low-density lipoprotein; Non-HDL, Non-highdensity lipoprotein. * Difference between groups was significant p < 0.05. the number of group visits attended between the 20-week and 12week texting programs.  Table 2). Between-group analysis showed no significant difference for either weight loss in kg or% weight loss between groups when the two texting programs were examined separately

| Lipids
When examining lipids, there was no difference in cholesterol, LDL, non-HDL, or triglycerides between baseline and final measurements in the control group, the 12-week texting, and 20-week texting groups (Table 2). There was no difference between groups.  Limitations with the texting program should be noted. First, standard charges for texts from cell phone carriers may have limited some participants from entering and did not permit us to randomize which participants enrolled in the texting program. This may have led to a selection bias in those who enrolled in the text messaging programs. Moreover, it is unknown how much of the message content was read, comprehended, and applied. Notably, some of our participants reported having difficulty reading the messages due to the small font. Further, there were troubleshooting issues including lack of notification of cell phone number change, and inadvertently stopping the texting program. A short survey administered to those who did not enroll in the texting program assessed reasons for declining participation. Reasons to decline participation included security concerns due to lack of encryption, severe physical disabilities, texting charges, and lack of fluency in text language. One-on-one sessions were offered to learn how to text before enrolling in the texting program, but it is unknown whether those lessons were enough for participants to gain texting competency.
In summary, group visits and group visit plus text messaging led to weight loss and improvement in HbA1c in African Americans and Hispanics with obesity in a low-income urban area with the effect being due to the group visit without the added benefit of a text messaging program. Our findings support the recommendation of Bennett and colleagues 26 that digital obesity treatments including text messaging should not be used as replacements for individual or group interventions, but could be used within comprehensive obesity treatment programs to deliver educational materials, offer tailored feedback and facilitate encounters with providers. Further studies are needed to explore the applicability of text messaging with lifestyle modification programs within a safety-net setting.