Tailoring Mobile Health Technology to Reduce Obesity and ImproveCardiovascular Health in Resource-Limited Neighborhood Environments: A Multi-Level, Community-Based Physical Activity Intervention

华盛顿特区心血管（CV）健康和需求评估的女性数据为未来体育活动（PA）干预队列的移动健康用户参与度提供了初步见解。在接受评估的女性中（99%的非裔美国人平均年龄为59（12）岁），90%的体重指数（BMI）被归类为超重或肥胖，其中30%为i类肥胖，19%为ii类肥胖，17%为iii类肥胖。在不同体重类别中，PA降低（p0.05），自我报告的久坐时间增加（p0.05）。尽管女性的舒张压和空腹血糖在体重类别中显著升高，但血压、胆固醇和血糖相对控制良好，其平均值与美国心脏协会心血管健康临界值的理想或中间水平一致。在30天的研究期间，参与研究的女性的pa监测系统依从性保持在60%以上，在研究期间，肥胖女性的依从性相似。因此，移动健康技术与CBPR策略的部署可以帮助目标PA改善资源有限社区非洲裔美国妇女的心血管健康。

Data on women in the Washington, DC Cardiovascular (CV) Health and Needs Assessment provided initial insights into mHealth user engagement for the future physical activity (PA) intervention cohort. Among women in the assessment (99% African American mean age=59 (12) years), 90% had a body mass index (BMI) categorized as overweight or obese, with 30% having Class-I obesity, 19% having Class-II obesity, and 17% having Class-III obesity. Across weight classes, PA decreased (p0.05) and self-reported sedentary time increased (p0.05). Although diastolic blood pressure and fasting blood glucose significantly increased across weight categories among women, blood pressure, cholesterol, and glucose were relatively well-controlled with mean values consistent with ideal or intermediate levels of the American Heart Associations CV health cut-points. PA-monitoring system compliance remained above 60% for the 30-day study period among women participating in the study, with similar compliance among women with obesity over the study period. Therefore, deployment of mHealth technology with CBPR strategies can help target PA for improving cardiovascular health among African American women in resource-limited communities. As a first step for the intervention, we gathered qualitative data to inform the development of a mobile app that promotes PA among African American women in Washington, DC. We recruited a convenience sample of African American women (N=16, age range 51-74 years) from regions of Washington, DC metropolitan area with the highest burden of cardiovascular disease. Participants used an app created by the research team, which provided motivational messages through app push notifications and educational content to promote PA. Subsequently, participants engaged in semi-structured focus group interviews led by moderators who asked open-ended questions about participants experiences of using the app. Focus groups were audio-recorded and transcribed verbatim, with subsequent behavioral theory-driven thematic analysis. Key themes based on the Health Belief Model and emerging themes were identified from the transcripts. Three independent reviewers iteratively coded the transcripts until consensus was reached. Then, the final codebook was approved by a qualitative research expert. In this study, 10 main themes emerged. Participants emphasized the need to improve the app by optimizing automation, increasing relatability (eg, photos that reflect the target demographic), increasing educational material (eg, health information), and connecting with community resources (eg, cooking classes and exercise groups). Involving target users in the development of a culturally sensitive PA app is an essential step for creating an app that has a higher likelihood of acceptance and use in a technology-enabled intervention. This may decrease health disparities in CVD by more effectively increasing PA in a minority population. To gain more insights into the utility of interventions targeting PA, we used epidemiologic data to examine PA as a mediator of the relationship between neighborhood social environment perceptions and depressive symptoms in the Jackson Heart Study, a prospective, community-based cohort study of African-American adults from Jackson, Mississippi (n=2209). Perceived social environment was defined by perceived neighborhood violence (higher score=more violence), neighborhood problems (higher score=more problems), and social cohesion (higher score=more cohesion). Depressive symptoms were measured by the Center for Epidemiologic Studies Depression Scale (CES-D) score. Multilevel modeling was used to estimate associations between each social environment factor and depressive symptom scores, adjusting for covariates. Multivariable linear regressions with bootstrap-generated 95% bias-corrected confidence intervals were estimated to test for significant unstandardized indirect effects between neighborhood social environment and depression with self-reported PA as a mediator, controlling for all covariates. Greater perceived neighborhood violence and problems were related to greater depressive symptoms. Neighborhood violence and problems were also indirectly related to depressive symptoms. Social cohesion was not directly or indirectly related to depressive symptoms. PA appears to mediate the relationship between perceived social environment and depressive symptoms. These results suggest that for African Americans, improving individuals neighborhood perceptions may be beneficial for increasing PA levels, which can influence psychological well-being and CV health. We also developed a standardized approach for defining valid wear time for commercial available PA trackers for use in the community-based, adaptive PA intervention. We examined two methods for defining a valid day from the Fitbit Charge 2. In Method 1, a valid day was defined as greater than or equal to 10 hours per day of wear time with heart rate data. Method 2 removed minutes without heart rate data, minutes with heart rate data less than or equal to 2 SDs below mean and less than or equal to 2 steps, and nighttime minutes. Within the context of pilot data from the PA intervention, we found that the new method (Method 2) resulted in significantly different total wear time than the more conventional Method 1. Additional studies are needed to understand the impact of new methods of processing PA tracker data because there are no gold standards for comparison. We used a mixed-method approach to examine the adoption of mHealth technology among African American women in the DC area community. Community members completed an informatics survey prior to participation in focus groups about their use of mobile technology and health apps. Based on survey data, we found that 69% reported using health-related apps mostly focused on physical activity and nutrition. The focus groups identified four overarching themes in the focus groups (user attachment, technology adoption, potential barriers and facilitators), where individual app tailoring could be a facilitator and software concerns could serve as a barrier to adoption of a mobile app for an mHealth intervention. Thus, early engagement of target end-users as part of a co-design and community-based participatory research process may help in creating tools for future mHealth interventions. Finally, we created a simulation model to test the effects of a place-tailored digital health app on PA levels and overweight/obesity prevalence among African American women in Washington, DC. The place-tailored app would help users identify PA locations and available recreational center classes in Washington, DC. This work was done in collaboration with the Public Health Informatics, Computational, and Operations Research (PHICOR) group and the City University of New York and builds on our prior work in creating the Virtual Population Obesity Prevention Model for Washington DC. Using this simulation model, we showed that a digital health app that helps identify recreation center classes would need high levels of app engagement to have significant increases in PA and reductions in overweight/obesity. In particular, at least 75% of women would need to be aware of the app, 75% of those aware of the app would have to download it, and 75% of those downloading the app would have to opt in for push notifications to see significant effects. However, the app does not overcome lack of access to recreation centers. Our findings demonstrate that this type of place-tailored, digital health app should be incorporated into multi-level interventions that also target the built environment and other social determinants of health to maximize its impact.