Leveraging Ecological Momentary Assessment Data to Characterize Individual Mobility: Exploratory Pilot Study in Rural Uganda.

BACKGROUND: The geographical environments within which individuals conduct their daily activities may influence health behaviors, yet little is known about individual-level geographic mobility and specific, linked behaviors in rural low- and middle-income settings. OBJECTIVE: Nested in a 3-month ecological momentary assessment intervention pilot trial, this study aims to leverage mobile health app user GPS data to examine activity space through individual spatial mobility and locations of reported health behaviors in relation to their homes. METHODS: Pilot trial participants were recruited from the Rakai Community Cohort Study-an ongoing population-based cohort study in rural south-central Uganda. Participants used a smartphone app that logged their GPS coordinates every 1-2 hours for approximately 90 days. They also reported specific health behaviors (alcohol use, cigarette smoking, and having condomless sex with a non-long-term partner) via the app that were both location and time stamped. In this substudy, we characterized participant mobility using 3 measures: average distance (kilometers) traveled per week, number of unique locations visited (deduplicated points within 25 m of one another), and the percentage of GPS points recorded away from home. The latter measure was calculated using home buffer regions of 100 m, 400 m, and 800 m. We also evaluated the number of unique locations visited for each specific health behavior, and whether those locations were within or outside the home buffer regions. Sociodemographic information, mobility measures, and locations of health behaviors were summarized across the sample using descriptive statistics. RESULTS: Of the 46 participants with complete GPS data, 24 (52%) participants were men, 30 (65%) participants were younger than 35 years, and 33 (72%) participants were in the top 2 socioeconomic status quartiles. On median, participants traveled 303 (IQR 152-585) km per week. Over the study period, participants on median recorded 1292 (IQR 963-2137) GPS points-76% (IQR 58%-86%) of which were outside their 400-m home buffer regions. Of the participants reporting drinking alcohol, cigarette smoking, and engaging in condomless sex, respectively, 19 (83%), 8 (89%), and 12 (86%) reported that behavior at least once outside their 400-m home neighborhood and across a median of 3.0 (IQR 1.5-5.5), 3.0 (IQR 1.0-3.0), and 3.5 (IQR 1.0-7.0) unique locations, respectively. CONCLUSIONS: Among residents in rural Uganda, an ecological momentary assessment app successfully captured high mobility and health-related behaviors across multiple locations. Our findings suggest that future mobile health interventions in similar settings can benefit from integrating spatial data collection using the GPS technology in mobile phones. Leveraging such individual-level GPS data can inform place-based strategies within these interventions for promoting healthy behavior change.

Acceptability and feasibility of mobile phone-based ecological momentary assessment and intervention in Uganda: A pilot randomized controlled trial.

Valid, reliable behavioral data and contextually meaningful interventions are necessary for improved health outcomes. Ecological Momentary Assessment and Intervention (EMAI), which collects data as behaviors occur to deliver real-time interventions, may be more accurate and reliable than retrospective methods. The rapid expansion of mobile technologies in low-and-middle-income countries allows for unprecedented remote data collection and intervention opportunities. However, no previous studies have trialed EMAI in sub-Saharan Africa. We assessed EMAI acceptability and feasibility, including participant retention and response rate, in a prospective, parallel group, randomized pilot trial in Rakai, Uganda comparing behavioral outcomes among adults submitting ecological momentary assessments (EMA) versus EMAI. After training, participants submitted EMA data on five nutrition and health risk behaviors over a 90-day period using a smartphone-based application utilizing prompt-based, participant-initiated, and geospatial coordinate data collection, with study coordinator support and incentives for >50% completion. Included behaviors and associated EMAI-arm intervention messages were selected to pilot a range of EMAI applications. Acceptability was measured on questionnaires. We estimated the association between high response rate and participant characteristics and conducted thematic analysis characterizing participant experiences. Study completion was 48/50 participants. Median prompt response rate was 66.5% (IQR: 60.0%-78.6%). Prior smartphone app use at baseline (aPR 3.76, 95%CI: 1.16-12.17, p = 0.03) and being in the intervention arm (aPR 2.55, 95% CI: 1.01-6.44, p = 0.05) were significantly associated with the top response rate quartile (response to >78.6% of prompts). All participants submitted self-initiated reports, covering all behaviors of interest, including potentially sensitive behaviors. Inconsistent phone charging was the most reported feasibility challenge. In this pilot, EMAI was acceptable and feasible. Response rates were good; additional strategies to improve compliance should be investigated. EMAI using mobile technologies may support improved behavioral data collection and intervention approaches in low and middle-income settings. This approach should be tested in larger studies.

Data Velocity in HIV-Related Implementation Research: Estimating Time From Funding to Publication.

BACKGROUND: Given available effective biomedical and behavioral prevention and treatment interventions, HIV-related implementation research (IR) is expanding. The rapid generation and dissemination of IR to inform guidelines and practice has the potential to optimize the impact of the Ending the Epidemic Initiative and the HIV pandemic response more broadly. METHODS: We leveraged a prior mapping review of NIH-funded awards in HIV and IR from January 2013 to March 2018 and identified all publications linked to those grants in NIH RePORTER through January 1, 2021 (n = 1509). Deduplication and screening of nonoriginal research reduced the count to 1032 articles, of which 952 were eligible and included in this review. Publication volume and timing were summarized; Kaplan-Meier plots estimated time to publication. RESULTS: Among the 215 NIH-funded IR-related awards, 127 of 215 (59%) published original research directly related to the grant, averaging 2.0 articles (SD: 3.3) per award, largely in the early IR phases. Many articles (521 of 952, 55%) attributed to grants did not report grant-related data. Time from article submission to publication averaged 205 days (SD: 107). The median time-to-first publication from funding start was 4 years. Data dissemination velocity varied by award type, trending toward faster publication in recent years. Delays in data velocity included (1) time from funding to enrollment, (2) enrollment length, and (3) time from data collection completion to publication. CONCLUSION: Research publication was high overall, and time-to-publication is accelerating; however, over 40% of grants have yet to publish findings from grant-related data. Addressing bottlenecks in the production and dissemination of HIV-related IR would reinforce its programmatic and policy relevance in the HIV response.

Patient-level barriers and facilitators to completion of patient-reported outcomes measures.

PURPOSE: To identify patient-level barriers and facilitators to completion of patient-reported outcomes measures (PROMs) in a hand and upper extremity clinic in Baltimore, Maryland. METHODS: We conducted 12 h of direct observation of PROM completion (October-November, 2020). Ethnographic observation memos were qualitatively analyzed for barriers and facilitators using rapid thematic analysis. Informed by observation findings, we conducted 17 semi-structured interviews with mixed-literacy patients, caregivers, and clinic staff to understand the patient experience when completing PROMs (November 2020-March 2021). We identified initial themes through inductive and deductive framework analysis and validated findings through subsequent interviews with member-checking. RESULTS: We identified nine patient-level factors that influence PROM completion: platform design, print literacy, health literacy, technology literacy, language proficiency, physical functioning, vision, cognitive functioning, and time. CONCLUSIONS: There are multiple distinct patient-level factors that affect PROM completion. Failure to consider these factors in PROM design and implementation may lower completion rates or prevent accurate completion, undermining PROM validity. Because certain factors affect minority populations at disproportionate rates, this may also contribute to existing health disparities.

Mobile Ecological Momentary Assessment and Intervention and Health Behavior Change Among Adults in Rakai, Uganda: Pilot Randomized Controlled Trial.

BACKGROUND: An extraordinary increase in mobile phone ownership has revolutionized the opportunities to use mobile health approaches in lower- and middle-income countries (LMICs). Ecological momentary assessment and intervention (EMAI) uses mobile technology to gather data and deliver timely, personalized behavior change interventions in an individual's natural setting. To our knowledge, there have been no previous trials of EMAI in sub-Saharan Africa. OBJECTIVE: To advance the evidence base for mobile health (mHealth) interventions in LMICs, we conduct a pilot randomized trial to assess the feasibility of EMAI and establish estimates of the potential effect of EMAI on a range of health-related behaviors in Rakai, Uganda. METHODS: This prospective, parallel-group, randomized pilot trial compared health behaviors between adult participants submitting ecological momentary assessment (EMA) data and receiving behaviorally responsive interventional health messaging (EMAI) with those submitting EMA data alone. Using a fully automated mobile phone app, participants submitted daily reports on 5 different health behaviors (fruit consumption, vegetable consumption, alcohol intake, cigarette smoking, and condomless sex with a non-long-term partner) during a 30-day period before randomization (P1). Participants were then block randomized to the control arm, continuing EMA reporting through exit, or the intervention arm, EMA reporting and behavioral health messaging receipt. Participants exited after 90 days of follow-up, divided into study periods 2 (P2: randomization + 29 days) and 3 (P3: 30 days postrandomization to exit). We used descriptive statistics to assess the feasibility of EMAI through the completeness of data and differences in reported behaviors between periods and study arms. RESULTS: The study included 48 participants (24 per arm; 23/48, 48% women; median age 31 years). EMA data collection was feasible, with 85.5% (3777/4418) of the combined days reporting behavioral data. There was a decrease in the mean proportion of days when alcohol was consumed in both arms over time (control: P1, 9.6% of days to P2, 4.3% of days; intervention: P1, 7.2% of days to P3, 2.4% of days). Decreases in sex with a non-long-term partner without a condom were also reported in both arms (P1 to P3 control: 1.9% of days to 1% of days; intervention: 6.6% of days to 1.3% of days). An increase in vegetable consumption was found in the intervention (vegetable: 65.6% of days to 76.6% of days) but not in the control arm. Between arms, there was a significant difference in the change in reported vegetable consumption between P1 and P3 (control: 8% decrease in the mean proportion of days vegetables consumed; intervention: 11.1% increase; P=.01). CONCLUSIONS: Preliminary estimates suggest that EMAI may be a promising strategy for promoting behavior change across a range of behaviors. Larger trials examining the effectiveness of EMAI in LMICs are warranted. TRIAL REGISTRATION: ClinicalTrials.gov NCT04375423; https://www.clinicaltrials.gov/ct2/show/NCT04375423.

Developing a protocol for adapting multimedia patient-reported outcomes measures for low literacy patients.

BACKGROUND: Self-administration of patient-reported outcomes measures (PROMs) by patients with low literacy is a methodologic and implementation challenge. There is an increasing emphasis on patient-centered care and wider adoption of PROMs to understand outcomes and measure healthcare quality. However, there is a risk that the use of PROMs could perpetuate health disparities unless they are implemented in an inclusive fashion. We present a protocol to adapt validated, text-based PROMs to a multimedia format (mPROMs) to optimize self-administration in populations with limited literacy. We describe the processes used to develop the protocol and the planned protocol implementation. METHODS/DESIGN: Our study protocol development was guided by the International Quality of Life Assessment (IQOLA) protocol for translating and culturally adapting PROMs to different languages. We used the main components of IQOLA's protocol to generate a conceptual framework to guide development of a Multimedia Adaptation Protocol (MAP). The MAP, which incorporates human-centered design (HCD) and takes a community-engaged research approach, includes four stages: forward adaptation, backward adaptation, qualitative evaluation, and validation. The MAP employs qualitative and quantitative methods including observation, cognitive and discovery interviews, ideation workshops, prototyping, user testing, co-creation interviews, and psychometric testing. An iterative design is central to the MAP and consistent with both the IQOLA protocol and HCD processes. We will pilot test and execute the MAP to adapt the Patient Reported Outcomes Measurement Information System (PROMIS) Upper Extremity Short Form for use in a mixed literacy hand and upper extremity patient population in Baltimore, Maryland. DISCUSSION: The MAP provides an approach for adapting PROMs to a multimedia format. We encourage others to evaluate and test this approach with other questionnaires and patient populations. The development and use of mPROMs has the potential to expand our ability to accurately capture PROs in limited literacy populations and promote equity in PRO measurement.

Rethinking retention: Mapping interactions between multiple factors that influence long-term engagement in HIV care.

BACKGROUND: Failure to keep people living with HIV engaged in life-long care and treatment has serious implications for individual and population-level health. Nested within a four-province study of HIV care and treatment outcomes, we explored the dynamic role of social and service-related factors influencing retention in HIV care in Zambia. METHODS: From a stratified random sample of 31 facilities, eight clinics were selected, one urban and one rural from each province. Across these sites we conducted a total of 69 in-depth interviews, including with patients (including pregnant women) engaged in-care (n = 28), disengaged from care (n = 15), engaged facility transferee (n = 12), and friends/family of deceased patients (n = 14). At the same sites we conducted 24 focus group discussions with a total of 192 lay and professional healthcare workers (HCWs). Two-day observations in each of the eight facilities helped triangulate data on operational context, provider relations and patient-provider interactions. We ordered and analysed data using an adapted version of Ewart's Social Action Theory. RESULTS: Three overarching findings emerged. First, the experience of living with HIV and engaging in HIV care in Zambia is a social, not individual experience, influenced by social and gendered norms and life goals including financial stability, raising family and living stigma-free. Second, patients and their networks act collectively to negotiate and navigate HIV care. Anticipated responses from social network influenced patients' willingness to engage in care, while emotional and material support from those networks influenced individuals' capacity to remain in HIV care. Lastly, health system factors were most influential where they facilitated or undermined peoples' collective approach to health service use. Participants living with HIV reported facilitation of both their initial and continued engagement in care where services involved social networks, such as during couples testing and community outreach. Conversely, service features that were poorly aligned with respondents' social reality (e.g. workplace obligations) hindered long-term engagement. CONCLUSIONS: This study moves beyond listing barriers or socio-ecological groupings, to explain how social and health systems interact to produce HIV care outcomes. Our findings challenge the implicit assumption of individual agency underpinning many retention studies to highlight the social nature of illness and healthcare utilization for HIV in Zambia. This understanding of collective action for accessing and remaining in HIV care should underpin future efforts to revise and reform HIV and potentially other chronic service models and systems.

Integration of Family Planning Services into HIV Care and Treatment Services: A Systematic Review.

Evidence on the feasibility, effectiveness, and cost-effectiveness of integrating family planning (FP) and HIV services has grown significantly since the 2004 Glion Call to Action. This systematic review adds to the knowledge base by characterizing the range of models used to integrate FP into HIV care and treatment, and synthesizing the evidence on integration outcomes among women living with HIV. Fourteen studies met our inclusion criteria, eight of which were published after the last systematic review on the topic in 2013. Overall, integration was associated with higher modern method contraceptive prevalence and knowledge, although there was insufficient evidence to evaluate its effects on unintended pregnancy or achieving safe and healthy pregnancy. Evidence for change in unmet need for FP was limited, although two of the three evaluations that measured unmet need suggested possible improvements associated with integrated services. However, improving access to FP services through integration was not always sufficient to increase the use of more effective (noncondom) modern methods among women who wanted to prevent pregnancy. Integration efforts, particularly in contexts where contraceptive use is low, must address community-wide and HIV-specific barriers to using effective FP methods alongside improving access to information, commodities, and services within routine HIV care.

A Preliminary Assessment of Rotavirus Vaccine Effectiveness in Zambia.

BACKGROUND: Diarrhea is the third leading cause of child death in Zambia. Up to one-third of diarrhea cases resulting in hospitalization and/or death are caused by vaccine-preventable rotavirus. In January 2012, Zambia initiated a pilot introduction of the Rotarix live, oral rotavirus vaccine in all public health facilities in Lusaka Province. METHODS: Between July 2012 and October 2013, we conducted a case-control study at 6 public sector sites to estimate rotavirus vaccine effectiveness (VE) in age-eligible children presenting with diarrhea. We computed the odds of having received at least 1 dose of Rotarix among children whose stool was positive for rotavirus antigen (cases) and children whose stool was negative (controls). We adjusted the resulting odds ratio (OR) for patient age, calendar month of presentation, and clinical site, and expressed VE as (1 - adjusted OR) × 100. RESULTS: A total of 91 rotavirus-positive cases and 298 rotavirus-negative controls who had under-5 card-confirmed vaccination status and were ≥6 months of age were included in the case-control analysis. Among rotavirus-positive children who were age-eligible to be vaccinated, 20% were hospitalized. Against rotavirus diarrhea of all severity, the adjusted 2-dose VE was 26% (95% confidence interval [CI], -30% to 58%) among children ≥6 months of age. VE against hospitalized children ≥6 months of age was 56% (95% CI, -34% to 86%). CONCLUSIONS: We observed a higher point estimate for VE against increased severity of illness compared with milder disease, but were not powered to detect a low level of VE against milder disease.