Baseline characteristics of participants in MEDLY Uganda: a stepped-wedge cluster randomized type 2 hybrid implementation trial of heart failure care in Uganda
J Schwartz, D Spiegelman, J L Davis, H Ross, S Wali, C Nalwadda, A Hagaman, D Cameron, R Hernandez-Ramirez, C Muhangi Ngonzi, S Christie, J Kayima, P Lwabi, J Cafazzo, I SsinabulyaAbstract
Background
Heart failure (HF) in sub-Saharan Africa is associated with poor outcomes. Prior prospective African HF cohorts have largely focused on hospitalized populations and relied primarily on clinical diagnostic criteria. Few studies have employed biomarker- and echo-confirmed HF definitions, longitudinal follow-up across decentralized sites, or implementation science frameworks to improve care delivery. ImpleMEntation of a Digital-first care deLiverY model for heart failure in Uganda (MEDLY Uganda) is a stepped-wedge cluster randomized type 2 hybrid effectiveness-implementation trial conducted at six regional referral hospitals (RRHs) in Uganda. Following baseline, sequential implementation strategies include a clinical care improvement bundle (“Core-HF”) and Core-HF plus a patient- and clinician-facing digital health bundle (“Medly Uganda”). Primary outcomes are composite of HF hospitalization and all-cause mortality (clinical) and Self-Care in Heart Failure Index (implementation).
Purpose
To describe the baseline demographic, clinical, biomarker, and treatment characteristics of participants enrolled in MEDLY Uganda.
Methods
We recruited adults (≥18 years) with suspected New York Heart Association (NYHA) Class II or III HF based on clinical (Framingham) criteria at six RRH outpatient departments. Participants with NT-proBNP <125 pg/mL were excluded unless ejection fraction (EF) was <50%. Baseline assessments included sociodemographics, comorbidities, NYHA class, medication use, echocardiography, NT-proBNP, healthcare expenditures, and self-care and quality-of-life measures. Descriptive statistics were used to characterize the cohort.
Results
547 (92%) of a target sample size of 596 participants have been enrolled. 351 (64%) are female and the mean age is 57 (+/-18) years. 321 (59%) are rural-dwelling, 331 (61%) are married, 365 (67%) are unemployed, and the highest education level for 278 (51%) was primary school. 360 (66%), 44 (8%), and 65 (12%) had been diagnosed with hypertension, diabetes, or hyperlipidemia, respectively. Mean body mass index was 26 kg/m2 and waist-to-hip ratio is 0.93 (+/-0.2). Mean systolic and diastolic blood pressure were 121 (+/-26) and 81 (+/-16) mmHg, respectively. 370 (67%) and 176 (32%) were NYHA Class II and III, respectively. Mean NT-proBNP was 5267pg/mL (range 125-35,000). 293 (54%) reported taking medication less frequently than prescribed.
Conclusion
Participants enrolled in MEDLY Uganda represent a rigorously-phenotyped, biomarker-confirmed, geographically disparate outpatient HF population in Uganda. This study uniquely combines objective diagnostic criteria with a longitudinal stepped-wedge cluster randomized hybrid implementation design evaluating sequential health system interventions. Baseline findings highlight the clinical profile of HF in regional Uganda and substantial opportunities to improve guideline-directed therapy, self-care, and health system performance.MEDLY Uganda site launchMEDLY Uganda logo