Back

Implementation and early outcomes of a dedicated diabetes and hypertension clinic in rural Haiti: A 24-month retrospective cohort study

Moise, K.; Larco, P.; Charlotin, E. D.; Mexis, K.; Jean-Baptiste, E.; Larco, N. C.

2026-07-27 public and global health
10.64898/2026.07.22.26358732 medRxiv
Show abstract

Noncommunicable diseases (NCDs), particularly diabetes and hypertension, are a growing cause of death in low-income countries. In Haiti, cardiovascular disease has overtaken HIV as the leading cause of adult death, and the isolated Grand'Anse Department had no dedicated chronic disease facility before this initiative. This retrospective single-center cohort study analyzed 1,298 encounters among 590 patients at a dedicated diabetes and hypertension clinic (the Centre Dr. Rene Charles) in Jeremie, Grand'Anse, from April 2024 to March 2026. Standardized assessments included serial blood pressure, fasting glucose, HbA1c, lipids, renal indices, and electrocardiography. Longitudinal change was modeled with linear mixed-effects models on calendar time, supplemented by paired first-to-last comparisons. The cohort was 74.1% female (mean age, 58.2 years), and nearly two thirds had primary-level education or none. Diagnoses were hypertension alone (52.9%), diabetes with hypertension (26.8%), and diabetes alone (20.3%). Establishing the clinic in this setting proved feasible, but 54% of patients attended only once, making early attrition the central finding. Among the minority who returned, exploratory within-patient analyses showed improvement in blood pressure: in mixed-effects models on calendar time, systolic pressure fell by 0.78 mmHg per month among patients with hypertension (95% CI, 0.36 to 1.21; P=0.005), the paired first-to-last decline was 9.0 mmHg (P<0.001), and control rose from 21.4% to 38.1% (P<0.001). These estimates come only from returners and are best read as hypothesis-generating. Glycemic control did not durably improve: despite a large paired fasting-glucose fall (52.5 mg/dL) and a borderline time trend (P=0.05), mean HbA1c remained 10.4%, with only 13.2% at target. Indication-driven screening of a clinically selected minority frequently detected end-organ damage, reflecting selective testing rather than cohort-wide prevalence. A dedicated NCD clinic can be established in one of the world's most resource-limited settings; the priorities now are patient retention, broader medication access, and durable health-system investment.

Matching journals

The top 10 journals account for 50% of the predicted probability mass.

50% of probability mass above

"Similar papers" are the closest papers from that journal in the model's embedding space. They show what the match is built on, but the ranking comes mostly from a classifier over the whole training set, not from these examples alone.