Abstract
BACKGROUND: Implementation research to improve hypertension control is scarce in Latin America. We assessed the effectiveness of an intervention aimed at primary care practitioners and hypertensive patients in a setting that provides integrated care through an accessible network of family practices. METHODS: We conducted in Cardenas and Santiago, Cuba, a controlled before-after study in 122 family practices, which are staffed with a doctor and a nurse. The intervention comprised a control arm (usual care), an arm with a component targeting providers (hypertension management workshops), and an arm with, on top of the latter, a component targeting patients (hypertension schools). To evaluate the effect, we undertook a baseline survey before the intervention and an endline survey sixteen months after its start. In each survey, we randomly included 1400 hypertensive patients. Controlled hypertension, defined as a mean systolic and diastolic blood pressure below 140 and 90 mmHg, respectively, was the primary endpoint assessed. We performed linear and logistic regression with a Generalized Estimating Equations approach to determine if the proportion of patients with controlled hypertension changed following the intervention. RESULTS: Seventy-three doctors, including substitutes, and 54 nurses from the 61 intervention family practices attended the provider workshops, and 3308 patients -51.6% of the eligible ones- participated in the hypertension schools. Adherence to anti-hypertensive medication improved from 42% at baseline to 63% at the endline in the intervention arms. Under the provider intervention, the proportion of patients with controlled hypertension increased by 18.9%, from 48.7% at baseline to 67.6% at endline. However, adding the component that targeted hypertensive patients did not augment the effect. Compared to patients in the control arm, the adjusted OR of having controlled hypertension was 2.36 (95% CI, 1.73-3.22) in the provider and 2.00 (95% CI, 1.68-2.37) in the provider plus patient intervention arm. CONCLUSIONS: The intervention's patient component remains to be fine-tuned. Still, we demonstrate that it is feasible to substantially improve hypertension outcomes by intervention at the primary care level, despite an already relatively high control rate.