Initial outcomes of a community health worker programme for patients with type 2 diabetes in refugee camps at the West Bank, occupied Palestinian territory: a retrospective, matched cohort study

Abstract

BackgroundNon-communicable diseases are the leading cause of mortality in the occupied Palestinian territory, and the burden of diabetes among refugees in the West Bank region of the territory has increased drastically in the past three decades. Community health worker programmes have been effective in controlling diabetes in many parts of the world, but data from the Middle East and north Africa regions and in settings of chronic violence are scarce. This study sought to assess the efficacy of diabetes control with a home-based community health worker programme, Health for Palestine, which included Palestinian refugee community health workers and patients in Aida and Al-Azza Camps in the occupied Palestinian territory. MethodsWe conducted a retrospective, matched cohort study, over 19 months, comparing the concentrations of haemoglobin A1c in individuals with diabetes cared for by Health for Palestine community health workers (n=30) and in non-health worker individuals with diabetes (n=90), all of whom were refugees who received care at UN Relief and Works Agency clinics. The Health for Palestine community health worker intervention, collectively termed social accompaniment, includes diet and exercise counselling, medication supervision and adherence support, vital signs and glucose monitoring, trauma-informed psychosocial counselling, facilitation of transportation to clinic, and supportive referrals to clinics or hospitals. We tested for significance with multivariable linear regression with robust standard errors, using an ordinary least squares regression of the outcome with and without prespecified covariates. FindingsHealth for Palestine community health workers had an average 1·4 absolute percentage point (95% CI 0·66–2·1) greater decline in haemoglobin A1c per year than the non-community health worker cohort, after adjustment for potential confounders including smoking status, hypertension, physical activity, and history of myocardial infarction or stroke (p<0·001). Potentially relevant unmeasured confounders include patient proximity to a UN Relief and Works Agency clinic, whether the control patient lived within a refugee camp or nearby, and traumatic events related to the occupation. The unadjusted reduction in haemoglobin A1c in Health for Palestine community health workers with diabetes was 1·2 points (from a baseline value of 8·4% to 7·2%). At baseline, 50% of individuals with community health workers and 47% of individuals without community health workers met the haemoglobin A1c target of <8% (p=0·83, Fisher's exact analysis); at the end of the study period, 77% of individuals visited by Health for Palestine community health workers and 46% of those without community health workers met the target (p=0·003, Fisher's exact analysis). InterpretationA community health worker intervention grounded in social accompaniment is associated with improved diabetes control in urban refugee camps and in communities experiencing chronic violence. A low exclusion cutoff for haemoglobin A1c (<6·5%) might underestimate the programme's effectiveness. Although community health worker programmes are uncommon in the occupied Palestinian territory and the greater region, these results suggest that such programmes can successfully augment existing efforts to improve diabetes control and outcomes. Funding1for3 and Lajee Center.