This study will examine whether an adapted version of ]he Diabetes Prevention Program (DPP) can be effective when delivered by community health workers (CHWs) in a poor urban community in South Africa. We will randomize existing groups that are part of an non-governmental organization's (NGO) chronic disease management program (anticipated cluster N = 54; anticipated individual N=540), to receive either the program or usual care (wait-list). The primary outcome analysis will compare percentage of baseline weight loss at Y1 between the program and usual care; however, after Y1 usual care participants will also receive the program and both groups will be followed for another year.

Inclusion Criteria:

• member of participating partner NGO "health club" (members are medically stabilized individuals with diabetes mellitus or hypertension referred for disease and lifestyle management)
• BMI greater than or equal to 25 kg per meter squared

Exclusion Criteria:

• unsafe level of blood pressure (greater than or equal to 160 (systolic) and greater than equal to 100 mm (diastolic)) at screening
• elevated blood sugar (A1C greater than 11) at screening
• being pregnant, breast-feeding or planning pregnancy within 2 years
• chronic use of oral steroid medication
• intellectual disabilities that would prevent ability to understand the program
• not intending to stay in the health club over the next 2 years