Merriam 2009.

Methods	Study design: randomized controlled trial Sampling frame: Latino clients of Greater Lawrence Family Health Center  Sampling method: randomization at household level Collection method: recruitment letter with telephone follow‐up; supplementary outreach via public access TV, Spanish radio, and bilingual newspapers. Clinic visits for anthropometric measures and blood draws for lab values; oral surveys in Spanish for diet intake, physical activity, and depression  Description of the community coalition: Greater Lawrence Family Health Center (GLFHC) serving 80% of local Latino population houses a REACH diabetes project. Collaborators in study plan and implementation included family health center, Latina  PI (physician at GLFHC), and Latina co‐PI from U Mass medical school, local Council on Aging/Senior Center, YWCA, and Mayor's health task force. Community co‐ordinators of project came from Lawrence Council on Aging and YWCA
Participants	Communities: Greater Lawrence, MA area (urban) Country: USA Ages included in assessment: 25 and older Reasons provided for selection of intervention community: diabetes prevalence estimated at 11.8% among Lawrence Latino adults vs 6.4% among non‐Hispanics statewide; population of target community ˜60% Latino; target community also with high rates of poverty and limited resources Intervention community (population size): population of community not reported. Study completed 9959 telephone screening invitation calls, 2638 completed screening calls. 949 Latino clients of GLFHC assessed for eligibility, 312 enrolled, and 162 allocated to intervention Comparison community (population size): See above — 150 allocated to usual care
Interventions	Name of intervention: Lawrence Latino Diabetes Prevention Project (LLDPP); similar to Centers for Disease Control Diabetes Prevention Program Theory: Social Cognitive Theory, Patient‐Centered Counseling Aim: To reduce risk factors for type 2 diabetes among high‐risk Latinos (those with > 30% risk of developing diabetes in next 7.5 years) Description of costs and resources: REACH project funded by CDC. First‐year costs per participant $661 Components of the intervention: provision of information in Spanish via multiple formats in 13 group sessions, provision of pedometers, 3 individual home visits for monitoring progress, cash incentives, and transportation support Start date: October 2004 Duration: 1 year
Outcomes	Outcomes and measures: weight change, HgA1, physical activity Time points: baseline, 6 months, 1‐year follow‐up
Notes	To address low English fluency/low Spanish literacy, information presentations included video novellas, colorful food photo cards, and cooking demonstrations adapting culturally familiar foods. Significant family members were welcome to attend group sessions. Intervention fidelity was promoted through extensive training by PIs of 3 Spanish‐speaking community individuals who implemented intervention components. Study authors note that they successfully reduced the costs of the model Diabetes Prevention Program by modifying the number of sessions from 20 to 13 and by substituting less costly screening methods for glucose testing Funding source: government
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Low risk	Randomized block design (at household level)
Allocation concealment (selection bias)	Unclear risk	Not described or stated
Baseline outcome measurement similar	Low risk	No significant differences
Baseline characteristics similar	Low risk	No significant differences
Blinding of outcome assessment (detection bias) All outcomes	Unclear risk	Only blood pressure and dietary intake recall described as measured "by personnel blinded to study condition"
Incomplete outcome data (attrition bias) All outcomes	Low risk	94% completion rate, no significant differences between groups
Blinding of participants and personnel (performance bias) All outcomes	Unclear risk	Not reported
Protection against contamination	Unclear risk	Not reported
Selective reporting (reporting bias)	Low risk	Main study outcomes reported
Other bias	High risk	Intervention "dose": Attendance at group sessions was low, dropped from 60% at 1st session to 20% at last session