Koniak‐Griffin 2015.

Study characteristics
Methods	Study design: RCT, 2 arms Geographic location: Los Angeles, California, USA Ethical approval: yes Recruitment setting: parent education centres, churches, laundromats, organisations providing basic services to children and families (e.g. ESL classes, job training, social services) Method of recruitment: recruitment was conducted in 4 consecutive intervention cycles. Trained recruiters gave small group and individual presentations providing an overview of study and programme announcements. Length of follow‐up: 9 months (3 months after programme completion) Dropouts: 59 participants were lost to follow‐up; 13 in the intervention group and 17 in the control group after 6 months and 11 in the intervention group and 18 in the control group after 9 months. A priori calculation of effect size/power?: yes
Participants	Description: low‐income Latina immigrants that are overweight Health topic Cardiovascular disease Inclusion criteria Self‐identification as Latina, 35 to 64 years, Spanish‐ and/or English‐speaking, overweight (BMI ≥ 25) Exclusion criteria History of impaired physical mobility, type 1 diabetes, uncontrolled hypertension, heart attack, stroke, health clearance was required for participants with type 2 diabetes or hypertension controlled by diet and/or oral medications Intervention group Lifestyle behaviour intervention, 'Mujeres Sanas y Precavidas (Healthy Women Prepared for Life)' (111 randomised and 98 analysed at 6‐month follow‐up, and 100 analysed at 9‐month follow‐up) Control group Safety/disaster preparedness educational programme (112 randomised and 95 analysed at 6‐month follow‐up, and 94 at 9‐month follow‐up) Note: authors report having conducted a modified intention‐to‐treat analysis using mixed‐effects models for repeated measures over time; 13 participants were excluded from physical activity analysis because they did not meet the accelerometer recording criteria. PROGRESS‐Plus Place of residence: urban Time living in host country (years), mean (SD), range (n = 204): 18.6 (8.3), 1 to 40 Race/ethnicity: Latinas Occupation: 74.6% unemployed Gender: female only Education (grade) (n = 220): 52.5% ≤ 8th grade, 33.6% 9th to 12th grade, 12.6% ≥ 13 years Socioeconomic status/income (annual): ≤ USD 20,000 54.7%, USD 20,001 to 40,000 28.7%, USD 40,001 to 75,000 16.6% Health insurance: 31.8% insured Social capital: 72.2% married/living with a partner, 27.8% divorced/widowed/single Age (years), mean (SD), range: 44.6 (7.9), 35 to 64 Health literacy (baseline) Not measured
Interventions	Intervention: lifestyle behaviour intervention "Mujeres Sanas y Precavidas" Theoretical framework: community‐based participatory research conceptual framework Description: the culturally targeted promotora‐led programme included group education plus individual teaching and coaching units about healthy lifestyle behaviours to reduce cardiovascular disease risks. Promotoras presented standardised content in pairs and showed an instructor‐led stretching and exercising DVD, produced by an official public health department. In coaching sessions, food and physical activity diaries of participants were discussed with promotors (inter alia). The intervention promoted four key messages: (1) healthy food choices, (2) portion control, (3) managing emotional eating and (4) increasing physical activity. Participants received a pedometer, a copy of the exercise video presented in the classes and culturally‐appropriate recipes. Intervention provider: trained promotoras Delivery method/mode: 8 weekly face‐to‐face group sessions lasting 2 hours, followed by 4 months of individual teaching and coaching sessions (4 face‐to‐face sessions and 4 phone calls) Language of delivery: language concordant Format: group‐based, individually tailored Setting/location: community setting, participants' home Consumer involvement: evaluated with a smaller sample of intervention participants Comparator Type: no health literacy intervention (attention placebo control) Description: 6‐month educational programme on safety and preparedness topics (e.g. in case of earthquakes) followed by the possibility of 8 individual teaching and coaching contacts where class content was reviewed in in‐depth discussions. After completion of the study, participants were offered 2 classes on key information about a promotora‐led health intervention ("Su Corazón, Su Vida").
Outcomes	Outcomes assessed in the study: knowledge of heart disease, physical activity*, dietary habits, body weight, height and waist circumference, blood pressure, blood lipids and glucose Outcomes considered in this review Health related knowledge (heart disease knowledge) Health behaviour (physical activity) Note: *prioritised outcome, category 'health behaviour' Methods of assessing outcomes Heart disease knowledge: 10‐item questionnaire adapted from a previous survey, true/false format (using statements, e.g. "Heart disease is the leading cause of death in woman"), 0 to 10, higher score is better Translation procedure: validated Spanish version Reliability/validity: α = 0.80 Note: "Items also assessed prevention behaviours and awareness that early treatment exists." Physical activity: Kenz Lifecorder Plus Accelerometer (Kenz, Nagoya, Japan), assesses vertical acceleration and counts of movement that are correlated with steady‐state oxygen consumption; participants wore the accelerometer during waking hours for 7 consecutive days at each physical activity data collection period Reliability/validity: validated tool Note: "The Lifecorder activity counts were converted into METS (1 MET = 3.5 mL/kg min), thus enabling classification of intensity according to accepted standards as well as measurement of steps". Participants received verbal and written instructions with illustrations on the devices. Note: a bilingual research assistant, blinded to participant’s group assignment, administered the questionnaires via face‐to‐face interviews. Language of assessment: Spanish Timing of outcome assessment: baseline, 6 months after randomisation (short‐term, immediately after programme was completed) and 9 months after randomisation (medium‐term, 3 months after programme was completed)
Health literacy	Definition: not reported Health literacy components addressed by the intervention Prerequisites and tools Knowledge Motivation Steps of information processing Understand Appraise Apply Health domain: prevention
Notes	Trial ID: NCT01333241 Funding: funding was obtained by the National Heart, Lung, and Blood Institute (R01 HL086931) and was part of a registered clinical trial. Additional notes: authors were contacted and asked for additional information (e.g. control groups' post‐intervention knowledge scores) but provision of data was not possible.
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Low risk	"Randomization was performed using a web‐based programme custom‐developed for this study. Participants were assigned to the Lifestyle Behavior Intervention or the control group in a 1:1 ratio using a block randomization procedure."
Allocation concealment (selection bias)	Low risk	The randomisation method indicates a low risk of bias.
Blinding of participants and personnel (performance bias) All outcomes	Low risk	Personnel and participants were not blinded due to the nature of the study. However, outcomes considered in this review were objectively measured.
Blinding of outcome assessment (detection bias) objective outcome measures	Low risk	Both physical activity and knowledge were objectively measured. No subjective judgement of personnel required.
Incomplete outcome data (attrition bias) All outcomes	Low risk	“Retention was 86.5% and 87.0% or the 6‐ and 9‐month evaluations, respectively. (...) The retention rates across groups were not statistically different” The attrition rate is lower than 20% and the differential loss between study groups is not significant. A modified intention‐to‐treat‐analysis was conducted for physical activity; a completers only analysis was performed for participant‐reported outcomes.
Selective reporting (reporting bias)	High risk	All prespecified outcomes reported at clinicaltrials.gov are reported in the published reports. However, results of the control group's knowledge assessment were not reported.