Hawkins 2014.

Methods	Randomised controlled trial.
Participants	68 women were randomised. Setting: Baystate Medical Center and Mercy Medical Center in Western Massachusetts, USA (recruited from April 2010 to August 2011). Inclusion criteria: Hispanic women aged 18 to 40 years, with a gestational age of < 18 weeks, who were overweight or obese (pre‐pregnancy BMI ≥ 25 kg/m²) and who self‐reported participating in < 30 minutes of moderate‐intensity activity per week. Exclusion criteria: history of type 2 diabetes, hypertension, heart disease or chronic renal disease; current medications that adversely influence glucose tolerance; not planning to continue the pregnancy to term; contraindications to participating in moderate‐intensity physical activity or a low‐fat/high‐fibre diet; self‐reported participation in ≥ 30 minutes of moderate‐intensity exercise on ≥ 3 days per week or ≥ 20 minutes of vigorous‐intensity exercise on ≥ 1 day per week; or multiple gestation (e.g. twins).
Interventions	Intervention group (n = 33 randomised) The intervention consisted of 6 in‐person behavioural counselling sessions and 5 telephone booster sessions delivered by bicultural and bilingual health educators, tailored for Hispanic women’s culture and context. All materials were available in Spanish and English and were written at a sixth‐grade reading level. Diet: women were encourage to decrease their intake of foods high in saturated fat, and to increase intake of dietary fibre (as recommended by the ADA). Health educators assessed readiness and preferences for change, consistent with the Stage of Change framework, and assisted women in developing dietary change goals. Women were provided with a low‐literacy pictured‐based food guide by which ethnic and other foods were classified based on GI/fibre content and saturated fat using the ‘traffic light’ colours and self‐monitoring logs. Activities in the follow‐up in‐person and telephone‐delivered booster sessions included review of logs, problem‐solving of challenges, introduction of new tailored materials and goal setting. Exercise: the physical activity during pregnancy guidelines of the ACOG (≥ 30 minutes of moderate‐intensity activity on most days of the week) were discussed. Women were encouraged to achieve the standards set in the guideline through increasing their walking and developing a more active lifestyle. Informed by responses to a 'Stage of Change Questionnaire', women were provided with a stage‐matched manual which included motivationally targeted materials combined with tip sheets on building social support for new behavioural patterns and strategies for overcoming barriers to physical activity. The health educators assisted the women in developing personalised physical activity goals. Women were provided with a digital pedometer and a physical activity log to track their progress Control group (n = 35 randomised) Women in the control group received standard care (no further details reported).
Outcomes	Data in meta‐analyses (or other data) for: GWG; behaviour changes associated with the intervention; relevant biomarker changes associated with the intervention; gestational age at birth; birthweight. Additional narrative text for: GDM; adherence to the intervention; views of the intervention.
Notes	Funding: "This work was supported by CDC/ASPH S3948". Declarations of interest:"None declared".
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Unclear risk	Quote: "eligible patients were randomized... by the health educators to either a lifestyle intervention or a standard care group. Randomization was stratified by age (< 30 years, ≥ 30 years) and pre‐pregnancy BMI (25–30 kg/m², ≥ 30 kg/m²with a block size of four".
Allocation concealment (selection bias)	Unclear risk	As above; no further information provided.
Blinding of participants and personnel (performance bias) All outcomes	High risk	Blinding of women and trial personnel not considered feasible in view of the intervention and control.
Blinding of outcome assessment (detection bias) All outcomes	Low risk	Quote: "Assessments were conducted by telephone, at baseline, mid‐pregnancy, and at 6 weeks postpartum by bilingual and bicultural interviewers blinded to the assigned intervention group". Though not clear whether clinical outcomes (such as GDM) were able to be assessed blind, we have judged risk of detection bias as low.
Incomplete outcome data (attrition bias) All outcomes	Low risk	Of the 33 women randomised to the intervention group, 30 (94%), 32 (97%) and 24 (75%) were available for the mid‐pregnancy, clinical outcome and postpartum assessments, respectively. Of the 35 women randomised to the control group, 29 (85%), 34 (97%) and 29 (85%) were available for the mid‐pregnancy, clinical outcome and postpartum assessments respectively. The losses at mid‐pregnancy and postpartum were associated with women being unable to be contacted via telephone; losses for clinical outcomes were associated with women being delivered off‐site.
Selective reporting (reporting bias)	High risk	Reporting of GDM is incomplete (only the number of cases across both groups in text) and a very limited number of clinical outcomes are reported.
Other bias	Low risk	No obvious sources of other bias identified.