Wei 2016.

Methods	Randomised controlled trial
Participants	120 women randomised Setting: Department of Endocrinology, Zhongda Hospital, the Affiliated Hospital of Southeast University, China from September 2011 to December 2012 Inclusion criteria: women between 24 and 28 weeks' gestation with a singleton pregnancy, GDM, and written informed consent. All women underwent a 75 g OGTT at 24 to 28 weeks according to the criteria of the American Diabetes Association (ADA). Based on the 1‐step approached recommended by the World Health Organization, ADA and IADPSG, women were defined as having GDM if they had at least 1 abnormally high plasma glucose out of the 3 measurements: fasting > 92 mg/dL (5.1 mmol/L), 1‐hour > 180 mg/dL (10.0 mmol/L), or 2‐hour > 153 mg/dL (8.5 mmol/L). Exclusion criteria: diagnosis of diabetes mellitus, previous treatment for GDM, presence of infection, or other severe metabolic, endocrine, medical or psychological co‐morbidities
Interventions	CGMS (n = 58*) There were 2 groups: CGMS early subgroup (n = 30): during gestational weeks 24 to 28; and CGMS late subgroup (n = 30): during gestational weeks 28 to 36. *There were 3 additional women who withdrew following allocation, it was unclear if they were allocated to the early or late subgroup. The CGMS group was monitoring using CGMS (Gold Medtronic MiniMed, Northbridge, CA, USA), with the sensor inserted into the upper outer buttock of the subjects. While women wore the sensor, glucose concentrations (bedtime, and 1 hour before the beginning of each meal) were also monitored using the Accu‐Chek meters and were input into the CGMS as calibration 4 times per day. Self‐monitoring (n = 62) All women Women were taught to perform self‐monitoring of blood glucose using Accu‐Chek Advantage meters (Roche Diagnostics, Manheim, Germany). Women were instructed to check their level 4 times per day (fasting, and 1 hour after the beginning of each meal) from the first visit at which they received the GDM diagnosis until birth (except during the period where the GGMS group used their device). Women underwent dietary counselling for a eucaloric diet with a low GI and low saturated fat levels (50% to 60% energy from carbohydrates; 25% to 30% from fat; 15% to 20% from protein; with energy intake distributed as equally as possible throughout the day, based on recommendations of the China Diabetic Association). Moderate intensity exercise was encouraged; 35 to 45 minute sessions, three times a week. A physiotherapist motivated the women to continue or start exercising, and provided written instructions, with walking, swimming and cycling recommended as types of exercise. Insulin treatment was administered under conditions of 2 fasting blood glucose values > 105 mg/dL (5.8 mmol/L), 2 1‐hour post‐prandial levels > 155 mg/dL (8.6 mmol/L), a 2‐hour post‐prandial level > 130 mg/dL (7.2 mmol/L), or a fasting blood glucose > 90 mg/dL (5.5 mmol/L) with at least 2 post‐prandial values > 141 mg/dL (7.5 mmol/L) according to ADA guidelines. Women received NPH insulin as an immediate acting insulin with an initial dose of 0.2 units/kg; if fasting levels were high, treatment was given before bed‐time; if post‐prandial levels were high, regular insulin or short acting insulin was given before meals. Follow‐up meetings were every 2 to 4 weeks until 28 weeks, fortnightly until 32 weeks, and weekly thereafter.
Outcomes	Review outcomes reported: caesarean section; use of additional pharmacotherapy (insulin use; dose); glycaemic control (HbA1c); hypoglycaemia; gestational weight gain (and excessive; inadequate; appropriate gain); views of intervention; large‐for‐gestational age; perineal mortality; stillbirth; neonatal death; preterm birth; macrosomia; small‐for‐gestational age; hypoglycaemia; gestational age at birth; birthweight
Notes	Funding: "This work was partially supported by the following foundation: the Social Development Project of JiangSu Province (No. SBE201170735, Wang SH)". Declarations of interest: "The authors declare no competing financial interests". Dates: September 2011 to December 2012
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Low risk	Quote: "randomly allocated … by a computer generated random number table".
Allocation concealment (selection bias)	Unclear risk	No details provided
Blinding of participants and personnel (performance bias) All outcomes	High risk	Not feasible to blind participants and personnel.
Blinding of outcome assessment (detection bias) All outcomes	Unclear risk	No detail provided; unclear how lack of blinding would have affected outcomes.
Incomplete outcome data (attrition bias) All outcomes	Low risk	120 women were randomised, and 106 were analysed; 58 were randomised to CGMS and 62 to self‐monitoring; overall 51/58 were analysed in the CGMS group and 55/62 in the self‐monitoring group. In the CGMS group, 2 withdrew, 1 had site discomfort, 3 were lost, and 1 discontinued; in the self‐monitoring group, 4 were lost, and 3 discontinued.
Selective reporting (reporting bias)	Unclear risk	No access to trial protocol to enable confident assessment of selective reporting.
Other bias	Low risk	No other obvious sources of bias identified.

Abbreviations 
 BMI: body‐mass index
 CGMS: continuous glucose monitoring system
 GDM: gestational diabetes mellitus
 GI: glycaemic index
 HbA1c: glycated haemoglobin

IADPSG: International Association of the Diabetes and Pregnancy Study Groups
 NICU: neonatal intensive care unit
 OGCT: oral glucose challenge test
 OGTT: oral glucose tolerance test
 PROM: preterm rupture of membranes