Charron‐Prochownik 2008.

Methods	Randomised controlled trial
Participants	53 girls were randomised Setting: diabetes clinic, Pennsylvania, USA; recruitment dates not clear Inclusion criteria: adolescent girls aged 16‐19.9 years of age, with type 1 diabetes Exclusion criteria: pregnancy during trial period (additional information provided by trial authors)
Interventions	Intervention groups READY‐Girls CD‐ROM group (n = 17): received a self‐instructional, developmentally appropriate, evidence‐based CD‐ROM and 1 comprehensive session before a routine diabetes clinic visit Book group (n = 16): received a self‐instructional, developmentally appropriate, evidence‐based book and 1 comprehensive session before a routine diabetes clinic visit Control group (n = 20): received standard care. Women were seen at routine diabetes clinic visits only
Outcomes	Process evaluation: timing, effort, ease of use, satisfaction Outcome measures: reproductive health and preconception care knowledge, beliefs, intention and behaviours, and metabolic control Outcomes assessed at baseline, post‐intervention and 3‐month follow‐up by questionnaires Note: Thurheimer 2016 pooled data from Charron‐Prochownik 2008 and Fischl 2010, and reported on general risk‐taking behaviours, condom use, and sexually transmitted infections
Notes	Funding: American Diabetes Association Clinical Research Award (1999 –2003), the General Clinical Research Center of Children’s Hospital of Pittsburgh (grant M01 RR00084), and the National Institutes of Health/National Institute of Nursing Research/Center for Research in Chronic Disorders (grant P30 NR03924) Declarations of interest: not reported
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Unclear risk	"randomized, controlled, repeated‐measures feasibility study."
Allocation concealment (selection bias)	Unclear risk	Not described
Blinding of participants and personnel (performance bias) All outcomes	High risk	Blinding of participants and personnel not feasible
Blinding of outcome assessment (detection bias) All outcomes	Unclear risk	Blinding of outcome assessors not specified. Data collection for some outcomes used self‐administered questionnaires. It was not clear how lack of blinding would affect the outcome assessment.
Incomplete outcome data (attrition bias) All outcomes	High risk	Data not reported for 1 participant from the CD‐ROM group and 5 participants from the control group for HbA1c at 3 months (differential proportion of missing data in already small cohort). Note: Thurheimer 2016 presented pooled data from Charron‐Prochownik 2008 and Fischl 2010, and reported on outcomes at 3 months for 136/141 (96%) participants across the 2 trials; Thurheimer 2016 reported that attrition of 28/136 (21%) participants occurred for 3‐month follow‐up; it is not clear how many participants were lost from each group, and how missing data were handled: "It was assumed that the data were missing at random as related to the developmental age of the subjects and not to the outcome variables being studied... Because the goal of the study was to examine the risk‐taking behaviors of these particular subjects rather than all adolescents in general, missingness is considered ignorable and therefore missing at random. This allowed for a more unbiased estimation to be made through observed data within each group... It was expected that the subjects would be independent from one study to the next, and since analysis of the same subjects was being tested longitudinally, the data were highly correlated."
Selective reporting (reporting bias)	Unclear risk	Insufficient information to assess, with no access to trial protocol. Results reported in figures (with no measures of variance), including HbA1c reported only as mean % change, with no measure of variance
Other bias	Unclear risk	Baseline characteristics not reported by group