Eleven studies were included (n=7,716). Eight studies were experimental (n=7058), including six randomised controlled trials (RCTs) (n=5,198) and three studies were quasi-experimental (n=658).
Contraceptive outcomes
Pregnancy/STD counselling in primary care versus general health counselling (one RCT): No statistically significant findings were associated with the intervention.
Contraceptive initiation in a non-family-planning setting (two studies): Contraceptive initiation in an urban STD clinic (versus less proactive care) was associated with significantly higher rates of contraceptive use at eight months (44 per cent versus 26 per cent, p<0.001, one RCT n=877). There was no statistically significant difference between the two groups at 12 months. Similarly, a quasi-experimental study set in a correctional facility reported higher rates of contraceptive use at four weeks in women receiving the intervention than in controls (39 per cent versus four per cent, p<0.05, n=224).
Immediate versus later initiation of hormonal contraception (two studies): One controlled trial found a significantly increased likelihood of starting a second pack of pills in the intervention group (88 per cent versus 74 per cent, p<0.05, n=250). Differences in the third cycle were not statistically significant. A small RCT (n=60) of early initiation of hormonal patches reported no statistically significant findings.
Advance provision of emergency contraception versus education/advice only (four studies): Two RCTs (n=3,247) and two experimental studies (n=1,860) all reported significantly higher rates of emergency contraception use in the intervention group over three- to 12-month follow up (p range <0.05 to <0.001). The rate and consistency of (non-emergency) contraceptive use did not differ significantly between the groups.
Reminder systems for injectable contraception (two studies): One quasi-experimental study reported that a postcard reminder system significantly increased the proportion of women receiving their repeat injection on time (76 per cent versus 64 per cent, p<0.05, n=184). However an RCT (n=250) using reminder letters reported no statistically significant findings.
Pregnancy (six studies): No studies found any statistically significant benefit from the intervention programme.