Strategies for improving postpartum contraceptive use: Evidence from non-randomized studies

Most postpartum women do not have the birth control they would like. Teens often get pregnant again within a year. In earlier work, we found few randomized trials on learning about postpartum family planning. Here, we looked at other types of studies that tested ways to improve postpartum use of birth control. We wanted to see if certain programs were related to more use of family planning after giving birth. | Until 3 November 2014, we ran computer searches for studies of programs to improve family planning among postpartum women. We wrote to researchers for missing data. Programs had to have contact within six weeks postpartum. The special program was compared with a different program, usual care, or no service. Our main outcomes were birth control use and pregnancy. | We found six studies with a total of 5143 women. Of three studies with pregnancy data, two showed fewer pregnancies in the treatment group compared to the control group. The programs in those studies were clinic counseling and community education. All studies showed the special program was related to more birth control use. In two studies, more women in the treatment group used a modern method of birth control than those in the control group. In another study, women in the treatment group were more likely to use pills or an IUD but less likely to use an injectable method. One study used a score for how well the birth control method usually worked. The methods of the treatment group scored higher than those of the control group. A study focused on IUDs showed more IUDs in the treatment group and less use of no method. Women in a health service program used birth control more often than those in a community education program or those getting standard care. Also, women in the health service group were more likely to use the lactation method. | We believe the data were very low quality for pregnancy and birth control use. The studies had problems in design, analysis, and reporting. Some did not adjust for factors that could affect the results. They had self-reported outcomes and used different measures for the outcomes. All studies had good follow-up times but most lost many women to follow up