UCSF

Background

Health care providers often tell women to wait until the next menses to begin hormonal contraception. The intent is to avoid contraceptive use during an undetected pregnancy. An alternative is to start hormonal contraception immediately with back-up birth control for the first seven days. Immediate initiation was introduced with combined oral contraceptives (COCs), and has expanded to other hormonal contraceptives. At the time of the initial review, how immediate start compared to conventional menses-dependent start was unclear regarding effectiveness, continuation, and acceptability. The immediate-start approach may improve women's access to, and continuation of, hormonal contraception.

Objectives

This review examined randomized controlled trials (RCTs) of immediate-start hormonal contraception for differences in effectiveness, continuation, and acceptability.

Search methods

In August 2012, we searched MEDLINE, CENTRAL, POPLINE, LILACS, ClinicalTrials.gov, and ICTRP for trials of immediate-start hormonal contraceptives. We contacted researchers to find other studies. Earlier searches also included EMBASE.

Selection criteria

We included RCTs that compared immediate start to conventional start of hormonal contraception. Also included were trials that compared immediate start of different hormonal contraceptive methods with each other.

Data collection and analysis

Data were abstracted by two authors and entered into RevMan. The Peto odds ratio (OR) with 95% confidence interval (CI) was calculated.

Main results

Five studies were included. No new eligible studies have been found since the review was initially conducted. Method discontinuation was similar between groups in all trials. Bleeding patterns and side effects were similar in trials that compared immediate with conventional start. In a study of depot medroxyprogesterone acetate (DMPA), immediate start of DMPA showed fewer pregnancies than a 'bridge' method before DMPA (OR 0.36; 95% CI 0.16 to 0.84). Further, more women in the immediate-DMPA group were very satisfied versus those with a 'bridge' method (OR 1.99; 95% CI 1.05 to 3.77). A trial of two immediate-start methods showed the vaginal ring group had less prolonged bleeding (OR 0.42; 95% CI 0.20 to 0.89) and less frequent bleeding (OR 0.23; 95% CI 0.05 to 1.03) than COC users. The ring group also reported fewer side effects. Also, more immediate ring users were very satisfied than immediate COC users (OR 2.88; 95% CI 1.59 to 5.22).

Authors' conclusions

We found limited evidence that immediate start of hormonal contraception reduces unintended pregnancies or increases method continuation. However, the pregnancy rate was lower with immediate start of DMPA versus another method. Some differences were associated with contraceptive type rather than initiation method, i.e., immediate ring versus immediate COC. More studies are needed of immediate versus conventional start of the same hormonal contraceptive.