Continuous compared with cyclic use of oral contraceptive pills in the Dominican Republic: a randomized controlled trial.

OBJECTIVE: To estimate whether continuous combined oral contraceptive pill (OCP) use leads to higher continuation and lower pregnancy rates over 12 months than cyclic use in a developing country setting. METHODS: We enrolled healthy women aged 18 to 30 years, in Santo Domingo, Dominican Republic. We randomly assigned women to cyclic or continuous use of OCPs. Participants made quarterly clinic visits for 12 months. During follow-up, we reviewed OCP adherence and continuation, side effects, and bleeding, and we tested for pregnancy. RESULTS: We enrolled 358 women (mean age, 22.7 years) and 335 (93.6%) completed the study. In intent-to-treat analyses, 77.6% of the continuous use group and 71.7% of the cyclic group continued OCPs at 12 months (P=.21). The main reason for OCP discontinuation in both groups was running out of OCPs or forgetting. Across all visits, 26.1% of women in the continuous use group and 22.3% of women in the cyclic group ever reported missing three or more OCPs in the past month (P=.43). In multivariable analyses, regimen was not associated with discontinuation, but both previous birth and perceived ease of use of OCPs decreased risk of discontinuation, whereas desire for reduced menstruation increased risk of discontinuation. Although more women reported amenorrhea or infrequent bleeding in the continuous use group, more women in the cyclic group found their bleeding patterns acceptable. Bleeding was not associated with discontinuation in multivariable analyses. Pregnancy rates at 12 months were similar (16.2% continuous and 17.4% cyclic). CONCLUSIONS: Continuous and cyclic OCP regimens were associated with similar 12-month continuation and pregnancy rates. Few factors predicted OCP discontinuation or pregnancy. CLINICAL TRIAL REGISTRATION: ClinicalTrials.gov, www.clinicaltrials.gov, NCT00570440. LEVEL OF EVIDENCE: I.

Identifying appropriate IUD candidates in areas with high prevalence of sexually transmitted infections.

BACKGROUND: The IUD is a highly effective, safe, inexpensive and long-lasting contraceptive. However, IUDs may increase PID risk during the early postinsertion period when inserted in women with cervical infections. We developed a simple algorithm to identify women at low risk of current sexually transmitted infection (STI) who are appropriate IUD candidates in regions with moderate or high STI prevalence. METHODS: We used data sets from family planning populations in Kenya, Zimbabwe, Jamaica and the United States to develop optimum algorithms. We then validated these algorithms using data sets from family planning populations in Thailand and Uganda. RESULTS: A simple unweighted algorithm based on age, living with partner, education, bleeding between periods and a behavioral risk score (number of sex partners, condom use) was the most useful. Adding clinical signs did not improve algorithm performance. Women categorized at low risk by this algorithm were at substantially reduced risks of cervical infection. Women identified at high STI risk had at least twice the risk as the overall clinic populations. Women in the moderate-risk group had STI risks similar to the overall clinic populations. CONCLUSION: Women categorized as low risk by the algorithm can be referred for IUD insertion while women categorized at high risk should not receive an IUD without further testing or treatment. Women in the moderate-risk group should be triaged based on the STI prevalence of the overall clinic population. A simple checklist has been developed to help providers estimate a client's risk of current STI and to guide appropriate triage.