| Methods |
Computer‐generated randomization; allocation concealment by prepackaged pill bottles with drug or identical‐appearing placebo. |
| Participants |
1485 women requesting IUDs from University College Hospital, Ibaden, Nigeria. Inclusion criteria: age 20 to 44 years and current menstruation. Exclusion criteria: history of ectopic pregnancy, pregnancy within 42 days, leiomyomata uteri, current salpingitis, uterine malignancy, sensitivity to tetracyclines, antibiotic administration within 14 days, impaired immune response, residence outside of Ibaden, or poor likelihood of follow up. All women were screened for gonorrhea and chlamydial infection. |
| Interventions |
Doxycycline 200 mg by mouth one hour before IUD insertion or an identical‐appearing placebo. |
| Outcomes |
Pelvic inflammatory disease diagnosed by Hager et al. criteria; unscheduled visits back to the clinic. |
| Notes |
Rates of pelvic inflammatory disease exceeded rates of unscheduled visits; study did not reach intended sample size (1800) |
| Risk of bias |
| Bias |
Authors' judgement |
Support for judgement |
| Allocation concealment (selection bias) |
Low risk |
Allocation concealment by prepackaged pill bottles with drug or identical‐appearing placebo |
