Table.

Randomized trials examining the effectiveness of screening in preventing pelvic inflammatory disease in nonpregnant women

Author/year	Study population	Methods	Randomization	Results
Oakeshott et al92	2529 sexually active female students, aged 16–27 years, recruited from 20 universities in London, UK.	Participants provided self-collected vaginal samples at baseline and completed a questionnaire on sexual health. CT was diagnosed using TMA. After one year, participants completed a questionnaire about possible symptoms of PID and sexual behavior over the past year. Medical records were collected for 17% and three doctors blinded to study group and CT status at baseline allocated reported PID cases into probable, possible, or no PID, using the CDC guidelines and Hager’s criteria.	98.6% (2529/2563) were randomly allocated for immediate CT screening and treatment (1259) or deferred screening (1186; vaginal samples analyzed after one year) using random number tables. 94% (2377/2529) were followed up after one year.	Incidence of PID was 1.3% (15/1191) for screened group and 1.9% (23/1186) for controls (RR 0.65, 95% CI 0.34–1.22). 9.5% (7/74) of CT positive women at baseline in control group developed PID. 1.6% (1/63) of CT positive women in screening group developed PID (RR 0.17, 95% CI 0.03–1.01).
Østergaard et al121	1700 sexually active female high school students from Denmark, aged ≥15 years.	Intervention group received a home sampling kit at baseline and questionnaire. Those who were CT-positive (diagnosed by TMA) were referred for treatment. The control group received the same questionnaire with an offer of free testing at a local clinic or doctor’s office. Home sampling kits were sent at follow-up to determine CT incidence. Follow-up questionnaire collected information on treatment or hospitalization for PID during the past year. Medical records were sought through the central Danish register for prescriptions related to PID.	5487 female high school students were randomized (1:1) by simple redeeming to a home sampling group (2603) or a usual care group (2884). 43% (2351/5487) responded to the baseline questionnaire. Among sexually active females 55% (930/1700) were followed for 1 year.	Incidence of PID was 2.1% in the intervention group and 4.2% in the control group (P = 0.045).
Scholes et al120	2607 sexually active female HMO enrollees aged 18–34 years from Washington State, US.	Self-administered questionnaires were mailed to single HMO enrollees to determine those at high risk of CT (based on age, race, gravidity, and sexual partner in the past 12 months). Emphasis was placed on contacting nonresponders who were assigned to the screening group. In the screening group swabs were tested for CT by ELISA. Specimens collected by cytobrush were tested by chlamydia cell culture. Control group (usual care) saw their providers as needed. Both groups received a follow-up questionnaire at 12 months and were asked about diagnosis of PID. Outpatient and inpatient databases were used to identify those participants with a diagnosis of PID. Medical records of those with possible cases of PID were reviewed by blinded abstracters to determine PID status (based on signs, symptoms, and laboratory findings).	57% (20,836/36,547) responded to the initial questionnaire and 85% (17,725) of responders were ineligible. Of the 13% eligible (2607), 1009 were randomly assigned (authors did not give randomization method) to the screening group and 1598 to the usual care group. 64% (645/1009) of women in the screening group were tested for CT. 76% were followed-up at 12 months.	Databases identified 57 participants diagnosed with PID. Medical records showed a clinical diagnosis of PID in 37 participants. Additional 5 were identified in chart review. 9 confirmed cases of incident PID in the screening group and 33 in the usual care group (RR 0.44, 95% CI 0.20–0.90).

Abbreviations: CT, Chlamydia trachomatis; TMA, transcription mediated amplification; PID, pelvic inflammatory disease; CDC, Centers for Disease Control; RR, relative risk; CI, confidence interval; HMO, health maintenance organization; ELISA, enzyme-linked immunosorbent assay.