CHLAMYDIA TRACHOMATIS (CT) is a prevalent and potentially devastating infection for women, with over 700,000 cases reported annually in the United States; however, up to 3 million total cases are estimated when underreporting and underdiagnosis are taken into account. 1 Because infection is most often asymptomatic, 2–4 untreated genital chlamydial infection can lead to pelvic inflammatory disease, 5–8 ectopic pregnancy, 7,9 and infertility. 9–12 Recent evidence has suggested that prior CT infection can be linked to ovarian cancer. 13
Detection of infection is challenging because most women do not develop symptoms that prompt clinical evaluation. Because many women annually attend gynecology or family planning (FP) clinics for routine Pap smears, contraceptive services, and reproductive health checkups, clinic-based CT screening programs were developed. 14,15 One of the first large-scale CT screening programs in the United States began in 1988 at FP clinics in the Pacific Northwest, U.S. Public Health Service (USPHS) Region X, which includes the states of Washington, Idaho, Oregon, and Alaska. 14–16 To both minimize costs and to increase the probability of detecting infection among asymptomatic women, selective screening criteria (SSC) were developed. 17 The SSC initiated in 1988 were based on risk factors for infection, including young age, clinical signs, having sex with an infected partner, sexual behaviors such as frequent multiple sex partners, and previous diagnosis of a sexually transmitted infection (STI). 18–21 Using SSC, women attending for annual pelvic examinations in publicly funded family planning clinics in Region X are routinely tested for CT. For ease of implementation in clinical settings outside sexually transmitted disease (STD) clinics and to allow for funding of diagnostic CT testing in non-STD settings, the selective screening criteria in Region X include 2 elements normally interpreted as indications for diagnostic testing and/or presumptive treatment: manifestations of cervicitis (mucopurulent secretion, cervical friability, and abnormal cervical ectopy) and evidence of recent exposure to CT.
Beginning in 1993, the Centers for Disease Control and Prevention (CDC) began administration of Infertility Prevention Projects based in each federal health region using selective screening criteria that were locally developed; this effort has been the cornerstone of the CDC’s programmatic efforts to reduce the morbidity and associated sequelae from this infection. Each of the 10 projects receives funding from CDC to administer chlamydia screening. Areas in the United States and other countries that have been performing clinic- and population-based screening for chlamydial infection have reported declines in prevalence, 22–24 and 2 randomized, controlled trials of screening have shown a 50% reduction in the diagnosis of pelvic inflammatory disease. 25,26
Many studies have evaluated the performance and cost-effectiveness of this CT control strategy, 27–30 and most have concluded that in settings of moderate prevalence, selective screening programs are cost-effective. 18 The most recent (1995) evaluation of Region X’s selective screening criteria used data from 1990 to 1993. 19 At that time, enzyme immunoassay (EIA) with direct fluorescence antibody (DFA) testing was the standard CT diagnostic test. Recently, highly sensitive nucleic acid amplification tests have been increasingly used and shown to be a factor in observed increases in prevalence in Region X, other parts of the United States, and in Sweden. 31–33 Some investigators have suggested that periodic reevaluation of selective screening should be performed when circumstances in the field change significantly. 18,34
The current reevaluation of the SSC in use in Region X centers on 3 questions: 1) What is the current performance of the selective screening criteria now used in Region X? 2) Have risk factors for the presence of chlamydial infection changed since the last SSC evaluation in 1995? 3) Can the performance of the selective screening criteria be optimized?
We analyzed 409,882 CT test records (excluding 4355 or 1% where test result or test type was unknown) for women attending 252 family planning, 123 STD and 251 “other” clinics, including community/migrant, college health, public health nursing, and adolescent clinics, in Region X from 1998 to 2000. STD clinics contributed 34,288 records, FP clinics 304,183 records, and “other” clinics 71,471 records. Women attending STD clinics were universally tested for infection. Women attending family planning and other clinics were tested according to the Region X SSC, which were: 1) age ≤24 years, or 2) age >24 years plus any 1 of the following: nonspecific cervicitis, mucopurulent cervicitis (MPC), friability of the cervix, cervical ectopy, clinician-diagnosed pelvic inflammatory disease (PID), patient-reported or clinician-assessed recent exposure to chlamydia or any other STD, including sex with a symptomatic sex partner in the past 60 days, C. trachomatis or any other STD in the prior 12 months, or new sex partner or 2 or more sex partners in the previous 60 days. Diagnostic tests used in Region X were EIA (SyvaMicrotrak), ligase chain reaction (LCR; Abbott LCx), nonamplified DNA probe (PACE2; Gen-Probe), and cell culture.
To define the performance of the current selective screening criteria in Region X, we analyzed all tests for the sensitivity and efficiency of the criteria. Sensitivity was defined as the percentage of positives detected, and efficiency was the percentage of tests that met the criteria. 35 Thresholds of 60% efficiency and 90% sensitivity were used as ideal benchmarks for criteria performance. The ideal scenario was to identify the most positives (a high percentage for sensitivity) while testing the fewest number of people (low percentage for efficiency). Sensitivities and efficiencies for clinic type and test type were calculated separately to explore variations in criteria performance in settings of lower prevalence or locations where use of the more sensitive nucleic acid amplification tests had increased.
To define risk factors associated with chlamydial infection, we analyzed data from women universally tested at STD clinics. Data were gathered on a standardized laboratory test request form in use across all clinics since 1993. Patients self-reported demographics and behavioral risks in response to questions posed by clinicians during the clinic visit. Clinicians reported results of cervical examinations using standard definitions within the program for mucopurulent endocervical discharge, cervical friability, abnormal cervical ectopy, and PID, each reported separately. 16 Laboratories reported test type and results. With CT positivity as the dependent variable, univariate odds ratios for each independent variable were calculated. Variables significant in univariate analysis to a level of P <0.05 were included in a multivariate logistic regression model (forward stepwise procedure with Wald statistical correction). The final model included a variable for type of diagnostic test to explore whether risk factors differed between those tested with amplified and nonamplified laboratory methods.
Using the results of the risk analysis, we developed 5 different sets of selective screening criteria to evaluate for sensitivity and efficiency using the benchmarks of 60% and 90%, respectively. Because no new variables were under consideration for inclusion in sets of SSC, the focus of the sensitivity and efficiency analysis was to assess whether the current SSC in Region X could be simplified. Therefore, the resultant SSC were modifications of the existing Region X criteria and could be applied retrospectively to our study population to address whether the current criteria could be optimized. Additionally, the criteria endorsed by the third U.S. Preventive Services Task Force (USPSTF) were also applied to our study population and compared with the performance of other selective screening criteria. This Task Force recommended screening for: 1) all sexually active women 25 years and younger; and 2) asymptomatic women over 25 years of age at increased risk, defined as new or multiple sex partners, prior history of an STD, or inconsistent condom use. 36
All analyses were performed using SPSS 8.0 statistical software (SPSS, Chicago, IL). Measures of significance were 2-sided and used a significance level of P <0.05.
Chlamydia positivity was highest in STD clinics (7.3%) and was approximately 4.0% in FP and other clinics (Table 1). Women less than 25 years of age comprised 70% of the FP clinic population and 64% of the “other” clinics’ population; women attending STD clinics were slightly older. EIA and LCR were the most frequently used laboratory tests. Almost 50% of tests in STD clinics were EIA, whereas in FP clinics, more than 45% of tests were LCR. The higher proportion of LCR tests within FP clinics reflects the wider dissemination of nucleic acid amplification tests in that setting. For all clinical settings, 97% of samples were clinician-acquired cervical swabs and 2.5% were urine samples (data not shown), reflecting the linkage of the screening program to annual pelvic examinations for women. Behavioral risks were common: 21% to 39% of the population (depending on clinic type) reporting a new sex partner in the last 60 days, and nearly one fourth of the STD clinic attendees reporting 2 or more sex partners over the same period. Not surprisingly, condom use was low; less than 25% of women used condoms at last sex, regardless of clinic type. Women who attended STD clinics were more likely to demonstrate clinical signs associated with increased likelihood of chlamydial infection such as mucopurulent cervicitis or cervical friability (Table 1).
Univariate analysis confirmed that the factors associated most strongly with chlamydial infection among STD clinic attendees were young age, clinical signs, and recent exposure to chlamydia or other STDs (Table 2). Over 80% of all women who tested positive were less than 25 years of age, regardless of clinic type (data not shown). Behavioral risks such as new sex partner, multiple sex partners, or history of chlamydial infection were associated with positivity, but not as strongly. Nonwhite race was also associated with infection, but Hispanic ethnicity was not. Women submitting urine samples were slightly more likely than those submitting cervical swabs to test positive, and those tested with culture were less likely to test positive than those tested with LCR, consistent with the lower sensitivity of culture in detecting CT (Table 2). When stratifying by test type, variables associated with infection did not differ between women tested with EIA and those tested with LCR (data not shown).
Factors associated with an increased risk of chlamydial infection in univariate analysis were explored in multivariate models. Among all women attending STD clinics, factors remaining significantly associated with infection, in order of relative strength, were young age, exposure to CT, sex with a symptomatic sex partner, cervical friability, and exposure to nongonococcal urethritis. Weaker, but statistically significant, associations were found with mucopurulent cervicitis, concurrent diagnosis of PID, exposure to Neisseria gonorrhoeae, multiple sex partners, and history of chlamydial infection in the last year. Women tested with either EIA or LCR were more likely to be positive than those tested with DNA probe or culture, but no statistically significant difference between EIA and LCR was found (Table 2).
Evaluation of the sensitivity and efficiency of selective screening criteria revealed that the current Region X criteria were more sensitive than the target benchmark, detecting 95.6% of infections, but were less efficient than the 60% benchmark, requiring testing 85.6% of women (Table 3). The sensitivity of these criteria remained high and exhibited little variation across all clinics (range, 94.5–97.5%). Even after stratifying by test type within clinical settings, the sensitivity of the current SSC in Region X showed little variation (range, 90.8–98.7%). The efficiency of the current criteria varied little between clinic types (range, 80.8–87.3%) and did not achieve our study’s efficiency benchmark of 60%. When exploring the variation by test type, for all test types except DNA probe, the range of the criteria’s efficiency was constant (82.6–89%). The efficiency of the current criteria was most variable when women were tested by DNA probe: 67.9% at other clinics, 88.1% at family planning clinics, and 91.0% at STD clinics. Although SSC were not used in STD clinics where universal testing was used, the criteria would have been no more efficient there than at sites that selectively screened.
Five sets of SSC developed from factors significantly associated with infection through univariate analysis and multivariate modeling showed marked differences in efficiency and sensitivity when compared with each other and with SSC from the USPSTF and those currently used in Region X. The sets of criteria were developed using an additive approach, beginning with just age (Table 3, criteria 3), which had good efficiency (56.1% in STD clinics, 65.1% in “other” clinics, and 70.7% in FP clinics), but reduced sensitivity, detecting approximately 85% of infections. The other criteria added additional components to increase the likelihood of detecting infection in women over 24 years of age using the evidence generated from the analysis of risk factors (Table 3, criteria 4–6). Combining testing, all young women and women over 24, if there were clinical signs, eg, cervicitis or ectopy (Table 3, criteria 4), increased the sensitivity of the criteria, detecting almost 90% of all positives, with a slight reduction in efficiency, requiring testing of 64% to 74% of women. There were only incremental differences on varying the selection criteria for women over 24 using combinations of clinical signs, risk behaviors, and/or recent exposure to an infected partner (criteria 5 and 6). An additional diagnostic model was developed (Table 3, criteria 7), because the current SSC in Region X allowed for diagnostic tests to be included, principally, cervical signs of infection or recent exposure. Not surprisingly, this model performed best in STD clinics, detecting 93.3% of positives, with the greatest efficiency, testing less than 70% of the women attending. However, the relative performance of this model was similar to the other criteria. The recently recommended criteria by the third U.S. Preventive Services Task Force had the lowest efficiency (Table 3), in part because the criteria recommend testing women who do not use condoms regularly, which comprised over 75% of our study population.
Our study confirmed that the factors most strongly associated with chlamydial infection are young age, clinical findings suggestive of infection, report of contact to CT, and reported sex with a symptomatic partner. Behavioral risk factors were weakly associated with infection, especially among women tested with more sensitive laboratory tests. Additionally, behavioral risk factors did not add efficiency to the selective criteria for screening. There were variations in the predictors’ performance by test type, but differences were slight and did not outweigh the consistent performance of young age, clinical findings, and recent exposure, as recent studies have confirmed. 37–43
With the exception of universal screening, none of the selective criteria developed from our models were able to detect >92% of CT-positives outside an STD setting. Our analysis found some added efficiency in including exposure variables, eg, exposure to CT or sex with symptomatic partner, and cervical findings as indications for screening older women. Although these items are usually considered indications for diagnostic testing and/or presumptive treatment in STD clinics and other clinical settings, the inclusion of signs of cervicitis or exposure to an STD in Region X SSC has been maintained over the years for ease in clinical decision-making in non-STD settings. Our results showed that there was very little difference in criteria efficiency or sensitivity when these items were used in a diagnostic model or as part of screening criteria (Table 3, criteria 6 and 7). Behavioral risks such as a new sex partner or multiple sex partners increased the population screened to nearly 80% while only yielding a slight increase in the proportion of infections found.
As an example, Figure 1 illustrates the predictive performance of changing the current SSC to items that reflect greater efficiency with minimal reduction in sensitivity, as suggested in the models in Table 3. In our study population, chlamydia positivity was 5.3% among females tested in FP clinics. The goal of SSC is to detect those women most likely to be positive. Age was the best predictor of infection. If that were the first criterion for screening, 85% of all positives (10,621 of 12,437) would have been detected, whereas testing two thirds of the female population (Fig. 1) over the 3-year period of our study. Using additional selection criteria for women over 24, eg, clinical signs or conditions, or recent exposure to chlamydia, 15% of this older age group would have been tested, but the positivity in this group was 4.6% (Fig. 1). Therefore, these additional criteria would have identified a subset of women at highest risk. The remaining population of women over 24 had the lowest positivity, 1.6%, which some have suggested is below the threshold to make screening for chlamydial infection cost-effective. 18,24,28–30,36 However, the number of positives missed would have been less than 1 positive per clinic per year.
The current selective screening criteria used in the Region X Infertility Prevention Project are very inclusive; 80% to 87% of women screened in the last 3 years satisfied these criteria. This region, as the first large-scale population-based chlamydia screening program implemented in the United States, has been using these criteria for nearly 10 years. Even with the marked reductions in prevalence observed in this population and the increasingly widespread use of highly sensitive nucleic acid amplification tests, the current SSC were sensitive, although not particularly efficient. They required testing over 85% of women in various clinics to capture over 95% of all positives. Although Miller and colleagues recently suggested that 50% efficiency and 80% sensitivity were good benchmarks for selectively screening women for chlamydial infection, 35 the Region X Infertility Prevention Project Advisory Committee argued that failing to detect nearly 20% of all positives was unacceptable (Committee minutes, January 2002, unpublished data).
For most of the screening scenarios investigated, criteria efficiency ranged from 56% to 95%, and criteria sensitivity ranged from 82% to 98% across the 3 clinical settings. The limited variation in criteria sensitivity likely reflects the relative stability over time of those factors associated with infection. The wide range of efficiency might suggest that criteria should be tailored to the clinical setting in which screening is occurring. In Region X, the SSC have always been consistent across all 4 states and non-STD clinics. This has facilitated program implementation and training of clinical staff through provision of clear, consistent messages and clinical protocols. Our study suggests that STD clinics might benefit most by using selective screening for chlamydial infection, although potential tradeoffs in reducing the comprehensive diagnostic services currently offered in many of these settings would need to be considered carefully.
Finally, our study showed that the current criteria in use in Region X could be optimized. A selective screening approach that first tests all women 24 years and younger, then tests women over 24 years if they have cervical signs implicative of infection or report recent exposure to CT or sex with a symptomatic partner (Table 3, 6), would have been over 10% more efficient with only a modest 5% decrease in sensitivity. Had this been used in the STD clinic population, efficiency would have increased over 25% while still detecting 92% of all positives.
There are at least 3 potential limitations with our evaluation. First, selection bias could have been introduced by excluding a small proportion of tests among the screened population, not accounting for women who were tested frequently during the 3-year study period or by clinics not testing eligible women. However, less than 1% of all tests performed on women were excluded; women who are repeatedly tested would generally be included in the screening criteria (mostly because of age), and the Project has had a high participation rate among clinics since the program’s inception in 1988. Second, almost all women in our study had a cervical sample used as the diagnostic specimen for CT; urine for nucleic acid amplification testing (NAAT) has become more widespread. However, the performance of NAAT performed on cervical samples is comparable to that on urine, 44 so use of urine as the diagnostic modality in this study is unlikely to have changed our conclusions. Finally, the poor performance of our behavioral risk factors could be reflective of not having asked the right questions to detect asymptomatic infections in women over 24. It is possible that there are other characteristics of the over 24-year-old population of women that could aid in the development of more SSC for this group.
Based on the results of our evaluation, the Region X Infertility Prevention Project has modified their selective screening criteria for women tested for chlamydial infection at family planning and other non-STD clinics and began testing women accordingly in January 2003. The criteria eliminated include the 2 items reflecting risk behavior (new sex partner or more than 1 sex partner in the past 60 days), recent exposure to a partner with an STI that is not CT, and history of an STI that is not CT. These modified selective screening criteria in Region X are: 1) test all women under 25 years old; and 2) for women ≥25 years of age, test for chlamydial infection if at least 1 of the following exists: signs of mucopurulent cervicitis, PID, reported exposure to chlamydia, sex with a symptomatic partner in the past 60 days, chlamydial infection in the last 12 months, PID, pregnancy, or planned IUD insertion. 16 * As funding for large-scale screening initiatives becomes increasingly strained, incremental efficiency gained in selective screening criteria should translate into program savings.
* Accurate pregnancy and IUD data were not available during the study period, but have been a part of the screening criteria in Region X based on standard gynecologic clinical standards recommended by the American College of Obstetrics and Gynecology. Cited Here...
1. Department of Health and Human Services. Sexually transmitted disease surveillance 2000. Atlanta: Division of STD Prevention, Center for Disease Control and Prevention, 2001.
2. McCormack WM, Alpert S, McComb DE, et al. Fifteen-month follow-up study of women infected with Chlamydia trachomatis.
N Engl J Med 1979; 300:123–125.
3. Schachter J, Stoner E, Moncada J. Screening for chlamydial infections in women attending family planning clinics. West J Med 1983; 138:375–379.
4. Brunham RC, Paavonen J, Stevens CE, et al. Mucopurulent cervicitis—The ignored counterpart in women of urethritis in men. N Engl J Med 1984; 311:1–6.
5. Westrom L, Joesoef R, Reynolds G, et al. Pelvic inflammatory disease and fertility. A cohort study of 1844 women with laparoscopically verified disease and 657 control women with normal laparoscopic results. Sex Transm Dis 1992; 19:185–192.
6. Hillis SD, Owens LM, Marchbanks PA, et al. Recurrent chlamydial infections increase the risk of hospitalization for ectopic pregnancy and pelvic inflammatory disease. Am J Obstet Gynecol 1997; 1:103–106.
7. Cates W Jr, Wasserheit JN. Genital chlamydial infections: Epidemiology and reproductive sequelae. Am J Obstet Gynecol 1991; 164: 1771–1781.
8. DeMuylder X, Laga M, Tennstedt C, et al. The role of Neisseria gonorrhoeae
and Chlamydia trachomatis
in pelvic inflammatory disease and its sequelae in Zimbabwe. J Infect Dis 1990; 162:501–505.
9. Miettinen A, Heinonen PK, Teisala K, et al. Serologic evidence for the role of Chlamydia trachomatis
, Neisseria gonorrhoeae
, and Mycoplasma hominis
in the etiology of tubal factor infertility and ectopic pregnancy. Sex Transm Dis 1990; 17:10–14.
10. Mabey DCW, Ogbaselassie G, Robertson JN, et al. Tubal infertility in the Gambia: Chlamydial and gonococcal serology in women with tubal occlusion compared with pregnant controls. Bull World Health Organ 1985; 63:1107–1113.
11. Tjiam KH, Zeilmaker GH, Alberda AT, et al. Prevalence of antibodies to Chlamydia trachomatis
, Neisseria gonorrhoeae
, and Mycoplasma hominis
in infertile women. Genitourin Med 1985; 61:175–178.
12. Robertson JN, Ward ME, Conway D, et al. Chlamydial and gonococcal antibodies in sera of infertile women with tubal obstruction. J Clin Pathol 1987; 40:377–383.
13. Ness RB, Goodman MT, Shen C, et al. Serologic evidence of past infection with Chlamydia trachomatis
, in relation to ovarian cancer. J Infect Dis 2003; 187:1147–1152.
14. Lossick J, DeLisle S, Fine D, et al. Regional program for widespread screening for Chlamydia trachomatis
in family planning clinics. In: Bowie WR, Caldwell HD, Jones RP, et al., eds. Chlamydial Infections: Proceedings of the Seventh International Symposium on Human Chlamydial Infections. Cambridge: Cambridge University Press, 1990:575–579.
15. Britton TF, Delisle S, Fine D. STDs and family planning clinics: A regional program for chlamydia control that works. Am J Gynecol Health 1992; 6:80–87.
16. Region X Infertility Prevention Project Manual. Seattle: Center for Health Training, 2002.
17. Handsfield HH, Jasman LL, Roberts PL, et al. Criteria for selective screening for Chlamydia trachomatis
infection in women attending family planning clinics. JAMA 1986; 255:1730–1734.
18. Marrazzo JM, Celum CL, Hillis SD, et al. Performance and cost-effectiveness of selective screening criteria for Chlamydia trachomatis
infection in women: Implications for a national chlamydia control strategy. Sex Transm Dis 1997; 24:131–141.
19. Marrazzo JM, Fine D, Celum CL, et al. Selective screening for chlamydial infection in women: A comparison of three sets of criteria. Fam Plann Perspect 1997; 29:158–162.
20. Han Y, Coles FB, Hipp S. Screening criteria for Chlamydia trachomatis
in family planning clinics: Accounting for prevalence and clients’ characteristics. Fam Plann Perspect 1997; 29:163–166.
21. Stamm WE. Chlamydia trachomatis
infections of the adult. In: Holmes KK, Sparling PF, Mardh PA, et al., eds. Sexually Transmitted Diseases, 3rd ed. McGraw-Hill Health Professions Division, 1999: 407–422.
22. Mertz KJ, Levine WC, Mosure DJ, et al. Trends in the prevalence of chlamydial infections—The impact of community-wide testing. Sex Transm Dis 1997; 24:169–175.
23. Herrmann B, Egger M. Genital Chlamydia trachomatis
infections in Uppsala County, Sweden, 1985–1993: Declining rates for how much longer? Sex Transm Dis 1995; 22:253–260.
24. Kretzschmar M, Welte R, van den Hoek A, et al. Comparative model-based analysis of screening programs for Chlamydia trachomatis
infections. Am J Epidemiol 2001; 153:90–101.
25. Scholes D, Stergachis A, Heidrich FE, et al. Prevention of pelvic inflammatory disease by screening for cervical chlamydial infections. N Engl J Med 1996; 334:1362–1366.
26. Ostergaard L, Andersen B, Moeller JK, et al. Home sampling versus conventional swab samples for screening of Chlamydia trachomatis
in women: A cluster-randomized 1-year follow-up study. Clin Infect Dis 2000; 31:951–957.
27. Howell MR, Quinn TC, Brathwaite W, et al. Screening women for Chlamydia trachomatis
in family planning clinics—The cost-effectiveness of DNA amplification assays. Sex Transm Dis 1998; 25: 108–117.
28. Townsend JRP, Turner HS. Analysing the effectiveness of chlamydia screening. J Oper Res Soc 2000; 51:812–824.
29. Welte R, Kretzschmar M, Leidl R, et al. Cost-effectiveness of screening programs for Chlamydia trachomatis
—A population-based dynamic approach. Sex Transm Dis 2000; 27:518–529.
30. Honey E, Augood C, Templeton A, et al. Cost effectiveness of screening for Chlamydia trachomatis
: a review of published studies. Sex Transm Infect 2002; 78:406–412.
31. Dicker LW, Mosure DJ, Levine WC, et al. Impact of switching laboratory tests on reported trends in Chlamydia trachomatis
infections. Am J Epidemiol 2000; 151:430–435.
32. Department of Health and Human Services. Chlamydia prevalence monitoring project annual report—2001. Atlanta: Division of STD Prevention, Centers for Disease Control and Prevention, 2002.
33. Gotz H, Lindback J, Ripa T, et al. Is the increase in notifications of Chlamydia trachomatis
infections in Sweden the result of changes in prevalence, sampling frequency, or diagnostic methods? Scand J Infect Dis 2002; 34:28–34.
34. Mosure DJ, Berman S, Dicker LW, et al. A re-evaluation of selective screening criteria for women attending family planning clinics in the US. In: Stephens RS, et al., eds. Chlamydial Infections: Proceedings of the Ninth International Symposium on Human Chlamydial Infection. San Francisco: International Chlamydia Symposium, 1998: 333–336.
35. Miller WC, Hoffman IF, Owen-O’Dowd J, et al. Selective screening for chlamydial infection: Which criteria to use? Am J Prev Med 2000; 18:115–122.
36. United States Preventive Services Task Force. Screening for chlamydial infection: Recommendations and rationale. Am J Prev Med 2001; 20:90–94.
37. Fenton KA, Korovessis C, Johnson AM, et al. Sexual behaviour in Britain: Reported sexually transmitted infections and prevalence genital Chlamydia trachomatis
seroprevalence. Lancet 2001; 358: 1851–1854.
38. Grun L, Tassano-Smith J, Carder C, et al. Comparison of two methods of screening for genital chlamydial infection in women attending general practice: Cross sectional survey. BMJ 1997; 315:226–230.
39. Gaydos C, Howell R, Pare B, et al. Chlamydia trachomatis
infections in female military recruits. N Engl J Med 1998; 339:739–744.
40. Hiltunen-Back E, Haikala O, Kautiainen H, et al. A nationwide sentinel clinic survey of Chlamydia trachomatis
infection in Finland. Sex Transm Dis 2001; 28:252–258.
41. Todd CS, Haase C, Stoner BP. Emergency department screening for asymptomatic sexually transmitted infections. Am J Public Health 2001; 91:461–464.
42. Penney G, Thomson M, Norman J, et al. A randomized comparison of strategies for reducing infective complications of induced abortions. Br J Obstet Gynaecol 1998; 105:599–604.
43. Ryan G, Abdella T, McNeeley G, et al. Chlamydia trachomatis
infection in pregnancy and effect of treatment on outcome. Am J Obstet Gynecol 1990; 162:34–39.
44. Black CM, Marrazzo J, Johnson RE, et al. Head-to-heat multicenter comparison of DNA probe and nucleic acid amplification tests for Chlamydia trachomatis
infection in women performed with an improved reference standard. J Clin Microbiol 2002; 40:3757–3763.