Sexually Transmitted Diseases:
Screening and Treating Chlamydia trachomatis Genital Infection to Prevent Pelvic Inflammatory Disease: Interpretation of Findings From Randomized Controlled Trials
Gottlieb, Sami L. MD, MSPH*; Xu, Fujie MD, PhD*; Brunham, Robert C. MD†
From the *Division of STD Prevention, Centers for Disease Control and Prevention, Atlanta, GA; and †British Columbia Centre for Disease Control, Vancouver, British Columbia, Canada
Conflicts of interest: None.
Correspondence: Sami Gottlieb, MD, MSPH, Department of Reproductive Health and Research, World Health Organization, 20 Avenue Appia, 1211 Geneva 27, Switzerland. E-mail: firstname.lastname@example.org.
Received for publication April 5, 2012, and accepted August 14, 2012.
Abstract: We critically reviewed randomized controlled trials evaluating chlamydia screening to prevent pelvic inflammatory disease (PID) and explored factors affecting interpretation and translation of trial data into public health prevention. Taken together, data from these trials offer evidence that chlamydia screening and treatment is an important and useful intervention to reduce the risk of PID among young women. However, the magnitude of benefit to be expected from screening may have been overestimated based on the earliest trials. It is likely that chlamydia screening programs have contributed to declines in PID incidence through shortening prevalent infections, although the magnitude of their contribution remains unclear. Program factors such as screening coverage as well as natural history factors such as risk of PID after repeat chlamydia infection can be important in determining the impact of chlamydia screening on PID incidence in a population. Uptake of chlamydia screening is currently suboptimal, and expansion of screening among young, sexually active women remains a priority. To reduce transmission and repeat infections, implementation of efficient strategies to treat partners of infected women is also essential. Results of ongoing randomized evaluations of the effect of screening on community-wide chlamydia prevalence and PID will also be valuable.
Over the past 2 decades, chlamydia control programs have been implemented in the United States and elsewhere to reduce the burden of chlamydia-associated pelvic inflammatory disease (PID) and longer-term sequelae such as infertility, ectopic pregnancy, and chronic pelvic pain.1,2 These programs have focused primarily on screening sexually active young women for infection, with varying degrees of emphasis placed on treating sex partners of infected women and screening for repeat infection.3 Several recent ecologic analyses have demonstrated encouraging declines in PID diagnoses in the setting of increased chlamydia screening efforts.4–6 However, making causal inferences from ecologic data can be problematic, and ecologic findings have not been consistent across all countries.7 In addition, the lack of concomitant decreases in chlamydia prevalence among screened populations4,8 highlights the complexities in understanding the impact of chlamydia control efforts.9
Randomized controlled trials of chlamydia screening to prevent subsequent PID have formed the basis for chlamydia screening recommendations.10 However, translating findings from these trials into realistic expectations of chlamydia screening benefits in real-world settings can be more complicated. The impact of screening may be shaped by 2 main groups of factors: (1) factors related to the natural history of chlamydial infection that affect screening benefits for individual women and (2) factors related to translation of individual screening benefits to outcomes in the population. The most recent systematic review11 of chlamydia screening studies for PID prevention included 2 randomized controlled trials12,13 published through 2008. Since then, 2 additional randomized controlled trials have been published.14,15 The newer trials not only provide additional clinical trial findings on the impact of screening but also provide further insight into important natural history and translational factors. In this article, we sought to update the previous review11 by critically evaluating the 4 randomized controlled trials of chlamydia screening together and to explore in-depth the factors affecting interpretation and translation of these data into public health prevention efforts.
REVIEW OF RANDOMIZED CONTROLLED TRIAL FINDINGS
The earliest randomized controlled trial of chlamydia screening to prevent PID, conducted by Scholes et al.12 in 1990 to 1992, had a major influence on developing chlamydia screening recommendations by the US Preventive Services Task Force and others.10,16 In this study, women aged 18 to 34 years who were registered with a Seattle area health maintenance organization were randomized to screening and control groups and then sent a survey to determine study eligibility based on a risk score. A total of 2607 high-risk women (risk score ≥3)12 were enrolled. Those in the screening group received a 1-time invitation for chlamydia screening, and those in the control group were not invited for screening; 64% of women in the screening group were actually tested for chlamydia. Pelvic inflammatory disease diagnosis in the ensuing year was confirmed by chart review. Women invited for 1-time chlamydia screening had a greater than 50% reduction in the risk of PID over the next year, compared with women in the control group (relative risk [RR], 0.44; 95% confidence interval [CI], 0.20–0.88). Although this study was a randomized controlled trial, a methodological limitation highlighted in several critiques11,17 centers on the fact that only 7% of the 36,547 women who were initially randomized were ultimately enrolled. The investigators reported more aggressive contacting of survey nonresponders from the screening group to expedite chlamydia testing appointments after determining eligibility. This may have been the reason for an observed 1:1.6 ratio between the screening and the control groups instead of the expected 1:2 ratio, which introduces the possibility of selection biases that could compromise randomization.
In 1997 to 1998, Ostergaard et al.13 conducted a cluster randomized trial of 17 high schools in a Danish county, in which students in intervention schools were offered 1-time screening of home-collected specimens for chlamydia testing, and students in control schools were given referral information about clinic-based testing. Less than half the students returned eligibility surveys. Of those enrolled, 93% of students in the intervention group underwent chlamydia testing versus only 8% of controls. Incident PID was assessed by self-report at 1 year, with confirmation by pharmacy records. The 1-time screening approach was associated with a halving of PID occurrence among female students over 1 year (4.2% vs. 2.1%, P = 0.045). This study had several limitations, for example, the outcome assessment was not blinded, and almost 50% of students were lost to follow-up. Nonetheless, its findings were similar to those from the trial by Scholes et al. regarding the magnitude of the effect of chlamydia screening on reducing PID incidence.
The randomized controlled trial with the most rigorous methodology was the Prevention of Pelvic Infection (POPI) trial conducted in 2004 to 2007 in London.14 In this trial, sexually active women 27 years or younger were recruited from sites at 20 universities and further education colleges. Upon enrollment, all 2529 women provided self-collected vaginal swabs. The swab specimens were then randomly assigned to screening and control groups. In the screening group, swabs were tested immediately, and women with positive test results were treated for chlamydia. The control group had deferred screening, that is, the swabs were stored and underwent testing 12 months later. All women were blinded to the study arm, told that their samples may not be tested for 1 year, and told to seek testing independently if at risk. Clinical PID incidence was assessed at 12 months through blinded chart review. Comparing PID incidence during the 12 months in the screening group versus controls, a decrease of 35% was observed, but the decrease was not statistically significant (RR, 0.65; CI, 0.34–1.22; P = 0.19). Among women whose swabs collected at baseline were positive for chlamydia, 1.6% (1/63) in the screening group, and 9.5% (7/74) in the control group developed PID in the ensuing 12 months (RR, 0.17; CI, 0.03–1.01; P = 0.07). Despite its methodological strengths, the major limitation of this trial was its lack of power. Study authors stated that they assumed a 2% PID incidence in controls for their initial sample size calculations (similar to what they ultimately observed and what had been observed in the trial by Scholes et al.). However, they altered sample size calculations when study recruitment was slow, assuming 3% PID incidence in controls, based on studies among higher-risk populations. Issues of power aside, the observed reduction in overall PID incidence with chlamydia screening in this trial was consistent with the results of the previous trials,12,13 albeit with a slightly lower effect size.
The final randomized controlled trial of chlamydia screening to reduce PID incidence was conducted in a county of Denmark, where registry data allowed assessment of outcomes across a defined population.15 In 1997, investigators randomly selected a group of 4000 women to be mailed a 1-time invitation for chlamydia screening using a mail-in home test kit. The remaining 11,459 women in the county served as controls receiving usual care. Twenty-nine percent of women in the screening group received testing through the intervention, and 9% of both screening group and control women were tested as part of usual care. Hospitalized PID in the ensuing year was defined by registry discharge diagnosis codes; outpatient PID was defined by receipt of doxycycline. Pelvic inflammatory disease incidence in this study was much lower than that in prior studies and did not differ between the control and the screening groups (0.51 and 0.45 per 100 person-years, respectively; hazard ratio, 1.12; CI, 0.70–1.79). Although this study applied a randomized controlled design on a population level, low uptake of the screening intervention and imprecise measurement of outpatient PID outcomes limit the ability to demonstrate a benefit related to screening. Difficulties in measuring the impact of chlamydia screening using clinical PID outcomes are not unique to this trial, however, because clinical diagnosis is both subjective and inaccurate and may be influenced by patient history.18
Taken together, these randomized controlled trials provide direct evidence that chlamydia screening and treatment can reduce the risk of PID for 1 year for individual women. Although the POPI trial was insufficiently powered to definitively assess the effect of chlamydia screening on incident PID, its findings were consistent with earlier studies. Nonetheless, the smaller effect size in the rigorously conducted POPI trial, along with methodological limitations of prior studies, suggests possible overestimation of expected screening benefits based on the earliest trials. Beyond the direct interpretation of results, however, these randomized controlled trials can also provide additional insights to help guide our thinking about chlamydia screening. We also interpreted the clinical trial findings in the context of chlamydia natural history and population-level translational factors to explore further what is known and unknown about chlamydia screening benefits and to highlight areas for continued research and programmatic focus.
NATURAL HISTORY FACTORS AFFECTING BENEFITS OF SCREENING FOR AN INDIVIDUAL WOMAN
Interpretation of these clinical trials both informs and is informed by an understanding of the natural history of chlamydia infection (Fig. 1). For an individual woman with chlamydia, the potential benefits of screening are dependent on how much “downstream” pelvic inflammation would still occur after the point at which the infection is detected by screening, and the potential for treatment to cure the infection at that point. Assuming effective treatment, the amount of downstream PID that could be averted depends on 3 main things: the overall risk of PID if the infection is never treated, when PID occurs during the course of chlamydial infection, and when the infection is detected. When the infection is detected depends on the duration of natural infection and the screening interval. Annual screening is less likely to provide benefit if PID develops very early in the course of incident infection than if it develops as a result of long-standing prevalent infection. The benefits of screening must also be weighed against potential harms. When women are screened and treated, they are returned back to the pool of women susceptible to another chlamydia infection sooner than they would have been had they not been treated. Thus, the potential harm of repeat infection needs to be considered, including whether immunopathologic responses to chlamydia are more likely during a repeat infection than during the initial infection.19 This may be a particular concern if little attention is paid to treating sex partners of infected women. The risk of repeat infection depends not only on the risk of reexposure but also on the degree to which protective immunity may develop after an initial infection and whether such immunity can be altered by treatment.20,21
Risk of PID From Untreated Chlamydial Infection
The risk of PID from untreated chlamydial infection is clearly difficult to assess.18 It is unknown how long infections have been present when detected, and once diagnosed, infections should be treated. However, the POPI trial, conducted before the national chlamydia screening program was implemented in England, had a unique design that allowed assessment of PID risk. Overall, 9.5% (CI, 4.7%–18.3%) of 74 women with untreated prevalent infection developed PID over 12 months. Because 43% of infected women were tested and treated independently of the study, 9.5% likely represents the lower bound of the estimate, because a substantial proportion of these women may not have had untreated infection followed for a full 12 months. Small numbers introduce some uncertainty around the estimate. However, the findings show that the risk of PID from untreated chlamydia infection is neither negligible, as suggested in one small but widely cited prior study,22 nor as high as 30%, as observed in another.23 In the trial by Scholes et al.,12 if the rate of PID in the control group is applied to the intervention group, 21 cases of PID would be expected, but only 9 were observed. Thus, 12 cases of PID seemed to be prevented by treating 44 women, suggesting that 27% of infections identified by screening would have resulted in PID without treatment. This proportion (27%; 12/44) is significantly higher than the 9.5% (7/74) observed in the POPI trial (P = 0.01). This is one reason that some have suggested that the magnitude of the effect in the trial by Scholes et al. may have been overestimated.17
Timing of PID Relative to Screening
The timing of PID relative to chlamydia acquisition is unknown, but the POPI trial shows that there remains a substantial risk of PID after the point at which prevalent infections are detected through screening. In addition, the trial shows that treatment of infections in the weeks after screening likely reduces the risk of PID among women who are infected with chlamydia (as opposed to all women in the screened population). Although the numbers of PID cases were small and the difference was not statistically significant, screening for and treating chlamydia reduced the risk of clinical PID by more than 80% (from 7/74 to 1/63) among chlamydia-infected women. The natural history findings that a nonnegligible proportion of women with prevalent infection go on to develop PID and that treatment reduces that proportion may be the strongest argument for screening and perhaps more important than the main findings of the trial (for which it was clearly underpowered).
Lastly, the POPI trial was conducted before screening was fully implemented in England. Thus, this trial may be especially likely to include infections of longer duration, compared with infections detected in a more mature screening program. The duration of chlamydial infection in the absence of treatment has not been completely elucidated,24 but available long-term studies suggest that a typical untreated infection can last a year or longer.22,25 As more and more women are screened, and screened more frequently, after initiation of a program, the average duration of infection would be expected to decrease. Thus, the POPI trial suggests that benefit remains even when screening long-standing prevalent infections. Nonetheless, the importance of PID from long-standing infection should not overshadow the role of PID also occurring during early, incident infection. In the POPI trial, we do not know how many incident chlamydia infections occurred during follow-up and thus cannot directly compare the risk of PID from incident infection versus prevalent infection; however, more cases of PID seemed to be related to incident chlamydia infection than to prevalent infection.
Importance of Repeat Infections
Of note, the one woman in the POPI trial who had PID after being treated for chlamydia at baseline tested positive for chlamydia at the time of PID, after reporting unprotected sex with multiple partners. This suggests a repeat infection, which is a common occurrence. The peak repeat infection rate has been estimated to be approximately 20% at 1 year for women.26 In addition, it has been hypothesized that screening and treatment programs might make repeat infections more likely, through reduction in population levels of protective immunity.20 Animal models show greater tissue destruction during repeat chlamydial infection compared with initial infection.19 In addition, available human epidemiologic studies show that the cumulative risk of PID increases with an increasing number of repeat chlamydia infections.27,28 However, it is not clear that the per-infection risk goes up with each infection. In the epidemiologic studies, the duration of infections was unknown, so the increased risk may just reflect a longer cumulative duration of infection. Are 2 short infections greater risk than 1 long infection? The answer to this question is still unknown and may not be completely straightforward. It is likely that the cumulative duration of infection and the number of repeated infections both play a role in the pathogenesis of chlamydial PID, but the relative importance of each has not been elucidated.
FACTORS INFLUENCING TRANSLATION OF SCREENING BENEFITS TO POPULATION IMPACT
In translating the PID prevention benefits of chlamydia screening for individual women to the benefits of screening across an entire population, several additional factors need to be considered (Fig. 1). The first variable to consider is the proportion of all PID that is attributable to chlamydia; that is, how much of the total PID burden could be eliminated even if all chlamydia were prevented? Clearly, the burden of chlamydia in the population also plays a role. Another factor is screening coverage in the population and the efficiency in targeting infected women, which reflects how well a screening program is implemented. Finally, an important consideration is how much screening (along with other chlamydia prevention efforts) reduces transmission in the population; that is, in addition to preventing PID in the women screened, how many cases of PID can be prevented by interrupting transmission and thus eliminating new incident chlamydia infections in the first place?
Proportion of PID Attributable to Chlamydia
Pelvic inflammatory disease has multiple potential infectious causes other than Chlamydia trachomatis, including Neisseria gonorrhoeae, Mycoplasma genitalium, and organisms associated with bacterial vaginosis. Thus, chlamydia prevention programs would only be expected to prevent the proportion of PID that is attributable to chlamydia. In the first 2 trials of chlamydia screening to prevent PID, in Seattle and Denmark, screening cut the number of incident PID cases in half.12,13 Even if all chlamydia-associated cases of PID were prevented by screening, a 50% decrease is larger than what might be expected based on earlier studies of chlamydia and PID. In a synthesis of 19 studies through the mid-1990s, chlamydia was implicated in approximately 30% of acute PID cases, as measured by detection of C. trachomatis at the cervix at the time of PID.29 This is more consistent with the decrease in PID observed in the POPI trial.14 However, most of these 19 studies were conducted when the prevalence of gonorrhea was markedly higher than it is now and when less sensitive chlamydia tests were used. In addition, C. trachomatis at the cervix may not always predict C. trachomatis in the fallopian tubes and vice versa.29 In the POPI trial, at least 42% of all PID in the study was associated with chlamydia: of 38 PID cases, only 26 had laboratory results, and 16/26 (62%) were positive. If any of the 12 people without tests had chlamydia, the proportion would be higher than 42%. The numbers are small; however, it is entirely possible that the proportion of PID related to chlamydia is higher than it used to be and varies by population.
The chlamydia screening trial by Andersen et al.15 highlights the importance of screening implementation and coverage in translating possible individual benefits of screening to a population. Only a minority of women in the study intervention group were tested for chlamydia, even with the additional effort to invite women to be screened. In the United States, national screening recommendations have been in place for almost 2 decades. Although chlamydia screening coverage has been increasing, it remains suboptimal. Health plan data (Healthcare Effectiveness Data and Information Set [HEDIS] measures) show that screening coverage was 43% among sexually active females in commercial health plans and 58% among Medicaid patients in 2010.30 However, HEDIS measures only assess insured women accessing care and use eligibility criteria that tend to overestimate coverage. An analysis using a more representative estimation of eligible women suggests that screening rates could be as low as half those estimated by HEDIS.31 It is important to consider that screening coverage in the overall population does not necessarily reflect the proportion of infected women who are reached. To maximize the benefit of chlamydia screening to reduce PID and other adverse outcomes, it will be important to expand chlamydia screening generally, but especially for groups with the highest prevalence of infection, such as sexually active adolescents and blacks.8
Effect of Screening on Chlamydia Transmission
The trials reviewed in this article assessed secondary prevention: whether screening and treatment can prevent existing infection from progressing to PID. However, chlamydia prevention programs can also have an impact through primary prevention: reducing transmission in the population and preventing new infections and associated sequelae in the first place. In the United States, trends in population-wide transmission have been indirectly assessed through national, population-based surveys estimating chlamydia prevalence over time.8 Between 1999 and 2008, in the era of screening recommendations, chlamydia prevalence decreased overall but did not change among 14- to 25-year-old females, the target of recommended screening.8 Prevalence is a function of both incidence and duration of infection. A screening program would be expected to decrease the average duration of infection; thus, chlamydia prevalence could stay the same if a decreased duration of infection were offset by an increase in incidence. In turn, PID incidence could be reduced by detecting and treating long-standing prevalent infection, even with no change in population-wide prevalence.
The POPI trial demonstrated that long-standing prevalent chlamydia infection is still an important contributor to PID incidence.14 However, the trial also highlighted the importance of incident chlamydia infections in overall PID burden. This may be especially important if arrested immunity contributes to an increase in new, repeat infections.20 Prevalence trends from the United States suggest that progress is being made, but more needs to be done to optimally interrupt chlamydia transmission dynamics. A recent modeling study demonstrates that treating male sex partners of infected women identified through screening is critical to interrupting chlamydia transmission in the population.32 Partner treatment is the only prevention intervention that not only contributes to reducing ongoing transmission but also reduces repeat infections. Expedited partner therapy, providing prescriptions or medications to a patient to take to his/her partner without examination of the partner, is endorsed by health officials as a useful and cost-effective strategy to increase partner treatment and decrease repeat infections.33,34
CONCLUSIONS AND FUTURE DIRECTIONS
Data from randomized controlled trials of chlamydia screening to prevent PID, taken together, offer evidence that chlamydia screening and treatment is a useful intervention to reduce the risk of PID among young women. The POPI trial, although underpowered to definitively assess screening impact, provides new natural history data on the risk and timing of PID that support the rationale for screening. This trial confirms that prevalent, perhaps even long-standing, chlamydia infection continues to yield PID risk but suggests that the magnitude of benefit to be expected from screening may have been overestimated based on earlier trials. It is likely that chlamydia screening programs have contributed to declines in PID incidence observed in ecologic studies through shortening prevalent infections, yet the magnitude of their contribution remains unclear. New dynamic modeling studies incorporating different possibilities for the timing of PID development may shed further light on the contributions of chlamydia screening to PID prevention.35
Several factors affect translation of the individual chlamydia screening benefits observed in randomized controlled trials to population-wide prevention. Exploring these factors highlighted several areas for future research and programmatic focus. First, our understanding of the proportion of PID that is attributable to chlamydia may be outdated. Updated estimates of the chlamydia attributable fraction should be sought not only for PID but also for tubal factor infertility and ectopic pregnancy, the ultimate targets of prevention efforts. Second, uptake of chlamydia screening is currently suboptimal, and expansion of screening among young, sexually active women remains a priority. Screening efforts would be enhanced by continued refinement of methods to measure screening uptake and other chlamydia control activities, along with disease outcomes.9 Finally, the POPI trial highlights the importance of interrupting chlamydia transmission to prevent new incident infections. Because partner treatment is essential to reducing both transmission and repeat infections, implementation of efficient strategies to treat partners, such as expedited partner therapy, is critical. Ongoing randomized evaluations in Australia36 and the Netherlands,37 assessing the effect of multiple rounds of chlamydia screening on chlamydia prevalence in addition to PID, will provide valuable data on the association between chlamydia screening efforts and interruption of ongoing transmission.
1. Centers for Disease Control and Prevention. Recommendations for the prevention and management of Chlamydia trachomatis infections, 1993. MMWR Morb Mortal Wkly Rep 1993; 42: 1–38.
2. Low N. Publication of report on chlamydia control activities in Europe. Euro Surveill 2008; 13: pii:18924.
3. Centers for Disease Control and Prevention. CDC Grand Rounds: Chlamydia prevention: Challenges and strategies for reducing disease burden and sequelae. MMWR Morb Mortal Wkly Rep 2011; 60: 370–373.
4. Scholes D, Satterwhite CL, Yu O, et al.. Long-term trends in Chlamydia trachomatis infections and related outcomes in a US managed care population. Sex Transm Dis 2012; 39: 81–88.
5. Owusu-Edusei K Jr, Bohm MK, Chesson HW, et al.. Chlamydia screening and pelvic inflammatory disease: Insights from exploratory time-series analyses. Am J Prev Med 2010; 38: 652–657.
6. Anschuetz GL, Asbel L, Spain CV, et al.. Association between enhanced screening for Chlamydia trachomatis and Neisseria gonnorhoeae and reductions in sequelae among women. J Adolesc Health 2012; 51: 80–85
7. Bender N, Herrmann B, Andersen B, et al.. Chlamydia infection, pelvic inflammatory disease, ectopic pregnancy and infertility: Cross-national study. Sex Transm Infect 2011; 87: 601–608.
8. Datta SD, Torrone E, Kruszon-Moran D, et al.. Chlamydia trachomatis trends in the United States among persons 14 to 39 years of age, 1999–2008. Sex Transm Dis 2012; 39: 92–96.
9. Chow JM. Measuring the uptake and impact of chlamydia screening programs—easier said than done. Sex Transm Dis 2012; 39: 89–91.
10. U.S. Preventive Services Task Force. Screening for chlamydial infection: U.S. Preventive Services Task Force Recommendation Statement. Ann Intern Med 2007; 147: 128–133.
11. Gottlieb SL, Berman SM, Low N. Screening and treatment to prevent sequelae in women with Chlamydia trachomatis genital infection: How much do we really know? J Infect Dis 2010; 201 (suppl 2): S156–S167.
12. Scholes D, Stergachis A, Heidrich FE, et al.. Prevention of pelvic inflammatory disease by screening for cervical chlamydial infection. New Engl J Med 1996; 334: 1362–1366.
13. Ostergaard L, Andersen B, Moller JK, et al.. Home sampling versus conventional swab sampling for screening of Chlamydia trachomatis in women: A cluster-randomized 1-year follow-up study. Clin Infect Dis 2000; 31: 951–957.
14. Oakeshott P, Kerry S, Aghaizu A, et al.. Randomised controlled trial of screening for Chlamydia trachomatis to prevent pelvic inflammatory disease: The POPI (Prevention of Pelvic Infection) trial. BMJ 2010; 340: c1642.
15. Andersen B, van Valkengoed I, Sokolowski I, et al.. Impact of intensified testing for urogenital Chlamydia trachomatis infections: A randomised study with 9-year follow-up. Sex Transm Infect 2011; 87: 156–161.
16. Centers for Disease Control and Prevention. Sexually transmitted diseases treatment guidelines, 2010. MMWR Morb Mortal Wkly Rep 2010; 59: 1–110.
17. Low N, Bender N, Nartey L, et al.. Effectiveness of chlamydia screening: Systematic review. Int J Epidemiol 2009; 38: 435–448.
18. Haggerty CL, Gottlieb SL, Taylor BD, et al.. Risk of sequelae after Chlamydia trachomatis genital infection in women. J Infect Dis 2010; 201 (suppl 2): S134–S155.
19. Darville T, Hiltke TJ. Pathogenesis of genital tract disease due to Chlamydia trachomatis. J Infect Dis 2010; 201 (suppl 2): S114–S125.
20. Brunham RC, Rekart ML. The arrested immunity hypothesis and the epidemiology of chlamydia control. Sex Transm Dis 2008; 35: 53–54.
21. Batteiger BE, Xu F, Johnson RE, et al.. Protective immunity to Chlamydia trachomatis genital infection: Evidence from human studies. J Infect Dis 2010; 201 (suppl 2): S178–S189.
22. Morre SA, van den Brule AJ, Rozendaal L, et al.. The natural course of asymptomatic Chlamydia trachomatis infections: 45% clearance and no development of clinical PID after one-year follow-up. Int J STD AIDS 2002; 13 (suppl 2): 12–18.
23. Stamm WE, Guinan ME, Johnson C, et al.. Effect of treatment regimens for Neisseria gonorrhoeae on simultaneous infection with Chlamydia trachomatis. N Engl J Med 1984; 310: 545–549.
24. Geisler WM. Duration of untreated, uncomplicated Chlamydia trachomatis genital infection and factors associated with chlamydia resolution: A review of human studies. J Infect Dis 2010; 201 (suppl 2): S104–S113.
25. Molano M, Meijer CJ, Weiderpass E, et al.. The natural course of Chlamydia trachomatis infection in asymptomatic Colombian women: A 5-year follow-up study. J Infect Dis 2005; 191: 907–916.
26. Hosenfeld CB, Workowski KA, Berman S, et al.. Repeat infection with chlamydia and gonorrhea among females: A systematic review of the literature. Sex Transm Dis 2009; 36: 478–489.
27. Hillis SD, Owens LM, Marchbanks PA, et al.. Recurrent chlamydial infections increase the risks of hospitalization for ectopic pregnancy and pelvic inflammatory disease. Am J Obstet Gynecol 1997; 176: 103–107.
28. Kimani J, Maclean IW, Bwayo JJ, et al.. Risk factors for Chlamydia trachomatis pelvic inflammatory disease among sex workers in Nairobi, Kenya. J Infect Dis 1996; 173: 1437–1444.
29. Paavonen J, Westrom L, Eschenbach DA. Pelvic inflammatory disease. In: Holmes KK, Sparling PF, Stamm WE, et al., eds. Sexually Transmitted Diseases. New York: McGraw Hill Medical, 2008: 1017–1050.
31. Heijne JC, Tao G, Kent CK, et al.. Uptake of regular chlamydia testing by U.S. women: A longitudinal study. Am J Prev Med 2010; 39: 243–250.
32. Kretzschmar M, Satterwhite CL, Leichliter JS, et al.. Effects of screening and partner notification on chlamydia positivity in the United States: A modeling study. Sex Transm Dis 2012; 39: 325–331.
33. Trelle S, Shang A, Nartey L, et al.. Improved effectiveness of partner notification for patients with sexually transmitted infections: Systematic review. BMJ 2007; 334: 354.
34. Gift TL, Kissinger P, Mohammed H, et al.. The cost and cost-effectiveness of expedited partner therapy compared with standard partner referral for the treatment of chlamydia or gonorrhea. Sex Transm Dis 2011; 38: 1067–1073.
35. Herzog SA, Heijne JCM, Althaus CL, et al.. Describing the progression from Chlamydia trachomatis and Neisseria gonorrhoeae to pelvic inflammatory disease: Systematic review of mathematical modeling studies. Sex Transm Dis 2012; 39: 628–637.
36. Low N, Hocking J. The POPI trial: What does it mean for chlamydia control now? Sex Transm Infect 2010; 86: 158–159.
37. van den Broek IVF, Hoebe CJPA, van Bergen JEAM, et al.. Evaluation design of a systematic, selective, Internet-based, chlamydia screening implementation in the Netherlands, 2008–2010: Implications of first results for the analysis. BMC Infect Dis 2010; 10: 89.
© Copyright 2013 American Sexually Transmitted Diseases Association