McNicholas, Colleen DO; Peipert, Jeffrey F. MD, PhD; Maddipati, Ragini MSW; Madden, Tessa MD, MPH; Allsworth, Jenifer E. PhD; Secura, Gina M. PhD, MPH
Sexually transmitted infections (STIs) carry significant and important implications for female reproductive health, including pelvic inflammatory disease (PID), ectopic pregnancy, chronic pelvic pain, and infertility.1–5 Age, race, history of STI, early age of first intercourse, number of lifetime sexual partners, and multiple sexual partners have been repeatedly shown to be associated with STIs. The Centers for Disease Control and Prevention (CDC) currently recommends routine CT screening for all sexually active women younger than 26 years and risk-based screening for women 26 years or older.6
Screening recommendations generally arise from disease prevalence data and are often tailored to patients with known risk factors. Accurate estimation of disease prevalence is difficult. Traditionally, statistics regarding prevalence are derived from large population studies such as the National Health and Nutrition Examination Survey (NHANES) or notifiable disease surveillance systems. These sources provide valuable information on an ongoing basis; most notably, they provide a generalizable assessment of disease prevalence. However, these sources can be dependent on positive test result reporting by clinicians and often draw from sentinel sites such as STI clinics, family planning clinics, and the National Job Training Program.
Accurate estimates of disease prevalence are difficult for a number of reasons. Screening modalities and protocols may change over time, making it difficult to determine whether changes in prevalence are attributable to an actual change in infection rates or a change in screening indications, testing method, and reporting. For example, Job Corps entrants from 2003 to 2007 were screened for Chlamydia trachomatis (CT) using different tests.7 Second, many infections are asymptomatic, and thus, adherence to screening guidelines highly influences the number of positive cases diagnosed. The CDC estimates that screening occurs in less than half of those who should be screened.8–11 Lastly, the existence of multiple diagnostic tests for each infection further complicates the issue because each of these tests carries its own sensitivity and specificity. For example, most clinicians still rely on traditional saline microscopy for trichomoniasis, which has been shown to have a sensitivity of approximately 60%, rather than the more modern and sensitive (95%) nucleic acid amplification or culture testing modalities.12–14
Positive cases reported to the CDC have allowed for estimation of disease prevalence. In 2010, more than 300,000 cases of Neisseria gonorrhoeae (GC) were reported for a rate of 100.8 per 100,000 people (0.10%). There are more than 1.3 million cases of CT in the United States for a rate of 426 per 100,000 people (0.43%).8,15 Trichomonas vaginalis (TV) prevalence is more difficult to estimate. Trichomoniasis is not a reportable infection, there are no screening guidelines, and the most commonly used diagnostic technique (microscopic evaluation) is inadequate. Despite these difficulties, the CDC has estimated that there are 2.3 million cases annually in United States, and data from the NHANES found an overall prevalence of 3.1%.14
In an effort to supplement prevalence data from large population-based, cross-sectional studies and surveillance initiatives, we implemented a universal screening protocol using highly sensitive and specific nucleic acid amplification tests for GC and CT and culture for TV in an urban population seeking no-cost contraception. We estimated the baseline prevalences of GC, CT, and TV in women at baseline enrollment and compared these rates with the national rates reported by the CDC.
MATERIALS AND METHODS
This is an analysis of data collected from participants at the time of enrollment into the Contraceptive CHOICE Project (CHOICE). All study related procedures were approved by Washington University in St Louis School of Medicine Human Research Protection Office before participant recruitment. The Contraceptive CHOICE Project is an observational cohort study of 9256 women residing in the St Louis area. The project was designed to promote the use of the most effective contraceptive methods and reduce unintended pregnancy in the region. The methods of CHOICE have been previously described16; we will provide a brief summary here. Participants were recruited through word of mouth, health care provider referrals, and at community family planning, abortion clinics, and university-based medical clinics. Women were eligible for CHOICE if they were 14 to 45 years old, not currently using a method of contraception or willing to try a new reversible contraceptive method, did not desire pregnancy for one year, did not have a hysterectomy or tubal ligation, and were currently sexually active with a male partner or planned to be sexually active in the next 6 months. All participants completed an in-person enrollment session that included standardized contraceptive counseling, informed written consent, a staff-administered baseline questionnaire, and STI screening. The baseline questionnaire included questions regarding demographic characteristics, reproductive health and contraceptive history, sexual behaviors with current partners, drug and alcohol use, and history of STI.
During enrollment, women completed self-collected vaginal swabs to test for CT, GC, and TV infections. DNA strand displacement analysis using BDProbeTec ET instrument (Becton Dickson, Sparks, MD) was used to detect CT and GC infections. T. vaginalis was detected using the InPouch TV culture (BioMed Diagnostics, White City, OR).
We examined the demographic and behavioral characteristics known to be associated with STIs and compared women who were tested positive for any STI with those who were tested negative for all infections. We used the χ2 test for comparisons of categorical variables and t test for continuous variables. We created a multivariable model using logistic regression to estimate the odds of any infection. We also created 2 multivariable models to estimate the odds of CT infection only and the odds of TV infection only. Independent predictors of CT infection and TV infection that were statistically significant in univariate analysis were included in the final multivariable models. All statistical analyses were completed using SAS Software (v.9.2.; SAS Institute, Cary, NC).
From August 2007 through September 2011, 9256 women were enrolled into the Contraceptive CHOICE Project. Table 1 summarizes the demographic and reproductive characteristics of the cohort. Greater than one third of the cohort (39%) was 26 years or older, and half (51%) identified themselves as African American or black. Nearly two thirds (65%) of the cohort reported at least some college or vocational education, and the majority (61%) were single or never married. Low socioeconomic status (SES) was common (58%) and was defined as those women who reported having trouble paying for housing, transportation, food, or health care, as well as those receiving public assistance such as food stamps, welfare, or unemployment.
Of the 9256 participants, results were available for 8347 (90%; Fig. 1). The remaining 909 include participants with incomplete results or who declined testing. There were 656 (7.9%) women in the cohort who were tested positive for 1 or more STI. Of the positive screens, there were 35 cases of GC for a prevalence of 0.4% (95% confidence interval [CI], 0.3–0.6), 260 cases of CT for a prevalence of 3.1% (95% CI, 2.8–3.5), and 410 cases of TV for a prevalence of 4.9% (95% CI, 4.4–5.4).
Demographic differences between those tested positive for 1 or more STI and those tested negative for all STIs are presented in Table 2. Of note, the STI-positive group was younger, with a mean age of 24.2 years, compared with participants tested negative (mean age, 25.3 years; P < 0.001). Black women were disproportionately represented in the group tested positive for 1 or more STI (81%). Those who were tested positive were also more likely to have never been married (76% vs. 60%, P < 0.001) and to be of low SES (73% vs. 56%, P < 0.001).
Multivariable analysis (Table 2), which included all factors found to be significant in the univariate model, identified age 18 to 21 years to be associated with a nearly 2-fold increased risk (odds ratio [OR], 1.9; 95% CI, 1.2–3.0) of any prevalent infection. Also significant were black race (OR, 4.0; 95% CI, 3.1–5.1), never having been married (OR, 2.7; 95% CI, 1.8–4.1), being partnered but not married (OR, 1.7; 95% CI, 1.1–2.7), and having less than or equal to a high school education (OR, 2.0; 95% CI, 1.4–2.8).
We looked specifically at the 656 women tested positive at baseline and evaluated whether they would have been screened under traditional age-based and risk-profile screening recommendations. Of the 656 positive screens, 188 (29%) were 26 years or older and outside the current age-based recommendation for annual screening. Of these 188 women, only 22 (12%) reported having more than 1 current partner, and 75 (40%) reported a history of GC, CT, syphilis, or HIV. Among the women who were tested positive, we identified 106 (16%) who were 26 years or older, reported no STI history, did not have multiple sexual partners, did not report vaginal symptoms, and, as a result, would not have been screened under the CDC guidelines.
Demographic and reproductive characteristics of those who were tested positive for CT (n = 260) and those who were tested positive for TV (n = 410) were also compared. N. gonorrhoeae was not included in this analysis because of the small number of positive cases. Younger women were more likely to be tested positive for CT. However, the distribution of TV prevalence was shifted toward older women. T. vaginalis became more prevalent than CT, beginning with the 22- to 25-year age group, but is most dramatically noted in the women older than 35 years (Fig. 2). Prevalence by race was also different. C. trachomatis prevalence in black participants was higher than in white women (4.6% vs. 1.7%). A more dramatic difference was seen in TV prevalence rates by race, with black women having a TV prevalence of 8.7% compared with 0.9% in white women.
Multivariable analysis confirmed these findings (Table 2). Younger age, both 14 to 17 and 18 to 21 years, were found to have a significantly higher risk for testing positive for CT at baseline (ORs, 6.7 [95% CI, 1.8–25.1] and 11.0 [95% CI, 3.3–36.6], respectively) compared with women aged 36 to 45 years (referent group). Age 14 to 17 years in the TV group was negatively associated with TV (OR, 0.4; 95% CI, 0.1–0.3). Black race also remained a significant predictor of both infections, however, with different magnitudes of effect. Black women were more likely to be tested positive for CT (OR, 2.2; 95% CI 1.6–3.1) and much more likely to be tested positive for TV (OR, 6.6; 95% CI, 4.5–9.7) than white women.
Accurately estimating STI prevalence is difficult but important. Prevalence data guide screening practices, which, in turn, facilitate earlier and more comprehensive treatment with the goal of reducing long-term sequelae from infection. The prevalence of STIs in our cohort was higher than that observed in national estimates. We found the prevalence of GC to be 0.4% compared with 0.1% reported by the CDC. Similarly, our prevalence of CT was much higher at 3.1% compared with 0.43% nationally. Lastly, overall prevalence of TV was found to be 4.9% compared with 3.1% from national estimates and as high as 8.7% among black women in our study population. Estimates from the NHANES data are more closely reflective of our findings, which report rates of GC at 0.3%, CT at 2.5%, and TV of 3.2%.17,18 It should be noted that St Louis has been shown to have an overall STI positivity rate that is higher than many other metropolitan areas. In 2011, the Missouri department of health reported rates of GC and CT for women residing in St Louis City as 611 per 100,000 and 1454 per 100,000.19
Of the 656 women tested positive for 1 or more STIs at baseline, 106 (16%) would not have been screened according to current guidelines. These are women who fell outside the age criteria for universal screening and would not have been considered for risk-based testing based on history or current sexual practices. This number confirms that STI prevalence can vary widely among populations, and knowledge of local and population-specific prevalence can be helpful for clinicians to determine how and when to apply national screening guidelines.
There are no current screening guidelines for TV despite its high prevalence. Based on our data, it seems that the risk profiles for TV and CT are different, and thus, screening practices should not focus solely on young sexually active women. Prevalence estimation of TV infection is particularly difficult. T. vaginalis is not a reportable disease, and the implications of TV infection are less well established than those of GC and CT. In addition, diagnostic testing is highly variable, with most clinicians still relying on saline microscopy (wet mount), which has low sensitivity. By using the specific and sensitive culture technique, we were able to confirm that TV was the most prevalent STI in our cohort. Screening for CT is clearly indicated to prevent the adverse reproductive outcomes of PID and tubal infertility20; however, the role of TV in the development of PID and infertility is less clear and deserves further study. Routine screening may be indicated if TV is implicated in the etiology of upper genital tract infection and adverse outcomes such as infertility, chronic pain, and/or ectopic pregnancy. Trichomoniasis has been identified as a risk factor for adverse pregnancy outcomes.21
There are several strengths to this study. We were able to implement a universal and consistent screening protocol that allowed for testing of a large cohort of women including women that would not have traditionally been considered for screening. Second, because this was a cohort of women enrolling into a contraceptive study, it was truly a screening protocol. Women were not presenting for care with STI-related symptoms or risk factors. Last, we were able to implement our screening using sensitive and specific diagnostic techniques.
Our study does have limitations. Findings from this study may not be generalizable. As previously mentioned, St Louis has higher rates of STI positivity than many other urban areas. However, previous studies have shown that despite differences in sampling techniques, CHOICE is largely reflective of both state (2006 Missouri Behavioral Risk Factor Surveillance System) and national (2006–2008 National Survey of Family Growth) surveys.22 Lastly, national data include women aged 15 to 65 years when determining rates of STIs. In our cohort, all women were aged 14 to 45 years. If national prevalence estimates were limited to reproductive-aged women, they may actually be more similar to our reported rates.
In summary, prevalence of GC, CT, and TV in our population was higher than national estimates. T. vaginalis was the most prevalent STI in our cohort, yet it is not a reportable STI and is not recommended as part of the routine screening guidelines. Black women are at a disproportionately higher risk for CT and TV. Lastly, age is an important epidemiologic difference between CT and TV infection: young age is associated with CT and older reproductive age with trichomoniasis.
1. Peipert JF. Clinical practice. Genital chlamydial infections. N Engl J Med 2003; 349: 2424–2430.
2. Sweet RL, Gibbs RS. Infectious diseases of the female genital tract. In: G.R. Sweet, ed. Pelvic Inflammatory Disease
. 4th ed. Philadelphia: Lippincott, Williams and Wilkins, 2002: 368–412.
3. Westrom L. Effect of acute pelvic inflammatory disease on fertility. Am J Obstet Gynecol 1975; 121: 707–713.
4. Weström L, Joesoef R, Reynolds G, et al. Pelvic inflammatory disease and fertility. A cohort study of 1,844 women with laparoscopically verified disease and 657 control women with normal laparoscopic results. Sex Transm Dis 1992; 19: 185–192.
5. Chow JM, Yonekura ML, Richwald GA, et al. The association between Chlamydia trachomatis
and ectopic pregnancy. A matched-pair, case-control study. JAMA 1990; 263: 3164–3167.
6. Workowski KA, Berman S. Sexually transmitted diseases treatment guidelines, 2010. MMWR Recomm Rep 2010; 59 (RR-12): 1–110.
7. Satterwhite CL, Tian LH, Braxton J, et al. Chlamydia prevalence among women and men entering the National Job Training Program: United States, 2003–2007. Sex Transm Dis 2010; 37: 63–67.
9. Hoover K, Bohm M, Keppel K. Measuring disparities in the incidence of sexually transmitted diseases. Sex Transm Dis 2008; 35 (12 suppl): S40–S44.
10. Hoover K, Tao G. Missed opportunities for chlamydia screening of young women in the United States. Obstet Gynecol 2008; 111: 1097–1102.
11. Hoover K, Tao G, Kent C. Low rates of both asymptomatic chlamydia screening and diagnostic testing of women in US outpatient clinics. Obstet Gynecol 2008; 112: 891–898.
12. Andrea SB, Chapin KC. Comparison of Aptima Trichomonas vaginalis
transcription-mediated amplification assay and BD affirm VPIII for detection of T. vaginalis
in symptomatic women: Performance parameters and epidemiological implications. J Clin Microbiol 2011; 49: 866–869.
13. Chapin K, Andrea S. APTIMA(R) Trichomonas vaginalis
, a transcription-mediated amplification assay for detection of Trichomonas vaginalis
in urogenital specimens. Expert Rev Mol Diagn 2011; 11: 679–688.
14. Sutton M, Sternberg M, Koumans EH, et al. The prevalence of Trichomonas vaginalis
infection among reproductive-age women in the United States, 2001–2004. Clin Infect Dis 2007; 45: 1319–1326.
15. Kent CK, Chaw JK, Kohn RP, et al. Studies relying on passive retrospective cohorts developed from health services data provide biased estimates of incidence of sexually transmitted infections. Sex Transm Dis 2004; 31: 596–600.
16. Secura GM, Allsworth JE, Madden T, et al. The Contraceptive CHOICE Project: Reducing barriers to long-acting reversible contraception. Am J Obstet Gynecol 2010; 203: 115.e1–115.e7.
17. Datta SD, Sternberg M, Johnson RE, et al. Gonorrhea and chlamydia in the United States among persons 14 to 39 years of age, 1999 to 2002. Ann Intern Med 2007; 147: 89–96.
18. Allsworth JE, Ratner JA, Peipert JF. Trichomoniasis and other sexually transmitted infections: Results from the 2001–2004 National Health and Nutrition Examination Surveys. Sex Transm Dis 2009; 36: 738–744.
20. Scholes D, Stergachis A, Heidrich FE, et al. Prevention of pelvic inflammatory disease by screening for cervical chlamydial infection. N Engl J Med 1996; 334: 1362–1366.
21. Cotch MF, Pastorek JG 2nd, Nugent RP, et al. Trichomonas vaginalis
associated with low birth weight and preterm delivery. The Vaginal Infections and Prematurity Study Group. Sex Transm Dis 1997; 24: 353–360.
22. Kittur ND, Secura GM, Peipert JF, et al. Comparison of contraceptive use between the Contraceptive CHOICE Project and state and national data. Contraception 2011; 83: 479–485.