Objective
The purpose of this study was to determine the prevalence of Chlamydia trachomatis (CT) and Neisseria gonorrhoeae (NG) in older women who are at increased risk of sexually transmitted infections (STIs) based on community STI prevalence. Additionally, we explored the associations between STI risk factors and CT/NG to determine the percentage of women who were over-screened.
Study Design
We conducted a retrospective chart review of women ≥25 years old who were either tested or screened for CT/NG during a gynecology visit at an urban teaching institution. Pregnancy and HIV infection were exclusion criteria. Descriptive statistics, univariate analyses, and logistic regression were performed.
Results
Of 658 eligible women, the median age of those positive for CT/NG was 30 years (range, 26–41 years). Chlamydia and gonorrhea prevalence was 1.7% (11/658 women) and 0.3% (2/658 women), respectively. All positive results were captured by testing women of any age who reported symptoms or an STI exposure and by screening women who were ≤40 years old. After adjustment of data for age, we found that symptomatic women were 3 times more likely to test positive for CT/NG (adjusted odds ratio, 3.4; 95% confidence interval, 1.1–10.3) and that STI-exposed women were 10 times more likely to test positive for CT/NG (adjusted odds ratio, 10; 95% confidence interval, 1.9–52.5). In asymptomatic non-STI–exposed women, nonmonogamous relationship ( P = 1.0), abnormal examination results ( P = 1.0), and previous STI ( P = .35) were not associated with CT/NG. Over-screening occurred in 21% of women (141/658), all of whom were menopausal, had a hysterectomy, or were >40 years old.
Conclusion
CT/NG prevalence among older women was low, even in a community of high STI prevalence. More than 20% of women could have avoided CT/NG evaluation without impacting the detection of positive results in our clinic cohort. Over-screening occurred among asymptomatic, non-STI–exposed women who were menopausal, had a hysterectomy, and were >40 years old.
There are an estimated 2.8 million chlamydia infections and 820,000 gonorrhea infections annually in the United States. The United States Preventive Services Task Force (USPSTF) recommends annual screening of all women <25 years old and women ≥25 years old who are at increased risk for acquiring sexually transmitted infections (STIs). Women are considered at increased risk of STI acquisition if they engage in high-risk sexual behaviors (ie, report new or multiple current sexual partners, exchange sex for money or drugs, use condoms inconsistently, or have sex while under the influence of alcohol or drugs). Additionally, USPSTF states that black women may be at increased risk of chlamydia and gonorrhea, irrespective of age or sexual behaviors, in communities of high STI prevalence; therefore, annual screening can be considered.
Chlamydia trachomatis (CT) and Neisseria gonorrhoeae (NG) infections can lead to significant acute and chronic morbidity among reproductive-aged women, including pelvic inflammatory disease, infertility, ectopic pregnancies, and chronic pelvic pain. Physicians and other health care providers play a critical role in STI prevention, particularly secondary prevention through screening. However, CT/NG screening without an indication in older women, other than living in communities of high STI prevalence, might lead to significant health care expenditures while yielding minimal clinical benefit. Furthermore, the clinical implications of untreated asymptomatic infections in nonreproductive aged women are likely different than in women of younger, reproductive ages.
Studies have demonstrated that health care providers under-screen young women, while perhaps over-screening older women. Using a commercial claims database of >34 million patients, Tao et al found that 65% of CT/NG tests were performed among women 25-44 years old. Within this older age group, only 56% of tests were prompted by symptoms or pregnancy. It is unclear whether the remaining 44% were screened unnecessarily, because this study did not analyze patient-level clinical data.
Given the lack of patient-level data among older women who live in high STI prevalent areas and the lack of direct evidence for guideline committees to recommend a specific age to stop screening, we sought to determine the prevalence of CT/NG in women ≥25 years old, with and without symptoms, who were examined at a gynecology clinic located in a community with high STI prevalence. In addition, we explored associations between STI risk factors and CT/NG to determine the percentage of women who were over-screened.
Materials and Methods
We conducted a cross-sectional study of women ≥25 years old who were tested for CT/NG in an outpatient gynecology resident clinic at Johns Hopkins Hospital in Baltimore, MD. Because of Baltimore’s high STI prevalence, our gynecology clinic follows the USPSTF recommendations that all patients, irrespective of age or symptoms, be tested for CT/NG (cotesting) annually. A list of consecutive clinic patients who were examined from Jan. 1, 2009, to Dec. 31, 2010, was obtained from the billing department and compared with a list of CT/NG test results that had been generated by the laboratory during that time frame. Electronic medical records were reviewed, and data were collected and entered into a database by 2 authors (J.A.J. and J.S.C). To ensure data integrity, 1 author (J.S.C.) reviewed 10% of completed entries. We used a standardized data collection form to collect demographic and clinical information (ie, age, race, marital status, signs, and symptoms at time of visit). If a patient had multiple gynecology clinic visits over the time frame, only data from the initial visit that had both a clinical note and CT/NG test result were used for data abstraction. Women who were pregnant or infected with HIV were excluded because the screening guidelines for these populations can differ from a healthy gynecologic population. Other exclusion criteria included patients with incomplete records (either missing CT/NG test result or clinical note) and patients with equivocal CT/NG test results. The Johns Hopkins University Institutional Review Board determined that the study met criteria for exempt review.
Chlamydia rates in Baltimore during 2009 and 2010 were 1221.3 and 1329.1 per 100,000 population, respectively, which are more than twice the national rates of 586.7 in 2009 and 610.6 in 2010. Gonorrhea rates in Baltimore during 2009 and 2010 were 449.7 and 509.7 per 100,000 population, respectively, which are 4-5 times the national rates of 98.1 in 2009 and 100.2 in 2010. Clinic prevalence of CT and NG were obtained from reports that were submitted to the City of Baltimore. The clinic prevalence of CT and NG for all ages, which ranged from 13–69 years, was 6.4% and 1.9%, respectively. The prevalence of CT and NG for women <25 years old was 11.6% and 3.7%, respectively. Cervicovaginal secretions that were collected by the provider during pelvic examination were sent to the Hopkins Clinical Laboratory Core for analysis with a nucleic acid amplification test (NAAT; APTIMA Combo 2 Assay; Hologic Gen-Probe Incorporated, San Diego, CA). Results are available through the electronic medical record and maintained in a database within the laboratory.
Definitions
Testing was defined as NAATs performed in women who reported genitourinary symptoms (ie, irregular bleeding; urinary urgency, frequency, or dysuria; discolored or foul smelling vaginal discharge, or vaginal irritation/itching); screening was defined as NAATs performed in women without genitourinary symptoms. Abnormal speculum examination findings included cervical motion tenderness, cervical friability, and uterine tenderness. History of an STI included patients who reported ever being diagnosed with chlamydia, gonorrhea, trichomoniasis, genital herpes, or syphilis. Because the year of previous STI diagnosis was not recorded consistently in the medical records, there is a possibility that some of these infections were remote or further removed from the time of the study clinic visit.
Data analysis
CT and NG tests were grouped and analyzed as 1 unit (CT/NG). Continuous variables were summarized by mean ± standard deviation or median with corresponding interquartile range. Differences between categoric variables were evaluated with the Fisher exact test. A Student t test with unequal variance was used for evaluation of continuous variables that followed a normal distribution. Nonnormal continuous variables were analyzed with a Wilcoxon rank sum test. All reported probability values were 2-tailed, and a value of < .05 was considered significant. Associations between STI risk factors and CT/NG test positivity were assessed with logistic regression models to calculate unadjusted odds ratios (ORs), adjusted ORs (aORs) that were controlled for age, and the corresponding 95% confidence intervals (CIs). Age was selected as a covariate a priori because younger age has been shown to be associated with CT/NG. The models were not adjusted for other risk factors because of the low frequency of positive CT/NG events. Additionally, we categorized women into 2 groups based on the presence or absence of symptoms and/or reported STI exposure. This categorization allowed us to analyze the data for women with symptoms and/or STI exposure who were tested separately from those women without symptoms or women with STI exposure who were screened. Testing all women, regardless of age, is accepted generally; however, the routine screening of older women is debatable. Inconsistent or no use of barrier contraception and number of sexual partners were not included, because these variables rarely were documented in the medical record. All analyses were performed with STATA statistical software (version 12; StataCorp, College Station, TX).
Results
There were 720 nonpregnant, HIV-negative women who ranged in age from 25-69 years and who were evaluated during the 2-year timeframe of the study. Sixty-two patients were excluded because of missing documentation (either clinical note or test result), which left 658 eligible patients (91.3%). Approximately 2% of patients (13/658) had positive CT/NG test results, all of which were among black women. Specifically, the prevalence of CT and NG was 1.7% and 0.3%, respectively.
In 2009, 268 tests were performed, and 3 women tested positive (1%); in 2010, 372 tests performed, and 10 women (2.7%) tested positive. The difference between year of testing was not statistically significant ( P = .16). Thirty-three percent of women were 25-30 years old; 35% of the women were 31-40 years old; and 32% of the women were ≥41 years old ( Figure ). Younger age was associated with a positive CT/NG test ( P = .046). The oldest patients who tested positive (total of 2) were 41 years old, both of whom reported genitourinary symptoms during their clinic visits. Other demographic and clinical characteristics are presented in the Table .
| Demographic | Total (n = 658), n (%) | Chlamydia or gonorrhea result, n (%) | P value | |
|---|---|---|---|---|
| Positive (n = 13) | Negative (n = 645) | |||
| Year of clinic visit, n (%) | .16 | |||
| 2009 | 286 (43.5) | 3 (23.1) | 283 (43.9) | |
| 2010 | 372 (56.5) | 10 (76.9) | 362 (56.1) | |
| Age, y a | 34 (29–43) | 30 (28–32) | 34 (29–44) | .046 |
| Black race, n (%) | 610 (92.7) | 13 (100) | 597 (92.6) | .61 |
| Single (not married), n (%) | 592 (90.2) | 12 (92.3) | 580 (90.2) | 1.0 |
| Ever pregnant, n (%) | 620 (94.2) | 13 (100) | 607 (94.1) | 1.0 |
| Monogamous, n (%) | 643 (97.7) | 11 (84.6) | 632 (98) | .033 |
| Sexually transmitted infection exposed, n (%) | 13 (2) | 2 (15.4) | 11 (1.7) | .025 |
| Previous sexually transmitted infection, n (%) | 347 (52.7) | 5 (38.5) | 342 (53.0) | .40 |
| Symptomatic, n (%) | 190 (28.9) | 7 (53.9) | 183 (28.4) | .045 |
| Current hormonal contraception use, n (%) | 106 (16.1) | 3 (23.8) | 103 (16) | .45 |
| Menopausal, n (%) | 76 (11.6) | 0 | 76 (11.8) | .38 |
| Hysterectomy, n (%) | 24 (3.7) | 0 | 24 (3.7) | 1.0 |
| Tobacco use, n (%) | 263 (40.0) | 5 (38.5) | 258 (40.0) | 1.0 |
| Illicit drug use, n (%) | 120 (18.9) | 2 (15.4) | 118 (18.1) | 1.0 |
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
