Chlamydia screening in a metropolitan Atlanta primary care clinic.

Background: Chlamydia is the most commonly reported infectious disease in the United States, yet many women at risk for chlamydia infection are not being tested.

Methods: We conducted a chlamydia screening study (Phase 1) and retrospective medical chart review (Phase II) of 199 women from 16 to 40 years of age attending a primary care clinic in metropolitan Atlanta.

Results: Two (1%) of the 199 study participants tested positive for chlamydia during Phase 1. Phase II medical chart reviews indicated that only 35% of study participants had been tested for chlamydia at least once in the previous 15 months. Three of these individuals tested positive, yielding a prevalence of 5.6%. All infections occurred in women 30 years of age or younger.

Conclusion: The results suggest that chlamydia screening is low and that recommended guidelines are not being followed. Strategies are needed to increase the rates of screening and adherence to these guidelines to ensure that chlamydia is detected before complications occur.

Key Words: chlamydia, screening, sexually transmitted disease 

**********

The most frequently reported infectious disease in the United States is caused by the Chlamydia trachomatis bacterium. (1) Each year an estimated 4 million new cases are diagnosed and over 2 billion dollars are spent treating chlamydia and chlamydia-associated conditions. (2-6) Failure to detect and treat chlamydia may lead to pelvic inflammatory disease, ectopic pregnancy, infertility, and an increased susceptibility to human immunodeficiency virus (HIV). (1,4,6)

Approximately 75% of infected women are asymptomatic and do not realize they are infected, (6) making routine testing the most, if not the only, effective way of identifying a high percentage of chlamydia cases before complications occur. Screening programs in asymptomatic populations have been shown to reduce the occurrence of chlamydial infections, particularly in areas where programs have been in place for several years. (1,4) In both public and private settings, however, many women at risk for chlamydia are not regularly screened, (4,7) often because of missed opportunities. A recent study found that only one third of surveyed physicians responded that they would test asymptomatic sexually active teenage women for chlamydia during a routine gynecologic examination. (7) Another study found that only 16% of female patients and 1.5% of male patients aged 12 to 19 years enrolled in a managed care organization (MCO) had been tested for chlamydia during a 2-year period. (8)

Chlamydia testing rates and reporting are likely to expand with the recent adoption of chlamydia screening as a performance measure for the Health Plan Employer Data and Information Set (HEDIS), a widely used set of measures to assess the quality of managed care performance. (9) HEDIS 200l, which reports on managed care performance in calendar year 2000, includes a combined risk-age screening measurement to assess the percentage of sexually active women ages 16 to 26 who were continuously enrolled in the MCO during the measurement year and were tested at least once for chlamydia. National HEDIS 2001 data indicates that 24% of sexually active females ages 16 to 20, and 21% of sexually active females ages 21 to 26, were tested for chlamydia. (10)

The Centers for Disease Control and Prevention (CDC) recommends annual chlamydia screening of sexually active females less than 20 years of age and annual screening for women 20 years of age or older with one or more risk factors (eg, lack of barrier contraception or new and/or multiple sex partners). (5) The CDC has also suggested that public health programs may obtain comparable sensitivity and test a similar proportion of female clinic patients by screening all sexually active women younger than 30 years of age. (4)

The objectives of this study were to estimate the prevalence of chlamydia infection in women ages 16 to 40 years of age attending a primary care clinic (Phase I) and to determine chlamydia-screening practices at the clinic via a retrospective medical chart review (Phase II). This study was approved by the Institutional Review Board of The Prudential Center for Health Care Research (predecessor of the USQA Center for Health Care Research).

Methods

Phase I: Screening Study

During Phase 1, we conducted a chlamydia screening study in the obstetrics and gynecology (OB/GYN) unit of a primary care clinic in metropolitan Atlanta. The clinic served patients with all types of health insurance coverage, including managed care. Screening took place during regular clinic hours 2 days of each week in February and March 1999, and all women ages 16 to 40 years were eligible to participate. At the beginning of each screening day, nurses received a computerized list of patients they would see that day; the list included each patient's name, date of birth, and medical identification number. Research assistants identified eligible patients on each nurse's list based on the patient's date of birth, and nurses informed all eligible women about the study and invited them to participate. Patients did not receive an incentive if they participated in the study.

Specimen Collection and Analysis

Urine-based Ligase Chain Reaction (LCR) was used for all tests. During the patient visit, the physician obtained verbal informed consent from the patient and noted this in the patient's medical chart. A first-void urine specimen (15-20 ml) was collected from each participant and labeled with the collection date, the patient's date of birth, and study ID number. No additional information was collected about the participants during the screening phase of the study. Urine specimens were cooled to 2 to 8[degrees]C within 10 minutes of collection and the temperature was checked hourly using a digital laboratory thermometer (accuracy of [+ or -] 1[degrees]C) during collection hours. Specimens were stored for up to 2 days in a temperature-monitored refrigerator before transport to the University of Alabama at Birmingham Infectious Diseases Laboratory.

Specimens were processed and analyzed according to the instructions of the manufacturer (Abbott Diagnostics, Abbott Park, IL). In brief, 1 ml urine was pipetted in a 1.5-ml tube and centrifuged for 15 minutes. Urine pellets were frozen at -20[degrees]C, subsequently resuspended in 1 ml urine resuspension buffer, and the tubes were placed in a heating block (95-100[degrees]C) for 15 minutes. The samples were tested using the LCx test with 100 ml processed urine transferred to individual LCx unit-dose tubes containing 100 ml LCx assay mixture for amplification (K Smith, Division of Infectious Diseases Laboratory, Department of Medicine, University of Alabama at Birmingham, personal communication, September 1999).

Study participants were informed of their test results within 7 to 10 days. Patients with a negative result were notified confidentially by mail, and clinic staff telephoned patients with a positive result and a follow-up appointment was made at that time. Women with a positive result were treated at the clinic with the pharmaceutical regimen of their physician's choice.

Phase II: Chart Review

Within 60 days of Phase I, we conducted a retrospective review of the study participants' medical charts to collect information on the presence of chlamydia-related conditions (defined for the purposes of the study as pelvic pain, infertility, vaginal discharge, dysuria, dyspareunia, ectopic pregnancy), obstetric status, and sexual activity. Date of birth had been obtained on all participants during Phase I. In addition, we documented the frequency of chlamydia screening (ie, routine screening in asymptomatic patients) and chlamydia testing (ie, diagnostic testing in patients presenting with chlamydia symptoms) in the 15 months before Phase I. If a chlamydia-related condition (eg, ectopic pregnancy or pelvic pain) was documented during the office visit, we considered it to be diagnostic chlamydia testing; if the patient was pregnant and the test was administered during an obstetric workup, the test was considered routine obstetric screening; lastly, the test was considered routine screening if the medical chart did not indicate obstetric-related screening or a chlamydia-related condition. All chlamydia tests at the clinic were performed using nonamplified probe assays (Gen-Probe, Inc., San Diego, CA).

Analysis

The prevalence of chlamydia was estimated by two measures: 1) the number of positive cases identified during Phase I (screening phase) over the total number of study participants and 2) the number of positive cases identified during Phase II (retrospective chart review) over the total number of participants tested for chlamydia. We examined the reason for previous chlamydia testing (diagnostic, routine screening, and obstetric testing) and the characteristics of women previously tested and those not previously tested. Age stratification analyses were performed to assess the value of using an age-based screening strategy.

Results

Phase I

Of the 298 women approached about the study, 199 (67%) agreed to participate (Fig. 1). Two (1.0%) of the 199 study participants tested positive for chlamydia during Phase I.

[FIGURE 1 OMITTED]

Phase II

We reviewed the medical charts o1" 152 (76%) of the 199 screening study participants. The remaining 47 medical charts were not reviewed either because a written waiver allowing a review of their chart had not been obtained (n = 35) or the participant was new to the clinic and had not been available for previous testing (n = 12).

Of the 152 participants whose charts were reviewed, 98 (64%) had not been tested for chlamydia in the 15 months before the screening phase of the study. The remaining 54 (35%) study participants had been tested at least once during that time and, of those, three (5.6%) tested positive for chlamydia. Two of the positive cases were identified during routine screening, (ie, medical records did not indicate a chlamydia-related condition) and one tested positive during diagnostic testing (ie, medical records indicated a chlamydia-related condition). When those previously tested were stratified by age, 52% of women ages 16 to 26 had been previously tested, compared with 35% overall.

Table 1 shows the age and clinical characteristics of study participants tested during Phase I and those not previously tested. There were no statistically significant differences between the two groups. The age of study participants and of those who tested positive for chlamydia both during Phases I and II are summarized in Table 2. All infections occurred in women 30 years of age or younger.

Discussion

The prevalence of chlamydia in our study population was 1% during the screening phase (Phase I) and 5.6% tested positive during the previous 15 months (Phase II). In this study, 52% of women ages 16 to 26, and 35% in all age groups, were tested for chlamydia over a 15-month period. This screening rate is much higher than national HEDIS 2001 data, which reported an average 22.5% screening rate for women ages 16 to 26. Our higher rate may be partially explained by the fact that HEDIS data were based on a combination of administrative data and chart review, while our rates are based solely on chart review. In addition, HEDIS reviews a 12-month period, while our rate is based on a 15-month period. Also we did not collect information on sexual activity.

There may have been differences between the women who agreed to participate in the study and those who were approached about the study but did not participate, reflecting a more proactive approach to health care by the group of women who agreed to participate in the study compared with those who did not. A major limitation of determining screening rates in this study population is the small sample size. In addition, 47 of 199 medical charts were not available for review; thus, we were not able to assess previous testing for our total study population.

In this study, all infections occurred in women age 30 or younger. This finding is consistent with previous studies, which report that chlamydia infection is highest among younger women and that selected screening missed only a small percentage of infections, (5,9,11) suggesting that age-based screening may be a way to target the highest prevalence subgroups in a low prevalence setting.

We found urine-based LCR easy to use in a busy clinic environment, particularly because providers routinely collect a urine sample during OB/GYN visits, so the additional time burden of specimen collection was minimal. Traditionally, chlamydia screening has required an endocervical culture, a potentially time-consuming procedure for providers, particularly in widespread screening programs, and invasive for patients. Although there is no difference between the specificity of LCR and culture (both are approximately 100% specific), LCR is approximately 94% sensitive, in contrast to endocervical culture which has a sensitivity of 70 to 85%. (12) In addition, urine-based LCR tests are able to identify the 5 to 30% of women with chlamydia who are infected in the urethra only and would not be diagnosed by an endocervical culture. (12) LCR can also be used with cervical specimens, maintaining the benefit of enhanced LCR sensitivity. A potential disadvantage of urine-based DNA amplification screening is that women who might need a pelvic examination may not receive one at the time of screening. Rather than replace pelvic examinations, providers should use urine-based screening as an opportunity to screen both those receiving a pelvic examination and those who may need more frequent screening.

DNA amplification testing is more expensive than traditional screening methods. When the high cost of chlamydia-associated sequelae is considered, however, DNA amplification may be a cost-saving method because of its greater sensitivity. In a cost-benefit analysis, Howell et al (13) found DNA amplification to be more cost-effective than other analytic techniques, even at prevalence levels of 3.15%. On the basis of Howell et al's criteria, DNA amplification in our study would not have been considered cost-effective for universal screening but may have been cost-effective for screening of women younger than 30 years of age. Additional research is needed to determine whether routine screening is cost-effective for younger women in a primary care setting.

Conclusions

The results suggest that chlamydia screening is low and that recommended guidelines are not always followed. Strategies are needed to increase rates to ensure that chlamydia is detected before complications occur. Given the value of chlamydia testing to patient health, it is beneficial to find reliable and cost-effective ways to screen for chlamydia. High sensitivity, urine-based testing may prove a valuable time and cost-saving tool in facilitating widespread testing programs with the objective of identifying chlamydia infections before complications set in. Our experience is consistent with research suggesting that selective screening of younger women may be a sensible approach. Age-based screening or other criteria may be useful to promote targeted screening of the highest-prevalence subgroups within low-prevalence populations.

Table 1. Characteristics of stud' participants tested in
the 15 months before Phase I and those not tested

                                       No. (%)

                           Patients tested    Patients not
                            in 15 months      tested in 15
                               before         months before
                               Phase I           Phase I
Characteristic              (n = 54) (a)      (n = 98) (a)

Age (yr)
  16-20                        9 (17%)           2 (2%)
  21-25                        8 (15%)          14 (14%)
  26-30                       19 (35%)          24 (24%)
  31-35                       12 (22%)          35 (36%)
  36-40                        6 (11%)          23 (23%)
Chlamydia-related
    conditions
  Yes                          9 (17%)          12 (12%)
  None noted in medical       45 (83%)          85 (87%)
    chart
  Unknown                      0 (0%)            1 (1%)
Obstetric status
  Patient                     17 (31%)          18 18%)
  Nonpatient                  36 (66%)          80 (82%)
  Unknown                      1 (2%)            0 (0%)
Sexually actives (b)
  Yes                         25 (46%)          54 (55%)
  Not noted in medical        29 (54%)          45 (45%)
    chart

(a) No statistical differences between those tested in the 15 months
prior to Phase I and those not previously tested.

(b) Patient was considered sexually active if documented in physician
notes, if actively using barrier/oral contraceptives, or physician
notes document that patient attempting pregnancy.

Table 2. Chlamydia infection by age of study
participant (n = 199)

Age      No. in age    No. participants with
(yr)       group         positive test (a)

16-20        17               1 (0.5)
21-25        30               2 (1.0)
26-30        57               2 (1.0)
31-35        57               0 (0.0)
36-40        33               0 (0.0)
Total       199               5 (2.5)

(a) Includes data from Phases I and II.

Key Points

* The most frequently reported infectious disease in the United States is from the Chlamydia trachomatis bacterium, yet many women at risk for chlamydia are not regularly tested.

* Up to 75% of women with chlamydia are asymptomatic and do not realize that they are infected, making routine testing the most effective way of identifying a high percentage of cases before complications occur.

* Practical strategies to increase testing rates are needed to ensure that chlamydia is detected before complications occur.

Acknowledgments

We thank Drs. Arnold Cohen and Joanne Armstrong for reviewing earlier versions of the manuscript and the staff at the study clinic location for their generous support throughout this project.

References

(1.) Centers for Disease Control and Prevention. Sexually Transmitted Disease Surveillance, 1997. Atlanta, GA, U.S. Department of Health and Human Services, Division of STD Prevention.

(2.) Washington AE, Katz P. Cost of and payment source for pelvic inflammatory disease: Trends and projections, 1983 through 2000. JAMA 1991; 266:2565-2569.

(3.) Institute of Medicine. The Hidden Epidemic: Confronting Sexually Transmitted Diseases. Washington, DC, National Academy Press, 1996.

(4.) Centers for Disease Control and Prevention. Chlamydia trachomatis genital infections, United States 1995. MMWR Morb Mortal Wkly Rep 1997;46:193-198.

(5.) Centers for Disease Control and Prevention. Recommendations for the prevention and management of Chlamydia trachomatis infections, 1993. MMWR Recomm Rep 1993;42(RR-12):1-35.

(6.) Centers for Disease Control and Prevention. Chlamydia in the United States. Atlanta, GA, U.S. Department of Health and Human Services, National Center for HIV, STD and TB Prevention. Available at: http://www.cdc.gov/ nchstp/dstd/Fact/Sheets/chlamydia_facts.htm. Accessed July 11, 2003.

(7.) Cook RL, Wiesenfeld HC, Ashton MR, et al. Barriers to screening sexually active adolescent women for chlamydia: A survey of primary care physicians. J Adolesc Health 2001;28:204-210.

(8.) Burstein GL, Snyder MH, Conley D, et al. Adolescent chlamydia testing practices and diagnosed infections in a large managed care organization. Sea Transm Dis 2001;28:477-483.

(9.) Mangione-Smith R O'Leary J, McGlynn EA. Health and cost benefits of chlamydia screening in young women. Sex Transm Dis 1999;26:309-316.

(10.) National Committee for Quality Assurance. The State of Managed Care Quality, 2001: Chlamydia Screening in Women. Available at: http:// www.ncqa.org/somc2001/chlamydia/somc_2001_chl.html. Accessed May 2, 2003.

(11.) Masse R, Laperriere H, Rousseau H, et al. Chlamydia trachomatis cervical infection: Prevalence and determinants among women presenting for routine gynecologic examination. CMAJ 1991;145:953-961.

(12.) Black CM. Current methods of laboratory diagnosis of Chlamydia trachomatis infections. Clin Microbiol Rev 1997;10:160-184.

(13.) Howell MR. Quinn TC, Brathwaite W, et al. Screening women for Chlamydia trachomatis in family planning clinics: The cost-effectiveness of DNA amplifications assays. Sex Transm Dis 1998;25;108-117.

From the Emory Center on Health Outcomes and Quality, Atlanta, Rollins School of Public Health, Atlanta, GA (forrncrly The Prudential Center for Health Care Research[R], which became the USQA Center for Health Care Research[TM] and in 2001 was transferred to Emory to form the Emory Center on Health Outcomes and Quality. (Ms. McPhillips-Tangum is currently with CMT Consulting, Decatur, GA; Dr. Franks is currently with the Division of Prevention Research and Analytic Methods, Centers for Disease Control and Prevention, Atlanta, GA.)

We certify that none of the authors had affiliations or involvement with any organization or entity with a direct financial interest in the subject matter discussed in this article. We have no potential conflicts of interest to disclose. This study was funded in part by Aetna, Inc. The opinions expressed and conclusions reached are solely those of the authors and do not necessarily represent those of Aetna.

Reprint requests to Natalia V. Oster, MPH, Department of Health Policy and Management, Emory Center on Health Outcomes and Quality, Rollins School of Public Health, Emory University School of Medicine, 1518 Clifton Road, 6th floor, Atlanta, GA 30322. Email: [e-mail]

Accepted September 17, 2002.

Copyright [c] 2003 by The Southern Medical Association

0038-4348/03/9609-0863