ELECTRONIC ARTICLE |

,
,||
* Robert Wood Johnson Clinical Scholars Program, University of North Carolina at Chapel Hill
Department of Medicine, University of North Carolina at Chapel Hill, Chapel Hill, North Carolina
Department of Pediatrics, University of North Carolina at Chapel Hill, Chapel Hill, North Carolina
|| Epidemiology, University of North Carolina at Chapel Hill, Chapel Hill, North Carolina
| ABSTRACT |
|---|
|
|
|---|
Objectives. To determine CT and GC infection prevalence in 2 pediatric private practices.
Design. Cross-sectional study.
Setting. Two pediatric private practices in suburban North Carolina.
Patients. Convenience sample of patients aged 15 to 24 years who were seen from August 1998 through June 1999.
Main Outcome Measures. Prevalence of CT and GC infection.
Results. Of 1114 eligible patients, 803 (72%) completed questionnaires and provided urine specimens tested for CT and GC infection using ligase chain reaction assays. Mean age was 17.1 years (standard deviation: 1.8). Most participants were female (63%), white (87%), and from highly educated families (64% of their mothers graduated from college). Sexual activity was reported by 41%. Prevalence of CT infection in all participants was 0.9% (confidence interval [CI]: 0.4%1.8%); in sexually active participants, 2.1% (CI: 0.9%4.3%); in sexually active females, 2.7% (CI: 1.0%5.7%); and in sexually active males, 0.9% (CI: 0.0%5.1%). One case of GC infection was found.
Conclusions. The prevalence of CT and GC infection in this private practice population was much lower than reported in other settings. Screening recommendations may need to be reassessed if other low prevalence populations are found.
Key Words: Chlamydia trachomatis Neisseria gonorrhoeae pediatrics primary health care diagnosis/screening practice guidelines sexually transmitted diseases adolescence
Abbreviations: CT, Chlamydia trachomatis GC, Neisseria gonorrhoeae STD, sexually transmitted disease CDC, Centers for Disease Control and Prevention LCR, ligase chain reaction CI, confidence interval
| INTRODUCTION |
|---|
|
|
|---|
Both the US Preventive Services Task Force and the Centers for Disease Control and Prevention (CDC) recommend routine screening of sexually active female adolescents and high-risk women for CT infection.10,11 The CDC also recommends screening sexually active adolescent and young adult women for GC infection, and screening sexually active adolescent and young adult men for both GC and CT infection in settings in which the prevalence is 2% or higher.11 Despite these recommendations, compliance has been less than ideal. A survey of the screening practices of pediatricians and family medicine physicians in the region in which our study was conducted, a region with high incidence of CT and GC infection in STD and family planning clinics, showed that only 25% of physicians screened sexually active adolescent women annually.14 Reasons for not screening included low estimates of sexual activity among adolescent patients and the belief that the patient population was not at risk.15
Information on the prevalence of GC and CT infection in primary care settings is strikingly unavailable. Five studies of the prevalence of CT in primary care settings published since 1989 reported figures for women only.5,6,17,18,19 Two of the 5 combined data from primary care sites with specialty sites such as family planning and obstetrics/gynecology clinics,18,19 and only 3 of the studies collected data on women under 18 years old.5,6,19 Of those 3, only 1 provided prevalence figures for age strata under 25 years (13% in 1519 year olds and 6% in 2024-year- olds),6 and used highly selective enrollment criteria for participants under 18 years old. One study reported a GC infection prevalence of 2.3% in 15- to 44-year-old patients of private physicians offices and 3% in patients aged 15 to 19 years old from a combination of primary care and specialty sites.20 These 6 studies form the basis of current knowledge of CT and GC prevalence in primary care settings, and all reported prevalences higher than the 2% figure accepted as the cost-effectiveness threshold for CT and GC screening. None of these studies reflect CT or GC prevalence among adolescents and young adults seen in general clinic settings.
We conducted a study of the prevalence of CT and GC infection in patients of 2 pediatric private practices. Our hypothesis was that the prevalence of CT and GC in pediatric private practice settings was similar to that in other settings.
| METHODS |
|---|
|
|
|---|
Participating Practices
The study was conducted in 2 suburban primary care pediatric private practices in North Carolina. Each practice had 1 satellite office, and all physicians were general pediatricians. By report, both practices screened sexually active female patients for CT and GC infection at the time of their annual pelvic examination. Asymptomatic males were not screened.
Procedures
Between August 1998 and June 1999, research assistants were posted in the waiting rooms of the participating practices 3 to 5 days a week, Monday through Friday, during daytime business hours. All persons aged 15 through 24 years who presented during those times were invited to participate. Patients were excluded if they had taken systemic antibiotics within the previous 30 days, had already participated, or could not read and understand the consent form or questionnaire, which were written in English.
After providing written consent, each participant completed a confidential questionnaire in a private room. The survey asked reasons for the clinic visit, age, gender, self-identified race/ethnicity, highest education completed, mothers education, sexual history, and where they would seek health care if concerned about STD infection. A first-void urine specimen (the first 15 mL of urine stream) was collected from each subject, refrigerated at the site, and transported to the testing laboratory at the end of each day. Some specimens were frozen at -20°C before testing if testing could not be performed within 3 calendar days. All specimens were tested within 7 calendar days of collection and at the same laboratory. Urine specimens were tested for CT and GC DNA using the ligase chain reaction (LCR) assay (LCx, Abbott Laboratories, Abbott Park, IL). Positive test results were reported to the principal investigator, who informed the participants clinician, who contacted and treated the participant.
Data Analysis
Data entry and analyses were performed using Stata 6.0 (Stata Corp, College Station, TX).21 The results of the LCR assays, gender, race/ethnicity, and sexual history questions were analyzed as dichotomous variables using the Pearson
2 test. The Fisher exact test was used to compare CT and GC infection prevalence by gender and race. The Wilcoxon rank-sum test was used to compare prevalence by age. The preferred site of STD care by gender and sexual history was analyzed using Pearson
2 test. A 2-sided P value of less than .05 was considered to indicate statistical significance. Binomial exact methods were used to calculate 95% confidence intervals (CI). Multivariate analysis was not performed because of the low number of positive cases.
| RESULTS |
|---|
|
|
|---|
Of 1114 subjects eligible to enroll in the study, 803 (72%) consented to participate, completed the questionnaire, and provided a urine specimen. The mean age of participants was 17.1 years (standard deviation: 1.8 years). They were predominately female (63%) and white (87% white; 8% black; Table 1). More than half reported that their mothers had graduated from college, suggesting that participants were primarily from higher socioeconomic strata. A total of 331 (41%) participants reported ever "having sex." There was no difference in gender between participants and nonparticipants. Participants were slightly older than nonparticipants (17.1 years vs 16.7 years [P = .004]).
|
|
Symptomatic Versus Asymptomatic Infections
Of the 7 participants who tested positive, only 1 was seeking health care for STD symptoms (a male with urethral discharge). Of the 6 asymptomatically infected participants, 3 were women seeking routine pelvic examinations. These infections would have been detected under usual circumstances because the participating practices screened all sexually active female patients for CT and GC infection during routine annual examinations. The other 3 asymptomatically infected participants were female patients screened only because they participated in the study.
| DISCUSSION |
|---|
|
|
|---|
Potential reasons for this surprisingly low prevalence of CT infection among sexually active youth merit discussion. One potential explanation for this finding is that urine LCR assays for CT have lower sensitivity than endocervical and urethral swab tests for CT, although the prevalence we found in sexually active female participants is still lower than the prevalences that other researchers have reported using the same urine-based technology.2,4,7,8,20,24,25,28,35,37 Testing technology is not a likely explanation for the low prevalence we found in sexually active male adolescents, because urine LCR tests function as good or better than any other tests available for use with male patients.28
We did not collect data to compare sexual behaviors among sexually active participating and nonparticipating groups and therefore cannot exclude the possibility that participants were at lower STD risk than nonparticipants because of factors such as number of partners and condom use. Other possible reasons for the low prevalence we found may be socioeconomic status. Several authors have shown that poverty and low socioeconomic status are characteristics associated with increased risk of STDs, although this is less true for CT than other STD infections.1,29,30 Most of our participants came from highly educated families living in an affluent area. Although education and affluence are not biologically active characteristics for protection from STDs, they may be markers for decreased STD risk, such as delayed onset of sexual activity, use of safer sexual behaviors, limited numbers of partners, and/or consistent use of condoms. Additionally, STD risk may have been influenced by already low prevalences of CT and GC infection within the participants sexual networks.
Another possible reason for these low prevalences may relate to antibiotic use in general health care. Eighteen percent (354/1987) of patients approached to participate were excluded because they had taken antibiotics within the previous 30 days. We did not ask the reason for antibiotic use, but did request the names of the antibiotics. Of the 269 persons who provided a name, 146 (54%) had taken an antibiotic that would have effectively treated an uncomplicated CT infection.30 The high number of patients ineligible because of antibiotic use within the past month raises the question that frequent courses of antibiotics prescribed to our participants over the past many months to years for common diagnoses such as acne and respiratory infections may have resulted in unintentional treatment of asymptomatic CT infections.
The low prevalence of CT infection we found in female adolescents also may be the result of successful screening programs. Participating practices reported routine annual screening of sexually active female patients, and although we did not verify screening practices used before the study, the success of these screening programs may have reduced prevalences to the low levels we found. Other studies have shown significant declines in prevalence when screening programs have been consistently implemented.5,8,16,22,23,24,25 It is also possible that study participants may have gone to other sites for reproductive health care (eg, Planned Parenthood, family planning, and STD/health department clinics) and participated in effective screening programs at these sites, as suggested by our finding that 50% of sexually active participants preferred alternative sites of care for STD concerns (data not shown). This possibility would not explain the low prevalence we found in males because, to our knowledge, there was no routine screening program for male adolescents in the region in which our study was performed. Other limitations include that our sample was small, regional, and relatively homogeneous. Our 28% refusal rate is similar to other studies of STDs in adolescents.24 Regardless of the explanation for the low prevalence of CT infection, the fact that it was lower than expected calls for more research because of potential screening implications. Although routine universal screening of sexually active adolescents and young adults has been shown to have significant benefits,5,8,16,22,23,24,25 it is expensive and not without potential harms.28 The proportion of false-positive tests increases as prevalence falls. The LCR assay used in this study is believed to have excellent performance characteristics with a sensitivity in excess of 90% and specificity greater than 97%.28,38,39 If we assume that the LCR assay for CT infection has a sensitivity of 91% and specificity of 99.5% in a population with 2% prevalence, approximately 29% of positive specimens are false positives. If test specificity is <99.5%, this number would be even greater. False positives may result in substantial psychological stress, have negative impacts on romantic relationships, cause unnecessary stigmatization, and be especially significant for adolescents who find it difficult or even dangerous to inform partners (and perhaps parents) of their infections.
| CONCLUSION |
|---|
|
|
|---|
| ACKNOWLEDGMENTS |
|---|
We thank the patients and staff of Chapel Hill Pediatrics, PA, Chapel Hill, North Carolina, and Raleigh Pediatrics and Associates, PA, Raleigh, North Carolina, for their participation in the study. We also thank John L. Schmitz, PhD, and the staff of the McLendon Clinical Laboratories at the University of North Carolina for performing the LCR assays, and David F. Ransohoff, MD, for helpful review of this manuscript.
| FOOTNOTES |
|---|
Address correspondence to Dana Best, MD, MPH, General Pediatrics and Adolescent Medicine, Childrens National Medical Center, 111 Michigan Ave, NW, Washington, DC, 20010-2970. E-mail: dbbest@cnmc.org
| REFERENCES |
|---|
|
|
|---|
This article has been cited by other articles:
![]() |
Screening for Chlamydia and Gonorrhea in Private Practice Journal Watch (General), December 11, 2001; 2001(1211): 4 - 4. [Full Text] |
||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||