Newman, Lori M. MD*; Warner, Lee PhD, MPH†; Weinstock, Hillard S. MD, MPH*
IN 2004, NONVIRAL SEXUALLY TRANSMITTED DISEASES (STDs) accounted for 3 of the 5 most frequently reported diseases in the United States.1 The annual direct and indirect cost of 5 major nonviral STDs has been estimated at approximately $7 billion.2 Although costs associated with all STDs are shared by many different sectors, publicly funded STD clinics account for a large proportion of reported STDs and their associated costs in the United States, including approximately half of syphilis cases and one third of gonorrhea cases.3
Although testing, diagnosis, and treatment are among the primary functions of publicly funded STD clinics, the clinic visit also represents a unique epidemiologic opportunity for identifying patients at increased risk for acquiring subsequent STD. Previous research suggests patients attending an STD clinic most likely to be diagnosed with subsequent infection include those who are younger, are of minority race/ethnicity, and have been diagnosed with an STD at an earlier visit.4–15 These studies of STD incidence have largely been conducted in a single locale5–8,11,17,14,15 or in controlled research environments with active STD screening and follow-up.6,7,11,16 There have been fewer multisite analyses of STD incidence and predictors of STD incidence from longitudinal, observational clinic studies that use routinely collected clinic data.10
We examined routinely collected demographic, behavioral, and clinical information from a multisite retrospective cohort study of patients attending an STD clinic17 to assess the frequency of nonviral STD diagnoses in patients attending an STD clinic and to identify characteristics of patients most likely to be diagnosed with a nonviral STD at a subsequent visit. In this analysis, we illustrate the potential for STD clinics to use initial STD diagnosis and other routine patient information to target STD prevention messages and programs to a patient population at increased risk of returning with an STD at a subsequent visit.
Materials and Methods
We analyzed secondary data from a retrospective cohort study of HIV incidence among patients attending STD clinics in 7 cities between 1991 and 1999. The methods, briefly described here, have been described in detail elsewhere.17 For the present analysis, patient medical records were reviewed for clinic sites that participated during the last 3 years of the study (1997–1999). The analysis involved the records of patients seeking care for any reason other than follow-up visit at 5 publicly funded STD clinics in 4 cities (Atlanta, Denver, Houston [2 clinic sites], and New Orleans) from January 1997 through December 1998. To ensure that follow up was uniform for all patients attending clinics during this period, patient records were reviewed both at the initial clinic visit and during the subsequent 12-month period to ascertain frequency of STD diagnoses. Demographic data, risk behaviors, and clinical and laboratory information were abstracted by study staff from routinely collected records of all clinic visits for these patients occurring during the 12-month period after the initial clinic visit.
Initial STD was defined as diagnosis of at least one nonviral STD at the time of the patient’s first clinic visit in 1997 or 1998 or within 1 month of the first visit. STD diagnoses occurring less than 30 days after the initial visit were considered to be initial STDs for the purposes of this analysis, because they may have reflected infections diagnosed at the time of the first visit. STD diagnoses at a subsequent visit were defined as at least one nonviral STD diagnosis from 1 to 12 months after the initial visit.
The nonviral STD diagnoses included in this analysis were: chlamydia, gonorrhea, mucopurulent cervicitis, nongonococcal urethritis (NGU), pelvic inflammatory disease, primary or secondary syphilis, and trichomoniasis. Although diagnoses of viral STDs (e.g., genital warts, genital herpes, and HIV) as well as other STDs (early or late-latent syphilis, syphilis of unknown duration, or genital ulcer disease) also may be markers of subsequent sexual risk and influence infection with nonviral STDs, these were not included as a result of difficulties distinguishing subsequent from initial infection. All clinical diagnoses in this analysis were extracted as recorded by clinicians in the patient record.
All patients were followed for 12 months after their initial visit. Records for clinic visits throughout this period were examined for the presence of nonviral STD diagnoses. Patients could be diagnosed with multiple STDs at a single visit or could have multiple diagnoses at multiple visits. Bivariate analysis was conducted to evaluate association of STD at a subsequent visit with selected demographic, behavioral, and clinical factors. Multivariable logistic regression was conducted to assess factors independently associated with an STD at a subsequent visit separately for men and women to account for differences associated with sex. All models included characteristics found to be significant in bivariate analyses as well as characteristics previously demonstrated to be important predictors of STD. Results were considered statistically significant at P ≤0.05 and all statistical tests were 2-tailed.
Altogether, 64,463 patients attended participating STD clinics at least once during 1997 or 1998 (range by city = 11,411–22,416). Overall, 59% were men (94% of whom were heterosexual), 46% were 20 to 29 years old, 70% were black, and 60% had symptoms of a possible STD at the initial visit (Table 1). Overall, 21,848 patients (34%) were diagnosed with one or more initial STDs (range of prevalence by city = 30–38%, data not shown). For men, overall initial STD prevalence was 36.9%, with the most common diagnoses being gonorrhea (17.0%), NGU (15.7%), and chlamydia (6.7%). For women, overall initial STD prevalence was 29.6%, with chlamydia (10.8%), gonorrhea (9.2%), and trichomoniasis (7.8%) being the most commonly diagnosed STDs. In the 12-month period after their initial visit, 4518 patients (7.0%) were diagnosed with a subsequent STD at one or more clinic visits (range by city = 5.2–8.3%). Like initial STD, overall subsequent STD was higher among males than females (7.7% vs. 6.1%, P <0.01).
In bivariate analysis, being diagnosed with an initial STD was strongly associated with being diagnosed with an STD at a subsequent visit. When considering all nonviral STDs, the probability of subsequent STD diagnoses was significantly higher among patients with an initial STD diagnosis compared with those without, both overall (12.0% vs. 4.4%, P <0.01, data not shown) and by sex (13.7% vs. 4.1% for males [P <0.01]; 9.0% vs. 4.8% for females [P <0.01]).
Similar associations between initial and subsequent STDs were also found in both sexes for individual STDs (Table 2). For example, males diagnosed with gonorrhea at their initial visit were more than twice as likely to have diagnoses of gonorrhea at a subsequent visit compared with men diagnosed with chlamydia or NGU initially or men without STD initially (10.3% vs. 4.3% and 1.6%, respectively). Similarly, men diagnosed with chlamydia or NGU initially were more likely to have diagnoses of chlamydia or NGU at a subsequent visit compared with men diagnosed with gonorrhea initially or men without an STD initially (10.2% vs. 5.6% and 2.8%, respectively). Approximately two-thirds of the men seen for gonorrhea, chlamydia, or NGU at a subsequent visit were men who had been diagnosed with these respective infections at the initial visit.
Similar associations between the type of initial and subsequent STD diagnoses were found among women. Women diagnosed with gonorrhea initially, for example, were more likely to have gonorrhea at a subsequent visit compared with women with chlamydia initially or women without an initial STD diagnosis (5.0% vs. 3.6% and 1.3%, respectively). Women diagnosed with chlamydia initially were likewise more likely to have a diagnosis of chlamydia at a subsequent visit compared with women with gonorrhea initially or women without an initial STD diagnosis (4.8% vs. 3.6% and 1.6%, respectively). In contrast with men, however, only approximately one-third of the women seen for a diagnosis of gonorrhea or chlamydia at a subsequent visit were women who were diagnosed with these infections at their initial visit.
Overall, 1257 of 4518 patients (27.8%) with an STD at a subsequent visit in the 12-month study interval came back by the third month, 57.8% by the sixth month, and 81.8% by the ninth month (data not shown). At each monthly interval, the number of patients diagnosed with an STD at a subsequent visit was greater among patients who had an initial STD compared with patients who did not despite patients with an initial STD representing only approximately one-third of the population (Fig. 1). This difference was most marked in the second and third months after the initial visit, when patients with initial STDs accounted for 84% and 63% more STD-associated visits than patients without initial STDs and decreased over time. In the 11th or 12th month after the initial visit, for example, patients with initial STDs accounted for only 18% and 26% more such visits than patients without initial STDs, respectively.
The proportion of patients with a subsequent STD also decreased with increasing patient age (Fig. 2) regardless of whether patients were diagnosed with an STD at their initial visit. For example, among patients less than age 20, 9.8% were diagnosed with an STD at a subsequent visit overall (13.8% of those with an initial STD; 6.8% of those without) in contrast to 4.1% overall among those aged 40 and over (9.6% of those with an initial STD and 2.4% of those without). Across all age groups, the proportion of patients diagnosed with an STD at a subsequent visit was significantly higher among patients with an initial STD than among patients without.
Multivariable regression analyses revealed several factors to be significantly associated with an STD at a subsequent visit. Among men, subsequent STD was significantly higher among patients who were black race/ethnicity or had an initial STD (Table 3). Other characteristics that were significantly, albeit more moderately, associated with a subsequent STD among men included: being symptomatic at the initial visit, reporting exchange of sex for money or drugs, reporting sex with men, and being less than age 20. Report of injection drug use was not found to be a significant predictor of subsequent STD.
Among women, multivariable analyses revealed that report of exchange of sex for drugs or money had the strongest, statistically significant association with STD at a subsequent visit (Table 3). Other factors significantly associated with subsequent STD included initial STD, being less than age 20, black race/ethnicity, and being symptomatic at the initial visit. Report of injection drug use was of borderline significance.
The data from this analysis were then used to examine the theoretical impact of using characteristics associated with an STD at a subsequent visit as selection criteria for an STD prevention intervention. For example, if a clinic had the resources to provide intensive STD risk counseling for all patients (100%), the clinic could avert the greatest proportion possible of subsequent STDs (100%) (Table 4). However, because in most settings, intensive counseling of all patients is not feasible, a clinic might provide intensive counseling for only patients under age 20 (16% of the population), which would have the potential to influence 22% of the population who had an STD at a subsequent visit. Table 4 outlines other strategies that involve intermediate proportions of a clinic population. For example, targeting persons with an initial STD (34% of the population) could potentially influence 58% of the population diagnosed with a subsequent STD.
Our analysis of STD clinic patient records in 4 urban cities found that approximately one in 3 patients were diagnosed with a nonviral STD at the initial clinic visit. Moreover, approximately one in 14 of these patients were diagnosed with an STD within the subsequent 12-month period. Although these figures likely underestimate the actual STD incidence resulting from limitations discussed below, this large number of patients returning with a new infection within a relatively short time period clearly represents a population for targeting limited STD prevention resources.
Our analysis suggests that STD clinics can use initial diagnosis and other routine patient information to help identify individuals with increased likelihood of returning to the clinic with a subsequent STD. Consistent with other studies, we found that patients who were younger5,9,17,14,15,18 or diagnosed with an STD at their initial clinic visit6,8–10,12,13,15,16 were approximately twice as likely to be diagnosed with an STD at a subsequent visit as compared with their counterparts.
Publicly funded STD clinics provide a critical resource for providing prevention services to patients at increased risk for STD acquisition. Although a full range of prevention services ideally would be provided during a patient’s initial clinic visit, it may not be feasible or practical for high-volume STD clinics to provide this level of service for all patients at the first visit. Therefore, it is important that STD clinics be able to prioritize how and to whom in-depth prevention activities could be provided.
STD clinics with adequate resources for providing prevention services to all patients could similarly benefit from identifying subpopulations that may require more intensive resources than others.8,12 For example, Gunn et al outlined criteria for identifying clients for 3 different levels of prevention and risk-reduction counseling based on recent STD history.8 The data in this analysis similarly support providing basic STD health education to all patients and also more intensive individual risk-reduction counseling to patients at highest risk for being diagnosed with an STD at a subsequent visit (e.g., patients with an initial STD). Other potential interventions may include screening for comorbid conditions such as substance use or mental health issues and making referrals, expedited partner therapy, or recommending repeat STD testing or counseling visits.
This article provides a practical example of how selection of different patient criteria yields intervention opportunities with varying numbers of patients. Depending on resources available, intervention criteria could be selected to maximize intervention impact. Although we selected initial STD and age in our example, investigators could conduct similar exercises with other characteristics found to be important, feasible, and acceptable criteria locally (such as sexual orientation or exchange of sex).
Our analyses of the timing of subsequent infections indicated that patients return to the clinic with subsequent infection throughout the next 12-month period. Moreover, at any point during this time period, patients with an initial nonviral STD were more likely than those without to have a nonviral STD at a subsequent visit. This finding suggests the need for repeat testing of patients who have been diagnosed with an initial nonviral STD. Because the subsequent STDs in this retrospective cohort study likely represent primarily symptomatic patients, a repeat testing visit could also identify subsequent asymptomatic infections that are not typically diagnosed through passive follow-up.
This analysis could not determine if subsequent infections represented persistent infection (because of patient failure to take recommended therapy or infection with a resistant organism), reinfection with the same organism from an untreated partner, or new infection from a new partner through high-risk sexual activity. Examination of the disease-specific data in this analysis showed that male patients diagnosed with gonorrhea at the initial visit had almost a 10-fold greater chance of returning with gonorrhea than male patients without an initial STD diagnosis. However, male patients with gonorrhea at the initial visit also had a 2-fold greater chance of returning with chlamydia than male patients without an initial STD diagnosis. A similar situation was observed in patients with chlamydia at the initial visit. These data suggest that subsequent infections were not merely the result of persistent infection or reinfection of the same organism and more likely represented continued participation in subsequent high-risk sexual behavior.
Our findings are generally consistent with those reported from other studies despite differing methodologies. For example, Gunn et al found a 8% subsequent STD (bacterial and first episode viral) infection rate in a study in San Diego of 2576 patients attending an STD clinic followed passively for 12 months.8 Fortenberry et al found that 41% of 236 adolescents attending the county STD clinic or community-based adolescent health clinics were diagnosed with an STD during scheduled follow-up visits within 12 months, a figure higher than the 10% reported in this study for individuals under age 20 returning without scheduled follow-up.6 Rietmeijer et al also found in a retrospective cohort study of patients attending a Denver STD clinic that incident chlamydial infection was diagnosed more than twice as often among patients diagnosed with chlamydia at baseline compared with those who were not (20.2% vs. 9.3%).14 Richert et al reported that the highest risk of returning with an STD was among patients diagnosed with an STD at their initial visit.12
Our findings are subject to some limitations. First, we suspect that our estimate of the 1-year incidence of nonviral STD, although substantial, is a minimum estimate of the true STD incidence. Although we were able to identify STDs diagnosed at participating clinics, we were unable to ascertain STDs that may have been diagnosed in settings other than the participating STD clinic, including the private sector. Also, we were unable to ascertain subsequent STDs occurring among patients who did not return for care on their own (e.g., patients with asymptomatic infection). Recent data from Project Respect, a randomized, controlled trial of HIV counseling in patients attending an STD clinic in which patients were routinely tested at 3, 6, 9, and 12 months, estimated that 69% of new infections were asymptomatic.19 Viral STDs diagnosed after the initial visit also were not included in this analysis because we could not distinguish with certainty that these represented subsequent versus initial infections.
Second, because these data were from an observational clinic study that used routinely collected clinic data, our findings are subject to other limitations. For example, diagnostic criteria and laboratory techniques as well as ascertainment of demographic and behavioral factors may have varied across sites. Documentation of prevention interventions at the time of the clinic visit was not available and may have varied among patients and across sites. Furthermore, because this analysis was a convenience sample of public STD clinics in large urban areas, our findings may have limited generalizability to other clinical settings.
These limitations are largely offset by a number of strengths of this study. First, unlike previous studies, this analysis of subsequent STD includes a large number of patients from multiple geographic locations, with 2 cities (Atlanta and New Orleans) having gonorrhea, chlamydia, and syphilis rates consistently ranking among the top 10 cities in the country and 2 cities (Denver and Houston) having STD rates closer to the U.S. average rates.20 Second, this analysis was able to evaluate data for multiple nonviral STD diagnoses and provide a more complete picture of the STD burden experienced by these clinics. Third, this analysis importantly reflects the actual behavior of patients attending STD clinics rather than patients in an artificial study intervention setting.
Our analyses, consistent with those published previously, demonstrate that a substantial number of patients seen in STD clinics return with subsequent STD infections despite previously having been seen at an STD clinic. This suggests that opportunities for STD clinics to intervene with individuals at high risk of acquiring repeat STDs may not have been fully exploited. STD clinics could use initial STD diagnosis and other readily available and routinely collected patient information to prioritize interventions for patients at greatest risk of returning with a subsequent STD or to encourage such patients to follow up for repeat testing.
1. Centers for Disease Control and Prevention. Notice to readers: Final 2004 reports of notifiable diseases. Morb Mortal Wkly Rep 2005; 54:770.
2. Institute of Medicine. The Hidden Epidemic: Confronting Sexually Transmitted Disease. Washington, DC: National Academy Press, 1997.
3. Centers for Disease Control and Prevention. Sexually transmitted disease surveillance, 2004. Atlanta: US Department of Health and Human Services, September 2005.
4. Burstein GR, Gaydos CA, Diener-West M, et al. Incident Chlamydia trachomatis
infections among inner-city adolescent females. Sex Transm Infect 2001; 77:26–32.
5. Ellen JM, Hessol NA, Kohn RP, et al. An investigation of geographic clustering of repeat cases of gonorrhea and chlamydial infection in San Francisco, 1989–1993: Evidence for core groups. J Infect Dis 1997; 175:1519–1522.
6. Fortenberry JD, Brizendine EJ, Katz BP, et al. Subsequent sexually transmitted infections among adolescent women with genital infection due to Chlamydia trachomatis, Neisseria gonorrhoeae, or Trichomonas vaginalis
. Sex Transm Dis 1999; 26:26–32.
7. Golden MR, Whittington WL, Handsfield HH, et al. Effect of expedited treatment of sex partners on recurrent or persistent gonorrhea or chlamydial infection. N Engl J Med 2005; 352:676–685.
8. Gunn RA, Fitzgerald S, Aral SO. Sexually transmitted disease clinic clients at risk for subsequent gonorrhea and chlamydial infections: Possible ‘core’ transmitters. Sex Transm Dis 2000; 27:343–349.
9. Hillis SD, Nakashima A, Marchbanks PA, et al. Risk factors for recurrent Chlamydia trachomatis
infections in women. Am J Obstet Gynecol 1994; 170:801–806.
10. Mosure DJ, Berman S, Kleinbaum D, et al. Predictors of Chlamydia trachomatis
infection among female adolescents: A longitudinal analysis. Am J Epidemiol 1996; 144:997–1003.
11. Oh MK, Cloud GA, Fleenor M, et al. Risk for gonococcal and chlamydial cervicitis in adolescent females: Incidence and recurrence in a prospective cohort study. J Adolesc Health 1996; 18:270–275.
12. Richert CA, Peterman TA, Zaidi AA, et al. A method for identifying persons at high risk for sexually transmitted infections: Opportunity for targeting intervention. Am J Public Health 1993; 83:520–524.
13. Richey CM, Macaluso M, Hook EW. Determinants of reinfection with Chlamydia trachomatis
. Sex Transm Dis 1999; 26:4–11.
14. Rietmeijer CA, Van Bemmelen R, Judson FN, et al. Incidence and repeat infection rates of Chlamydia trachomatis
among male and female patients in an STD clinic: Implications for screening and rescreening. Sex Transm Dis 2002; 29:65–72.
15. Thomas JC, Weiner DH, Schoenbach VJ, et al. Frequent re-infection in a community with hyperendemic gonorrhoea and chlamydia: Appropriate clinical actions. Int J STD AIDS 2000; 11:461–467.
16. Orr DP, Johnston K, Brizendine E, et al. Subsequent sexually transmitted infection in urban adolescents and young adults. Arch Pediatr Adolesc Med 2001; 155:947–953.
17. Weinstock H, Sweeney S, Satten GA, et al. HIV seroincidence and risk factors among patients repeatedly tested for HIV attending sexually transmitted disease clinics in the United States, 1991 to 1996. J Acquir Immun Defic Syndr 1998; 19:506–512.
18. Burstein GR, Zenilman JM, Gaydos CA, et al. Predictors of repeat Chlamydia trachomatis
infections diagnosed by DNA amplification testing among inner city females. Sex Transm Infect 2001; 77:26–32.
19. Peterman, TA, Tian, L, Metcalf, CA, et al. High incidence of new asymptomatic infection following treatment for chlamydia, gonorrhea, or trichomoniasis. International Society for Sexually Transmitted Disease Research, Amsterdam, The Netherlands, July 10–13, 2005:Abstract TP-159.
20. Centers for Disease Control and Prevention. Sexually Transmitted Disease Surveillance, 1999. Atlanta: US Department of Health and Human Services, September 2000.