Gunn, Robert A. MD, MPH*†; Lee, Marjorie MPH†; Oh, Christina MPH‡; Brodine, Stephanie MD‡
From the *Division of STD Prevention, Centers for Disease Control and Prevention, Atlanta, Georgia; †Public Health Services, HIV, STD, and Hepatitis Branch, Health and Human Services Agency, San Diego, California; and ‡Graduate School of Public Health, San Diego State University, San Diego, California
The authors thank Rita Perry and Jody Thomas for manuscript preparation and Thomas Peterman, MD, Richard Kahn, MPH, and Akbar Zaidi, PhD, for helpful suggestions.
This program evaluation was supported in part by an appointment (Dr. Gunn) to the research participation program at the Centers for Disease Control and Prevention (CDC) administered by the Oak Ridge Institute for Science and Education through an interagency agreement between the US Department of Energy and CDC.
Correspondence: Robert A. Gunn, MD, MPH, STD Control Officer, HIV, STD and Hepatitis Branch, 3851 Rosecrans St (P501C), San Diego, CA 92110. E-mail: firstname.lastname@example.org.
Received for publication November 27, 2006, and accepted January 31, 2007.
Objective: To describe the serologic test for syphilis (STS) prevalence among STD clinic clients, determine the correlation between STS prevalence trends and reported community-diagnosed primary and secondary (P&S) case incidence, and evaluate the usefulness of STS prevalence monitoring as a component of syphilis surveillance.
Study: During the period 1985–2004, 21,4336 STS were done among STD clinic clients and a variety of STS prevalence measures were evaluated.
Results: From 1985–1991, 10.2% of STS were positive, which declined to 5.6% during 1992–2004. Overall, STS positivity (≥1:8) and male positivity (≥1:8) trends were correlated with reported community-diagnosed P&S case incidence and case incidence in men (r = 0.58 and r = 0.81, respectively). Male STS positivity (≥1:8) began increasing in 2001, 1 year before the increase in syphilis incidence in men, which began in the latter half of 2002 and occurred mostly among men who have sex with men.
Conclusion: In a syphilis outbreak in men who have sex with men, STS prevalence (≥1:8) among male STD clinic clients was a useful measure of syphilis case incidence trends and may provide an early warning for a subsequent increase in community-diagnosed case incidence.
FOR MANY DECADES, THE NUMBER of reported primary and secondary (P&S) syphilis cases has been monitored to determine time, place, and person incidence trends of syphilis in the United States. After the widespread use of penicillin to treat syphilis in the 1940s and 1950s, a major decrease in syphilis incidence occurred and the downward trend continued into the mid-1990s, with a few interruptions along the way—an increase among men who have sex with men (MSM) in the late 1970s and early 1980s, and a crack-cocaine-related outbreak among primarily heterosexual blacks in the late 1980s and early 1990s.1–3 In the mid-1990s, nationwide incidence was at an all-time low and in 1999 the Centers for Disease Control and Prevention officially launched a national syphilis elimination plan.4 However, as the plan was unfolding, an upturn in syphilis began, occurring almost exclusively among MSM, and spread throughout the country, involving MSM in most metropolitan areas.5–8
In March 2000, the CDC held a meeting of STD consultants to review syphilis surveillance methods and develop recommendations. The consultant group recommended that monitoring serologic tests for syphilis (STS) prevalence could be incorporated into routine surveillance to estimate disease burden and trends, identify populations with high rates of infection, and evaluate case-reporting surveillance data.9 The recent decline and subsequent increase of syphilis over the last decade provided an opportunity to examine empirical data regarding the relationship between STS prevalence and reported community-diagnosed case incidence. This report provides an evaluation of various STS prevalence measures among STD clinic clients and the correlation with reported community-diagnosed P&S syphilis cases in San Diego County for the 20-year period 1985–2004. Because the current outbreak among MSM began in the latter half of 2002, we also wanted to determine whether the STS positivity rate among STD clinic male clients predicted the increase in reported community-diagnosed P&S syphilis incidence in men.
Reported P&S syphilis cases and rates were obtained from syphilis annual reports, and the proportion of these cases that were initially diagnosed in the STD clinic (information source STD clinic) was available since 1992. Following a routine protocol, the San Diego County STD Clinic, as in most STD clinics, routinely obtained STS for nearly all clinic clients, at least annually. STD clinic serologic testing data for 1985 through June 1991 were available in a summary table format, whereas data from July 1991 through 2004 were available as line listed test data, which included STS titer, in an electronic database maintained by the San Diego County Public Health Laboratory.
STS screening positivity was defined as all reactive screening tests (VDLR or RPR) divided by the denominator of total STS screening tests that were submitted from the county STD clinic. STS confirmed positivity was defined as all reactive screening tests that were confirmed as syphilis infection with a confirmatory test (MHATP, TTPA) divided by the same denominator (total STS screening tests). Person prevalence for both of these measures was defined as all persons in a given year with a positive STS divided by the number of persons tested that year, expressed as either a screening test prevalence or a confirmed test prevalence. Person prevalence was compared with screening positivity during a 2-year period (2001–2002).
Because high-titer screening tests are more likely to indicate recent infection and thus may correlate more closely with incident reported P&S syphilis case rates, all positive screening tests were grouped by titer as ≥ 1:8, ≥1:32, and ≥1:64. Because the increase in syphilis that began in the latter half of 2002 was among MSM, we compared titer specific STS positivity trends with the reported case incidence in men who were diagnosed at community sites other than the STD clinic (community-diagnosed cases). We also determined the STS screening positivity rate by demographic and risk group characteristics of tested clients.
Descriptive analysis was done using SPSS 11.0 for Windows software and STS positivity rates were calculated for each year. STS positivity was compared with reported community-diagnosed P&S case rates, using simple linear regression.
During the 20-year period 1985–2004, there were 214,336 interpretable STS tests done among STD clinic clients and 16,681 were positive (7.8% STS screening positivity rate) with 12,376 confirmed positive (5.8% STS confirmed positivity rate). The screening tests STS positivity rate during 1985–1991 was 10.2% with an 8.0% confirmed positivity rate. During the more recent years (1992–2004), when electronic line listed data were available for each year and reported syphilis incidence was lower, the overall STS positivity rate decreased to 5.6% with a 3.6% confirmed positivity rate. The biologic false-positivity rate was 1.2% (STS screening test positive, confirmatory negative) and the STS screening test positive, confirmatory test not done, rate was 0.8% (usually a confirmatory test not done when doing STS titer follow-up) (Table 1). Thus, during the 1992–2004 period, among the 6317 positive screening STS, 65% were confirmed positive, 22% were biologic false-positives, and 13% were most likely tests performed to follow titers (Table 2). During 1992–2004, the overall STS positivity rate was 5.6% with a 1.3% positivity rate for a titer of ≥1:8, 0.5% for ≥1:32 and 0.3% for ≥1:64 (Table 3). Overall, STS with a titer of <1:8 accounted for 77% of all STS positives.
There was little difference between STS positivity and person-based prevalence. During a 2-year period (2001–2002), STS screening positivity was slightly higher than person-based prevalence using both all STS positive and STS confirmed positive numerators (STS screening positivity 4.0% vs. 3.6% person-based prevalence, STS confirmed positivity 2.9% vs. 2.7%, person-based prevalence).
The temporal trends of STS positivity and STS confirmed positivity for total tests were very similar (r = 0.88, data not shown). Thus, STS positivity, which is easier to calculate compared with confirmed positivity or person prevalence, was used in trend comparisons. Overall, reported P&S syphilis cases increased during 1987–1991 (an outbreak among blacks associated with crack-cocaine use),2 which then declined to near elimination levels in 1996–2001,10 followed by a marked increase in 2002–2004, almost exclusively among MSM (Fig. 1).
The temporal trends of total STS positivity and positivity with a titer ≥1:8 were correlated with reported community-diagnosed P&S case rates (r = 0.58 and r = 0.72, respectively, Fig. 2). However, a change in direction from a decreasing trend to an increasing STS positivity trend occurred in 2001, whereas the community-diagnosed case rate and overall cases increased slightly in the latter half of 2002 and markedly increased in 2003 (Figs. 1 and 2). Comparing the STS positivity rates in men with community-diagnosed case incidence in men showed a similar pattern with the strongest correlation (r = 0.81) being with the ≥1:8 titer trend (Fig. 3). STS trends by titer among MSM showed little change through 2001, but all began increasing in 2002 and continued through 2004 (Fig. 4).
During 1992–2004, 812 cases of P&S syphilis were reported: 269 (33%) were diagnosed in the STD clinic (information source was recorded as “STD clinic”) and a decreasing trend in the proportion of cases that were diagnosed in the STD clinic was evident (1992–1996 [38%], 1997–2000 [30%], 2001–2004 [28%]). In addition, during the most recent period (2001–2004), 25% of reported P&S syphilis cases not initially diagnosed in the STD clinic (information source “not STD clinic”) were seen in the STD clinic for confirmation or follow-up and those syphilis serology results were included in the STD clinic STS database. Thus, the most recent STD clinic STS prevalence data included STS results from 53% of the countywide reported P&S cases (28% information source STD clinic, 25% information source not STD clinic, but follow-up/confirmation titer(s) done in STD clinic). From the perspective of examining the correlation of STS prevalence trends with community-diagnosed P&S cases, 25% of STS results from community-diagnosed cases were contained in the STD clinic STS prevalence trends. The same analysis for the earlier periods was not done (impractical manual visual matching procedure and the 25% overlap of the STD clinic STS positivity and community-diagnosed cases is assumed for the entire 1992–2004 period).
STS positivity was higher among women and persons of color and increased with increasing age. It was also higher in persons with a history of injecting drugs, commercial sex workers, and MSM (Table 4).
The STS positivity rate (7.8%) observed over a 20-year period in this STD clinic attending population was clearly elevated compared to what has been observed in the general population (0.81%),11 but is similar to the positivity rate seen in jails12,13 and drug treatment centers.13 The finding that STS positivity increased with increasing age and was higher among women and among minority populations is also consistent with other observations.8 STS prevalence was also higher among MSM and injecting drug users—2 groups with higher syphilis and STD prevalence.
The proportion of positive STS tests confirmed was 65% with a higher percent (78%) confirmed during the most recent years which was offset by a decrease in the proportion of screening tests for which the confirmatory test was not done (Table 2). This finding was the result of a laboratory policy change to reflexly do a confirmatory test on all positive screening tests. Thus, laboratory policies for confirmation testing will affect STS confirmed rates and suggests that screening test positivity is a more stable measure for monitoring trends. The proportion of STS positives that were biologic false-positives (22%) was similar to that found in another study (23%) among jail detainees in the South.13 The biologic false-positivity rate (1.2%) and proportion of total positives (22%) were similar during the most recent 8 years in this STD clinic evaluation suggesting that this artifact is relatively constant and does not substantially affect temporal trends in STS positivity rates.
This evaluation showed that overall STS positivity among STD clinic clients was moderately correlated (r = 0.58) with reported community-diagnosed P&S case incidence, even though a large proportion (77%) of STS positivity was accounted for by persons with low titers (<1:8) which could represent background prevalence of many years accumulation. However, STS ≥1:8 titer positivity was more highly correlated (r = 0.72) and served as a better trend measure of reported community-diagnosed P&S case incidence. STS ≥1:8 titer positivity is probably more indicative of recent infection and was reported in 2 other studies of prevalence monitoring, which showed a correlation between STS prevalences obtained in jails12 and jails plus other sites13 and case incidence.
During the MSM syphilis outbreak in San Diego, in which reported cases of syphilis in men began increasing in the latter half of 2002, the overall correlation between total male STS-positivity and reported community-diagnosed case incidence in men was low (r = 0.37). However, it was much higher among those with STS titer ≥1:8 (r = 0.81), which suggests that, similar to all cases, using a titer ≥1:8 provides a better measure for prevalence monitoring (Fig. 2). Titers of ≥1:64 are usually small in number and temporal trends can vary considerably. Visually, the total STS and all 3 titer group trends appeared to change in direction from decreasing to increasing in 2001. However, in prior years positivity had increased in 1 year only to decrease in the subsequent year. Thus, meaningful changes in trends may require 2 years or more of observation. However, an increase in positivity in 1 year, especially if larger than that seen in previous years, could possibly indicate an early warning that an increase in reported cases is imminent.
The findings from this evaluation are limited to San Diego county syphilis morbidity trends and STS positivity from STD clinic clients, and the correlation between STS positivity trends and community-diagnosed syphilis incidence was somewhat expected. In other communities, variability of patients' source of care (STD clinic vs. other providers), provider serologic testing patterns, and stage of diagnosis of early infections (P&S vs. early latent) will likely affect trends and correlations with syphilis incidence. Estimating STS positivity prevalence from multiple sites, such as jails, HIV care clinics, university laboratories, major local laboratories, and family planning services, as well as in STD clinics, may be helpful in establishing a composite community STS positivity trends. However, considering that monitoring STS ≥1:8 titer prevalence in an STD clinic or other site(s) on a semiannual basis would be relatively easy to do, a rising prevalence in the light of a low stable reported case incidence should be of concern and some follow-up would be appropriate. Follow-up may uncover clusters of unreported cases or aberrations in case reporting surveillance. However, additional evaluations from STD clinics and from other screening sites in other communities are needed to establish the usefulness of adding STS prevalence monitoring to current reported case-based syphilis surveillance.
1. Nakashima AK, Rolfs RT, Flock ML, et al. Epidemiology of syphilis in the United States, 1941–1993. Sex Transm Dis 1996; 23:16–23.
2. Gunn RA, Montes JM, Toomey KE, et al. Syphilis in San Diego County 1983–1992: Crack Cocaine, prostitution, and the limitations of partner notification. Sex Transm Dis 1995; 22:60–66.
3. Rolfs RT, Goldberg M, Sharrar RG. Risk factors for syphilis: Cocaine use and prostitution. Am J Public Health 1990; 80:853–857.
4. Centers for Disease Control and Preservation. The National Plan to Eliminate Syphilis from the United States, Atlanta, GA: US Department of Health and Human Services; 1999.
5. Centers for Disease Control and Prevention. Primary and secondary syphilis—United States, 2002. MMWR Morb Mortal Wkly Rep 2003; 52:1117–1120.
6. D'Sousa G, Lee JH, Paffel JM. Outbreak of syphilis among men who have sex with men in Houston, Texas. Sex Transm Dis 2003; 30:872–873.
7. Ciesielski CA. Sexually Transmitted Diseases in men who have sex with men: An epidemiologic review. Curr Infect Dis Rep 2003; 5:145–152.
8. Kahn RH, Heffelfinger JD, Berman SM. Syphilis outbreaks among men who have sex with men: A public health trend of concern. Sex Transm Dis 2002; 29:285–287.
9. Centers for Disease Control and Prevention. Recommendations for Public Health Surveillance in the United States. Atlanta, GA: US Department of Health and Human Services; 2003.
10. Gunn RA, Harper SL, Borntrager DE, et al. Implementing a syphilis elimination and importation control strategy in a low-incidence urban area: San Diego County, California, 1997–1998. Am J Public Health 2000; 90:1540–1544.
11. Hahn RA, Magder LS, Aral SO, et al. Race and the prevalence of syphilis seroreactivity in the United States population: A national sero-epidemiologic study. Am J Public Health 1989; 79:467–470.
12. Kahn RH, Scholl DT, Shane SM, et al. Screening of syphilis in arrestees: Usefulness for communitywide syphilis surveillance and control. Sex Transm Dis 2002; 29:150–156.
13. Finelli L, Farley TP, Gibson JJ, et al. Prevalence monitoring in syphilis surveillance: Results from a multicenter research program. Sex Transm Dis 2002; 29:769–774.