Michaud, Joshua M. MHS*; Johnson, Sheridan M.†; Ellen, Jonathan MD*
*Department of Pediatrics, Johns Hopkins University School of Medicine, Baltimore, Maryland; and †Baltimore City Health Department, Baltimore, Maryland
This work was supported by the Centers for Disease Control Syphilis Elimination funding for the Baltimore City Health Department.
The authors collected data during routine public health activities of disease control and analyzed data for program evaluation. This activity was therefore designated as public health practice and nonresearch. In accordance with the Code of Federal Regulations, Title 45, Part 46, The Public Service Act, human subjects review is not required for public health nonresearch activities.
The authors thank the Disease Intervention Staff and Supervisors of the Baltimore City Health Department for their daily dedication to the tough job of interviewing cases and finding disease. They also thank Baltimore’s Health Commissioner Peter Beilenson for his longstanding, consistent support of this initiative.
Correspondence: Joshua M. Michaud, MHS, 2700 Lighthouse Point East, Suite 210, Baltimore, MD 21224. E-mail: email@example.com
Received for publication September 15, 2003,
revised November 19, 2003, and accepted November 25, 2003.
Background: Traditional syphilis control tools could be limited in high-risk groups in which the disease is concentrated. Novel programmatic methods such as tracking and targeting sex partner meeting venues could be valuable.
Goal: The goals of this study were to determine if syphilis cases’ sex partner meeting venues are geographically different than their residences and to determine the characteristics of identified meeting places.
Results: For cases diagnosed from September 2001 to December 2002 with geocodable data, only 9% of meetings took place in the same census block group as residence, and mean and median distance from residence to meeting place was 1.73 and 1.03 miles. The most common meeting location type overall was a street or corner, but differed by risk behaviors.
Conclusion: Baltimore syphilis cases in general met sex partners outside their immediate neighborhoods. Meeting locations could provide new targets for syphilis control interventions.
IN AREAS WHERE SYPHILIS is not predominated by cases among men who have sex with men (MSM), the disease tends to be concentrated in populations engaging in high-risk behaviors such as prostitution and illicit drug use. 1–3 Individual members of these populations tend to be difficult to find using the traditional syphilis control tool of partner notification as a result of poor locating information on sexual and social contacts. 4–6 As a result of this important limitation, there is a need for novel tools for syphilis control.
One possibility is to target interventions to sex partner meeting venues. These venues have previously been shown to be important for gonorrhea transmission. 7 Common meeting venues, especially those associated with high-risk activities, could supplement traditional case-finding techniques by identifying previously unknown cases. For example, in Baltimore City, a syphilis intervention targeting meeting venues was implemented with some success. 8 These areas could also serve as good targets for primary prevention activities.
The Baltimore City Health Department’s (BCHD) Syphilis Elimination Program has developed and implemented a sex partner meeting venue tracking and response component in its syphilis control program. The purpose of implementing this program was to increase the health department’s ability to find previously unknown syphilis cases and to provide an opportunity for primary prevention in areas frequented by persons at high risk for syphilis infection.
To provide evidence that sex partner meeting venues can be an effective intervention target, it is important to determine whether sex partner meeting locations are separate and distinct from residence. If cases meet their partners in their neighborhoods of residence, then interventions could simply be targeted at geographic areas where clusters of cases’ residences are found. If this is not true, it opens the possibility of targeting meeting locations separately.
The first objective of this study was to determine whether meaningful meeting data could be collected on syphilis sex partner meeting venues. The second objective was to determine if syphilis cases tend to meet their sex partners in venues outside their residential neighborhoods. The third objective was to describe meeting venues and determine if they vary by risk behaviors.
The BCHD introduced an enhanced syphilis interview, including meeting venue information, in November 2000. The population for this study was all interviewed early syphilis cases (primary, secondary, or early latent) diagnosed between September 1, 2001, and December 31, 2002, who were residents of Baltimore City. All information was gathered by trained BCHD disease intervention specialists, recorded on an interview form, and entered into an Access database. Cases were asked if they engaged in one of the following behaviors during the last 12 months: a male who had sex with a male (MSM), intravenous drug use, exchange of sex for drugs, working as a prostitute, or picking up a prostitute. Those persons who self-reported engaging in at least 1 of these behaviors was identified as “high risk.”
Enhanced interview data could not be collected from all interviewed cases because some cases moved to other jurisdictions, were unlocatable, refused the interview, or unavailable for another reason. BCHD limited data collection on meeting venues to only those venues where a new sexual partner was met in the 6 months before syphilis diagnosis. This was a programmatic decision implemented because we felt that only recent meeting venues were likely to be useful for disease intervention. All residences with full address and all meeting venues with a proper address, intersection, or specific landmark were geocoded using MapInfo GIS software. 9
We first compared the demographics and risk behaviors of those cases providing at least one meeting venue with those not providing this data. Then, for those who provided at least one meeting venue, we compared the census block group and census tract of their residence with the block group and tract of each sex partner meeting venue and calculated the straight-line distance between their residence and meeting venue(s).
Four hundred twenty-two Baltimore residents were diagnosed with early syphilis during the study period. Three hundred forty-four (82%) were interviewed, and of those interviewed, 281 (82%) had enhanced interview data available. One hundred thirty-six (48%) individuals with enhanced interview data provided no meeting place information: 116 (84%) as a result of no new sexual partners within the last 6 months, 10 (8%) because they refused, and the remaining 10 (8%) for other reasons.
Table 1 compares individuals providing meeting venue information with those not providing this information. In the univariate analysis, those who provided meeting venues were significantly more likely to be male and white, and were more likely to report risk behaviors. In the multiple logistic regression model, age group and each risk behavior except intravenous drug use were independently associated with providing venue information.
Of the 281 residential addresses, 260 (93%) were geocodable. Cases resided in 170 unique census block groups and 109 unique census tracts within Baltimore. There were 181 geocodable meeting venues, and 116 individuals had both a geocodable residential address and geocodable meeting venue. These 116 individuals reported a total of 166 meeting venues. These meeting venues were found in 109 unique census block groups and 76 unique census tracts. Only 15 of 166 meetings (9%) occurred in the same census block group, and 62 of 166 (37%) occurred in the same census tract as a case’s residence.
Table 2 presents meeting venue data. The overall mean and median straight-line distance from residence to sex partner meeting venue was 1.73 miles and 1.03 miles, respectively, with a range of zero (meeting venue was their residence) to 7.43 miles. Significant differences in mean distance between residence and meeting venue were found for males versus females (2.0 vs. 1.1 miles, P <0.01) and self-reported HIV-positives versus -negatives (2.7 vs. 1.6 miles, P = 0.05), but there was no significant difference in average distance between white and black, younger (<25 years) and older (25+ years), or by risk behavior (data not shown in Table 2). The street or corner was the most common meeting venue overall, but differences by risk of the case were found as presented in Table 2.
Through its adoption of an expanded interview focusing on sex partner meeting venues where high-risk activities are reported, the Baltimore Syphilis Elimination Project has added a “risk-space” approach to the city’s syphilis control program. The term risk-space in this context refers to identifiable geographic locations where high-risk activity associated with syphilis transmission occurs. This concept has formed the basis for the city’s outbreak response plan.
This study provides evidence that valuable data can be collected on syphilis sex partner meeting venues. Although meeting venue data could not be collected from all new cases of syphilis, the comparison in Table 1 indicates that meeting venue information was collected from a subset of individuals more likely to engage in high-risk activities. This is exactly the group that we are interested in targeting so the missing data are not of substantial concern, although missing information it is likely to be unavoidable.
This data indicates that Baltimore syphilis cases often meet sex partners outside their own neighborhoods. This supports the observation that syphilis transmission has been concentrated among persons engaging in high-risk activities, and that these activities tend to occur in certain areas of the city.
The Internet was not reported as a sex partner meeting venue in Baltimore during the timeframe of this study, even among MSMs. This finding runs contrary to reports from other jurisdictions which have experienced a large number of MSM cases who report using the Internet to identify partners. 10,11 More recently, however, some MSM cases in Baltimore have reported use of the Internet to meet partners (data not shown).
A limitation to using sex partner meeting venue data is that only a proportion of infected cases will provide geographically identifiable venues. Also, geocoding meeting venues is impossible without specific location information, and data could be lost because a case might be unwilling or unable to provide a specific cross-street, address, or landmark.
It is also difficult to know how to define relevant neighborhoods. Although census block groups were originally designed as relatively homogenous divisions approximating a “neighborhood,” mobility within the inner city and emigration from the city frequently will change how a neighborhood is defined by residents. We attempted to overcome this problem by looking at 3 distinct geographic measures (census tracts, block groups, and straight-line distance), which all provide evidence that syphilis cases do travel a considerable distance to meet sexual partners. Also, we did not collect information about mobility of the cases, which could cloud interpretation of these results if “residence” is relatively mobile for a large percentage of the study population.
Notwithstanding these limitations, this data provides some evidence that tracking sex partner meeting venues can be valuable because they are likely to be distinct from residences. Syphilis research and control programs not already doing so should take sex partner meeting venues into consideration when designing and implementing disease intervention activities.
1. Rolfs RT, Goldberg M, Sharrar RG. Risk factors for syphilis: cocaine use and prostitution. Am J Public Health 1990; 80: 853–857.
2. Farley TA, Hadler JL, Gunn RA. The syphilis epidemic in Connecticut: Relationship to drug use and prostitution. Sex Transm Dis 1990; 17: 163–168.
3. Greenberg J, Schnell D, Conlon R. Behaviors of crack cocaine users and their impact on early syphilis intervention. Sex Transm Dis 1992; 19: 346–350.
4. 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.
5. Cowan FM, French R, Johnson AM. The role and effectiveness of partner notification in STD control: A review. Genitourin Med 1996; 72: 247–252.
6. Kohl KS, Farley TA, Ewell J, Scioneaux J. Usefulness of partner notification for syphilis control. Sex Transm Dis 1999; 26: 201–207.
7. Potterat JJ, Rothenberg RB, Woodhouse DE, Muth JB, Pratts CI, Fogle JS. Gonorrhea as a social disease. Sex Transm Dis 1985; 12: 25–32.
8. Michaud JM, Ellen JM, Johnson S, Rompalo AR. Responding to a community outbreak of syphilis by targeting sex partner meeting location: An example of a risk-space intervention. Sex Transm Dis 2003; 30: 533–538.
9. MapInfo, version 7.0. Troy, NY: MapInfo Corp, 2002.
10. Klausner JD, Wolf W, Fischer-Ponce L, Zolt I, Katz MH. Tracing a syphilis outbreak through cyberspace. JAMA 2000; 284: 447–449.
11. McFarlane M, Bull SS, Rietmeijer CA. The Internet as a newly emerging risk environment for sexually transmitted diseases. JAMA 2000; 284: 443–446.