Zwahlen, Marcel PhD*; Spoerri, Adrian MSc*; Gebhardt, Martin PhD, MPH†; Mäusezahl, Mirjam MD, MPH†; Boubaker, Karim MD†; Low, Nicola MD, MFPH*
PUBLIC HEALTH SURVEILLANCE IS THE ongoing systematic collection, analysis, interpretation, and dissemination of data regarding a health event for use in public health action to reduce morbidity and mortality and to improve health.1 Surveillance for sexually transmitted infections (STIs) should provide information for action to control their spread.2 Surveillance systems can collect information about all cases of disease diagnosed in any setting (comprehensive surveillance) or from a sample of healthcare providers, usually thought to provide services for populations at high risk of disease (sentinel surveillance).3 In the United States and Sweden, surveillance for STIs is intended to capture all diagnosed cases.4,5 In France and Germany, only selected practitioners or laboratories report cases.6,7 Surveillance for STIs in the United Kingdom combines elements of both systems reporting all cases of diagnosed infection but from only one source, sexually transmitted disease clinics.8 Ideally, surveillance systems should provide consistent information according to time, place, and person that allow interventions to be targeted appropriately.
Switzerland is a high-income, developed European country (population 7.4 million) with an insurance-based health system.9 Surveillance systems for STIs collect both basic demographic data from laboratory reports, which cover the entire population, and behavioral surveillance data from dermatovenereology clinics, which permit more detailed analyses of behavioral trends in high-risk populations.10,11 There are no national guidelines about screening or diagnosis of STIs. The objective of this study was to describe and compare time trends from different surveillance systems for chlamydia, gonorrhea, and syphilis in Switzerland.
Materials and Methods
We analyzed anonymous data about cases of bacterial STIs reported from 1997 to 2003 from the 3 sources of surveillance data for STIs in Switzerland,12 all of which are collated by the Swiss Federal Office of Public Health.10 The data were collected in accordance with the Swiss law on epidemics (1970) and regulations on the reporting of infectious diseases (“Meldeverordnung”). Ethical committee approval was therefore not required. In this article, we refer to laboratory-diagnosed episodes of Neisseria gonorrhoeae as gonorrhea, Chlamydia trachomatis as chlamydia, and Treponema pallidum as syphilis infections.
From 1987, laboratories registered with the Swiss Federal Office of Public Health have been legally obliged to report positive results of tests for notifiable pathogens. Notifiable sexually transmissible agents included N. gonorrhoeae, C. trachomatis, and T. pallidum,10 but T. pallidum was removed from the list in 1999.13 The number of tests performed is not reported. There are approximately 400 hospital and private laboratories in Switzerland, and approximately 200 of these report cases of any notifiable pathogen each year. Between 1997 and 2003, 81 laboratories reported cases of gonorrhea in 1 or more years and 88 reported chlamydia cases. Laboratory reporting is incomplete, but approximately 60% and 70% of all diagnosed gonorrhea and chlamydia cases in Switzerland, respectively, are thought to be captured by the system.14 Laboratory case reports include information about age, sex, canton of residence, and year of birth. For the period covered by this study, available data about diagnostic methods show that chlamydia was diagnosed by nucleic acid amplification test. Testing methods for gonorrhea and syphilis did not change during the study period. Repeated infections in the same person and multiple diagnoses of the same episode in different settings cannot be identified.
Swiss Network of Dermatology Polyclinics
Dermatology polyclinics provide specialist diagnosis and treatment services for STIs and dermatologic problems. The Swiss Network of Dermatology Polyclinics comprises 6 clinics in public university or city hospitals in Basel, Bern, Geneva, Lausanne, and Zurich (2 clinics). These clinics see 100 to 200 patients per day, with Geneva reporting the largest number. The Zurich clinics have the highest proportion of patients presenting with STIs (approximately 10% of all patients).11 Patients are usually self-referred, but general practitioners and other specialists can also make referrals. Screening of asymptomatic patients is not routine and testing is done according to clinical judgment. Approximately 15% of patients are men who have sex with men, 45% are of non-Swiss nationality (of whom two thirds are from other European countries), and 10% are women.11 We have used data from this system to report trends in chlamydia, gonorrhea, and syphilis in men only.
The network started contributing to national surveillance in 1989 as part of a European project to monitor the prevalence of HIV among patients with STIs.11 The clinics collect patient-based data on 14 laboratory-diagnosed (chlamydia, gonorrhea, chancroid, syphilis, HIV, trichomoniasis) and clinically diagnosed (genital herpes simplex, urethritis, proctitis, mucopurulent cervicitis, pelvic inflammatory disease, genital warts, pediculosis pubis, and scabies) STIs and infestations. Since 1990, all patients diagnosed with a first episode of infection in any 90-day period have been offered a test for HIV and behavioral risk data have been recorded.10,14 In addition to sociodemographic data (sex, age, marital status, nationality, canton of residence, and educational level), multiple infections diagnosed in one individual at the same clinic visit can be identified, but individuals presenting for multiple consultations cannot be identified. The total number of people being tested for STIs is not recorded.
Swiss Sentinel Surveillance Network
The Swiss Sentinel Surveillance Network (Sentinella) provides information about STIs diagnosed in women by gynecologists. Starting in 1986, a nonrandom rolling sample of 150 to 250 general practitioners, internists, pediatricians (and gynecologists since 1995) has reported weekly morbidity data on infectious and noninfectious diseases.15 These physicians comprise approximately 3% of all family practitioners for each specialty and cover approximately 3.5% of all consultations in Switzerland. Since 1995, 25 to 35 private gynecologists have participated in the network. Data were requested to supplement the limited information about STIs in women that was available from the Swiss Network of Dermatology Polyclinics. Women have open access to gynecology services through the compulsory health insurance system and more than half of women aged 15 to 34 years see a gynecologist each year.16 Testing for STIs takes place according to clinical judgment. The Sentinella system record episodes of chlamydia, trichomonas, and herpes simplex, and includes a category for all other infections, but these are not specified. Case reports include information about laboratory diagnosis and year of birth. The number of tests performed was not recorded. Multiple infections at different consultations in the same person cannot be identified. In this report, we include chlamydia diagnoses only.
Table 1 summarizes the numbers of cases of chlamydia, gonorrhea, and syphilis reported from each reporting source from 1997 to 2003. In the Swiss Network of Dermatology Polyclinics, these 3 infections accounted for 29% (721 of 2477) of all episodes of STIs in men. From 1997 to 2003, 52 gynecologists participated in the gynecologist part of the Swiss Sentinel Surveillance Network for 1 or more years. Chlamydia diagnoses accounted for 26% (300 of 1167) of all reported STIs.
Numbers of chlamydia infections, but not denominator data, are reported by all 3 systems. From 1997 to 2003, laboratories reported a total of 18,853 cases of chlamydia (13,223 in women, 5286 in men, 344 of unknown sex) (Table 1). The number of cases with known sex reported decreased from 2573 in 1997 to 2093 in 1999 but increased to 3449 in 2003, an overall increase of 31%, 65% in men and 25% in women. There were 32 of 88 laboratories that reported chlamydia cases in all years from 1997 to 2003. The proportion of cases from these laboratories was 85% (2255 of 2665) in 1997 and 76% (2642 of 3494) in 2003.
Among cases in which age was recorded, the greatest number of chlamydia cases in women was in 20 to 24 year olds (31.5% [3965 of 12,574) and 25 to 29 year olds (23.9% [3003 of 12,574]) (Fig. 1). Across all years, cases in 15 to 19 year olds ranked fourth after 30 to 34 year olds. The number of cases reported from this age group has, however, increased so that it ranked third in 2003 (18.5% [438 of 2370]). In all other age groups numbers of cases between 1997 and 2003 were stable or declined. For men, very similar numbers of infections were reported from 25 to 29 (22.2% [1046 of 4713]) and 30 to 34 (22.9% [1078 of 4713]) year olds (Fig. 1). The lowest number of infections was in 15 to 19 year olds throughout the study period (3.6% [168 of 4713]).
In the sentinel populations, the numbers of chlamydia cases were low. In the Swiss Network of Dermatology Polyclinics, 14 to 40 male cases per year were diagnosed with no clear time trend. Of 165 cases reported from 1997 to 2003, 134 (81.2%) were in heterosexual men. The age distribution of these cases was similar to that of laboratory-reported cases, but numbers were small. In the Sentinella system, there was no clear trend in chlamydia diagnoses over time (Table 1). The age distribution was similar to that of laboratory reported cases. Fifteen- to 19-year-old women accounted for only 7.3% of cases (22 of 300).
Numbers of cases of gonorrhea, but no denominator data, are available from laboratory reports (men and women) and the Swiss Network of Dermatology Polyclinics (men). Laboratory reports included 2817 gonorrhea cases from 1997 to 2003 (503 in women, 2262 in men, 52 unknown) (Fig. 2). The number of cases with known sex reported increased each year from 259 in 1997 to 528 in 2003, an overall increase of 104%, 119% in men and 48% in women. Of 81 laboratories reporting gonorrhea cases, 17 reported cases throughout the study period. The proportion of cases reported by these laboratories was 76% (205 of 269) in 1997 and remained stable until 2002 (73% [390 of 532]) but decreased to 64% (341 of 531) in 2003.
Among cases for which age was available, the greatest number of cases in women was in 20 to 24 year olds (25.2% [116 of 461]) and 25 to 29 year olds (21.0% [97 of 461]). Numbers of cases in 15 to 19 year olds (15.8% [73 of 461]) and 30 to 34 year olds (16.3% [75 of 461]) were similar. In men, 30 to 34 year olds accounted for the largest number of cases (25.8% [532 of 2061]) followed by 25 to 29 (19.9% [411 of 2061]) and 35 to 39 year olds (18.0% [371 of 2061]). The Swiss Network of Dermatology Polyclinics reported between 31 and 79 male gonorrhea cases per year during the study period. There was a slight upward trend overall, which was mainly the result of an increase in cases reported from Zurich. Of a total of 318 cases, 137 (43.1%) were in men who have sex with men. The age distribution was similar in heterosexuals and men who have sex with men, but numbers of cases in each age group were small.
Syphilis cases were only reported by the Swiss Network of Dermatology Polyclinics. The number of syphilis cases of all stages increased by 127%, from 22 cases in 1997 to 50 in 2003; this was the same as the number of gonorrhea cases. Primary and secondary syphilis accounted for 72% (136 of 189) of all syphilis cases. Of 189 male cases of syphilis (6.7% of all diagnoses in men), 57.1% (108 of 189) were in heterosexuals and 41.3% (78 of 189) in homo-/bisexual men (orientation was unknown in 3 cases). The proportion of syphilis cases in gay and bisexual men increased over time from 9% (2 of 24) in 1997 to 48% (24 of 50) in 2003. HIV infection was more frequent among men with syphilis (15.9%) than with any other diagnosis (4.7%).
Behavioral surveillance data collected by the Swiss Network of Dermatology Polyclinics showed that over the study period, men who have sex with men accounted for an increasing proportion of diagnoses, from 18.2% in 1997 to 25.5% in 2003. There were no marked trends over time for either heterosexuals or men who have sex with men in the reported numbers of partners in the past 6 months, frequency of condom use with casual and regular partners, acquisition of infections from sex workers, or the nationality of men with diagnosed infections.
The Swiss surveillance system for STIs includes elements of both comprehensive reporting (laboratory reports) and sentinel systems (gynecology and dermatology clinic patients). This allowed us to compare trends in case reports about the same infection collected from multiple sources. All 3 components of the surveillance system report numbers of cases of STIs but not the numbers of tests done. National laboratory reports of C. trachomatis and N. gonorrhoeae increased between 1997 and 2003, but cases of chlamydia from men attending dermatology clinics and women attending private gynecologists did not increase and gonorrhea cases in men attending dermatology clinics only increased modestly. Syphilis reports from men attending dermatology clinics increased during the study period but were not recorded in any other system.
Comparison of Trends From Surveillance Systems
The finding of different trends in reports of STIs in laboratory and clinical reports in Switzerland is new. Between 1988 and 1994, reports of gonorrhea and syphilis from both laboratories and the Swiss Network of Dermatology Polyclinics fell.10 These trends were attributed in part to the success of STOP-AIDS, the Swiss national AIDS prevention campaign.17 Trends in reported numbers of cases of infection and their age distribution without denominator data are difficult to interpret as previously noted in Switzerland.14,18,19 The trends observed could reflect a combination of increased testing, a change in the population being tested, increased test sensitivity, changes in reporting, and true changes in morbidity. More widespread testing and reporting might have contributed to the laboratory trends in gonorrhea and chlamydia, because more laboratories have registered with the Swiss Federal Office of Public Health. This is suggested by the declining proportion of cases reported from laboratories that reported cases consistently throughout the study period. The trends observed in laboratory reports are also consistent with other European countries, where rates of bacterial STIs have increased recently.5,6,8
The small steady numbers of chlamydia infections and the observed age distribution reported from the Sentinella and dermatology polyclinic systems probably represent diagnostic testing. In this situation, few tests are done in teenagers of either sex, older women having cervical smears are more likely to be tested, and many male cases of chlamydia are clinically diagnosed as nongonococcal urethritis. This probably accounts for the low male-to-female ratio for diagnoses of chlamydia as compared with gonorrhea. Underascertainment is suggested by comparing population rates of laboratory-reported chlamydia in Switzerland (48.1 cases per 100,000 in 2003) with data from countries with more complete surveillance, for example, Sweden (300 per 100,000 in 2003).5 The self-reported lifetime incidence of STIs in Swiss adults (9.5% in 2000; André Jeannin, IUMSP, Lausanne, personal communication) is, however, similar to that elsewhere in Europe (11–13% in the United Kingdom).20 A previous study conducted among gynecologists in the Sentinella system found a chlamydia prevalence of 2.8% among 15- to 34-year-old women attending for checkups. Extrapolating this prevalence to all women in Switzerland, the authors estimated that, as a result of underdiagnosis and underreporting, laboratory reports represented only 5% of all chlamydia infections in women.16
Implications for Surveillance of Sexually Transmitted Infections in Switzerland
The attributes of the Swiss surveillance systems have previously been evaluated14 using published guidelines.1 The simplicity, flexibility, and acceptability of all 3 components of the system were found to be acceptable. Laboratory reports were assessed as being the most representative source of data with high sensitivity for symptomatic infections. The coverage of sentinel populations (men attending dermatology polyclinics and women attending gynecologists) was acknowledged to be low. This study confirms that, although the Sentinella system is one of the largest existing primary care surveillance systems,21 it is not an efficient source of data on STIs; the system includes few gynecologists and we found that each reported an average of less than one case of chlamydia per year. The numbers of cases of gonorrhea reported from dermatology polyclinics was also low and, by 2003, equaled by reports of syphilis. This unusual pattern reflects the specialism of the dermatology clinics; STIs are a small part of their case load, but this preferentially includes cases of syphilis, particularly those with skin manifestations.22 Of interest, numbers of gonorrhea cases did increase in 2 clinics that consider themselves to be sexually transmitted disease clinics.11 Men with urethral discharge might increasingly present to urologists or general practitioners, although we cannot confirm this because little is known about sexual health-seeking behavior in Switzerland.
This study has shown that trends in bacterial STIs at a national level reported by 3 different components of the Swiss surveillance system from 1997 to 2003 were not consistent. The data were also difficult to interpret because of the lack of denominator data and the absence of consistent guidelines for testing for sexually transmitted infections. Findings from this study, reported to the Swiss Federal Office of Public Health,12 have also contributed to a review of surveillance for STIs and HIV in Switzerland. Changes to the system include: from 2006, laboratory reporting for T. pallidum has been reintroduced, and laboratory surveillance for gonorrhea is being enhanced by the collection of additional details from the diagnosing physician; data from the Swiss Network of Dermatology Polyclinics and Sentinella system will stop contributing to the national surveillance system; denominator data for HIV tests will be reported by laboratories from 2007; and a national center for STIs will be established. This study shows that all STI surveillance systems should be evaluated regularly to ensure that they fulfill their public health role.
1. Centers for Disease Control and Prevention. Updated guidelines for evaluating surveillance systems: Recommendations from the guidelines working group. MMWR Morbid Mortal Wkly Rep 2001; 50:1–51.
2. Catchpole MA. The role of epidemiology and surveillance systems in the control of sexually transmitted diseases. Genitourin Med 1996; 72:321–329.
3. Thacker SB, Stroup DF. Future directions for comprehensive public health surveillance and health information systems in the United States. Am J Epidemiol 1994; 140:383–397.
4. Sexually Transmitted Disease Surveillance, 2002. Department of Health and Human Services. Atlanta: Centers for Disease Control and Prevention, 2003.
5. Smittskyddsinstitutet. Verksamhetsberättelse och epidemiologisk årsrapport 2003. Sexuellt överförda infektioner/STI [in Swedish]. Solna, Smittskyddsinstitutet, 2004:49–51.
6. Herida M, Sednaoui P, Goulet V. Gonorrhoea surveillance system in France: 1986–2000. Sex Transm Dis 2004; 31:209–214.
7. Bremer V, Marcus U, Hofmann A, Hamouda O. Building a sentinel surveillance system for sexually transmitted infections in Germany, 2003. Sex Transm Infect 2005; 81:173–179.
8. The UK Collaborative Group for HIV and STI Surveillance. Focus on Prevention. HIV and Other Sexually Transmitted Infections in the UK in 2003. London, Health Protection Agency Centre for Infections, 2004.
9. Herzlinger RE, Parsa-Parsi R. Consumer-driven health care: Lessons from Switzerland. JAMA 2004; 292:1213–1220.
10. Paget WJ, Zimmermann HP. Surveillance of sexually transmitted diseases in Switzerland, 1973–1994: Evidence of declining trends in gonorrhoea and syphilis. Soz Praventivmed 1997; 42:30–36.
11. Paget WJ, Batter V, Zwahlen M. The Swiss Network of Dermatology Policlinics HIV prevalence study: Rationale, characteristics and results (1990–1996). Soz Praventivmed 1999; 44:1–7.
12. Low N, Spoerri A, Zwahlen M. Die sexuell übertragenen Infektionskrankheiten in der Schweiz, 1997–2003. Bulletin Bundesamt für Gesundheit 2005:844–849.
14. Paget WJ. The Surveillance and Epidemiology of Sexually Transmitted Diseases in Switzerland. PhD thesis. Basel: University of Basel, 2002.
15. Swiss Federal Office of Public Health. Sentinella 1999–2000. Annual Report of the Swiss Sentinel Surveillance Network. Bern: Bundesamt für Gesundheit, 2002.
16. Paget WJ, Zbinden R, Ritzler E, et al. National laboratory reports of Chlamydia trachomatis seriously underestimate the frequency of genital chlamydial infections among women in Switzerland. Sex Transm Dis 2002; 29:715–720.
17. Dubois-Arber F, Jeannin A, Meystre-Agustoni G, et al. Evaluation de la strategie de prévention du VIH/SIDA en Suisse:Septième rapport de synthèse 1999–2003. Lausanne: Institut universitaire de médicine sociale et préventive, Raisons de Santé, 2004:40a.
18. Klaue K, Spencer B, Balthasar H. Santé sexuelle et reproductive en Suisse. Lausanne: Institut universitaire de médecine sociale et préventive, Raisons de santé, 2002:85.
19. Balthasar H, Spencer B, Addor V. Indicateurs de santé sexuelle et reproductive en Suisse. Lausanne, Schweizerisches Gesundheitsobservatorum, Documents de travail, 2003:3.
20. Fenton KA, Korovessis C, Johnson AM, et al. Sexual behaviour in Britain: Reported sexually transmitted infections and prevalent genital Chlamydia trachomatis infection. Lancet 2001; 358:1851–1854.
21. Bollag U, Capkun G, Caesar J, Low N. Trends in primary care consultations for asthma in Switzerland, 1989–2002. Int J Epidemiol 2005.
22. Lautenschlager S. Sexually transmitted infections in Switzerland: Return of the classics. Dermatology 2005; 210:134–142.