GONORRHEA IS THE SECOND MOST common sexually transmitted infection (STD) after chlamydia among reportable STDs in Canada.1 The continuing high incidence of gonorrhea, coupled with the worldwide resurgence of this STD during the past decade,2–4 have renewed interest in the determinants of gonococcal (GC) infection. The cornerstone of an effective STD control program is the prompt detection, treatment, and risk reduction counseling of cases and their sexual partners. Despite these efforts, many patients return for subsequent treatment of the same or other STD.
An STD reinfection signals continued poor sexual health, which is inextricably entwined with the effectiveness of risk-reduction counseling and partner management. Reinfections also strain healthcare resources by requiring additional case management and thereby add to the existing high costs of managing STDs. In Canada, for example, the annual direct and indirect cost of GC infection and its associated sequelae in 1991 was estimated to range from $30 million to $74 million.5 In addition to the economic burden, repeat STDs can increase the risk of infection-related complications and adverse long-term sequelae. For example, women with repeat GC infections are at elevated risk of tubal scarring, a pathology associated with pelvic inflammatory disease and ectopic pregnancy.6,7 Equally important to disease morbidity is the facilitation by GC infection of human immunodeficiency virus (HIV) transmission from an infected partner.8
Individuals with frequent STD reinfections have been implicated in the persistence of disease in the population, given that reinfection is a marker of high-risk sexual behavior.9–13 Of the limited number of studies on GC reinfection, some show a relationship between gender, age, ethnicity, socioeconomic status and repeat infection.10,12–16 Other studies contend that multiple partnerships and unsafe sex practices are equally important predictors of reinfection.14,16,17
Interventions aimed at curtailing the recent rise in GC infections may be enhanced by defining the determinants of reinfection. To this end, the objectives of this study were to examine the extent and frequency of GC reinfection in the province of Alberta, Canada, and to investigate the association between reinfection and patient characteristics in a cohort of patients with STDs.
Gonorrhea Notification and Treatment in Alberta
Laboratories and healthcare providers are required under the Public Health of Alberta to report all positive cases of GC to the provincial Ministry of Health, Alberta Health and Wellness (AHW). Healthcare providers use the standardized STD notification form to document a client’s demographic information, date and location of visit, reason for visit, clinical and laboratory findings, diagnoses, and treatment. Information from the STD notification form is entered into the computerized AHW STD surveillance database that, since its implementation in 1997, has allowed automated tracking of the infection histories for all STD cases reported in Alberta. Since 1991, the first-line treatment for uncomplicated GC in adults has been a directly observed single dose of oral ciprofloxacin and is available free of cost to the patient through the universal healthcare system. Patients are treated presumptively or on confirmation of laboratory diagnosis of GC infection, as appropriate.
Study Population and Design
We examined a retrospective cohort of all laboratory-confirmed GC cases in Alberta reported by STD clinics operated by regional public health departments and by other settings (e.g., hospitals, family planning clinics, and general practices). Entry into the cohort was defined as the date of first diagnosis of Neisseria gonorrhoeae in an anogenital specimen by positive culture or polymerase chain reaction (PCR) between January 1, 1991, and December 31, 2003. Systematic surveillance data collection of GC cases began in 1991. Laboratory culture of GC was the main method used to identify infection between 1991 and 1997. Beginning in 1997, specimens were predominantly tested using a nonamplified nucleic acid test (NAAT), Gen-Probe PACE 2 (Gen-Probe Inc., San Diego, CA). From 2000, additional specimen screening was performed by PCR using Roche Cobas Amplicor with results confirmed using Roche Light Cycler (Roche Diagnostics, Basel, Switzerland). By 2003, 75% to 80% of all specimens were being processed by NAAT or PCR.
Patient information was retrieved from the AHW STD surveillance database for GC cases reported since 1997 (year of database implementation) and from STD clinic charts for cases before 1997. Subject linkage was based on matching of at least 2 of the following: name, date of birth, and residential address. Variables of interest were age, gender, ethnicity, marital status, sexual practice (i.e., heterosexual, homosexual, bisexual), place of treatment (STD clinic, other clinic setting), number of sexual partners in the past 3 months, and concurrent STDs. Ethnicity was self-reported by patients or as assessed by the healthcare provider. Canadian aboriginals included individuals of North American Indian, Métis, or Inuit origin. Sexual practice was inferred from patient reports about the gender of their sexual partners along with information about the anatomic site of specimen collection.
A reinfection was defined as an episode of GC detected at least 14 days after a previously recorded infection for patients tested by culture methods or at least 4 weeks after the previous infection for those tested by NAAT or PCR. The choice of 4 weeks was guided by the fact that NAAT is able to detect nonviable nucleic acids posttreatment for periods shorter than 4 weeks, resulting in false-positive tests. For each case, we ascertained whether the patient had a subsequent infection during the study period and that treatment was received. A GC episode reported within these timeframes was removed from the analysis because it was assumed to be either a duplicate report of a single episode or a persistent infection resulting from treatment failure.
The follow-up period was defined as the time between a previously treated GC infection and the next incident infection or the end of the study period, whichever occurred first. Subject follow up was censored at December 31, 2003. Patients who had no record of a reinfection were assumed to be free of subsequent infection throughout the follow-up period. Based on our criterion for exclusion by NAAT or PCR testing, we further assumed that patients reporting a baseline infection within 4 weeks from the end of the follow-up period would not be permitted adequate time for a potential reinfection to occur and were thus excluded from the analysis. The time between infections was calculated as the difference between the dates of treatment as recorded in the STD notification form or clinic chart. All reinfections for a patient during the follow-up period were considered in the analysis because this method better represents the risk trajectory of patients experiencing multiple infections compared with examining only a single reinfection.
The procedures for this study were approved by the University of Alberta Health Research and Ethics Board.
Pearson’s χ2 test was used to compare differences in categorical variables and Student t test for normally distributed continuous variables. Median times to reinfection were compared by Wilcoxon rank sum test. Results were considered statistically significant at P ≤0.05 (2-tailed). Rates of reinfection were estimated by dividing the number of incident GC episodes during follow up by the number of person-years at risk. Time to reinfection was stratified by age to compare younger (<25 years) with older patients (≥25 years).
Bivariate and multivariate logistic regression models were developed to examine risk of reinfection using generalized estimating equations account for the correlation of multiple infections within subjects.18 Variable significant at P <0.20 in the bivariate analysis were further examined in multivariate models controlling for potential confounders such as age, gender, number of sexual partners, year of infection, number of reinfections, and their interaction with other covariates. Variables with >50% missing data (e.g., concurrent STDs) were excluded from the final model but were examined in reduced models. The final model was chosen with a manual stepwise elimination of variables using the likelihood ratio.
The cohort was comprised of 5,701 patients reporting 6,269 GC infections between January 1, 1991, and December 31, 2003. Of the cohort, 460 individuals (8% of the cohort) were identified with reinfections, whereas 5,241 patients were nonreinfected during the study period. In addition to their baseline infection, the 460 reinfected patients reported 568 subsequent GC infections, ranging from one (81% of reinfected patients) to 5 reinfections (0.4%) and which together represented 16% (1,028 of 6,269) of the total GC episodes during the study period.
Stratum-Specific Reinfection Time and Rates
Table 1 compares reinfected and nonreinfected subjects at baseline. The median follow-up time for the cohort was 40.9 months (range, 0.5–151 months). Approximately 57% of reinfections occurred within 1 year and the median time to reinfection was 9.2 months (interquartile range [IQR], 3.9–24.4).
In examining the time to first reinfection, stratification by baseline age revealed that patients younger than 25 years of age had a shorter reinfection time compared with those who were older (Table 2). The median first reinfection time for younger patients was 7.9 months (IQR, 3.2–20.5) and 11.4 months (IQR, 3.3–31.3) for older patients. Significant differences between age groups were also noted among males, ethnic groups, homo-/bisexual patients, and in those treated in an STD clinic.
As shown in Table 3, the 568 reinfections in the cohort were observed over 24,230 person-years of follow up, for an overall reinfection rate of 2.34 per 100 person-years (95% confidence interval, 2.09–2.59). The highest rates of reinfection were among single, common law, black, and homo-/bisexual patients.
Predictors of Reinfection
Results from regression analyses are shown in Table 4. As a result of a large proportion of patients (approximately two thirds of the cohort) for whom data on concurrent STDs was unavailable, we opted not to adjust the final multivariate model for this variable. However, when we examined a reduced model that controlled for concurrent STDs, the same variables as the larger model remained predictive of reinfection. Independent predictors of reinfection in the multivariate model were black, aboriginal and other ethnicity, homo-/bisexual practice, and treatment at an STD clinic. Those who were married or in a common law relationship were significantly less likely to become reinfected compared with single individuals. No significant interactions between covariates were observed.
Approximately 16% of GC infection episodes reported in this study were attributed to a small proportion of individuals (8% of cohort) who had multiple infections during follow up. These reinfected individuals were distinguishable by specific demographic (ethnicity, marital status) and behavioral characteristics (STD clinic attendance, sexual practice), which can be useful markers for targeted prevention.
The overall median reinfection time of 9 months posttreatment resonates the brief reinfection period found in other studies.11,14–16 Reinfection suggests that risk behaviors were likely unmodified or that safe sexual behaviors were inconsistently practiced during the interim. Although modifiable behaviors such as sexual partner monogamy can reduce STD reinfection, long-term maintenance of protective behaviors has been found to vary by population subgroups.15,19,20 Among young patients with STDs, for example, posttreatment abstinence from high-risk sexual practices has been shown to decrease with time, whereby rates of inconsistent condom use and partner turnover often return to pretreatment levels.21,22 This may be one explanation for the shorter reinfection time found among patients younger than 25 years of age in this study and other studies.9,14,15,22,23 Further examination by age group also revealed that the most vulnerable patients were young males, homo-/bisexual, and STD clinic attenders. However, we found no differences in the reinfection rate between these age groups, which suggest that although younger patients with GC may become infected more quickly than their older counterparts, both age groups have the same probability of reinfection. This finding was supported by the lack of association with age in the regression analysis.
Higher rates and risk of reinfection were predicted by ethnic group, marital status, sexual practice, and place of treatment. The ethnic disparities in GC reinfection emphasize the continuing population health inequities associated with STDs,9,10,12 despite universal access to free testing and treatment services in Canada. The differential reinfection risk of GC among black and aboriginal individuals compared with white patients suggests that several factors may play a role in the higher risk observed in the former. Data from the United Kingdom and the United States indicate that GC infection among persons of black or black Caribbean origin is associated with higher turnover of sex partners, a history of STDs, and more frequent STD clinic attendance compared with whites and Asians.24,25 The elevated risk of GC reinfection among blacks compared with other ethnic groups has also been supported by other studies,9,10,12,13,15,22 which suggest that sexual networks and sexual mixing patterns may be pertinent to the risk of GC transmission. Although less data exist on Canadian aboriginals, the higher rates of STDs in this group has been previously associated with being male (a possible marker of greater mobility outside the community of residence) and having multiple sexual partners.26
The lower risk of reinfection among married persons and those living in common law relationships might reflect the stability of sexual partnerships compared with those of individuals who are single, among whom higher partner turnover and greater number of partners likely contributes to GC infection and reinfection. Stable sexual partnerships may also afford better partner management as a result of the ease of locating partners as well as motivating sex partners to seek treatment.
Homo-/bisexual patients (mostly men who have sex with men [MSM]) had a 2-fold higher risk of reinfection compared with heterosexuals, which confirms previous findings of elevated GC risk among MSM compared with heterosexual men and women.10,15 It has been suggested that the recent rise in risky sexual behavior and STDs in general, and in particular among MSM, is the result of complacency regarding HIV infection in an era of more effective treatments for HIV infection.3,27,28
Finally, clients of STD clinics represent a higher risk group for repeat GC infections compared with those attending other clinical settings such as general practice or family planning clinics. In a population-based study in the United Kingdom, risk factors for STD infection such as sex with multiple partners, having homosexual partners, and injection drug use were more common among STD clinic attendees compared with patients who sought treatment in other settings.29 These findings are suggestive of a group of patients with STDs for whom greater vigilance and more intensive interventions are needed. An understanding of how such core groups of patients with STDs, namely individuals characterized by high-risk behaviors for infection,9–11 are central to the maintenance of GC infection in relation to public health control efforts is worthy of further investigation.
The results of this study call for more patient-specific and culturally sensitive interventions that consider the unique determinants of GC infection in different subgroups. First, although sex education is mandatory in Alberta schools, considerable variability exists in its implementation, particularly in First Nations communities. This inconsistency may be a determinant of sexual risk-taking among young, sexually active persons (W. Yacoub, personal communication, 2005) and could be addressed in public health policies.
Second, offering testing in nonclinical settings, improving partner referral services for those who test positive for GC, and developing methods for patient-delivered therapy to sex partners can help reduce reinfections.20,23,30,31 Patient management should involve greater encouragement of sexual abstinence until all sex partners are treated, counseling on behavior change such as the reduction in the number of sexual partners and consistent condom use, and a consideration of the patient’s perceptions of GC risk and vulnerability to infection.
Finally, routine rescreening of high-risk patients may be an option for more effective control of GC reinfection. Currently, Canadian STD patient management recommendations32 promote intensified patient education and partner notification for repeat patients but do not advocate rescreening. This recommendation should be reassessed in light of the reinfection time and the key predictors for GC reinfection identified in the current and other similar investigations.14–17 Given that among patients with multiple STDs, more frequent reinfections often occur among those initially infected with GC,15,17,23 a prudent period for rescreening can be considered for selected individuals at 6 to 12 months after an initial GC infection. Rescreening offers the opportunity to shorten the period of infectivity and thereby interrupt the cycle of disease transmission to untreated partners. The success of rescreening can be improved by adequate patient follow up, by facilitating retesting using patient-obtained specimens, and by laboratory technologies that perform well with noninvasive specimens such as urine. Further study is needed to assess the efficacy and cost-effectiveness of various targeted rescreening strategies.
Several limitations must be considered in interpreting the results of this study. First, patients lost to follow up, including those seeking subsequent STD care outside Alberta, would have overcontributed to follow-up time and underestimated the rate of reinfection. However, we expect the impact of losses to follow up to be minimal given that all reported STD episodes for a patient treated in Alberta are captured by the centralized provincial STD surveillance system. Second, as a record-based study, we were unable to examine other important sexual risk behaviors (e.g., partner turnover rates, consistency of condom use), which may help define the array of confounders for GC reinfection. Third, undiagnosed infections, cases of self-medication, and resolution of infection from antibiotic treatment of a different condition could contribute to detection bias and thereby lower our estimated rates of reinfection. On the other hand, because a significant proportion of patients attending Alberta STD clinics present for and undergo screening for other STDs regardless of the presence or absence of symptoms, this practice may contribute to earlier detection of GC infection and, thus, to a higher estimated rate of reinfection.
In summary, we identified a specific subsample of patients with GC whose higher risk of reinfection was predicted by distinct demographic and behavioral characteristics. Gonorrhea reinfection reflects the inadequacy of patient management and warrants such efforts to be strengthened through improved prevention and curative services for patients with STDs and their sexual partners who are at elevated risk of repeat infection.
1. Public Health Agency of Canada. Reported Cases and Rates of Notifiable STI From January 1 to June 30, 2004 and January 1 to June 30, 2003. Sexual Health and Sexually Transmitted Infections Section, 2006. Available at: www.phac-aspc.gc.ca/std-mts/stdcases-casmts/
. Accessed March 21, 2006.
2. Hansen L, Wong T, Mathias M. Gonorrhea resurgence in Canada. Int J STD AIDS 2003; 14:727–731.
3. Rietmeijer CA, Patnaik JL, Judson FN, Douglas JM Jr. Increases in gonorrhea and sexual risk behaviors among men who have sex with men: A 12-year trend analysis at the Denver Metro Health Clinic. Sex Transm Dis 2003; 30:562–567.
4. Hughes G, Andrews N, Catchpole M, et al. Investigation of the increased incidence of gonorrhoea diagnosed in genitourinary medicine clinics in England, 1994–6. Sex Transm Infect 2000; 76:18–24.
5. Goeree R, Gully PR. The burden of chlamydial and gonococcal infection in Canada. In: Ronald AR, Peeling RW, eds. Prevention of Infertility. Research Studies of the Royal Commission on New Reproductive Technologies. Ottawa: Minister of Supply and Services Canada, 1993:29–76.
6. Melly MA, Gregg CR, McGee ZA. Studies of toxicity of Neisseria gonorrhoeae
for human fallopian tube mucosa. J Infect Dis 1981; 143:423–431.
7. Platt R, Rice PA, McCormack WM. Risk of acquiring gonorrhea and prevalence of abnormal adnexal findings among women recently exposed to gonorrhea. JAMA 1983; 250:3205–3209.
8. Laga M, Manoka A, Kivuvu M, et al. Non-ulcerative sexually transmitted diseases as risk factors for HIV-1 transmission in women: Results from a cohort study. AIDS 1993; 7:95–102.
9. Ellen JM, Hessol NA, Kohn RP, Bolan GA. 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.
10. Gunn RA, Fitzgerald S, Aral SO. Sexually transmitted disease clinic clients at risk of subsequent gonorrhea and chlamydia infections: Possible ‘core’ transmitters. Sex Transm Dis 2000; 27:343–349.
11. Bernstein KT, Curriero FC, Jennings JM, Olthoff G, Erbelding EJ, Zenilman J. Defining core gonorrhea transmission utilizing spatial data. Am J Epidemiol 2004; 160:51–58.
12. Kinghorn GR, Pryce D, Morton RS. Repeated gonorrhea in Sheffield: The size of the problem, epidemiologic significance, and personal characteristics of repeaters. Sex Transm Dis 1982; 9:165–169.
13. Brooks GF, Darrow WW, Day JA. Repeated gonorrhea: An analysis of importance and risk factors. J Infect Dis 1978; 137:161–169.
14. Orr DP, Johnston K, Brizendine E, Katz B, Fortenberry JD. Subsequent sexually transmitted infection in urban adolescents and young adults. Arch Pediatr Adolesc Med 2001; 155:947–953.
15. Hughes G, Brady AR, Catchpole MA, et al. Characteristics of those who repeatedly acquire sexually transmitted infections: A retrospective cohort study of attendees at three urban sexually transmitted disease clinics in England. Sex Transm Dis 2001; 28:379–386.
16. Mehta SD, Erbelding EJ, Zenilman JM, Rompalo AM. Gonorrhoea reinfection in heterosexual STD clinic attendees: Longitudinal analysis of risks for first reinfection. Sex Transm Infect 2003; 79:124–128.
17. Fortenberry JD, Brizendine EJ, Katz BP, Wools KK, Blythe MJ, Orr DP. 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.
18. Zeger SL, Liang KY. Longitudinal data analysis for discrete and continuous outcomes. Biometrics 1986; 42:121–130.
19. Diclemente RJ, Wingood GM, Sionean C, et al. Association of adolescents’ history of sexually transmitted disease (STD) and their current high-risk behavior and STD status: A case for intensifying clinic-based prevention efforts. Sex Transm Dis 2002; 29:503–509.
20. Shain RN, Perdue ST, Piper JM, et al. Behaviors changed by intervention are associated with reduced STD recurrence: The importance of context in measurement. Sex Transm Dis 2002; 29:520–529.
21. Fortenberry JD, Brizendine EJ, Katz BP, Orr DP. Post-treatment sexual and prevention behaviours of adolescents with sexually transmitted infections. Sex Transm Infect 2002; 78:365–368.
22. Richert CA, Peterman TA, Zaidi AA, Ransom RL, Wroten JE, Witte JJ. A method for identifying persons at high risk of sexually transmitted infections: Opportunity for targeting intervention. Am J Public Health 1993; 83:520–524.
23. Oh MK, Cloud GA, Fleenor M, Sturdevant MS, Nesmith JD, Feinstein RA. Risk of gonococcal and chlamydial cervicitis in adolescent females: Incidence and recurrence in a prospective cohort study. J Adolesc Health 1996; 18:270–275.
24. Fenton KA, Mercer CH, McManus S, et al. Ethnic variations in sexual behaviour in Great Britain and risk of sexually transmitted infections: A probability survey. Lancet 2005; 365:1246–1255.
25. Centers for Disease Control and Prevention (CDC). Summary of notifiable diseases—United States, 2000. MMWR Morb Mortal Wkly Rep 2002; 49:i–xxii, 1–100.
26. Calzavara LM, Bullock SL, Myers T, Marshall VW, Cockerill R. Sexual partnering and risk of HIV/STD among aboriginals. Can J Public Health 1999; 90:186–191.
27. MacKellar DA, Valleroy LA, Secura GM, et al. Repeat HIV testing, risk behaviors, and HIV seroconversion among young men who have sex with men: A call to monitor and improve the practice of prevention. J Acquir Immun Defic Syndr 2002; 29:76–85.
28. Katz MH, Schwarcz SK, Kellogg TA, et al. Impact of highly active antiretroviral treatment on HIV seroincidence among men who have sex with men: San Francisco. Am J Public Health 2002; 92:388–394.
29. Fenton KA, Mercer CH, Johnson AM, et al. Reported sexually transmitted disease clinic attendance and sexually transmitted infections in Britain: Prevalence, risk factors, and proportionate population burden. J Infect Dis 2005; 191(suppl 1):S127–138.
30. Janssen RS, Holtgrave DR, Valdiserri RO, Shepherd M, Gayle HD, De Cock KM. The serostatus approach to fighting the HIV epidemic: Prevention strategies for infected individuals. Am J Public Health 2001; 91:1019–1024.
32. Golden MR, Whittington WLH, 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.