Trends in chlamydia positivity among heterosexual patients from the Victorian Primary Care Network for Sentinel Surveillance, 2007–2011

Chlamydia is the most commonly notified infectious disease in Australia, predominantly diagnosed among young heterosexual populations. The number of infections diagnosed each year in Victoria has increased steadily, from 11 149 in 2007 to 19 834 in 2012.1 In recent years, multiple interventions have been conducted nationally, aimed at increasing rates of testing for chlamydia.2-8 In combination with policy and practice change, these interventions have been successful in nearly doubling the number of chlamydia tests conducted in Australia from 161 550 in 2007 to 299 685 in 2012.9

Increased testing could explain rising notifications; however, it is important to determine whether chlamydia prevalence is also increasing. There is some evidence from clinic-based surveillance that the prevalence of chlamydia in the population is rising.10-13 However, existing studies have all been based on sexual health clinic attendees only, which account for less than 2% of chlamydia diagnoses in Australia.14 Further, only one of these studies included data on heterosexual clients from multiple clinics, and it did not report on or adjust trends for reported sexual behaviours.13

Sentinel surveillance through primary health care clinics offers an opportunity to collect information on those being tested for chlamydia and provide ongoing estimates of chlamydia positivity and correlates of infections detected. When testing rates are sufficiently high, chlamydia positivity in clinics can be an accurate indicator of chlamydia prevalence in the population being tested.15 We aimed to determine whether chlamydia positivity is increasing among heterosexual people tested for chlamydia at a number of sentinel surveillance clinics in Victoria, taking into account sexual behaviours.

The Victorian Primary Care Network for Sentinel Surveillance on blood-borne viruses and sexually transmitted infections (STIs) collects data on all clients tested for Chlamydia trachomatis at selected clinics treating people at high risk of STIs.16 Eight sites collect data on heterosexual clients; these sites include two sexual health clinics and six other primary care clinics providing services to young people at high risk. The latter include family planning, hospital, community and correctional facility clinics, which opportunistically screen patients for chlamydia based on the chlamydia risk factors of the individuals presenting. These clinics see caseloads of homeless and incarcerated youth, young pregnant women, young women with fertility issues and individuals referred by their infected partners.

Demographic and behavioural data were collected using a questionnaire from all patients undergoing chlamydia testing as part of standard clinical practice. The reason for the patient’s visit and STI test was asked. Responses were coded as: asymptomatic, patient-initiated; asymptomatic, doctor-initiated; STI symptoms present; other reason. Symptoms were not defined and may have been unrelated to chlamydia. Chlamydia polymerase chain reaction results were collected directly from the laboratories servicing the sentinel sites and were linked to questionnaire records.

For analysis, indeterminate chlamydia results were excluded. All test results from the same individual within 30 days were excluded as these may not reflect new tests. Subjects could have multiple tests across the study period (2007–2011). Patients aged less than 16 years, those reporting sex work, or those reporting any same-sex sexual partners in the past 12 months were excluded from analysis. Trends in test numbers were determined using linear regression. Changes from baseline (2007) in incidence of testing positive were assessed using three-level random-effects Poisson regression models, with clinic and subject treated as nested random factors to account for dependencies across these levels. Wald tests were undertaken to compare rates between subsequent years. Models were offset for total number of tests and adjusted for relevant covariates (age group, being born overseas, symptoms, more than one sexual partner in the past year, and inconsistent condom use). Covariates were included based on previously reported risk factor analyses.17 Differences in risk groups over time were explored by bivariate logistic regression analyses, with standard errors adjusted for clustering by clinic. Only a single record per person per year was included in risk group analysis. Analyses were conducted in Stata version 11 (StataCorp).

The project was approved by the Department of Health Human Research Ethics Committee. The individual ethics committees of all sentinel sites were consulted.

Over 5 years (2007–2011), there were 31 682 chlamydia tests among women and 23 771 tests among men. The number of people tested each year increased over time by an average of 498 women per year (95% CI, 126–870) and 467 men per year (95% CI, 258–676) (Box 1).

Fifty-three per cent of women and 32% of men tested were under 25 years of age. Among women, 54% of tests were conducted at a sexual health clinic and 46% at other sites; among men, 91% of tests were conducted at a sexual health clinic.

The observed increase in chlamydia positivity may be due to a true increase in chlamydia prevalence in the population or may be a result of the methods used in this surveillance system. System limitations mean that we cannot be certain that observed trends reflect population trends. The system only includes patients who sought medical attention and had chlamydia testing at one of the sentinel clinics. The selected sentinel sites may not be representative of the whole population and may result in a higher prevalence of chlamydia because of the nature of the clinics involved. Positivity varied by clinic type among women tested within the sentinel surveillance system (6.3% at sexual health centres and 4.1% at other clinics).17

Supporting the argument that the prevalence of chlamydia is increasing in Australian women, other studies in Australia have also shown an increase in chlamydia rates within their samples.11,13,14 While sensitive to rates of testing, surveillance data also support an increase in chlamydia prevalence.1 Possible hypotheses explaining a rise in chlamydia prevalence in the population might include decreasing levels of immunity,18,19 decreased rates of general antibiotic use,20,21 increases in treatment failure and reinfections22 and increased transmission rates. Increased transmission of chlamydia could be due to an increase in high-risk behaviours, such as higher numbers of sexual partners, increased reporting of casual sex partners and younger age of first sex, which have been shown in other studies in Australia.23,24 Our analysis also showed that several risk factors for chlamydia increased among people tested at sentinel sites between 2007 and 2011. The exception was a decline in reporting of inconsistent condom use; however, an overwhelming majority of respondents (85%) reported that they did not use condoms consistently.

In conclusion, we identified an increase in chlamydia positivity over time, particularly among young women. Explanations for this trend may include changes in behaviours, leading to increased risk of transmission or an increase in the number of women at high risk presenting for testing. The continued emphasis on young women is warranted, and chlamydia screening is important; however, public health measures need to go beyond testing to target high-risk sexual behaviours.

1 Number of chlamydia tests and percentage of positive results, by year and sex