Article Text

Download PDFPDF

Improving response rates for self collected urine samples
  1. J Hocking1,
  2. S Tabrizi2,
  3. D Jolley3,
  4. S M Garland4,
  5. C K Fairley5
  1. 1The Macfarlane Burnet Institute for Medical Research and Public Health, Melbourne, Australia
  2. 2Department of Microbiology, Royal Women’s Hospital, Melbourne, Australia
  3. 3School of Health Sciences, Deakin University, Melbourne, Australia
  4. 4Royal Women’s Hospital, Melbourne, Australia
  5. 5Melbourne Sexual Health Centre, Department of Public Health, University of Melbourne
  1. Correspondence to:
    Jane Hocking, Burnet Institute, GPO Box 2284, Melbourne 3001, Victoria, Australia;

Statistics from

Chlamydia trachomatis is the commonest bacterial sexually transmitted infection (STI) in Victoria, Australia, with the number of notifications increasing threefold in the past 8 years from 1287 in 1994 to 3977 in 2001.1 As infection with chlamydia is frequently asymptomatic, notification data underestimate population prevalence. Innovative study designs are necessary to investigate chlamydia prevalence and risk factors. We conducted a pilot study among women aged 18–32, to estimate the rate of response to a request to provide a mailed self collected urine specimen for chlamydia testing. Recruitment via mail was compared with recruitment via mail and follow up telephone contact.

Between March and May 2002, the names and addresses of 150 Victorian women aged 18–32 were randomly selected from the electoral roll. These were linked with the Electronic White Pages and telephone numbers obtained where possible, producing two groups: (1) women with telephone numbers identified, and (2) women without telephone numbers identified. All women were mailed a letter of invitation and an information leaflet. Women in group 2 were also mailed a reply paid participation form asking them to indicate whether they wished to participate.

Women in group 1 were telephoned after 1 week and consent sought to mail them a urine kit. Two reminder letters were sent to non-responders in group 2. Women testing positive were treated with azithromycin through their nominated doctor.

Participants provided 20 ml first void urine in the container provided. Specimens were tested for chlamydia by polymerase chain reaction.

Telephone numbers were found for 70 (47%) women. Among women in group 1, five (7%) were excluded because they were living overseas. Of the remaining 65 women, 35 (54%, 95% CI 41 to 66) agreed to participate and 29 (45%, 95% CI 32 to 57) provided a specimen. One case of chlamydia was diagnosed giving a prevalence of 3.4% (95% CI 0.1 to 17.8) in this group (table 1). Among women in group 2, 11 (14%) were excluded because they were not living at their registered address. Of the remaining 69 women, 16 (23%, 95% CI 14 to 35) agreed to participate and 14 (20%, 95% CI 12 to 32) provided a specimen. No cases of chlamydia were diagnosed.

In this pilot study we showed recruitment via mail and telephone had a significantly higher response than mail alone (45% v 20%, p=0.002). This suggests that telephone communication will increase response in population based chlamydia research that uses mail contact as the principal recruitment tool. Although the method of recruitment was not randomly allocated, the 25% difference in response is unlikely to be explained by differences between the two groups.

A response rate of 45% for those recruited via mail and telephone compares well with results obtained in similar overseas studies.2–,4 However, unlike our study that used the electoral roll as the sampling frame, these studies used a primary healthcare sampling frame, not available in Australia. As we were only able to locate telephone numbers for 47%, an alternative sampling frame would be necessary for future research using mailed, self collected specimens.

Table 1

Response rates by age and method of recruitment


Jane Hocking and this project were supported by the Victorian Health Promotion Foundation (VicHealth).

JH, conducted the pilot study and drafted the letter; ST, conception and design particularly with reference to specimen collection and conducted all chlamydia testing; DJ, conception and design particularly with reference to population sampling and statistical methods and reviewed and made revisions to the letter; SG, conception and design of study, assisted with the ethics application and reviewed and made revisions to the letter; CF, conception and design, assisted with the ethics application and revised letter critically for important intellectual content.


View Abstract

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.