RATES OF SEXUALLY TRANSMISSIBLE infections (STIs) have been increasing in Australia1 and other parts of the world2–4 over the past 5 years. Chlamydia trachomatis (chlamydia) is the most commonly notified STI in the United States and Australia, with 41,306 notifications (203.2 rate per 10,000 population) received in Australia in 2005.1,2 Partner notification (PN) is a well-established public health tool for the effective control of STIs.5 In Australia, general practitioners (GPs) diagnose most patients with chlamydia6 but receive little assistance with PN. Furthermore, with about 19,000 GPs in Australia,7 it is likely that most diagnose only a few cases per year and are unlikely to have ready access to partner letters or client brochures.8
It is possible that GPs would find PN easier if they had ready access to partner notification letters. We have previously shown that 70% of GPs in Victoria are connected to the Internet in a way that it is immediately accessible,8 and that 57% of clients diagnosed with an STI would like access to a partner letter.9 Our aim was to determine if a website providing electronic access to a printable partner letter would assist GPs to undertake PN for patients with chlamydia.
Materials and Methods
This study was carried out in Melbourne, Australia. GPs were surveyed before and after the introduction of an intervention in 2 laboratories in Victoria. We arranged for the intervention to occur in laboratories that served about half of general practices in our study. The intervention was text below positive chlamydia results that provided a website address with treatment guidelines, a printable client brochure, and a printable partner letter for patients with chlamydia (index case) to pass on to exposed sexual partners.
The sampling frame for this study was GPs in Victoria. We sent preintervention surveys to 499 randomly selected GPs with a reply paid envelope on June 8, 2005. Those GPs who did not initially return the preintervention survey, were sent a second and third survey, 3 and 6 weeks later. On May 1, 2006, we sent postintervention surveys to the 233 GPs who returned the preintervention surveys using the same methodology described above. The survey included questions relating to demographics, GP chlamydia testing, treatment and partner notification practices, connectivity to the Internet and website usage, and usefulness in the postintervention survey for those exposed.
The Website Intervention
The initial stage of the intervention included a password-protected website. The website address (www.mshc.org.au/secure) and a password were printed on all positive chlamydia results that were sent to GPs by 1 laboratory over a 6-week period from August 29, 2005, to October 14, 2005. Access to the website was not possible without a valid password between these dates. Website tracking software was used to obtain the number of GPs who gained access to our website using the password, the date and time of every login, the IP address of the computer used, and the number of times the website was viewed. Duplicate logins were removed. The password was removed from the website on October 15, 2005, and from that time onwards, the address alone was printed on the results by this laboratory (Fig. 1).
The website address was printed on all positive chlamydia results that were sent to GPs by the second laboratory from February 28, 2006. This laboratory used the text (partner letter, patient brochure, and treatment guidelines) from our original website and added it to their own website.10
The data collected in this study were entered onto the Statistical Package for Social Sciences, version 12 (SPSS Inc, Chicago). Then the postintervention surveys were divided into those GPs who were exposed to the website (GPs using first and second laboratories who diagnosed at least 1 patient with chlamydia during the period when the web address appeared on positive results from that laboratory) and those who had not diagnosed a case in the period or used another laboratory (unexposed). Two comparisons were undertaken. The first was a paired analysis that compared the preintervention and postintervention responses for GPs who did and did not have access to the website. A second analysis compared the postsurvey results among those GPs exposed and not exposed. A McNemar test was used to determine the difference in pre- and postintervention surveys among those GPs exposed and not exposed to the intervention during the survey. χ2 or t test were used to compare the responses from GPs who were and were not exposed in the postintervention survey.
Ethics approval was granted by the Alfred Hospital Melbourne and the University of Melbourne.
One hundred nineteen (24%) of 499 eligible GPs responded to the first round of the survey, 58 (15%) of 380 to the second round, and 56 (17%) of 322 responded to the third round. In total, of the 233 GPs (47%) who returned completed preintervention surveys, 59 used the first laboratory (25%), and 58 used the second laboratory (25%).
Three GPs who had returned preintervention surveys retired before the postintervention survey was sent and 3 moved away without providing a forwarding address. One hundred five of 227 (46%) eligible GPs responded to the first round of the postintervention survey, 36 (30%) of 122 to the second round, and 32 (37%) of 86 eligible GPs responded to the third round. In total, 173 GPs (76%) returned completed postintervention surveys.
The demographic and testing characteristics of GPs who were exposed and not exposed were similar and are shown in Table 1.
Website Results From Tracking Software
During the 6-week period that the website was accessible only by using a password, there were 59 unique IP addresses recorded using the password on the laboratory reports during which time 109 pathology results with this password were sent from the laboratory (59/109 [54%, 95% CI, 45%–64%]).
Use of Materials
There was a highly statistically significant increase in the use of printable partner letters from 13% (10/78) to 36%(28/78) (P = 0.0009) and brochures from 33% (26/78) to 54% (42/78) (P = 0.003) among those exposed to the website while there was no significant change in the use of printable partner letters from 4% (4/90) to 8% (7/90) (P = 0.45) and brochures from 27% (24/90) to 24% (22/90) (P = 0.78) among those not exposed to the website (Table 2). The proportion of GPs who reported practicing PN all of the time did not change in those exposed or not exposed but was greater than 93% in all groups.
Of the GPs exposed, 40 indicated that they saw the website address on positive results and 34 indicated that they used the website. Twenty-eight GPs (82%, 95% CI 66–93) found it to be very or somewhat useful.
GPs exposed to the website were also asked if they thought a website for other STIs would be helpful. Fifty-seven percent (45/79), 48% (38/79), 43% (34/79), and 81% (64/79) felt that a website to assist with PN for patients diagnosed with gonorrhoea, trichomoniasis, syphilis, and herpes, respectively, would be useful.
Our study showed that the addition of text with a website address to assist GPs with PN was commonly used and considered to be very useful. We are not aware of other studies that have addressed the issue of providing materials on the web through a link on positive results. These results are most applicable to countries like Australia, where STIs are diagnosed in primary care where the clinicians see relatively few cases in a given year and are unlikely to have resources such as partner letters available.
Our study had a number of limitations. First, it was an observational study and did not randomize the GPs to intervention. However, a randomized study would not have been feasible because there are too few laboratories. To lessen the effect of bias on the study results, we used a paired design that analyzed those exposed and unexposed to the website separately. Furthermore the baseline characteristics of the GPs who were exposed and not exposed were similar.
Our study also used self-reported practices by the GP as the outcome measure rather than assessing whether the clients themselves reported improved PN or, better still, looked at reinfection rates among those diagnosed with chlamydia, as has been done in other recent studies.11 We would argue, however, that for such a cheap and simple intervention that provides access to resources, it may not warrant size and cost of a study of the magnitude that would be necessary to demonstrate endpoints other than that the resources were used and were popular.
We did not find that the intervention significantly increased self-reported PN by the GPs. However, in our study, improved PN after the intervention was not statistically possible to assess because GPs almost invariably reported that they carried out partner notification 100% of the time. The high level of reported PN may relate to social desirability bias among GPs, or alternatively be unusually high in the group who volunteered to be part of this study. Notwithstanding this limitation, the significant increase in use of partner letters from 13% to 36% suggests that PN improved.
In Australia now, over 88% of GPs are computerized12 and we demonstrated that 66% of GPs in our study had immediate access to the web. Providing resources for conditions that they uncommonly diagnose were regularly used and popular in our study. We are working with Victorian laboratories to include these for all positive results of STIs where PN is recommended. Improving the detection of STIs in those at high risk of infection, such as partners of infected individuals, should contribute to other measures to reduce the prevalence of STIs.
2. Centers for Disease Control. Chlamydia in the United States. National Center for HIV, STD and TB prevention, Division of Sexually transmitted Diseases, 2006. Available at: http://www.cdc.gov/std/Chlamydia/default.htm
. Accessed July 2006.
5. Donovan B, ed. Australian Contact Tracing Manual, 2nd ed. Darlinghurst, New South Wales: Commonwealth Department of Health and Aging; 2002.
6. Keogh L, Hammond J, Mulvey G, et al. Contact tracing for sexually transmissible diseases in general practice in Victoria, Australia. Venereology 1998; 11:32–37.
8. Tomnay J, Gebert R, Fairley CK. A survey of partner notification practices among general practitioners and their use of an internet resource for partner notification for Chlamydia trachomatis
. Sex Health 2006; 3:217–220.
9. Tomnay JE, Pitts M, Fairley CK. Partner notification: Preferences of Melbourne clients and the estimated proportion of sexual partners they can contact. Int J STD AIDS 2004; 15:415–418.
11. Golden MR, Whittington W, Handsfield H, 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.
12. Henderson J, Britt H, Miller G. Extent and utilization of computerization in Australian general practice. Med J Aust 2006; 185:84–87.