To the Editor: In their study of a SAFE-style trachoma control program (which included Antibiotic treatment, Facial cleanliness, and Environmental improvement, but not Surgery) in a remote Australian community, Ewald et al suggest that no systematic SAFE trachoma control program exists in Australia.1

In the Kimberley region of Western Australia, the Kimberley Public Health Unit (KPHU) has coordinated a trachoma control program since 1989. The World Health Organization SAFE strategy has been implemented since 1996, as described in the Kimberley regional trachoma control guidelines and a peer-reviewed publication.2,3

The trachoma control program in the Kimberley has been delivered by a variety of environmental health, health promotion, community and clinical health professionals employed by State and local governments, and by community-controlled and other non-government organisations. We believe a coordinated regional approach is mandatory, because of the numerous organisations involved in program delivery.

The trachoma control program in the Kimberley continues to achieve good results. The prevalences of follicular trachoma during annual screening of school-aged children in the Kimberley have been published annually in the KPHU Bulletin.4 Since 1996, in accordance with the WHO SAFE strategy, communities with trachoma prevalences of less than 5% were not screened in subsequent years and did not contribute to regional prevalence data. Thus, the observed decrease in trachoma prevalence between 1996 and 2001 is likely to be greater than that shown in the Box.

We believe the Kimberley region is well placed to achieve the WHO aim of eradicating blinding trachoma by 2020.5 However, it is a concern that other regions of Australia with endemic trachoma infection may not conduct disease control activities in a coordinated manner because of lack of leadership in trachoma control or insufficient resources, or both. The community described by Ewald et al borders the Kimberley region and has strong cultural links with several Kimberley groups. The achievements in trachoma control in the Kimberley cannot be sustained in the long term without a nationally coordinated approach. We believe that it falls within the statutory responsibilities of State and Territory departments of health to ensure that environmental and clinical health services are coordinated to achieve trachoma control in Australia.

Trachoma in Kimberley children

Point prevalence of follicular trachoma among children in the Kimberley region aged 5–15 years at annual trachoma screening, 1991–2002

To the Editor: The important article by Ewald et al shows that offering azithromycin treatment to 70% of a remote community may be inadequate to control hyperendemic trachoma, even when combined with a health promotion campaign.1 Yet, significant short-term gains in similar locations have been achieved with as little as 20% of the population receiving azithromycin, where administration to children with trachoma and their household contacts was directly observed by health staff.2

In that study,2 we reported a 6-month follicle resolution rate in schoolchildren of 72%, followed, however, by a return of trachoma prevalence towards baseline levels over 12 months. The critical factors for short-term success with azithromycin appear to be appropriate selection of cases and contacts, plus, where possible, directly observed therapy to minimise reinfection from untreated cases. Sustainability of initial reduction in trachoma prevalence is problematic, and issues of how extensively and how often to screen and/or treat need to be determined in the Australian context. Similar sustainability considerations have arisen in community scabies programs.3

Ewald et al are correct to identify population mobility as a major issue, with trachoma likely to be reintroduced by untreated children entering a community where a treatment program has occurred. Hence the need for a coordinated regional approach. However, a regional approach to trachoma control does not necessarily mean a uniform approach, and it is vital to tailor programs to suit the capacity of communities and their degree of commitment to labour-intensive treatment and health promotion.

Our study suggested that directly observed twice-yearly azithromycin therapy, with a health promotional component, is likely to be preferable to an annual program.2 Mathematical modelling supports this more frequent dosing and, unlike in Africa, the cost of azithromycin should not be a constraining factor in Australia.4

Regardless of the strategy selected, if, after treatment, the prevalence remains hyperendemic (> 20%), then the outcomes from concurrent health promotion are compromised. Appropriately targeted and directly observed azithromycin therapy can help create conditions favourable for health promotion campaigns, which in turn prolong those gains by reducing trachoma transmission.5 Major issues for trachoma control in Australia are (i) who to screen and treat (with directly observed azithromycin therapy); (ii) how often to screen and treat; (iii) how to plan trachoma programs as regional initiatives; and (iv) who will fund, coordinate and implement trachoma programs.

In reply: These letters reinforce a number of important points about control of trachoma (and other endemic infections) in Australia. The questions of whom and how often to treat need refining through Australian experience. Long-term control needs multifaceted, intersectoral collaboration to alter environmental and behavioural conditions against disease transmission. Strategies should be sustained, regional (large as practicable); acknowledging, and guided by, Aboriginal kinship networks; and recommend observed drug treatment (which was negotiated for the final treatment in our study). A non-uniform approach could include more frequent treatment in hyperendemic communities, probably leading to less net use of antibiotic treatment.

Reports from the Kimberley Population Health Unit show a very mixed picture, with wide year-to-year fluctuations in prevalence in many communities. While hyperendemic communities remain in a region, the prevalence of trachoma may increase unnoticed in communities no longer screened because their prevalence has dropped below 5%. If not looked for, it is unlikely to be noticed. Further analysis, such as the graph provided by Johnson and Mak, is to be applauded in the context of a thorough analysis. When this happens, it will greatly strengthen the case for active trachoma control in other regions.

For trachoma prevention, and for many other reasons, we believe environmental health interventions are critical. These remain difficult to evaluate given the high mobility of people in Aboriginal communities. Reliable, long-term, regional environmental health and mobility data are needed as part of this broad issue.