• chlamydia
• screening

The recent pilot studies of chlamydia screening in Portsmouth and the Wirral show that there is a substantial burden of chlamydial infection in young women and that high uptake of screening and good coverage of the target population can be achieved.^1,2 This is important. However, the pilot studies do not demonstrate the effectiveness of chlamydia screening in reducing either morbidity or the prevalence of infection (nor were they designed to do this). In fact, further screening (in the recall study³) of the same target group in the same settings, approximately 16 months after the pilot screening had ended, shows no change in chlamydia prevalence: 11.2% (pilot) v 11.9% (recall) in the Wirral and 9.8% v 11.4% in Portsmouth. Opportunistic screening continued after the pilot in family planning clinics in the Wirral, but there has been no reduction in chlamydia prevalence (11.4% during March-August 2000 compared with 12.4% during March-August 2002).

It would be wrong to conclude that opportunistic screening does not work. The incidence of chlamydia in the United Kingdom appears to be rising and it may be that the prevalence found in the recall study would have been higher still in the absence of earlier screening and treatment. Thus, controlled studies are needed to determine effectiveness empirically. Economic modelling is important for assessing the long term effects of different screening scenarios but is of little value without reliable empirical data for which it cannot substitute.

The accompanying editorial⁴ addresses important questions about screening men, screening in primary care, and asks “what further evidence is required before national screening for all at-risk groups?” Screening both sexes is clearly a more expensive, but potentially more cost effective, strategy than screening women alone. Since the vast majority of general practices outside the pilot areas are not currently involved in any organised screening programme, the ideal opportunity now exists for a randomised evaluation of different strategies to determine the most cost effective approach to screening in general practice.

In reality, any screening programme in general practice would take considerable time to be introduced on a large scale. Phased introduction in the context of a randomised trial poses no ethical problems because the optimal approach (for example, women or both sexes? opportunistic or cyclical?) is unclear and no strategy has yet been shown to reduce chlamydia prevalence or morbidity in the United Kingdom.

We propose a trial in which general practices would be randomised to screening young women alone, screening young men and women, or to no defined screening programme. Effectiveness would be determined by comparison of chlamydia prevalence or associated morbidity across the three arms at follow up. Such a trial, combined with an economic evaluation of the different screening strategies, would provide the direct, robust evidence that is currently lacking but essential to achieve effective control of chlamydia in the United Kingdom through wise use of resources.