We read with interest the Editor's Choice entitled ‘A revolution in cervical screening’, and agree with many of the author's perspectives. However, on two points, it is unfortunate that the author's views are not consonant with the available literature on screening: test performance characteristics in low resource settings, and cost effectiveness. Dr Grant states that ‘in countries with limited health resources cervical screening by testing for the human papillomavirus on a single occasion may be effective in preventing invasive cancer of the cervix, and may be better and cheaper than regular visual inspection of the cervix with acetic acid [VIA]’. While we agree with the first part of his statement, the second is speculation, and is not supported by the current literature.
Rigorously designed studies in low resource settings reveal the test qualities of human papillomavirus testing, under conditions characteristic of routine service delivery, to be similar to that of visual inspection of the cervix with acetic acid—especially sensitivity1,2. No studies are currently available comparing the two in terms of reducing cancer and therefore we question the author's basis for claiming that human papillomavirus testing may be ‘better’.
Second, in most low resource settings, the likelihood of ‘regular’ testing for cervical cancer or precancer is very low. Rather, in almost all such settings, researchers and policymakers are predicting incidence reductions assuming once in a lifetime testing, maybe twice (in the best of circumstances, not more frequently than every five years). Published results of recent, rigorous computer-modelling exercises indicate that visual inspection of the cervix with acetic acid is either the most cost effective approach or could even result in a cost savings to the provider, with a predicted reduction in cancer mortality equivalent to or greater than that resulting from human papillomavirus testing3. To be sure, testing for infection with human papillomavirus or for protein-marker evidence of the sequelae of such infection will almost certainly play an important role in cervical cancer prevention worldwide. However, at present, unfortunately, human papillomavirus testing is too expensive for consideration in the lowest resource settings where the need for testing is the greatest. In addition, testing is not an end in itself, but only a means toward identifying who might need (or should be offered) treatment. Again, the literature indicates that providing an immediate link to treatment among those tested is critical in obtaining the greatest ultimate reduction in mortality.
Because visual inspection of the cervix with acetic acid is a ‘real-time’ test (i.e. the results are immediate), management decisions can be made without delay following testing—a crucial element in a cancer prevention programme, especially where problems related to loss-to-follow up are pervasive. Whether the programme is in East Asia or East London, when a test result involves a time lag, the likelihood that all necessary follow up steps will take places and, in particular, that the patient receives the recommended treatment or referral, is much lower than if the results are available immediately. Real-time testing would clearly improve the usefulness of human papillomavirus in any clinical setting. In summary, while human papillomavirus-related technologies are likely to play an important role in future cervical cancer prevention programme, to say that human papillomavirus testing may be better and cheaper than visual inspection of the cervix with acetic acid in low resource countries is both premature and inconsistent with currently available scientific evidence.