Informed participation in screening is essential

Editor—A paradox with early detection is that its popularity bears little relation to the magnitude of its benefits and harms. Sarah Stewart-Brown and Andrew Farmer's editorial about screening could therefore be dismissed by some people as public health negativism.1 There is sparse evidence that subjects suffer psychological morbidity, so why worry?

Consider what happens when we screen 250 000 women for cervical cancer. The aim is to help a subset of the 40 or so women who would die annually of cervical cancer among this number. It can only be a subset because not all cases are detectable through screening, and not all those detected are amenable to cure even if found early. Among those patients tested we find newly abnormal smears in some 15 500.2 All but a few hundred (that is, the number of incident serious cases during a screening round of five years) represent overdetection. In other words, over 15 000 women have abnormal cytological findings but would never have a problem from cervical cancer. Through screening we have given them a problem. Yet most of these women think they are lucky, believing that they would have died without screening.

Then there are women who have disease detected on screening, receive treatment, yet still develop recurrent disease.3 We have made no real difference to their destiny except to make them angry and litigious. Paradoxically they too are firm supporters of screening but believe that someone must have made a mistake in their case. Much the same happens with the women who develop symptomatic disease between screening tests. Their disappointed expectations are an inevitable part of screening, yet these women blame the screeners, not the screening. They will happily campaign for more screening in the belief that if we invested more we could make screening perfect. By offering screening to 250 000 we have helped a few, harmed thousands, disappointed many, used £1.5m each year, and kept a few lawyers in work. But the women believe that thousands of lives have been saved and that more could be if only so many “mistakes” were not made. Yet cervical cancer is theoretically an ideal disease for prevention through early detection.

Simplistic beliefs about the power, ease, and affordability of early detection are widespread. This is evident from both reactions to the giving of evidence based information to potential recipients4 and the uncritical acceptance of flawed conclusions based on uncontrolled observation of screening.5 It is pointless to ask about the harm done by screening until we have asked whether recipients have been given accurate information.