348 Peters.qxd

Background. Women with mild dyskaryosis are currently managed by six-month cytological surveillance. While there is good evidence that women suffer psychological distress on receipt of an abnormal test, and that this is amenable to educational intervention, it remains uncertain whether this distress is prolonged and, if so, how it should best be managed. Aim. To investigate whether a structured educational intervention containing a risk communication package impacts upon psychological sequelae associated with this surveillance. Method. A pragmatic cluster-randomized controlled trial during 14 months in 1995 and 1996, based in general practices in Avon and South Glamorgan, that compared the intervention with standard care. Follow-up was by postal questionnaire at six weeks and four months after the screening laboratory had reported the test result. The intervention was an invitation to attend the general practice to consult with a practice nurse trained to deliver the package. The main outcome measures were Spielberger state-anxiety, SF36 Mental Health dimension, four condition-specific questions regarding concerns about gynaecological health and timing of the repeat smear test, and attendance for the repeat test. Results. Of 514 eligible women, 270 were recruited, of whom 240 returned the six-week questionnaire and 181 returned the four-month questionnaire. On all but one outcome measure, the differences between the groups were not statistically significant. At six-week follow-up, the proportion who preferred the repeat test to be sooner than six months was statistically significantly higher among controls (74% versus 53%; 95% CI = 9% to 33%). At the four-month follow-up, the difference was 7% (95% CI = -7% to 21%). Conclusion. There appear to be high levels of anxiety during surveillance for mild dyskaryosis that were not reduced by the intervention. Given that a personally delivered educational intervention designed to reduce anxiety could be viewed as the best available practice, it is of concern that women in the intervention group demonstrated sustained anxiety over a four-month period. The research agenda therefore seems to return to the fundamental question of whether surveillance should be the management of choice.