Multifactorial audit of invasive cervical cancer: key lessons for the National Screening Programme.

AIMS To audit factors associated with the development of invasive cervical cancer. METHODS Twenty cases of invasive cervical cancer in one health district for 1991-93 were audited by multifactorial analysis. RESULTS The average age was 53 years with 20% (4/20) aged over 65 years. Of the patients, 45% (9/20) were identified by a cervical smear, with 40% (8/20) from the National Screening Programme (NSP) and 5% (1/20) opportunistically; 30% (6/20) had not received a smear, 10% (2/20) being aged under 65 and 20% 65 or over. Those with no smear had all been hospital patients during the previous five years. No response to a smear invitation occurred in 5%. In 20%, there had been a true negative smear two to five years previously. Inappropriate laboratory diagnosis or inappropriate clinical management occurred in 30% and 15% (3/20), respectively. In 20%, two or more factors were present in the same patient. CONCLUSIONS Why cervical cancer occurs after a true negative smear requires research and women aged over 65 with no smear must be targeted. Failsafe systems should incorporate inadequate smears and smear adequacy should be given priority in quality assurance (QA) and training. False negative reports must be minimised but accepted as an inherent part of the NSP and not an automatic indicator of poor laboratory performance. Comprehensive national QA standards are required, to which providers must conform. Postcoital bleeding is an absolute indication for a smear and, when appropriate, opportunistic smears offered at all hospital attendance. Cervical cancer audit must be guaranteed access to all clinical and laboratory information and be seen as a means to improve the effectiveness and quality of the NSP.