Abstract
Aim: Performance measures for Australian laboratories reporting cervical cytology are a set of quantifiable measures relating to the profile and accuracy of reporting. This study reviews aggregate data collected over the ten years in which participation in the performance measures has been mandatory.
Methods: Laboratories submit annual data on performance measures relating to the profile of reporting, including reporting rates for technically unsatisfactory specimens, high grade or possible high grade abnormalities and abnormal reports. Cytology-histology correlation data and review findings of negative smears reported from women with histological high grade disease are also collected. Suggested acceptable standards are set for each measure.
This study reviews the aggregate data submitted by all laboratories for the years 1998–2008 and examines trends in reporting and the performance of laboratories against the suggested standards.
Results: The performance of Australian laboratories has shown continued improvement over the study period. There has been a fall in the proportion of laboratories with data outside the acceptable standard range in all performance measures.
Laboratories are reporting a greater proportion of specimens as definite or possible high grade abnormality. This is partly attributable to an increase in the proportion of abnormal results classified as high grade or possible high grade abnormality. Despite this, the positive predictive value for high grade and possible high grade abnormalities has continued to rise.
Conclusion: Performance measures for cervical cytology have provided a valuable addition to external quality assurance procedures in Australia. They have documented continued improvements in the aggregate performance, as well as providing benchmarking data and goals for acceptable performance for individual laboratories.
Acknowledgements:
The authors would like to acknowledge the invaluable input by members of the committees involved in establishing and reviewing the performance measures during their evolution. The authors would also like to gratefully acknowledge the administrative assistance provided by the QAP clerical staff.