How good are skin cancer clinics at melanoma detection? Number needed to treat variability across a national clinic group in Australia

Background: The number needed to treat (NNT) is a key measure of the quality of melanoma diagnosis. There are few data on this measure from primary care skin cancer clinics in Australia. Objective: We sought to report the NNT from a large pathology database and examine several patient characteristics. Methods: We calculated NNT by doctor and clinic among 10,612 lesions, 6796 patients, 57 doctors, and 15 clinics from a pathology database. NNT was calculated with and without seborrheic keratoses. Results: Overall NNT was 30 (with seborrheic keratoses) and 23 (without seborrheic keratoses). Excluding the 4 doctors with NNT greater than 60, total NNT decreased from 30 to 21 and from 23 to 15, respectively (with and without seborrheic keratoses). NNT was higher for female patients and younger patients (<30 years). NNT varied by doctor from 0 to 192 and 117, respectively (with and without seborrheic keratoses). Limitations: Given the retrospective design, we were unable to examine doctor characteristics such as age, sex, medical training, and patient pressure to excise. Conclusions: Substantial variability in individual doctor NNT produced an overall NNT similar to that reported from mainstream general practice, and higher than specialist practice.