Although primary care physicians (PCPs) play a key role in skin cancer screening, their skills in detecting malignant tumours is suboptimal. To determine whether a short dermoscopy e-learning course (4 h) in skin tumour diagnosis for PCPs is non-inferior to a long course (12 h) in selective triage of skin lesions. Secondly, to evaluate whether regular refresher training sessions are necessary to maintain the PCPs' skills in the medium term. A randomized 2 × 2 factorial non-inferiority trial was conducted online over an 8-month period among 233 PCPs including 126 certified general practitioners, 94 PCPs in training, and 13 occupational physicians, all without prior advanced dermoscopy training. Participants were randomized 1:1:1:1 to receive short training and mandatory refreshers (n = 58), short training and optional refreshers (n = 59), long training and mandatory refreshers (n = 58), or long training and optional refreshers (n = 58). PCPs' skills were evaluated before training (T0), immediately after training (T1) to test the non-inferiority, and after 5 months (T2) to evaluate the impact of the refreshers. The primary endpoint was the difference in the change of score after short and long training. The non-inferiority margin was set at -28%. Among the 233 randomized participants, 216 (93%) completed T1 and 197 (84.5%) completed T2. For short versus long training, the primary endpoint was 1.392 (95% CI: 0.138; 2.645) in the per-protocol population (p < 0.001) and 1.016 (95% CI: -0.224; 2.256) in the modified intention-to-treat population (p < 0.001). After training, the type of refresher showed no impact on the score (p = 0.840). However, PCPs who completed all refreshers showed the best mean overall score at T2 (p < 0.001). These findings confirm that short dermoscopy e-learning is non-inferior in training PCPs to triage skin lesions compared to long training. After training, regular refreshers are important to maintain the PCPs' acquired skills over time.
Read full abstract