79 THE VALUE OF TELEDERMOSCOPY TO THE EXPERTISE OF GPS DIAGNOSING SKIN DISORDERS detection increases. This finding is supported by a Cochrane review on accuracy of dermoscopy that shows that this sensitivity increases with more clinical expertise [6]. What we know from a previous study is that 95.1% of the GPs learned from the teledermoscopy consultation [10]. Teledermoscopy provides the GPs with direct feedback on the correctness of their pre-diagnosis by the confirmation of the TD. However, the system does not provide any active feedback on the GP performance. Literature on audit and feedback mechanisms shows that feedback leads to minor improvements in professional practice, but the effect is influenced by the way in which the feedback is delivered [11]. New studies on analyzing and advancing the effect of the feedback mechanism incorporated in teledermoscopy consultations might lead to a higher GP learning curve. Due to the rising number of correct pre-diagnosis of GPs, the percentage of teledermoscopy consultations in which the TD provides the GP with ICD-10 diagnoses decreases over time from 93 percent in the first year to 82 percent in the third year. This decrease might imply advancement in GPs’ skin disorder diagnosing expertise. However, of the total teledermoscopy consultations included in this study, in the subsequent three years 6.6%, 9.2% and 6.3% were diagnosed as melanoma by the TD for which the GP did not provide a (correct) pre-diagnosis. In addition, two consultations which were pre-diagnosed by the GP as a benign neoplasm and a radiation-related disorder, were classified by the TD as melanoma or other malignant neoplasm of the skin (C43-C44). The high number of incorrectly diagnosed cases by GPs, indicates that the added value of teledermoscopy after three years is still very high. However, these cases were not histopathological proven. Also, fourteen ICD-10 diagnosis categories were given by the TD which were not pre-diagnosed by GPs at all. This could indicate that GPs are unfamiliar with these diagnoses. The results of this study reveal a potential learning effect of teledermoscopy on GPs’ skin disorder diagnosing expertise. An increase in the number of pre-diagnosed consultations by the GP, an increase in diagnosis sensitivity and a modest increase in positive predictive value after three years for specific diagnoses categories of teledermoscopy usage were seen. However, overall, GPs’ pre-diagnosis expertise of skin disorders appeared low in this study. Especially Benign neoplasms (D10-D36) and Other disorders of the skin and subcutaneous tissue (L80-L99) appear difficult to diagnose by the GPs since a prediagnosis for these disorders was often lacking. In this study, GPs were not obliged to fill in a pre-diagnosis, but the number of provided pre-diagnoses appears comparable to normal practice. In the study of Rijsingen et al. they assessed the quality of referral letters of GPs to the dermatologist of patients with skin tumors [4]. Their study showed that GPs do not always provide a diagnosis for suspicious lesions in referral letters to the dermatologist. A diagnosis was missing in 18.3% of the 4
RkJQdWJsaXNoZXIy MTk4NDMw