Evaluation of a self-administered questionnaire on hand dermatitis.

The purpose of the study was to evaluate a self-administered questionnaire on hand dermatitis that was developed to identify persons with hand dermatitis in epidemiological studies. A total of 109 nurses were subject to dermatological examination of the hands within 1 month of returning the questionnaire. 2 types of questionnaire diagnoses were made: a 'symptom-based' diagnosis and a 'self-reported diagnosis'. These were compared to the medical diagnosis of hand dermatitis. The prevalence of hand dermatitis in the 12 months before the study, based on the medical diagnosis, was 18.3%. The prevalence according to the symptom-based diagnosis and the self-reported diagnosis was 47.7% and 17.4%, respectively. The sensitivity and specificity of the symptom-based diagnosis were 100% and 64%, respectively. It is concluded that the symptom-based diagnosis can be used as screening instrument for the detection of cases in large study populations, if followed by dermatological examination of persons with a positive diagnosis. The sensitivity and specificity of the self-reported diagnosis were 65% and 93%, respectively. It is concluded that the self-reported diagnosis can be used to obtain a rough estimate of the prevalence, although comparison of prevalence figures between study populations may be distorted due to a difference in reporting of hand dermatitis. The results of the study illustrate the size of the differences in prevalence estimates that may arise as a result of differences in the definition and method of diagnosing hand dermatitis.