Dementia-specific risks of scabies: retrospective epidemiologic analysis of an unveiled nosocomial outbreak in Japan from 1989-90.

Background: Although senile dementia patients in long-term care facilities are at leading risk of scabies, the epidemiologic characteristics of this disease have yet to be fully clarified. This study documents the findings of a ward-scale nosocomial outbreak in western Japan from 1989-90, for which permission to publish was only recently obtained.

Methods: A retrospective epidemiologic study was performed to identify specific risk factors of scabies among patients with dementia. Analyses were based on a review of medical and nursing records. All inpatients in the affected ward at the time of the outbreak were included in the study. Observational and analytical approaches were employed to assess the findings.

Results: Twenty of 65 inpatients in the ward met the case definition of scabies. The outbreak lasted for almost 10 months and as a result, the spatial distribution of infections showed no localized patterns in the latter phase of the outbreak. The duration of illness significantly decreased after initiation of control measures (P = 0.0067). Movement without assistance (Odds Ratio [OR] = 11.3; 95% Confidence Interval [CI]: 2.9, 44.8) and moving beyond the room (but within the ward) (OR = 4.1; 95% CI: 1.4, 12.5) were significantly associated with infection, while types of room (Western or Japanese) and sleeping arrangement (on beds or futons laid directly on the floor) appeared not to be risk factors.

Conclusion: Univariate analysis demonstrated the importance of patients' behaviours during daily activities in controlling scabies among senile dementia patients. The findings also support previous evidence that catching scabies from fomites is far less common. Moreover, since cognitive disorders make it difficult for individuals to communicate and understand the implications of risky contacts as well as treatment method, and given the non-specific nature of individual contacts that are often unpredictable, real-time observations might help improve control practices.