A Clinical Prediction Model to Assist Screening Patients for Scabies in Primary Care

Area of Science:

• Dermatology
• Epidemiology
• Clinical Prediction Modeling

Background:

• Scabies diagnosis in primary care can be challenging.
• Accurate and accessible diagnostic tools are needed to support clinical decision-making.

Purpose of the Study:

• To develop and validate a clinical prediction model for scabies diagnosis.
• The model aims to assist in primary care screening.

Main Methods:

• Logistic regression analysis of scabies prevalence survey data.
• Predictors selected using best subsets regression and Akaike's information criterion.
• Internal validation performed using bootstrap resampling.

Main Results:

• Age and symptoms of itch (child and contacts) predicted clinical scabies diagnosis.
• Household insect bites, sores, or blisters predicted qPCR-based diagnosis.
• Models demonstrated good accuracy, discrimination, and calibration.

Conclusions:

• Logistic models incorporating itch symptoms accurately predict scabies status.
• These models show potential for improving scabies screening in community and primary care.
• The "simultaneous itch in the family" is a key predictor.