The PRONE score: an algorithm for predicting doctors' risks of formal patient complaints using routinely collected...

Area of Science:

• Medical Regulation
• Health Services Research
• Clinical Risk Management

Background:

• Medicolegal agencies are typically reactive, addressing substandard care after it occurs.
• Predicting individual clinician medicolegal risk is challenging, limiting proactive intervention.
• There is a need for reliable tools to identify doctors at risk of repeated patient complaints.

Purpose of the Study:

• To develop a simple, reliable scoring system for predicting Australian doctors' risk of recurrent patient complaints.
• To create a predictive algorithm applicable at the individual doctor level.

Main Methods:

• A national sample of 13,849 formal patient complaints against 8424 doctors in Australia was analyzed.
• Data from a 12-year period, collected by state health service commissions, were used.
• Multivariate logistic regression identified predictors of subsequent complaints within two years of an initial complaint.

Main Results:

• The PRONE (Predicted Risk Of New Event) score, a 22-point system, was developed.
• Key predictors include specialty, sex, number of prior complaints, and time since the last complaint.
• The PRONE score demonstrated good predictive performance (c-statistic=0.70), with strong validity and reliability.

Conclusions:

• The PRONE score is a valid method for assessing individual doctors' risk of attracting recurrent complaints.
• This tool can enable regulators to target quality improvement interventions proactively.
• The approach is potentially replicable in other complaint or malpractice handling agencies.