Identifying and acting on potentially inappropriate care? Inadequacy of current hospital coding for this task

Area of Science:

• Healthcare data quality
• Medical coding accuracy
• Health services research

Background:

• Australian healthcare policy increasingly relies on large government datasets to identify inappropriate care.
• Previous criticisms highlight poor correlation between government data and professional body data.
• Government data is collected retrospectively by clerical staff, unlike contemporaneous clinical data.

Purpose of the Study:

• To assess the accuracy of official hospital coding data for hyperbaric services.
• To compare government-collected coding data with actual clinical activity in a tertiary hospital.

Main Methods:

• Reviewed all hyperbaric-relevant coding data submitted by Royal Hobart Hospital (2010-2011).
• Compared coding data against original source documents detailing actual hyperbaric unit activity.
• Identified and categorized errors in diagnoses and procedures.

Main Results:

• 70% of patients treated with hyperbaric oxygen had coding errors (diagnoses and/or procedures).
• Identified errors included missing patients/treatments, additional patients/treatments, and incorrect codes.
• Coding errors were observed in other specialties, suggesting a systemic issue not unique to hyperbaric medicine or this institution.

Conclusions:

• Current government datasets are compromised and unfit for informing healthcare policy.
• Using these flawed datasets for policy debates is inappropriate.
• Urgent clinical engagement with hospital coding departments is recommended to improve data accuracy.