To examine whether admission-based predictors available within the electronic health record (EHR) can identify patients at increased risk for harmful diagnostic errors during hospitalization.
Approach:
Predictors Evaluated: The six prespecified predictors evaluated included clinical complexity, diagnostic uncertainty, and care team structure.
Key Findings:
Patients with two or more ambulatory encounters in the 14 days preceding admission had more than twice the odds of experiencing a harmful diagnostic error, while traditional markers of complexity did not show a statistically significant association.
Interpretation:
The association between repeated outpatient encounters and harmful diagnostic errors may indicate a risk signal rather than a causal mechanism.
Limitations:
The study was underpowered and vulnerable to false negative findings.
The distinction between prediction and causation remains unclear.
Conclusion:
The study highlights the transition from ambulatory to inpatient care.
