Potential Diagnostic Error for Emergency Conditions, Mortality, and Healthy Days at Home

There is limited evidence regarding the frequency of diagnostic errors and outcomes associated with them in emergency care in the US.

To examine rates of potential diagnostic errors and associated clinical outcomes among Medicare beneficiaries with emergency hospitalizations.

This cohort study examined a national sample of fee-for-service Medicare beneficiaries aged 65 years or older with emergency hospitalizations for 10 high-risk conditions from 2016 to 2019. Data were analyzed from December 20, 2022, to April 16, 2025.

Potential diagnostic errors in the emergency department (ED).

The percentage of hospitalizations with an ED discharge (potential diagnostic error) in the preceding 9 days was calculated. From this observed ED visit rate, an adjusted rate of potential diagnostic errors was calculated by subtracting the rate of background ED discharges among a comparison cohort of beneficiaries with similar baseline risk. To examine the association between potential diagnostic error and outcomes, multivariable linear regression models were specified for the outcomes: 30-day mortality and healthy days at home (HDAH). Potential diagnostic error was the exposure, and the models were adjusted for diagnosis, patient demographics, and chronic conditions.

Among 302 837 emergency hospitalizations, 13 550 (4.5%) were preceded by an ED discharge (ie, potential diagnostic error) within 9 days. For the comparison group with similar baseline risk, 1.2% (95% CI, 1.2%-1.3%) had an ED discharge in a random 9-day period. After accounting for this background use, the adjusted potential diagnostic error rate was 3.2% (95% CI, 3.1%-3.3%) for all conditions in aggregate but ranged from 2.1% (95% CI, 1.9%-2.4%) for spontaneous intracranial hemorrhage to 15.6% (95% CI, 11.9%-19.3%) for spinal abscess. Having a potential diagnostic error was associated with higher adjusted 30-day mortality for emergency admissions (15.7% vs 14.9%; point absolute difference, 0.8 percentage points, 95% CI, 0.2 to 1.4 percentage points; P = .007) as well as fewer 30-day HDAH (13.5 days vs 15.0 days; difference, -1.4 days; 95% CI, -1.6 to -1.3 days; P < .001), although associations varied among conditions.

In this cohort study of Medicare beneficiaries aged 65 years and older with emergency hospitalizations, the adjusted potential diagnostic error rate was modest overall but varied by condition. Potential diagnostic errors were associated with worse outcomes for emergency conditions overall, with associations again varying among individual conditions.