Increase in the Admission of Stroke Mimics after Change in Emergency Shifts

In 2010, changes were made to the shift pattern of neurology residents for cover in the Emergency Department at a university hospital. This resulted in a decrease in the number of emergency hours worked by neurology specialists and allowed for a natural quasi-experiment.

We aimed to evaluate if changes to the number of emergency hours worked by neurology residents and specialist neurologists, (intervention) altered the number or pattern of admitted stroke mimics (SMs).

Observational retrospective study from January 2007 to December 2013. Time of intervention was set as August 2010. We used a segmented linear regression - ARIMA - to evaluate changes in the temporal pattern of admitted SMs. A statistical correlation between the number of emergency hours worked by neurology residents and the number of admitted SMs was calculated using the Pearson Correlation Coefficient.

Of the 2,552 patients admitted to the stroke unit, 290 were SMs (11.4%). After August 2010, there was an increase in the number of admitted SMs (p = 0.003). After 2010, the most frequent SM diagnosis changed from a psychiatric condition to peripheral vertigo. A positive correlation was found between the number of hours worked primarily by neurology residents and the number of admitted SMs (Pearson correlation coefficient = 0.94; p = 0.002).

Changes in the pattern of Emergency Department shifts were associated with an increase in the rate of admitted SMs and with a higher number of mimics with a final diagnosis of peripheral vertigo.