Response to Letter to the Editor, regarding: Observation units as substitutes for hospitalization or home discharge

We appreciate the thoughtful comments of Drs. Suri and Baugh, which reiterate several important limitations of our study. ¹ We recognize that our study was an imperfect attempt to assess the clinical use of observation units. Chest pain patients were the subject of our published analysis because chest pain is the most common diagnosis in observation, ² so there were a large number of these patients in the NHAMCS dataset. Although not included in the manuscript due to smaller sample size, we applied the same analysis to syncope and cellulitis patients and found that the majority of these patients admitted to observation would have been discharged home had the unit not been available. While we agree that appropriateness of hospitalization is best decided by the bedside clinician, both clinical and non-clinical factors can influence clinician triage decisions. ^3,4 Our study suggests that the availability of an observation unit may have an effect on the variability in clinical decision making.

The recent growth in observation units is partly attributable to policy changes, including the Readmission Reduction Program and the Two Midnight Rule, that are intended to reduce overutilization of hospitalizations. Although our study is by no means definitive, it points to a possibility of an unintended consequence of such policies: the potential for overutilization related to observation units. We agree with Drs. Suri and Baugh that our results should not be over-interpreted to suggest that observation units are inherently problematic. Nonetheless, as our profession moves toward accountability for value provided, we must be conscious of the potential for overuse. We believe more definitive studies are needed to address both the benefits and unintended consequences of the growth in observation units.