Study finds no correlation between reduced readmissions, increased observation stay rates

Although hospital readmissions rates began falling faster after the implementation of the Patient Protection and Affordable Care Act (ACA) (P.L. 111-148), there was no observed association between increased use of observation services and reduction in readmissions. In a New England Journal of Medicine (NEJM) article, researchers found that hospitals did not attempt to achieve reductions under the Hospital Readmissions Reduction Program (HRRP) by increasing the use of observation stays as feared by program critics.


Section 3025 of the ACA established the HRRP, which reduces payments made to acute care hospitals that have excess readmissions for patients with certain conditions. Readmissions are costly for the Medicare program, adding an estimated $17 billion in expenditures that the program considers avoidable. The HRRP penalized hospitals for having higher readmission rates than expected within 30 days of discharge. The conditions first included were acute myocardial infarction, heart failure, and pneumonia. Later, total hip or knee replacement and chronic obstructive pulmonary disease (COPD) were added.

Observation usage

Some believed that hospitals would simply place patients who returned in observation status instead of readmitting them to avoid the penalty. The researchers reviewed stays for the initial three conditions included in the program and identified readmissions after 30 days of discharge, as well as whether observation services were used within 30 days. The results showed that while monthly readmission rates were decreasing before the ACA, they began decreasing faster after enactment for both target and nontargeted conditions, particularly in the first six months. Prior to the ACA, observation service use was rising “significantly” and in a similar pattern for both targeted and nontargeted conditions, and continued to rise through the analysis period.

The study revealed that despite a rise in observation services, there was no correlation between the change in the readmission rate and the use of observation status. In addition, although readmission rates fell quickly at first following the HRRP, the article’s authors theorized that hospitals made significant changes during this period but were unable to keep the reduction rate high long-term. They also postulated that while observation status use continued to rise, this may have occurred due to factors unrelated to the HRRP, such as confusion over what recovery audit contractors would deem an appropriate inpatient stay.