Medicare Observation Stays: When is a Hospitalization Not a Hospitalization?

July 2012

A recent USA Today article by Susan Jaffe examines hospitals’ growing use of observation stays in lieu of fully admitting Medicare patients.  This phenomenon has drawn the attention of journalists and consumer advocates because of the significant consequences of observation stays for beneficiaries’ out-of-pocket costs.  In a recently completed HCFO grant, Lan Zhao, Ph.D. with Social and Scientific Systems, has also investigated the frequency and implications of observation stays.

Observation stays are hospital visits that provide physicians with the opportunity to evaluate whether a patient should be admitted for an inpatient stay. They can take place in an inpatient setting, but because observation stays are considered outpatient services, they are reimbursed under Part B of Medicare rather than Part A.  The Centers for Medicare and Medicaid Services recommends that observation stays should last no more than 48 hours. However, the agency does not require that hospital admit or discharge patients after that period.

Under Part A, patients usually pay only a $1,156 deductible (in 2012) for each hospitalization.  In an observation stay, however, they are liable for the Part B copayments for each service they receive.  Furthermore, observation stays do not count toward the three-day prior hospitalization requirement for Medicare coverage of inpatient care at a skilled nursing facility (SNF). Jaffe’s USA Today article focuses particular attention on patients’ pharmaceutical bills noting that because Medicare does not necessarily pay for observation stay patients’ maintenance drugs for chronic conditions like diabetes or high blood pressure, hospitals are free to charge patients any amount.  Her interviews also suggest that many patients do not realize they were not formally admitted until they receive their hospital bills.

In her HCFO study, Zhao compared 2001 and 2006 Medicare claims data, confirming that there was a significant increase in the number of observation stays (in total and per beneficiary) with some substitution of observation for 1-day inpatient stays.  She also found evidence that the length of stay for observation patients increased over time.  Looking at out-of-pocket expenses, Zhao found the financial burden of observation stays fell on a small percentage of all patients receiving these services, but that the expenses incurred by those beneficiaries could be substantial.  Less than 10 percent of patients with observation stay had out-of-pocket expenses that were higher than the deductible they would have paid for an inpatient hospitalization. While the median out-of-pocket expense for observation stay patients was just under $1,200 in 2006, some beneficiaries had expenses of almost $20,000.

Because of the potentially high out-of-pocket burden for observation stay patients subsequently admitted to a SNF, Representative Joe Courtney (D-CT) introduced legislation in 2010 (HR 5950) that would change Medicare policy to allow any part of an observation stay exceeding 24 hours to count toward the requirement of a three-day hospital stay for Medicare coverage of a SNF stay.  Zhao estimated that this policy change would have cost Medicare an additional $50 million in 2006.  Additional information about Zhao’s study can be found on HCFO’s website.