Does Observation Status Undercut Claims that Hospital Readmissions Have Been Reduced?  MedPAC Says No, Research Says Yes.

As required by the Affordable Care Act, the Medicare Hospital Readmissions Reduction Program (HRRP) reduces Medicare payment rates, by up to 3%, to hospitals that readmit patients with certain specified conditions within 30 days.[1]  The financial penalty is applied only when the patient’s initial hospital stay and subsequent hospital stay are both identified as inpatient.  However, if a patient is classified as an outpatient in observation status for either the initial or subsequent hospital stay, the penalty is not imposed. 

Does observation status affects claims about reduced hospital readmissions?  Research focusing on commercial payers finds that observation status completely offsets gains made in reducing hospital readmissions; the Medicare Payment Advisory Commission (MedPAC) finds that observation status has little effect on rehospitalization.

Research Study

A review of patient-level claims data, 2007-2015, for 350 commercial payers finds that apparent reductions in hospital readmissions are offset by hospitals’ increased use of observation status.  In other words, researchers find no change in the rate of hospital readmissions when the data include observation patients who return to the hospital for care within 30 days.[2]

Reducing rehospitalization of patients is a public policy goal that is viewed as reflecting a measure of quality.  However, as noted above, rehospitalization is defined as inpatient hospital status followed within 30 days by another inpatient hospital stay.  When either or both hospital stays are not classified as inpatient, a patient returning to the hospital for care is not considered to have been rehospitalized.  Accordingly, observation status for either the initial or subsequent hospital stay, or both, leads to patients not being counted in rehospitalization data.  Researchers document that excluding patients in observation status, which is considered an “outpatient” status, skews readmission data.

Researchers found that although the rate of rehospitalization for inpatients declined from 17.8% to 15.5% (a 2.3% decrease) between 2007 and 2015, “the rate of readmission after an observation stay increased from 10.9% to 14.8%” (a 3.9% increase) during the same nine-year period.[3]  When the researchers counted observation patients’ initial and returning stays as  hospital admissions, the reduction in readmissions declined to 1.2 per 1000 hospitalizations, “which suggests that there has been virtually no change in all-cause readmissions.”[4]

Researchers offer various explanations for why patients in observation return to the hospital for additional care.  A first explanation is that the severity of illnesses in observation patients has increased over time.  This explanation is consistent with researchers’ finding that there has been a large increase in the number of patients classified as observation, rather than inpatient.  (Indeed, the researchers found that in 2015, 14% of commercially-insured patients with emergency department visits were hospitalized and that 57% were admitted as inpatients while 43% were hospitalized as observation patients.[5])

A second explanation is that hospitals do not offer comprehensive care-coordination or discharge planning services to patients in observation status, as they offer inpatients.  A third explanation is that hospitals want to discharge observation patients in 48 hours, leading to premature discharges.[6]  Finally, hospitals are subject to regulatory consequences for engaging in poor discharges for inpatients while they “face no repercussions for unsafe or poorly handled discharges from observation care that may lead to repeat hospitalizations.”[7]

The researchers conclude that “excluding observation stays from readmission measures means that information on more than 400,000 additional unscheduled hospitalizations will be omitted from readmission measures each year.”[8]  They recognize that all patients need high-quality care transitions, whether they are inpatients or observation patients, and that repeated observation stays may reflect similar concerns about quality of care transitions as inpatient readmissions.[9]

MedPAC Report

As mandated by the 21^st Century Cures Act, MedPAC examined how HRRP affected readmissions, observation stays, and emergency department (ED) visits (as well as mortality rates).[10]  MedPAC acknowledges that “The decline in readmission rates coincided with increases in the rate of observation and the rate of ED use”[11] and that “The joint timing of a decline in inpatient admissions with an increase in observation stays and ED visits suggests that there was some substituting of outpatient care for inpatient care.”[12]  Nevertheless, on several bases, MedPAC concludes that HRRP has not caused increased use of observation status.

Citing national data from 2010 to 2016, MedPAC reports, “The faster growth in ED visits and observation stays for those without a recent admission to the hospital allows us to conclude that the readmission policy was not likely the driver behind the ED and observation growth experienced.”[13]  MedPAC suggests that other policies that went into effect during the same general period as HRRP – audits by Recovery Audit Contractors and the two-midnight rule – also influenced the increased use of observation stays.[14]

MedPAC also looks at data at the hospital level.  Although “hospitals with above-average declines in readmissions did tend to have increases in observation and ED use,” MedPAC concludes that “only a small share of the increase in observation and ED use was related to the HRRP.”[15]  Additional evidence at the hospital level cited by MedPAC is that although rehospitalization rates for conditions covered by HRRP were greater than rehospitalization rates for conditions not covered by HRRP, there was no difference in observation stays or ED visits for patients in covered and non-covered conditions.[16]

Conclusion

Research with commercial payers finds that observation status outweighs claimed reductions in rehospitalizations.  MedPAC recognizes some correlation between HRRP and increases in observation status and ED use, but discounts arguments of cause and effect.  This seems counterintuitive.

There is no question, however, that observation status has a significant negative effect on Medicare beneficiaries, especially those who need care in a skilled nursing facility after they leave the hospital and are required to pay for that care entirely out-of-pocket.[17]

The Improving Access to Medicare Coverage Act of 2017 (S. 568/H.R. 1421), which counts all time in the hospital, whether “inpatient” or “outpatient,” would resolve the problem of observation for most beneficiaries, especially those who are most vulnerable.

June 21, 2018, T. Edelman