When it comes to payment incentives in U.S. healthcare, it seems as though every step in the right direction also leads to some strange sort of unintended consequence. The publication of a new study in The New England Journal of Medicine seems to offer a case in point. “Excluding Observation Stays from Readmission Rates—What Quality Measures Are Missing,” authored by Amber K. Sabbatini, M.D., M.P.H., and Brad Wright, Ph.D., in the May 31 issue of NEJM, sheds light on a strange corner of hospital care delivery.
The subject is complex and nuanced, and has to do with the decision to admit a patient for an observation stay in a hospital, rather than a normal inpatient stay. Observation stays are treated as outpatient events, and compensated for as such, at lower rates than regular inpatient stays; but at the same time, observation stays are also not counted as admissions for purposes of counting readmissions that might come later. And therein lies the rub.
As Drs. Sabbatini and Wright note, “Readmissions that occur soon after hospital discharge are thought to reflect the quality of care transitions in particular. Policies that aim to reduce readmissions have thus had the beneficial effect of offering incentives to health systems to pay greater attention to care coordination and discharge processes for patients leaving the hospital. Despite some evidence that these efforts are working—readmissions have decreased for both conditions targeted by the HRRP and other conditions—we believe current readmission measures are missing a key component of the quality equation by failing to consider the effect of observation stays.”
Indeed, the authors note, “Observation stays represent a substantial and growing proportion of unscheduled U.S. hospital admissions. Hospital observation was designed to provide a period of treatment or monitoring for patients who are neither well enough to be immediately discharged nor clearly in need of inpatient care, usually after an emergency department (ED) evaluation. Yet as payers have discouraged short inpatient hospitalizations in an effort to reduce low-value care, providers are increasingly relying on observation stays as an alternative. Although some hospitals use specific care pathways and dedicated units for observation care, in most cases patients hospitalized under observation are treated on the same general medical wards, have similar medical needs, and receive equivalent clinical services, even though observation stays are billed as outpatient services.”
And, they note, “The increasing use of observation stays has important ramifications for the assessment of hospital quality using readmission measures. First, observation stays are not considered index events — patients hospitalized under observation and then discharged are not followed to determine whether and when they are readmitted. Second, because the observation stays are considered outpatient services, such stays among patients who have recently been hospitalized do not count toward a hospital’s 30-day readmission rate. Therefore, unscheduled hospitalizations billed as observation stays are missing from both the numerator and the denominator of the readmission-rate equation, which omits critical information about the quality of care transitions for many patients hospitalized for acute conditions.”
Studying outcomes from observation stays versus normal inpatient stays, they found that,” In 2015, a total of 14 percent of patients with index ED visits were hospitalized. Of those patients, 57% were admitted as inpatients and 43 percent were hospitalized under observation. In keeping with prior research, we found that the rate of hospital readmission within 30 days after discharge from an inpatient hospitalization decreased from 17.8 percent to 15.5 percent (a 2.3 percentage point absolute decrease, or a 12.9 percent relative decrease) during the 9-year study period (see graph). However, the rate of readmission after an observation stay increased from 10.9 percent to 14.8 percent (an increase of 3.9 percentage points, or 35.8 percent). Although the majority of readmissions after observation stays were for inpatient care, the increase in readmissions appears to be driven by repeat observation stays, which nearly doubled in frequency, from 3.6 percent in 2007 to 6.9 percent in 2015.”
The bottom line? There is a big set of unintended consequences here, as provider leaders shift some admissions to observation stays, in an effort to lessen their chances of being punished for excessive avoidable readmissions. But, as these researchers point out, the coordination of care turns out to be, not surprisingly, less rigorous and robust in observation stays, which means that some of those patients ultimately do end up back in regular inpatient hospital stays, at slightly higher rates. And that is, in practical terms, a conundrum.
As the authors state, “We believe the need for high-quality care transitions after an unscheduled hospitalization does not depend on whether a hospital can bill for that admission as an observation stay or an inpatient admission. All patients with an acute condition require timely and coordinated care. Moreover, there is no reason to think that a repeat observation stay is any less preventable or less reflective of the quality of care transitions than an inpatient readmission. Although repeat observation stays cost payers less than inpatient readmissions, they still represent excess costs for the health care system and are meaningful for patients.”
What’s more, they say, “Both public and private payers are making substantial investments in value-based care arrangements, and hospital readmissions have been an important focus of this movement.1 Yet several questions remain regarding the best way to handle observation stays in assessing readmissions. For example, should observation stays and inpatient admissions be combined in one measure, or would separate measures be more appropriate? Are there any potential unintended consequences of including observation stays in readmission measures? Would an analysis of hospital-specific, risk-adjusted readmission rates produce similar results? Although additional research is needed to understand and address the effect of observation stays on readmission measures, our findings indicate that a large and increasing proportion of hospital care transitions are falling out of the current quality equation.”
So it will be important, going forward, for clinician and administrative leaders in hospital organizations, to begin to use data to analyze their outcomes around observation stays versus regular inpatient stays; and they will need to rethink their patterns of observation-stay admission, in order to better coordinate care for their patients and improve outcomes. In other words, there really are no shortcuts—in hospital care management, or anything else.