New Study: CMS’s New 2-Midnight Rule Increases Hospital’s Use of Observation Status

Final rules published in August 2013 by the Centers for Medicare & Medicaid Services (CMS) created time-based rules for determining inpatient status in an acute-care hospital.[1]  Under the rules' two-midnight presumption, a physician should order an inpatient admission if the physician expects that the patient's stay in the acute care hospital will be at least two midnights.[2]  In the preamble to the final rules, CMS expresses the hope that the new rules will "reduce the frequency of extended observation care when it may be inappropriately furnished."[3]<  That hope seems unlikely to be realized.

Applying the two-midnight rule retrospectively to all observation and inpatient encounters at the University of Wisconsin Hospital and Clinics between January 1, 2012 and February 28, 2013, physician researchers with the University of Wisconsin School of Medicine and Public Health found that application of the two-midnight rule would actually have increased the hospital's use of observation status.[4]  Looking only at Medicare beneficiaries treated at their hospital in the 14-month period, the physicians found that 7.4% of inpatient encounters would have been switched to outpatient if the two-midnight rule had been in effect. 

The study found that non-clinical criteria such as time of patient appearance at the hospital and day of the week would affect patient status under the new rule.  The physicians found that patients admitted after 4:00 p.m. would have been admitted as inpatients 31.2% of the time, while patients arriving at the hospital before 8:00 a.m. would have been designated inpatients only13.6% of the time under the new rule.  Similarly, patients arriving at the hospital on a weekend would have been admitted to inpatient status 26.5% of the time, compared to 22.6% of patients admitted on a weekday.  The physicians concluded, "[non-clinical] external factors, such as the time of day and specific day (weekday vs. weekend) of hospitalization, impact the likelihood of achieving a ≥2-midnight stay." 

Finding little overlap in the diagnosis codes for their short-stay inpatients and observation patients, the physicians concluded that inpatients staying less than 2 midnights "are [clinically] different from observation patients" at their hospital, contrary to the Inspector General's finding that short-term inpatients and observation patients are clinically the same.[5]  This finding at the University of Wisconsin Hospital led the researchers to conclude that "[less-than-2]-midnight inpatients should not be arbitrarily reclassified as observation based solely upon LOS [length of stay]" as the new rule suggests.

Finding that most diagnosis codes within observation "were the same regardless of LOS," the physicians concluded, "LOS does not reliably differentiate clinically different observation populations that merit different insurance coverage (Medicare Part A for ≥2-midnight encounters, Medicare Part B for <2-midnight encounters)."

The researchers opined that loss of "billable inpatient encounters under the new rules" might lead hospitals to change practices in order to increase admissions so that they exceed the two-midnight threshold.  The physicians described the changes as "defensible under audit given the ambiguities of the rule . . . surrounding time of day of admission, weekend and transfer hospitalizations, and the fact that common observation ICD-9 codes are similar across LOS."  They conclude that the unintended consequence of the two-midnight rule might be to "drive up hospital LOS, reduce efficiency, and increase the overall cost of care."  They call on CMS to return to the "original intent" of observation status – "to determine if a patient can safely return home after a short period of additional care" – and to classify as full inpatients all patients "beyond this narrow scope" whose conditions require an intensity and level of service provided by an acute care hospital.

For more information on the problem of observation status, see: https://www.medicareadvocacy.org/medicare-info/observation-status/.