Dementia Defined Differently Across Hospitals, SNFs

Abstract

Close to half of Medicare beneficiaries diagnosed with dementia by a hospital find their diagnosis has changed to either mild or no cognitive impairment upon admission to a skilled nursing facility, a retrospective cohort analysis has shown. The study was conducted by Brian Downer, PhD, an assistant professor of rehabilitation sciences at the University of Texas Medical Branch in Galveston, and his colleagues (J Am Med Dir Assoc 2017;18:726–728). They used Medicare Part A and Minimum Data Set (MDS) 3.0 data samples recorded between 2013 and 2014 for nearly 2 million Medicare beneficiaries 65 years or older who were admitted to an SNF within 3 days of discharge from an acute hospital stay and had at least a 5-day MDS assessment. Data for individuals with delirium, stroke, or coma were excluded. Dementia was defined according to the appropriate ICD-9-CM codes. In all, dementia was diagnosed in the hospital for 13.4% of the total beneficiaries examined. These patients were typically men in their mid-80s with mental symptoms, aggressive behaviors, limitations on their activities of daily living, or comprehension issues as compared with peer beneficiaries who did not have a dementia diagnosis. These patients were also more likely than those without a dementia diagnosis to have been rated according to the Cognitive Performance Scale (CPS). About 92% of all beneficiaries with a hospital dementia diagnosis also had an active SNF dementia diagnosis, compared with about 13% of those with no hospital dementia diagnosis. About 60% of all beneficiaries were considered by SNFs at admission to be cognitively intact, whereas 22% were considered by SNFs to have mild impairment. Just over 15% were considered at SNF admission to be moderately impaired, and 3.4% were diagnosed as severely impaired. Among beneficiaries who were given a dementia diagnosis at hospital discharge, nearly 18% were considered by an SNF to be cognitively intact; nearly 26% were considered to have only mild cognitive impairment, 45% were reclassified as moderately impaired, and 11.3% were classified as severely impaired. About 65% of those without a hospital-based dementia diagnosis were considered cognitively intact upon SNF admission, but 21.4% were reclassified as having mild impairment, about 11% as moderately impaired, and 2.2% as severely impaired. The findings could indicate a need for standardized cognitive assessments that can be performed in a variety of contexts; for example, the MDS does not screen for mild cognitive impairment. This could have implications for care coordination in the elderly as well as for accountable care organizations that rely on population health data, because not all Medicare recipients receive or complete the BIMS. When that happens, a CPS score can be calculated using items from the MDS for patients, such as cognitive skills used for daily decision making, ability to make themselves understood, and the ability to feed themselves. “Beneficiaries classified as cognitively intact and mildly or moderately impaired by the Cognitive Function scale are frequently able to complete the BIMS, whereas those unable to make themselves understood or unable to complete the BIMS require a CPS score to be calculated,” the authors wrote. “These findings provide evidence that a hospital diagnosis of dementia might not always reflect cognitive status on admission to an SNF,” the authors wrote. Their findings suggest research also is needed into the degree of consistency across a variety of conditions commonly occurring in patients discharged to SNFs. Whitney McKnight is a freelance writer based in Chevy Chase, MD. For years it has no longer surprised me when a new nursing home admission has “severe” or “advanced” dementia on their hospital problem list, yet they turn out to be fully conversant, oriented to person and place, and able to discuss their medical problems in reasonable detail. Obviously, the hospital has a lower bar for calling someone “severely” demented than we do. We know very well that some emergency room physicians and hospitalists also seem to have very lax criteria for diagnosing (and treating) urinary tract infections or even urosepsis that to us would be considered asymptomatic bacteriuria, as well as dehydration, which I have seen as a diagnosis in patients with a blood urea nitrogen of 15, creatinine of 0.7, sodium of 139 and urine specific gravity of 1.010. It’s hard to understand why these diagnoses get assigned, and they sometimes become chart lore or serve as the basis for regulatory or civil actions, e.g., negligence lawsuits or survey deficiencies based on dehydration. Whenever possible, and as constructively as possible, we should attempt to educate our colleagues at the “big house” when our mutual patients are labeled with inappropriate diagnoses. —Karl Steinberg, MD, CMD, HMDC Editor in Chief