Premium
Association Between Estimated Mortality Risk and Measured Quality of Care in Older Adults
Author(s) -
Gotanda Hiroshi,
Ganz David A,
Wenger Neil S
Publication year - 2018
Publication title -
journal of the american geriatrics society
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 1.992
H-Index - 232
eISSN - 1532-5415
pISSN - 0002-8614
DOI - 10.1111/jgs.15491
Subject(s) - medicine , confidence interval , reimbursement , medical expenditure panel survey , health care , demography , relative risk , risk assessment , absolute risk reduction , gerontology , health insurance , computer security , sociology , computer science , economics , economic growth
Quality measures based on processes of care (e.g., breast cancer screening) are frequently used to compare performance of providers and healthcare organizations in a value‐based reimbursement system, but using these measures to make performance comparisons could penalize clinicians caring for older adults with multiple comorbidities. These clinicians may more frequently withhold recommended care based on limited prognosis, patient preference, or potential adverse effects and, as a result, may incur financial penalties because of poor measured performance. Alternatively, they may need to perform additional administrative work to exclude patients from quality measurement. To examine the potential performance handicap associated with caring for complex populations, we cross‐sectionally examined the association between broad outpatient processes of care and 5‐year mortality risk in a nationally representative older adult sample (≥ 65) from the 2002 to 2015 Medical Expenditure Panel Survey (n=26,006). In adjusted analyses, performance score based on process‐of‐care measures was significantly worse in the high‐risk group (5‐year mortality risk ≥50; adjusted performance score 67.4%; adjusted absolute difference 6.7%, 95% confidence interval (CI)=5.3–8.1%, p<.001) and in the intermediate‐risk group (5‐year mortality risk 25–49%; adjusted performance score 71.7%; adjusted absolute difference 2.4%, 95% CI=1.7–3.1%, p<.001) than in the low‐risk group (5‐year mortality risk <25%); adjusted performance score 74.1% following <25%. The trend was overall constant across types of care processes and was more prominent in the younger age brackets of our sample. These results suggest that use of process‐of‐care measures can lead to disproportionate burden for clinicians and organizations primarily taking care of complex populations.