Hospital-Level Analysis of proportion of Patients with Alzheimer’s and Dementia on 30-Day Total Episode Payments and Readmissions

Objective: Surgical procedures are adversely associated with quality of life and well-being of patients with Alzheimer’s and dementia (AD), and cost of care. Among all patients seen for surgery, patients with AD are amongst the most readmitted patients to hospital. There has been a paucity of evidence on hospital-level proportion of patients with AD on readmission risk and cost. We hypothesized that hospitals with higher proportions of AD patients are more likely to have higher readmission rates and ultimately, total 30-day episode payments.

Study Population: We utilized administrative claims between January 2012 and June 2017 from the Michigan Value Collaborative (MVC), a Blue Cross Blue Shield of Michigan (BCBSM) Collaborative Quality Initiative (CQI) including BCBSM Preferred Provider Organization (BCBSM-PPO) and Medicare Fee-For-Service (FFS) patients across 31 different medical and surgical services for 77 hospitals in the Michigan. We identified all patients with AD with an evidence of a diagnosis code via any healthcare encounter throughout their enrollment during the study period.

Methods: Using the Medicare Fee Schedule to perform price standardization, 30-day episode payments were calculated. All payments were risk adjusted using multivariable logistic regression using Hierarchical Condition Categories (HCCs), age, gender, insurance type, and prior 6 month payments. To account for potential skew in the standardized, risk adjusted payment distribution, payment winsorization was performed at the 99^th and 1^st percentiles. We calculated hospital-specific unadjusted readmission rates, proportion of all patient episodes with evidence of an AD patient, and proportion of aggregate hospital-specific episode payments that were related to an AD patient. All hospitals were ranked by their readmission rate and Pearson’s correlation coefficients were calculated to examine associations between readmission rate, AD hospital-specific proportions, AD proportion of total episode payments, and total 30-day episode payments.

Results: There were significant correlation among 30-day readmission rates, hospital-specific proportion with AD, and 30-day total episode payment. Patients treated in hospitals in the top quartile for AD patients were 1.79 times more likely (95% CI: 1.534, 2.089) to be readmitted within 30 days compare to patients in the lowest AD proportion quartile. Hospital-specific 30-day readmission was significantly associated with hospital-specific proportion of AD patients and 30-day total episode payments (Pearson’s correlation of 0.73 and 0.64, p < 0.0001, respectively). Hospital-specific 30-day total episode payment was somewhat correlated with proportion of patients with AD (Pearson’s correlation = 0.46, p < 0.0001). There was a substantial range in the difference between AD and non-AD patients in risk adjusted hospital-specific 30-day total episode payments across hospitals (Range: -$386, $4,847).

Conclusion: Hospitals’ readmission rates as a result of surgical encounters and inpatient admissions are strongly associated with the proportion of admitted patients with AD. In defining hospitals’ readmission penalties, the Center for Medicare and Medicaid Services should adjust for proportion of patients with diagnosis of AD. Additionally, critical examination of hospital-specific factors (modifiable and non-modifiable) affecting differences in total episode payments between those with and without high proportion of AD patients could create efficiency in treatment and cost containment.