Outcomes of Common Major Surgical Procedures in Older Adults With and Without Dementia

Introduction Dementia is associated with poor outcomes after surgical procedures.1-4 However, the types of major surgical procedures commonly performed in older adults with dementia and the surgical outcomes have not been characterized in a national sample. This cross-sectional study was conducted to identify commonly performed major surgical procedures in patients with dementia and compare in-hospital mortality, home discharge, and length of stay (LOS) among patients with and without dementia.


Introduction
Dementia is associated with poor outcomes after surgical procedures. [1][2][3][4] However, the types of major surgical procedures commonly performed in older adults with dementia and the surgical outcomes have not been characterized in a national sample. This cross-sectional study was conducted to identify commonly performed major surgical procedures in patients with dementia and compare in-hospital mortality, home discharge, and length of stay (LOS) among patients with and without dementia.

Methods
We analyzed data from the Premier Healthcare Database, an all-payer, hospital-based, administrative and billing database that contains approximately 25% of inpatient admissions in the United States.
We identified adults 65 years or older who underwent major surgical procedures, as defined by the Agency for Healthcare Research and Quality, between January 1, 2016, and March 31, 2018. Dementia was defined using the Chronic Conditions Data Warehouse algorithm. 5 We estimated the odds ratios (ORs) and 95% CIs for in-hospital mortality and home discharge and the mean difference in LOS between patients with and without dementia using generalized estimating equations for logistic and linear regressions, respectively, to adjust for age, sex, race, admission source, Charlson Comorbidity Index, and procedure type and to account for clustering within hospitals. The outcome analysis was performed for the top 10 procedures combined and separately. Analyses were performed using SAS, version 9.4 (SAS Institute Inc). This study qualified for exemption from the Brigham and Women's Hospital Institutional Review Board; the Premier Healthcare Database provided the researchers with a deidentified data set that did not have patient or hospital identifiers. The Partners Human Research Committee waived patient informed consent because this study was minimal risk, the waiver of informed consent did not increase the risk to human participants, and the research could not be practically conducted without a waiver of informed consent. This study followed the Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) reporting guideline for crosssectional studies.

Results
The study population included 164 551 patients with dementia and 2 320 668 patients without dementia who underwent major surgical procedures at 804 hospitals. Patients with dementia were older (mean [SD] age, 81 [7] vs 74 [7] years), more likely to be women (57.4% vs 50.9%), more likely to be African American individuals (9.7% vs 7.0%), and more likely to have undergone an emergent operation (60.8% vs 25.8%) than those without dementia. Similar proportions of patients were treated at teaching hospitals (48.0% vs 47.3%) and urban hospitals (89.2% vs 89.8%). Of total surgical procedures, 6.4% were performed in patients with dementia ( Table 1). The most commonly performed procedure was hip or femur repair (16.5%) in patients with dementia and knee arthroplasty (8.6%) in patients without dementia.
Except for operations classified as other therapeutic procedures, the difference in LOS was the largest for breast and skin operations (mean difference, 2.59 days; 95% CI, 2.12-3.05 days).