医学
倾向得分匹配
队列
入射(几何)
查尔森共病指数
选择偏差
关节置换术
急诊医学
队列研究
共病
回顾性队列研究
物理疗法
内科学
外科
物理
光学
病理
作者
Chloe C. Kimball,Christine Nichols,Joshua G. Vose
出处
期刊:Orthopedics
[Slack Incorporated (United States)]
日期:2019-11-26
卷期号:43 (1): 36-41
被引量:22
标识
DOI:10.3928/01477447-20191122-01
摘要
Increased focus on reducing the cost of total knee arthroplasty (TKA) has driven greater interest in performing the procedure in an outpatient setting. This study used exact matching to compare clinical and economic outcomes following TKA in an outpatient vs rapid recovery inpatient setting. This study used a nationally representative commercial database. Patients were grouped into 2 cohorts: same-day outpatient TKA (surgery center or hospital outpatient) or short-stay inpatient TKA ("rapid recovery"; length of stay 1 day or less). Only patients discharged home under self-care or with health care were included. Exact 1:1 matching was performed on clinical and demographic characteristics to control for potential case-selection bias by choice of care setting. Prior to matching, 969 outpatients and 8101 rapid recovery inpatients met selection criteria. The outpatient cohort was younger (median age, 58 vs 61 years), predominantly female (56% vs 51%), and less comorbid (Charlson Comorbidity Index score of 0: 84.2% vs 74.0%) vs the rapid recovery cohort. Post-match, 863 patients were available in each cohort. The outpatient cohort exhibited a significantly lower incidence of opiate use (80.4% vs 90.7%; P<.001) and minor complications (2.8% vs 5.8%; P=.002). Incidence of major complications (5.2% vs 6.7%, P=.173) and 90-day readmissions (5.1% vs 7.3%, P=.064) were equivalent. The outpatient median 90-day episode payment was $6824 lower (22%) per patient ($24,749 vs $31,573, respectively; P<.001). This study suggests that among carefully selected patients undergoing outpatient TKA, outcomes are equivalent, if not improved, at a lower payor cost compared with a rapid recovery inpatient setting. [Orthopedics. 2020; 43(1):36-41.].
科研通智能强力驱动
Strongly Powered by AbleSci AI