Association Between Hospital Readmission and Acute and Sustained Delays in Functional Recovery During 18 Months After Elective Surgery: The Successful Aging after Elective Surgery Study

Abstract

Objectives: To examine the effect of hospital readmission on functional recovery after elective surgery in older adults.

Design: Prospective cohort of individuals aged 70 and older undergoing elective surgery, enrolled from June 2010 to August 2013.

Setting: Two academic medical centers.

Participants: Community-dwelling older adults (N = 566; mean age ± standard deviation 77 ± 5) undergoing major elective surgery and expected to be admitted for at least 3 days.

Measurements: Readmission was assessed in multiple interviews with participants and family members over 18 months and validated against medical record review. Physical function was assessed according to ability to perform instrumental activities of daily living (IADLs) and activities of daily living (ADL), Medical Outcomes Study 12-item Short-Form Survey Physical Component Summary score, and a standardized functional composite.

Results: Two hundred fifty-five (45%) participants experienced 503 readmissions. Readmissions were associated with delays in functional recovery in all measures of physical function. Having two or more readmissions over 18 months was associated with persistent and significantly greater risk of IADL dependence (relative risk (RR) = 1.8, 95% confidence interval (CI) = 1.5-2.3) and ADL dependence (RR = 3.3, 95% CI = 1.7-6.4). Degree of functional impairment increased progressively with number of readmissions. Readmissions within 2 months resulted in delayed functional recovery to baseline by 18 months, and readmissions between 12 and 18 months after surgery resulted in loss of functional recovery previously achieved.

Conclusion: Readmission after elective surgery may contribute to delays in functional recovery and persistent functional deficits in older adults.

Keywords: aging; cohort study; physical function; readmission; surgery.

Figure 1

Change from baseline functional recovery with time, to 18 months. Change in the SAGES functional composite is given in panel A, with horizontal gray line corresponding to zero change. The proportions of individuals exhibiting no IADL or ADL dependencies are depicted in panels B and C, respectively, with the zero line corresponding to no change from baseline. Mean change in SF-12 PCS is depicted in panel D.

Figure 2

Model-estimated probability of IADL independence as a function of time for an individual matching typical SAGES demographic and health history profile, under four different hypothetical hospitalization experiences. Dark black symbols and lines describe the trajectory associated with no readmissions, while other lines depict various patterns of readmission over 18 months. Visits at which hospitalizations are recorded are demarcated by triangles with the same shading as the corresponding trajectories.

Figure 3

SAGES functional composite at baseline and 18 months follow-up, among individuals who have 18 months’ data, stratified by number of readmissions experienced to 18 months. Means and 95% confidence intervals are shown. Differences at baseline are modest and statistically nonsignificant, but are greater at 18 months.