The Effect of Rehabilitation on 3-Year Readmission and Mortality Risk in Elderly Patients with Hip Fracture Surgery– A Retrospective Cohort Study based on Longitudinal Health Insurance Database of 1-Million Beneficiaries

Autor: Yi-Cheng Chu, 朱奕承
Rok vydání: 2019
Druh dokumentu: 學位論文 ; thesis
Popis: 107
Background: Hip fracture represents a worldwide challenge in geriatric health care, because it would lead to functional decline and disability. However, previous studies mostly recruited the subjects from a single or a few medical institutions, rather than through a representative national database. Furthermore, to our best knowledge, no study investigated the complete rehabilitation course encompassing both inpatient and outpatient rehabilitation phases for the elderly patients with hip fracture surgery. It is necessary to conduct a study to explore the effect of rehabilitation on 3-Year readmission and mortality risk in elderly patients with hip fracture surgery through the national health database. This purposes of this were 1) to explore the utilization of rehabilitation in elderly patients with hip fracture surgery in Taiwan; 2) to explore the effect of rehabilitation on 3-year readmission or mortality (Event 1, E1) as well as mortality (Event 2, E2) risk in elderly patients with hip fracture surgery; 3) to explore the subgroups effects in terms of sex, surgery type, age, length of stay, and comorbidity (CCI); 4) to conduct a classification analysis for readmission causes. Methods : This study utilized the Longitudinal Health Insurance Database of one-million beneficiaries to investigate the relationship between rehabilitation and 3-year readmission and mortality risk in elderly patients with first-ever hip fracture surgery. The rehabilitation included 3 models: model 1 - whether received rehabilitation or not (yes/no), model 2 (pathway): none/ inpatient only/ outpatient only/ both inpatient and outpatient rehabilitation, and model 3 (times): none/1-3 times/ over 4 times. The study participants were confined to 4,522 patients aged 65 or above who admitted to the hospital due to hip fracture and received surgery between January 2001 and December 2008. Among those patients, 1,947 (43.1%) patients received rehabilitation (including inpatient and/or outpatient rehabilitation) while the other 2,575 (56.9%) patients didn’t. Each patient was followed up for 3 years. Chi-square test and one-way ANOVA were used to compare baseline characteristics between rehabilitation group and non-rehabilitation group. Multivariate Cox-proportional hazards regression models with adjustment for age group, sex, whether osteoporosis, surgery type, initial length of stay, comorbidity (CCI), and hospital characteristics were conducted to investigate the effects of rehabilitation on the risks of readmission or mortality (Event 1, E1)/ mortality (Event2, E2) during the follow-up period. Results: Among 1,947 patients who received rehabilitation, 1,496(76.8%) patients received inpatient rehabilitation, 752(38.6%) patients received outpatient rehabilitation, while 274 patients received both inpatient and outpatient rehabilitation. The majority (98.8%) of those who received rehabilitation got the treatment by physical therapists. Relevant to patients without rehabilitation, patients receiving rehabilitation had a lower risk of 3-year mortality (E2) (Hazard ratio, HR=.76; 95% CI=.66–0.88); with model 2 and model 3 revealed the same trend (HR=.63-.83, 95% CI=.49-.97). Particularly, “received both inpatient and outpatient rehabilitation” or ”rehabilitation of 4 times and above” could lower the E2 more effectively (HR=.63-.64, 95% CI=.46-.89). In contrast, no significant effect of rehabilitation on the 3-year “readmission or mortality” (E1) risk was found except for the increased risk with mode of “inpatient rehabilitation only” (HR=1.25, 95% CI=1.12-1.40). Subgroups analyses revealed that rehabilitation was beneficial to lower the 3-year mortality risk for most subgroups, especially in patients receiving operative internal fixation, aged 70 years or above, male, or patients with comorbidity. Discussion and Conclusions: This study demonstrated that the use of rehabilitation was associated with reduction of 3-year mortality risk (E2), and a dose-response relationship was found given the larger effect for mode of “receiving both inpatient and outpatient rehabilitation” and “rehabilitation of 4 times or above”. On the contrary, we found that received “only outpatient rehabilitation” was associated with increased 3-year “readmission or mortality” risk (E1). It was speculated that those patients who sought for rehabilitation intervention after discharge tended to have poor functional recovery which might inevitably increase their risks of subsequent “readmission or mortality”. Further study is anticipated to confirm the findings and explore the underlying reasons. We suggest that orthopedists should routinely refer their patients for rehabilitation intervention during acute hospitalization period, and continue to receive outpatient rehabilitation if necessary to ensure a better outcome. As a secondary data analysis from a medical claims database, the “patient selection bias” resulted from the non-randomized assignment of treatments and missing of some important covariates in the Cox-regression models were unavoidable. It is anticipated to incorporate the database from “Post-acute Care Plan” and “Long-term Care Plan 2.0” to examine the effect of rehabilitation and different models of rehabilitation on elderly patients with hip fracture surgery more thoroughly in the future.
Databáze: Networked Digital Library of Theses & Dissertations
Externí odkaz: