Frailty in hospitalized adults

UNCG Author/Contributor (non-UNCG co-authors, if there are any, appear on document)
Deborah Ann Lekan (Creator)
The University of North Carolina at Greensboro (UNCG )
Web Site:
Debra Wallace

Abstract: The purpose of this cross-sectional, retrospective, descriptive study was to characterize frailty in hospitalized adults 55 years of age and older admitted to medical units at one large academic medical center during a 15-month time frame and determine if level of frailty on admission predicted length of stay (LOS) and 30-day readmission. Frailty is a syndrome characterized by multisystem physiologic dysregulation due to intrinsic and extrinsic stressors resulting in decreased compensatory reserve and ability to effectively respond to destabilizing health events. Stressors associated with hospitalization may increase risk for frailty or accelerate its development. Frailty is a significant concern as it is associated with morbidity, functional decline, long LOS, readmission, institutionalization, and mortality. There is scant research on frailty in acutely-ill hospitalized adults, especially those ¡Ý 65 years of age. Understanding frailty in this population is imperative because frailty is potentially preventable, treatable, and reversible. Frailty was operationalized as 14 evidence-based frailty components defined by 26 indicator variables. Frailty components were Nutrition, Weakness, Fatigue, Chronic Pain, Dyspnea, Falls, Vision, Depression, Cognition, Social Support, low Hemoglobin, low Albumin, high C-reactive protein (CRP) or hs-CRP, and abnormal WBC count. Each frailty component was scored as one point if at least one indicator variable was present on admission, and summed to derive a Frailty Score, where a higher Frailty Score suggests greater level of frailty (range, 0 to 14). Sociodemographic, clinical, and laboratory data were retrieved from the electronic medical record through web-based data query tools and record review (N = 278). Mean age was 70.2 (SD = 1.3; range, 55¨C98), slightly over half were female, 64% were White, one-third were Black. The mean comorbidity count was 13 (SD = 4.56; range. 1¨C26) and medication count was 12 (SD = 5.2; range, 0¨C31). The most prevalent frailty components (> 81%) were Fatigue, Weakness, Nutrition, Hemoglobin, Albumin, and CRP or hs-CRP. The mean Frailty Score was 9.03 (SD = 1.98; range, 2¨C13). Multiple linear regression was performed with 20 predictor variables and the Frailty Score and then with 14 of the 20 predictor variables that were significant in bivariate linear regression with the Frailty Score using the ENTER and STEPWISE method. All multiple regression models yielded seven significant predictor variables. Six predictors were common to all models: comorbidity, acute pain, ADL assistance, urinary incontinence, Braden Scale score, current tobacco use. In multiple regression with 20 predictors, age was a significant predictor however in multiple regression using ENTER and STEPWISE for 14 predictors, female gender was significant but not age. Results from STEPWISE regression yielded seven significant predictors that explained 27% of the variance in the Frailty Score (adj. R2 = .266, df (14, 263), F = 8.163, p = .000). Mean LOS was 9.92 days (SD = 9.58; range, 1¨C72; median, 7; mode, 5). Simple linear regression for the Frailty Score and log10 transformed LOS was statistically significant (adj. R2 = .090, df (1, 276), F = 29.293, p = .000). Twelve percent experienced 30-day readmission. Logistic regression conducted for the Frailty Score and 30-day readmission was not statistically significant (X 2 = 4.121, df (5), p = .532; ¦Â coefficient = .100, df (1), 95% CI = .913¨C1.1337, p = .307). The Frailty Score characterized this hospitalized population as acutely ill with high comorbidity, symptom burden, nutrition deficits and evidence of physiologic vulnerability and inflammation. Study findings have implications for nursing practice, interdisciplinary collaboration, education, research, and public policy.

Additional Information

Language: English
Date: 2013
Frailty, Stress, Hospitalization, Elderly, Middle-aged, C-reactive protein, Braden Scale
Frail elderly $x Care $z United States
Aging $x Health aspects $z United States
Aging $x Physiological aspects

Email this document to