Archives of Gerontology and Geriatrics 59 (2014) 175–180
Contents lists available at ScienceDirect
Archives of Gerontology and Geriatrics journal homepage: www.elsevier.com/locate/archger
Changes in the clinical features of older patients admitted from the emergency department Isabelle De Brauwer a,b,*, William D’Hoore a, Christian Swine a,c, Fre´de´ric Thys d, Claire Beguin e, Pascale Cornette a,b a
Institute of Health and Society (IRSS), Universite´ catholique de Louvain, Clos Chapelle-aux-Champs n830 B1.30.15, 1200 Brussels, Belgium Department of Geriatric Medicine, Cliniques universitaires Saint Luc, Avenue Hippocrate n810, 1200 Brussels, Belgium Department of Geriatric Medicine, Centre hospitalier universitaire Mont-Godinne, Avenue G. Therasse n81, 5530 Yvoir, Belgium d Department of Emergency Medicine, Cliniques universitaires Saint Luc, Avenue Hippocrate n810, 1200 Brussels, Belgium e Medical Information and Statistics, Cliniques universitaires Saint Luc, Avenue Hippocrate n810, 1200 Brussels, Belgium b c
A R T I C L E I N F O
A B S T R A C T
Article history: Received 19 December 2013 Received in revised form 28 February 2014 Accepted 17 March 2014 Available online 26 March 2014
Demographic changes and healthcare reforms may impact the proﬁle of hospitalized older persons. In this study, we sought to compare the characteristics of two prospective cohorts recruited at a ten-year interval (1999, n = 253–2009, n = 355). They included older patients (75 years) admitted through the emergency department for at least 48 h in acute non-geriatric wards in the same university hospital. The exclusion criteria were patients who were admitted directly to the intensive care unit, who were dependent for all 6 Activities of Daily Living (ADL), who had recently suffered from a major stroke, or whose with a life expectancy of less than 3 months. Median age was higher in 2009 than in 1999 (83 vs. 81; p = 0.020), with a higher proportion of those aged 85 years and over (p = 0.026). Patients in the 2009 cohort were less likely to live in a nursing home (p = 0.018), more dependent for the basic ADL (p < 0.001), more independent for the instrumental ADL (p < 0.001). They were more likely to have fallen in the previous year (p < 0.001). They took more medications (p < 0.001). Their length-of-stay was shorter (p < 0.001), but they were more likely to be discharged to a rehabilitation center (p < 0.001). They underwent more early re-admissions (p = 0.020) and similar 3-month functional decline (p = 0.614). In conclusion, within a decade, the social, functional and medical characteristics of older patients admitted to hospital have changed signiﬁcantly. In view of the high consumption of in-patient services by this population, hospitals must adapt to these rapid changes. ß 2014 Elsevier Ireland Ltd. All rights reserved.
Keywords: Older patients Acute hospital Clinical features changes
1. Introduction Demographic changes, together with the epidemiologic transition in the aging population, have a major impact on care needs and represent a signiﬁcant challenge for healthcare systems. New models of care are being developed to address the needs of a growing population of older persons, especially those of the oldest old. Older people are more likely to suffer from chronic diseases, they have more comorbidities and associated polymedication, and are at higher risk for functional decline. Their needs and expectations are therefore different to the general population
* Corresponding author at: Institute of Health and Society (IRSS), Clos Chapelleaux-Champs n830 B1.30.15, 1200 Brussels, Belgium. Tel.: +32 2 764 31 86; fax: +32 2 764 34 70. E-mail address: [email protected]
(I. De Brauwer). http://dx.doi.org/10.1016/j.archger.2014.03.005 0167-4943/ß 2014 Elsevier Ireland Ltd. All rights reserved.
and require a multidimensional and integrated approach, focusing on functional health promotion and prevention of diseases in primary care and long-term institutional settings (Stuck & Lliffe, 2011; Tinetti & Fried, 2004). Older patients also pose a speciﬁc challenge in the hospital setting. The number and proportion of older people admitted in acute hospital settings are increasing and they contribute signiﬁcantly to the number of hospital discharges and hospital days (Hall, DeFrances, Williams, Golosinskiy, & Schwartzman, 2010; Landefeld, 2006; Parker, Fayadevatan, & Lee, 2006). Furthermore, acute hospital stays represent a critical episode for frail older adults, whose risk of complications is high. Preventing and managing the functional consequences of an acute decompensation of chronic disease requires a continuous process of care that takes into account patients’ pre-morbid functional status and geriatric features, and promotes early functional rehabilitation (Ellis & Langhorne, 2005; Landefeld, 2006).
I. De Brauwer et al. / Archives of Gerontology and Geriatrics 59 (2014) 175–180
At the same time, economic constraints have led European countries to reform their health care systems. Together with rapid technological and scientiﬁc innovation, reforms have led to hyperspecialized wards, a reduction in the number of acute hospital beds and in length of stays, and an increase in outpatient services, e.g. day care hospitalization (Rechel, Wright, Edwards, Dowdeswell, & McKee, 2009). In this context, the speciﬁc needs of the older population could be not met while their hospitalization rate is rising (Ekdahl, Linderholm, Hellstro¨m, Anderson, & Friedrichsen, 2012; Sin Dan & Tu Jack, 2000). In order to anticipate the needs of older people and allocate appropriate services to this population, both demographers and economists continue to debate the trends in health and functional status of this population (Freedman, Martin, & Schoeni, 2002; Robine & Michel, 2004). Although data are available regarding these issues in the community, little is known about changes in the proﬁle of the older hospitalized population. The aim of this present study was to address this question by comparing the geriatric characteristics of older patients in two cohort studies conducted respectively in 1999 and in 2009.
2. Method This is a ﬁrst step analysis of two prospective observational cohort studies, conducted for the construction (Cornette, 2005) and validation of a screening tool for functional decline. 2.1. Design, setting and patients Data were collected in the medical and surgical wards of the same urban Belgian university hospital (960 beds), named Cliniques Universitaires Saint-Luc, over a period of 9 months in 1999 and 13 months in 2009. To be eligible, patients had to be aged 75 years or older and admitted for at least 48 h through the emergency department. We excluded patients who were admitted directly to the intensive care unit, who were admitted to the acute geriatric ward, who were dependent for all 6 ADL, who had recently suffered from a major stroke, or whose life expectancy was estimated to be shorter than 3 months. These exclusion criteria were related to the primary aim of the studies.
2.3. Three-month follow-up Three months after discharge, all patients were followed up by phone for basic ADL status, residency, rehospitalization and mortality. 2.4. Statistical analyses The research assistant (IDB) coded and recorded data using NCSS and PASS statistical software, version 2004. Descriptive reports include mean, standard deviation (SD) and median for continuous variables and percentages for categorical and ordinal variables. We used Wilcoxon Rank tests to compare continuous data and chi-square tests to compare categorical and ordinal data. We deﬁned the threshold of statistical signiﬁcance at 0.05. 2.5. Ethical considerations The study was approved by the hospital’s medical ethics committee. Informed consent was obtained from the patient or caregiver in case of cognitive impairment. 3. Results In 1999, on the 1196 patients eligible, 315 were included in the Cliniques Universitaires Saint-Luc. The great majority of patients who were not included were for logistical reasons; few refused (n = 32) and 58 were lost of follow-up during hospital stay or at discharge. Fifty-three patients were excluded because of age 70–74 years and 9 because of complete basic ADL dependency (6-point KATZ). In the 1999 cohort, 253 patients were analyzed. Fig. 1 summarized the inclusion process for the 2009 cohort from which ﬁnally, data from 355 patients were analyzed. Participant features are summarized in Tables 1–3. Participants in the 2009 cohort were older (82.9 vs. 82.0, p = 0.020), with a greater proportion of patients aged over 85 (38% vs. 29%, p = 0.026) (Table 1). Subgroup analyses were thus n=1790 hospitalized 75+
2.2. In-hospital data collection Within 48 h of admission, one resident in geriatric medicine (PC (1999 cohort), IDB (2009 cohort)), assessed participants’ cognitive function and pre-hospital functional status, and collected socio-demographic and medical data. The same questionnaire was used in 1999 and in 2009. To assess cognitive function, we used a short version of Folstein MMSE (21 points) (Folstein, Folstein, & McHugh, 1975). In patients with overt cognitive impairment, further data were collected by interviewing their caregiver. Pre-hospital functional status was deﬁned as two-weeks prior to admission. We used dichotomous KATZ index (0–6) (Covinsky et al., 2000) to evaluate dependence in six basic ADL, i.e. bathing, dressing, walking, toileting, continence and feeding, and dichotomous LAWTON index (0–7) (Lawton & Brody, 1969) to determine performance in seven instrumental ADL, i.e. telephoning, shopping, preparing meals, doing housework, using transportation, managing ﬁnances and taking medication. Higher scores indicate higher levels of independence. The International Standard Classiﬁcation of Education 2011 was used to assess educational level. A level 5 or more was deﬁned as higher education level. We collected diagnosis at discharge based on ICD9-CM codes.
383 included patients
EXCLUSION REASONS Complete bADL dependencies (n=152) GEMU (n=233) ICU (n=249) LOS < 48h (n=110) Major strokes (n=46) Not met, practical reasons (n=184) Palliative cares (n=44) Part of intervention group (RCT, primary aim) (n=239) Refusal (n=150) EXCLUSION REASONS Complete bADL dependencies (n=2) Geriatric intervention (n=2) ICU within 12h (n=1) LOS < 48h (n=10) LOS > 3months (n=1) Not met within> 48h (n=1) Palliative cares (n=4) Secondary refusals (n=6)
356 included patients
n=1 Secondary refusal at 3-month (n=1)
Final cohort: n=355
bADL: basic Activity of Daily Living; GEMU: Geriatric Evaluation and Management Unit; ICU: Intensive Care Unit; LOS: Length Of Stay; RCT: Randomized Control Trial;
Fig. 1. Inclusion Tree 2009 cohort.
I. De Brauwer et al. / Archives of Gerontology and Geriatrics 59 (2014) 175–180
Table 1 Socio-demographic characteristics, functional data, in-hospital and 3-month post-discharge features. 1999 (n = 253
2009 (n = 355)
Mean (SD) [median] or n (%)
Mean (SD) [median] or n (%)
Age, years 85+ Sex, male Community-dwelling Higher education UNESCO MMSE (0–21)a ADL, Katz (0–6)b 6 4–5 1–3 iADL Lawton (0–7)c 6–7 5 3–4 0–2 Fall within previous year Medications, n
82.0 (5.3) [81.0] 72 (29) 100 (40) 218 (86) 52 (21) 17.7 (3.1) [19.0]
82.9 (5.2). [83.0] 133 (38) 160 (45) 327 (92) 138 (39) 17.7 (3.9) [19.0]
143 (57) 79 (31) 31 (12)
128 (36) 173 (48) 54 (15)
69 (27) 48 (19) 87 (34) 49 (19) 96 (38) 4.2 (2.1) [4.0]
228 (64) 36(10) 46 (13) 45 (12) 213 (60) 5.9 (3.1) [6.0]
LOS, daysd In-hosp. mortality Discharge, residency Same New nursing home Rehab
14.2 (10.8) [11.0] 10 (4)
10.0 (6.8) [8.0] 15 (4)
199 (79) 8 (3) 23 (9)
237 (70) 11 (3) 85 (25)
3 Month-mortality 3 Month-residency Same New NH 3 Month-functional decline FDe post-rehab Re-hosp. within 3 m
201 (87) 17 (7) 73 (32) 13 (18) 44 (19)
257 (83) 24 (8) 105 (34) 40 (41) 83 (27)
0.020* 0.026* 0.173 0.018*