The Gerontologist 48:537-541 (2008)
© 2008 The Gerontological Society of America
Hazards of Hospitalization: Residence Prior to Admission Predicts Outcomes
Susan M. Friedman, MD, MPH1,
Daniel A. Mendelson, MD, MS1,
Karilee W. Bingham, RN, BS1 and
Robert M. McCann, MD1
Correspondence: Address correspondence to Susan M. Friedman, MD, MPH, Assistant Professor of Medicine, University of Rochester, 1000 South Avenue, Box 58, Rochester, NY 14620. E-mail: susan_friedman{at}urmc.rochester.edu
 |
Abstract
|
|---|
Purpose: Previous studies investigating adverse outcomes of hospitalized elders have focused on community-dwelling patients. Given the rapid growth of populations living in other settings, such as assisted living facilities, it is important to understand whether these patients are at higher risk of experiencing specific adverse outcomes during hospitalization, so that interventions can be developed to reduce risk. Methods: This is a prospective, observational study of 212 sequential patients admitted during a 1-month period in 2006 to a 38-bed Acute Care for Elders unit in Rochester, New York and followed until discharge. We categorized the patients by residence prior to admission (i.e., community, assisted living, and nursing home). Our outcome categories were: worsening function, delirium, depression, falls, pressure sores, and nursing home admission. Results: After adjusting for multiple characteristics, we found that patients admitted from assisted living facilities were at substantially higher risk than those admitted from the community for functional decline and falls. Patients from nursing homes had a trend toward increased risk for these outcomes, but the trend did not reach statistical significance. More than three fourths of assisted living facility residents were discharged to a nursing home after hospitalization, with a relative risk of 9.41 (p <.001) versus community-dwellers for this outcome. Implications: People who are admitted to the hospital from assisted living facilities are at high risk for falls and functional decline during hospitalization. Assisted living residents are at a particularly high risk of nursing home admission following hospitalization. Targeted preventive programs should be developed with a goal of reducing risk in this vulnerable population.
Key Words: Falls • Frailty • Function • Long-term care
It is well documented that the hospitalization of older adults is associated with a high incidence of multiple adverse outcomes, including functional decline, delirium, and falls (Covinsky et al., 2003; Creditor, 1993; Hirsch, Sommers, Olsen, Mullen, & Winograd, 1990; Hitcho et al., 2004; Inouye, Viscoli, Horwitz, Hurst, & Tinetti, 1993; Inouye et al., 1990). Previous studies evaluating outcomes of hospitalization of older adults suggest that those individuals who are more frail and disabled at baseline are at higher risk of adverse outcomes than their more robust and highly functioning counterparts (Lefevre et al., 1992; Sager et al., 1996). However, these studies have predominantly involved community-dwelling individuals (Covinsky et al., 2003; Inouye et al., 1993). It is therefore unclear whether, and to what extent, those persons who come from other settings are at higher risk for adverse outcomes of hospitalization, and which settings are associated with the highest risk for specific adverse outcomes.
The number of residents in assisted living facilities has increased in recent years, and that trend is expected to continue (McCormick & Chulis, 2003). Understanding the risk that hospitalization poses to these individuals becomes progressively more important as that group grows.
The identification of high-risk patients by place of residence prior to hospitalization provides a simple way of targeting at admission those individuals who are most vulnerable to experiencing adverse outcomes during hospitalization. Programs can then be developed to target those patients at highest risk; this, in turn, might reduce adverse outcomes.
|
Methods
|
|---|
Setting
We evaluated patients in the Acute Care for Elders (ACE) unit of Highland Hospital. The 38-bed medical unit is located in a 268-bed community hospital, incorporating the original ACE model features, namely, patient-centered nursing care, a prepared environment, early discharge planning, and medical care review (Counsell et al., 2000; Landefeld, Palmer, Kresevic, Fortinsky, & Kowal, 1995). Stated goals of this unit include preserving function, maintaining comfort and dignity, and minimizing the impact of hospitalization. Because of a referral network of approximately 20 nursing homes and 15 assisted living facilities, the unit cares for a mix of frail and less frail patients, allowing for a comparison of outcomes between these subpopulations. Patients older than 70 years of age, private patients of a geriatrician, and patients from an assisted living facility or a nursing home are given preference for admission to this unit.
Patients
As part of an outcomes-management program and to assess clinical outcomes, we followed until discharge all patients who were admitted to the ACE unit during the 1-month period from June 3, 2006 to July 3, 2006. Of the 212 patients admitted during this time, 25.9% (n = 55) were from nursing homes, 13.2% (n = 28) were from assisted living facilities, and 60.8% (n = 129) were admitted from the community (Table 1)<--CO?1-->. We define assisted living facilities as adult-living communities that have at least some meals and assistance with activities of daily living (ADLs) available. Average length of stay during this period (5.2 days) was similar to the average length of stay in the unit for the year (5.6 days), as well as the national average for medicine admissions for individuals older than 65 years of age (5.1 days; see Johnson, 2007).
Assessment
Using standardized collection tools, the patients' primary care nurses evaluated the patients daily for each outcome. Outcomes of interest were ADLs, delirium, depression, falls, and pressure sores. Nurses received training on data collection by means of two in-service sessions before data collection began, a follow-up in-service session after data collection had started, and individualized, daily feedback from the study coordinator based on form completion. We chose assessment tools that could be completed by the patients' primary nurses in a timely fashion, without requirement for additional evaluation or questioning of patients. The nurses collected data for 1,104 of the 1,107 (99.7%) patient days during the study period.
We had the nurses assess the presence of delirium by means of the confusion assessment method (Inouye et al., 1990), which is a four-question tool developed to provide a standardized approach for clinicians who are not psychiatrically trained to identify delirium quickly and accurately (Inouye, 2003). It has a sensitivity of between 94% and 100% and a specificity of 90% to 95%, with high interobserver reliability.
We had depression evaluated by means of the Hammond Signs of Depression Scale (Hammond, O'Keeffe, & Barer, 2000). This screening tool was developed as a tool that does not rely on verbal communications, and it was validated in an acute geriatric unit. It has a sensitivity of 90%, a specificity of 72%, a positive predictive value of 0.69, and a negative predictive value of 0.96. A score of 3 points or more out of 6 is considered positive.
The nurses assessed the patients to be either independent or not independent for each of seven ADLs, namely, walking, transferring, dressing, eating, bathing, grooming, and toileting.
Falls were reported daily. Pressure sores were noted as present or absent.
In addition to clinical data, the nurses collected information regarding demographics (age, race, and gender) and discharge status. We categorized the patients' place of residence prior to admission into community, assisted living, and nursing home.
Data Analysis
We obtained approval from the Research Subjects Review Board at the University of Rochester to analyze and present these data. We report outcome incidence for those patients who did not already have the target outcome and were thus considered at risk on the day of admission. We calculated the proportion of those individuals who declined in function by using the 148 patients who were not already dependent in all seven of the ADLs at baseline, and removing from further analysis 2 additional patients who died. We considered those individuals who declined by one or more ADLs during admission to have declined. Similarly, we calculated the proportion of the 190 patients without delirium on admission who developed it while in the hospital, and the proportion of the 179 patients without depression on admission who developed it. We considered all patients to be at risk for pressure sores and falls on admission.
We performed two logistic regressions for each outcome to assess the relative risk (RR) of site as a predictor. The first did not adjust for other patient characteristics. The second adjusted for age, race, gender, and number of ADL dependencies on admission, to evaluate the independent contribution of prior residence to each outcome.
|
Results
|
|---|
Table 1 presents demographic characteristics of the 212 patients admitted to the unit during the month. There was a bimodal distribution of ADL dependencies, with 22% of the patients having no dependencies on admission, and 47% having six or seven dependencies.
We found that 30.1% of those individuals who were not completely dependent in ADLs at baseline declined, 19.5% developed delirium following admission, 34.6% developed depressive symptoms, 3.3% fell, and 5.2% developed new pressure sores. Prior to adjusting the data, we found that patients from assisted living facilities and nursing homes were at higher risk of functional decline (RR = 6.34 and 3.33, respectively, with p <.05 for each) than were community dwellers (Table 2)<--CO?2-->. Residents of assisted living facilities were at higher risk for depression (RR = 2.64, p =.04) and falls (RR = 15.36, p =.02), and nursing home residents were at higher risk for developing delirium after admission (RR = 2.81, p =.01), than were their community-dwelling counterparts. Furthermore, residents of assisted living facilities were at a much higher risk of being discharged to a nursing home than were those patients who were admitted from the community (RR = 15.64, p <.0001).
After adjusting for age, race, gender, and ADL dependence on admission, we found that residents admitted from assisted living facilities were at substantially higher risk than individuals from the community for functional decline (RR = 5.19, p =.005) and falls (RR = 16.67, p =.03). Residents from nursing homes had a trend toward increased risk for these outcomes that did not reach statistical significance. After adjusting for baseline characteristics, we found that individuals from assisted living facilities were more than nine times more likely to be admitted to a nursing home following hospitalization than were their community-dwelling counterparts (p <.001). Risks for onset of delirium, depression, and pressure sores were not independently related to initial place of residence, after we adjusted for patient characteristics.
|
Discussion
|
|---|
Patients who are admitted to the hospital from assisted living facilities and from nursing homes are at a higher risk for several adverse outcomes than are patients who are admitted from the community. These findings are particularly important, because previous interventions for hospitalized older adults, such as geriatric evaluation units and ACE units (Counsell et al., 2000), have specifically targeted community-dwelling populations. Considerations should be made for how these interventions might best be adapted to the more frail populations found in alternate living situations.
After adjusting for other patient characteristics, we found that residents admitted from an assisted living facility were more than 16 times as likely to fall, and 5 times more likely to experience functional decline, than their community-dwelling counterparts. The implications of this decline can be seen immediately. Three fourths of assisted living residents were discharged to nursing homes. Although many were admitted for rehabilitation following hospitalization, with the intent to return to their previous residence, it is not known how many will actually return. It is clear that this population represents a highly vulnerable subgroup, which might benefit from intensive interventions to prevent further functional decline while in the hospital.
As the number of frail older adults in residential alternatives to nursing homes continues to increase, the number of these individuals who are hospitalized will add to the absolute risk, and thus the importance of developing interventions for this population. A recent study reports that the proportion of Medicare beneficiaries residing in facility-like alternatives to nursing homes, including assisted living facilities, rose from 16% in 1996 to 30% in 2001 (McCormick & Chulis, 2003). There were 32 million Medicare beneficiaries, with 359 hospital discharges per 1,000 beneficiaries in 1998 (Eggers & Greenberg, 2000). Conservatively estimating that individuals in facility-like alternatives to nursing homes are at the same risk for hospitalization as other Medicare beneficiaries, this translates to over 3 million hospitalizations per year in which such residents are placed at risk.
The findings of increased risk of functional decline and nursing home admission in assisted living residents confirm observations in other studies that this is a population in transition. Assisted living residents have been shown to have an annual mortality rate of 14.4 per 100 residents, an annual rate of nursing home transfer of 21.3 per 100 residents, and a rate of worsening morbidity of 22.7 per 100-day quarter per 100 residents (Zimmerman et al., 2005). Assisted living residents who are admitted to a nursing home following hospitalization tend to have more functional impairment than those who are admitted from other settings (Aud & Rantz, 2005).
It is unclear from this analysis whether these outcomes are a direct effect of the hospitalization, or a result of an at-risk population of individuals who are acutely ill, or both. There is considerable evidence from previous literature that hospitalization itself may lead to adverse outcomes (Binder et al., 2003; Creditor, 1993; Leff et al., 2005; Thomas & Brennan, 2000), although to our knowledge this has not been studied in a controlled fashion. The distinction is an important one in determining potential prevention strategies. If the outcomes are the result of hospitalization, then interventions to reduce or prevent hospitalization in these populations would be appropriate; interventions such as the "hospital at home" might be well targeted to this population (Leff et al.). If, in contrast, the outcomes are the result of an acutely ill, at-risk population, then interventions to target specific risk factors in hospitalized high-risk patients, such as the Hospital Elder Life Program (Inouye, Bogardus, Baker, Leo-Summers, & Cooney, 2000), might reduce adverse outcomes.
Some limitations of this study should be noted. We obtained these data from one unit of one hospital over 1 month, and generalizability may therefore be limited. Furthermore, the tools we used for evaluation were screening tools. The tool we used for the measurement of ADLs, in particular, is a coarse tool, with patients reported to be either independent or not independent. Thirty-six of the nursing home patients required assistance with all ADLs on admission and were therefore eliminated from the population at risk for developing declines. These results may, therefore, underestimate the declines seen in the nursing home group and other individuals who were highly dependent at baseline. Future studies evaluating these populations should use a tool with greater sensitivity to changes in function at the lower end of the functional spectrum.
Although we adjusted for demographic information, as well as functional status on the day of admission, we did not have information through this outcomes-management program about a diagnosis of dementia (Sands et al., 2003), functional status prior to hospitalization (Narain et al., 1988; Walter et al., 2001), or other risk factors that have been shown to adversely impact hospital outcomes. Therefore, we cannot conclude from these data whether place of residence is an independent risk factor for adverse outcomes, or whether it is merely a marker of a group at high risk. In general, patients admitted to the hospital from noncommunity settings have a higher prevalence of dementia and poor functional status (McCormick & Chulis, 2003), and those admitted to the nursing home from an assisted living facility have a higher prevalence of dementia and depression than those from the community (Aud & Rantz, 2005). Future studies examining the risks of patients who are admitted from noncommunity settings should adjust for these and other risk factors.
Because this was a small study, confidence intervals for RRs were wide. This was seen particularly for outcomes with low rates, such as falls and pressure sores. However, despite a small sample size, multiple outcomes showed statistically significant differences between groups. The precision of the RRs is therefore limited and should be confirmed by future studies.
In summary, patients who are admitted to the hospital from settings other than the community, and particularly those who are admitted from assisted living facilities, are at high risk for adverse outcomes during hospitalization. More fully understanding the root causes of this vulnerability, and what can be done to reduce its effects, is an important issue requiring further investigation. Future studies should address whether interventions prior to or during hospitalization may ameliorate the risk of decline, and the risk of institutionalization that may result.
|
Footnotes
|
|---|
Dr. Friedman received support for this study from Grant 5K23AG19545, a K23 grant, from the National Institute on Aging (NIA). The NIA had no role in the design and conduct of the study; collection, management, analysis, and interpretation of the data; or preparation, review, or approval of the manuscript. Dr. Friedman had full access to all of the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis. The authors do not have any financial interest in the subject matter or materials discussed in the article. 
We thank Dr. William Hall for his review and feedback on this manuscript.
1 Department of Medicine, University of Rochester, New York.
William J. McAuley, PhD
Received for publication June 14, 2007.
Accepted for publication September 13, 2007.
|
References
|
|---|
- Aud, M. A., & Rantz, M. J., (2005). Admissions to skilled nursing facilities from assisted living facilities. Journal of Nursing Care Quality, 20, 16-25.[Medline]
- Binder, E., Kruse, R., Sherman, A., Madsen, R., Zweig, S., & D'Agostino, R., et al (2003). Predictors of short-term functional decline in survivors of nursing home acquired lower-respiratory tract infection. Journals of Gerontology Series A: Biological Sciences and Medical Sciences, 58A, M60-M67.
- Counsell, S. R., Holder, C. M., Liebenauer, L. L., Palmer, R. M., Fortinsky, R. H., & Kresevic, D. M., et al (2000). Effects of a multicomponent intervention on functional outcomes and process of care in hospitalized older patients: A randomized controlled trial of acute care for elders (ACE) in a community hospital. Journal of the American Geriatrics Society, 48, 1572-1581.[Medline]
- Covinsky, K. E., Palmer, R. M., Fortinsky, R. H., Counsell, S. R., Stewart, A. L., & Kresevic, D., et al (2003). Loss of independence in activities of daily living in older adults hospitalized with medical illnesses: Increased vulnerability with age. Journal of the American Geriatrics Society, 51, 451-458.[Medline]
- Creditor, M. C., (1993). Hazards of hospitalization of the elderly. Annals of Internal Medicine, 118, 219-223.[Abstract/Free Full Text]
- Eggers, P. W., & Greenberg, L. G., (2000). Racial and ethnic differences in hospitalization rates among aged Medicare beneficiaries, 1998. Health Care Financing Review, 21, 91-105.[Medline]
- Hammond, M. F., O'Keeffe, S. T., & Barer, D. H., (2000). Development and validation of a brief observer-rated screening scale for depression in elderly medical patients. Age and Ageing, 29, 511-515.[Abstract/Free Full Text]
- Hirsch, C. H., Sommers, L., Olsen, A., Mullen, L., & Winograd, C. H., (1990). The natural history of functional morbidity in hospitalized older patients. Journal of the American Geriatrics Society, 38, 1296-1303.[Medline]
- Hitcho, E. B., Krauss, M. J., Birge, S., Claiborne Dunagan, W., Fischer, I., & Johnson, S., et al (2004). Characteristics and circumstances of falls in a hospital setting: A prospective analysis. Journal of General Internal Medicine, 19, 732-739.[Medline]
- Inouye, S., van Dyck, C., Alessi, C., Balkin, S., Siegel, A., & Horwitz, A., (1990). Clarifying confusion: The confusion assessment method. Annals of Internal Medicine, 113, 941-948.[Medline]
- Inouye, S., Viscoli, C., Horwitz, R., Hurst, L., & Tinetti, M., (1993). A predictive model for delirium in hospitalized elderly medical patients based on admission characteristics. Annals of Internal Medicine, 119, 474-481.[Abstract/Free Full Text]
- Inouye, S. K., (2003). The confusion assessment method (CAM): Training manual and coding guide. New Haven, CT: Yale School of Medicine.
- Inouye, S. K., Bogardus, S. T., Jr., Baker, D. I., Leo-Summers, L., & Cooney, L. M., Jr. (2000). The hospital elder life program: A model of care to prevent cognitive and functional decline in older hospitalized patients. Hospital Elder Life Program. Journal of the American Geriatrics Society, 48, 1697-1706.[Medline]
- Inouye, S. K., Wagner, D. R., Acampora, D., Horwitz, R. I., Cooney, L. M., Jr., & Hurst, L. D., et al (1993). A predictive index for functional decline in hospitalized elderly medical patients. Journal of General Internal Medicine, 8, 645-652.[Medline]
- Johnson, S., (2007). National average length of stay for medicine admissions. Solucient database, individual report, accessed August. 13, 2007.
- Landefeld, C. S., Palmer, R. M., Kresevic, D. M., Fortinsky, R. H., & Kowal, J., (1995). A randomized trial of care in a hospital medical unit especially designed to improve the functional outcomes of acutely ill older patients. New England Journal of Medicine, 332, 1338-1344.[Abstract/Free Full Text]
- Lefevre, F., Feinglass, J., Potts, S., Soglin, L., Yarnold, P., & Martin, G. J., et al (1992). Iatrogenic complications in high-risk, elderly patients. Archives of Internal Medicine, 152, 2074-2080.[Abstract]
- Leff, B., Burton, L., Mader, S. L., Naughton, B., Burl, J., & Inouye, S. K., et al (2005). Hospital at home: Feasibility and outcomes of a program to provide hospital-level care at home for acutely ill older patients. Annals of Internal Medicine, 143, 798-808.[Abstract/Free Full Text]
- McCormick, J. C., & Chulis, G. S., (2003). Growth in residential alternatives to nursing homes: 2001. Health Care Financing Review, 24, 143-150.[Medline]
- Narain, P., Rubenstein, L. Z., Wieland, G. D., Rosbrook, B., Strome, L. S., & Pietruszka, F., et al (1988). Predictors of immediate and 6-month outcomes in hospitalized elderly patients. The importance of functional status. Journal of the American Geriatrics Society, 36, 775-783.[Medline]
- Sager, M. A., Franke, T., Inouye, S. K., Landefeld, C. S., Morgan, T. M., & Rudberg, M. A., et al (1996). Functional outcomes of acute medical illness and hospitalization in older persons. Archives of Internal Medicine, 156, 645-652.[Abstract]
- Sands, L. P., Yaffe, K., Covinsky, K., Chren, M. M., Counsell, S., & Palmer, R., et al (2003). Cognitive screening predicts magnitude of functional recovery from admission to 3 months after discharge in hospitalized elders. Journals of Gerontology A: Biological Sciences and Medical Sciences, 58A, M37-M45.
- Thomas, E. J., & Brennan, T. A., (2000). Incidence and types of preventable adverse events in elderly patients: Population based review of medical records. British Medical Journal, 320, 741-744.[Abstract/Free Full Text]
- Walter, L. C., Brand, R. J., Counsell, S. R., Palmer, R. M., Landefeld, C. S., & Fortinsky, R. H., et al (2001). Development and validation of a prognostic index for 1-year mortality in older adults after hospitalization. Journal of the American Medical Association, 285, 2987-2994.[Abstract/Free Full Text]
- Zimmerman, S., Sloane, P. D., Eckert, J. K., Gruber-Baldini, A. L., Morgan, L. A., & Hebel, J. R., et al (2005). How good is assisted living? Findings and implications from an outcomes study. Journals of Gerontology B: Psychological Sciences and Social Sciences, 60B, S195-S204.
TOP
Abstract
Methods
