Abstract
Surveys of different age groups in the community suggest that the elderly, generally defined as persons older than 65, consume less alcohol and have a lower prevalence of alcohol-related problems than younger persons [1]. Results of a comprehensive community study carried out by Dilling and Weyerer [2,3] in Germany corroborate this evidence. The prevalence of alcohol abuse and dependence decreased from 4.5% among 45–64 years-olds to 2.4% among those 65 years of age and older. Rates for men were higher than those for women in all age groups.
There are a variety of explanations for the lower prevalence of alcohol problems among the elderly. Due to the much higher rate of mortality among alcoholics, only few attain old age. Metabolic changes caused by age decrease the elderly person's tolerance of alcohol. Health complaints occurring in old age, as well as chronic illness, force consumption of alcohol to be reduced. Alcohol consumption among the elderly may be underestimated. Questionnaires used to screen for alcoholism may be inappropriate for the elderly, who may not exhibit the social, legal and occupational consequences of alcohol abuse that are otherwise used to detect alcoholism. Furthermore, the consequences of heavy drinking can be mistaken for medical or psychiatric conditions otherwise common among the elderly. Such consequences may include cognitive impairment, depression, insomnia, poor nutrition, congestive heart failure, and frequent falls [4].
In contrast to most community studies, surveys conducted in healthcare settings have found the prevalence of problem drinking among the elderly to be on the increase [1]. Six percent to 11% of elderly patients admitted to hospitals exhibit symptoms of alcoholism, as do 20% of elderly patients in psychiatric wards [5]. The abuse of alcohol is also a common and frequently neglected problem among nursing home residents. The prevalence of problem drinking in nursing homes ranges from 2.8% to 49%, depending in part on survey methods [6,7].
In Germany, about 5% of the population 65 years of age and older live in residential and nursing homes. To date no representative data on alcohol problems among this population have been available. The current study aims to close this research deficit.
Aims
Based on data collected from 20 randomly selected homes for the elderly in the city of Mannheim, we examined the following questions: (i) how frequent are alcohol problems among persons in residential and nursing homes; (ii) which demographic and clinical features do residents with an alcohol problem exhibit; (iii) are alcohol problems among home residents a consequence of institutionalisation; (iv) does alcohol consumption increase the risk of falling?
Method
When the study commenced in 1994, there were 26 residential and nursing homes in the city of Mannheim. We drew a random sample of 20 homes, which we studied in succession between 1994 and 1997.
All residents residing in these homes on a fixed day were studied over a period of 6 months. On the basis of nursing documentation, the sociodemographic features and medical diagnoses made upon admission to the home, as well as the daily medication of the residents, were compiled. Our research team classed all medical diagnoses according to ICD-10 criteria. In addition, the home staff filled out a standardised assessment sheet for each resident on which functional impairment [8], conspicuous behaviour [9], mental disorders [9], alcohol consumption (distinguished as none, moderate, abuse/dependence), frequency of falls and medical care were recorded. The assessments pertained to the current status of the residents and were repeated 3 months and 6 months later. Details of this assessment sheet, which was based on established scales, are available from the authors. Alcohol abuse/dependence was rated when ICD-10 criteria were fulfilled. Moderate alcohol use was rated in current drinkers who did not meet ICD-10 alcohol abuse/dependence criteria.
In order to determine the home staffs' capability to make such assessments, we carried out a reliability study in one home. The results were quite satisfactory. Agreement among the assessments made by the home staff was good, with kappa values between 0.45 and 0.64 resulting for individual problem areas [10].
Results
Demographic features, alcohol problems and medical diagnoses at the time of admission
With one exception, all 1922 residents of the 20 homes participating in the study were included; 299 (15.6%) were from the residential and 1623 (84.4%) from the nursing care sectors. After the 6-month observation period, complete data for 70% of the original sample of residents were available for the entire period. The most important reasons for dropping out over the 6-month course of the study included death (10.7%) and inpatient hospital stays (5.9%).
The residents studied were mainly female (77.1%) and those older than 65 years (91.0%). The average age was 81.1 years. The mean duration of residence in the home was 3.9 years.
Proceeding on the basis of ICD-10 medical diagnoses made upon home admission, the residents demonstrated a high degree of physical and mental morbidity. In both the nursing and residential homes, cardiovascular and pulmonary diseases were most frequent, affecting 42.6% of all residents. These were followed by cerebrovascular diseases (29.6%), musculoskeletal diseases (29.3%), diabetes (25.9%) and high blood pressure (24.1%). Eighty percent of the home residents were moderately or severely limited (Barthel Index score: 0–90) in their activities of daily living [8].
Prevalence of alcohol diagnoses (ICD-10; F 10) at the time of home admission
Compared to home residents without a diagnosis of alcohol problems at the time of home admission, those residents with such a diagnosis were younger and more often single or divorced. A large number of those with alcohol problems, namely two-fifths, were transferred from mental hospitals to the homes (Table 2).
Features of residents with and without an alcohol diagnosis (ICD-10: F 10) at the time of home admission
Prevalence of alcohol problems (cross-sectional data)
If one considers alcohol consumption at the time of the cross-section, according to the home staff 3.4% of the home residents had abused alcohol once or on several occasions in the previous 4 weeks (Table 3). Here, too, the rate for men (6.9%) was significantly higher (p < 0.001) than that for women (2.3%). Compared with the medical diagnoses at the time of home admission, the prevalence of abusive alcohol consumption receded during the course of the home stay.
Alcohol consumption (staff assessment) among residents in old age homes
The prevalence rates for the individual homes deviated considerably. Some institutions had not admitted anyone with an alcohol problem nor was any current alcohol abuse/dependence recorded. At the other extreme were institutions in which over one-third of the residents had been medically diagnosed as having alcohol problems and of whom a large number (approximately 17%) continued their abuse of alcohol. Such a high prevalence of alcohol problems, however, was observed only in privately operated nursing homes. In publicly funded homes, the number was always below 10%. Our data indicate clearly that residents in privately operated homes are more often admitted (40.7%) from psychiatric hospitals compared to residents in publicly operated homes (9.0%). The reason for the willingness of privately operated homes to admit such a difficult clientele that had often been refused admission to other homes might lie in the fact that higher rates are paid for their care.
Prognosis of alcohol problems
Table 4 demonstrates that the medical diagnosis at the time of home admission is a valuable predictor of the later drinking behaviour in the home. Approximately one-quarter of those who had been medically diagnosed as having an alcohol problem continued to consume moderate amounts of alcohol, without abuse, after admission to the home. Another quarter continued the abuse of alcohol after their admission to the home, and the remaining 50% were completely abstinent. Abusive drinking behaviour was observed in only 1.7% of those without any medically diagnosed alcohol problem at admission, while 73.7% of this group were completely abstinent after arrival.
Alcohol use disorders among home residents at admission and later (mean length of stay = 3.9 years)
Alcohol problems and behaviour disorders
The relationship between current alcohol abuse/dependence and other conspicuous behaviour can be seen in Table 5: irritability, uncooperative behaviour, aggression and suicidal behaviour were relatively frequent among residents with alcohol abuse/dependence during the previous 4 weeks as assessed by the home staff. No significant difference was observed between the groups with regard to sleep disorders, depression, anxiety and delusions/hallucinations. Mainly owing to their lower average age, the residents with an alcohol problem exhibited fewer dementing disorders than residents without such a problem.
Behaviour problems among residents in old age homes with and without alcohol abuse/dependence
Alcohol problems and risk of falling
Falls are an important cause of morbidity and dependency in old age. Within a period of 6 months, 34.1% of all residents fell at least once. Controlling for potential confounding variables (age, sex, mobility impairment, impaired vision, activities of daily living impairment, dementia, depression, use of psychotropic drugs, multimedication), the risk of falling was significantly elevated (Odds ratio: 2.65; p < 0.01) among residents with alcohol problems.
Discussion
The results corroborate the findings from other studies wherein residents of old age homes constitute a group at risk of alcohol abuse and dependence. Of the 1922 home residents studied, 7.4% had already been medically diagnosed as having an alcohol problem at the time of their admission to the home (19.1% of whom were men and 3.8% women). Based on the assessment by home staff, the prevalence of current abuse/dependency of alcohol was lower at 3.4%. This may be due to the scant availability of alcohol in the homes as well as to the substantial mental and physical impairments of the home residents.
Abusive drinking in the absence of any corresponding admission diagnosis was observed for only a small number of residents. The abuse/dependence of alcohol, assessed primarily among younger male home residents, was long-term and was more the reason for rather than the consequence of home admission.
Due to their significantly elevated risk of falling and their concomitant problematical behaviour, such as suicidal behaviour, irritability and aggression residents with alcohol abuse/dependence constitute a particular burden to the staff. Staff of privately operated homes, which admitted in large numbers residents with medically diagnosed alcohol problems or other mental disorders, were especially challenged in Mannheim. This point is critical insofar as the nursing staff had not generally been trained to provide this type of care. Nor had most of the primary care physicians who had been the usual source of medical care for this resident clientele. A relatively high number, namely two-fifths, of the alcohol patients had been transferred to the homes from psychiatric hospitals. Unfortunately, special institutions for addicts do not figure in the care of elderly home residents. This care deficit has been strongly criticised by Bühringer et al. [11]. In Germany almost all of the institutions that do provide care to addicts focus their attention on the younger and middle-aged groups in the wake of the use of illegal drugs.
Home staff frequently complain that psychiatric hospitals often release elderly patients to homes without having provided them with adequate therapy. The homes generally lack the personnel resources, both in terms of number and training, to deal with them. Disregarding the lack of adequate facilities, many nursing homes nonetheless seem to have become a ‘secret’ source of institutional care for people who are suffering from severe mental disorders and/or alcohol problems.
This study has some methodological imperfections. The accuracy of the admitting physicians' diagnoses has not been formally assessed, but previous research indicates it would be highly likely that they would be more prone to underestimate than overestimate the prevalence of alcohol problems in new admissions. Clearly, further prospective research on alcohol and the elderly living in residential care is required. Our study indicates the potential importance of alcohol as a contributing factor to the need for institutional care and as a potential cause of increased dependency and need for assistance in this population.
Footnotes
Acknowledgements
The authors wish to thank the home residents for their participation; the heads of homes and nurses for their kind cooperation; and the colleagues who administered the interviews. This study was funded by a grant from the German Ministry of Education and Research (Project-Nr. 01 EB 9401).