Nursing homes are a relatively common service option for individuals with mental retardation. Braddock and associates (1) reported that more than 50,000 such individuals lived in nursing homes in 1989. Amado and colleagues (2) reported a similar figure, approximately 46,000, for a 1987 survey. The differences probably derive from different study methods. Both studies clearly note that nursing homes are a major residential option for persons with mental retardation. Longitudinal data presented by Amado and colleagues (2) indicated that this phenomenon was evident for at least a decade before their survey.
Considerable concern exists about the appropriateness of nursing home placements for individuals with mental retardation. Lakin and associates (3) reviewed data from the National Center for Health Statistics and concluded that, based on medical and physical needs, less that 25 percent of persons with mental retardation were appropriately served in nursing homes. Similarly, Davis and coworkers (4) concluded that only 10 percent of individuals with mental retardation who lived in Illinois nursing homes were appropriately placed. It has been suggested that although nursing homes may provide medical, nursing, and personal care services, they have been historically unprepared to meet the social, recreational, and habilitative needs of persons with mental retardation.
The use of various psychotropic medications to treat individuals with mental retardation has long been a point of controversy. It is estimated that between 40 and 50 percent of persons with mental retardation who live in institutions receive some sort of psychotropic medication, and that between 25 and 35 percent of those in community programs receive such medication (5). A previous study found that medication use seemed to be somewhat higher in more medically oriented residential facilities (6). Although variability is evident across settings, it is clear that the use of psychotropic medications is a significant component of treatment protocols for individuals with mental retardation.
The purpose of this study was to examine the use of psychotropic medications among persons with mental retardation who live in Oklahoma nursing homes. The importance of this question derives from the large number of individuals with mental retardation who live in nursing homes, as well as the large number who receive some sort of psychotropic medication.
Subjects for this investigation were 1,056 individuals who were receiving services in 1995 through the Oklahoma mental retardation service system. All subjects were living in nursing homes, and they constitute the entire population of individuals with mental retardation funded through the Oklahoma mental retardation service system to live in nursing homes.
The study included 460 males and 596 females. The mean±SD age was 60±16.9 years, with a range from two years to 95 years. The typical individual had been living in the nursing home for 13.6±9.26 years; the longest period of residence was 39 years.
Oklahoma uses the Developmental Disabilities Quality Assurance Questionnaire (7) annually to assess all consumers in the mental retardation service system and gather information about them. This assessment is administered by interviewers contracted by the state, and it includes major sections on adaptive behavior, living site conditions, health, use of medications, social interactions, community integration, challenging behavior, service planning, and consumer satisfaction. The questionnaire does not include a specific question on psychiatric diagnosis; however, it does ask whether the consumer has experienced any mental health problems within the past year.
Medication use is specifically addressed by a series of items. Interviewers review records and ask staff about the types and amounts of medications prescribed for each consumer. It was necessary to aggregate classes of drugs for descriptive purposes. The classes of medications recorded were antipsychotics, antidepressants, anxiolytics, anticonvulsants, sedative-hypnotics, stimulants, lithium, blood pressure drugs often used to treat behavior disorders, and drugs used to treat obsessive-compulsive disorder.
The initial step in the data analysis involved simply tabulating the patterns of medication use. As t1 shows, 31.8 percent of the 1,056 subjects were receiving antipsychotic medication. The most frequently used antipsychotic medication was thioridazine. Anxiolytics were prescribed for 16 percent, while 6.1 percent received antidepressant medication. Diazepam was the most frequently prescribed anxiolytic, and amitriptyline was the most frequently prescribed antidepressant. In addition, anticonvulsant medication, such as divalproex and carbamazepine, was prescribed to alter behavior for 7 percent of the subjects.
As noted, antipsychotic medication was the most frequently prescribed psychotropic for the 1,056 subjects. Within this broad classification, thioridazine was the most frequently prescribed medication, accounting for 45.5 percent of the antipsychotic prescriptions. Chlorpromazine and haloperidol accounted for 21.5 percent and 18.1 percent, respectively. It should be noted that 92.7 percent of the prescriptions for antipsychotics were described as being for behavior control.
The next step in the data analysis was to attempt to identify factors that might predict whether an individual with mental retardation in a nursing home would receive antipsychotic medication. To do so, use of antipsychotic medication was coded “yes-no,” and we attempted to predict this dichotomous variable from the total score on the adaptive behavior subscale of the instrument, age, length of nursing home residence, sex, presence of mental health problems (present-absent), and item scores for physical violence, property damage, rebellious behavior, disruptive behavior, profane-hostile language, self-injurious behavior, hyperactive tendencies, and screams-yells-cries inappropriately.
A linear combination of six variables was able to explain approximately 14.8 percent of the variance in use of antipsychotic medication. A multiple R of .385 (F=30.258, df=6, 1,045, p<.001) was obtained using the six variables. They were mental health problems, physical violence to others, adaptive behavior total score, screams-yells-cries inappropriately, hyperactive tendencies, and age. t2 summarizes the findings of this analysis. Variables that did not enter the equation were length of nursing home residence, sex, property destruction, disruptive of others' activities, profane or hostile language, rebellious, and self-injurious.
As might be expected, persons who were reported to have mental health problems were more likely to receive antipsychotic medication. In addition, higher scores on items measuring challenging behavior (indicating a greater presence of these behaviors) were associated with a greater tendency to receive antipsychotic medication. Older individuals and individuals with relatively greater adaptive skills were also more likely to receive antipsychotic medication.
Our final analysis involved a retrospective comparison of current dosages with those used in 1991. We randomly selected 100 current nursing home residents with mental retardation and compared their 1991 and 1995 prescriptions for antipsychotic medication. This process resulted in a sample of 48 persons, largely due to missing data from the 1991 database. We chose to limit this analysis to antipsychotic medication because it is the most frequently used medication and because it is possible to create chlorpromazine-equivalence units for comparison across most antipsychotic medications (8).
Chlorpromazine-equivalence units were created for each individual in 1991 and 1995, and data were submitted to a correlated sample t test. The mean daily dose in 1991 was 195.73 mg, and the mean daily dose in 1995 was 187.44 mg. The apparent decline in the amount of medications prescribed was not statistically significant.
It would appear that a relatively high percentage of individuals with mental retardation who live in nursing homes receive various forms of psychotropic medications. According to these data collected in Oklahoma, almost a third of the persons with mental retardation who live in nursing homes receive neuroleptics. Fewer individuals take anxiolytics, antidepressants, and other forms of medication that can alter behavior. Although the average dose was only about 200 chlorpromazine-equivalence units per day, the rate of antipsychotic use for these individuals was exceeded only by the rate at intermediate care facilities for mentally retarded persons in Oklahoma, where 42.9 percent of the residents were receiving antipsychotic medication (6).
The presence of behavior problems or mental health problems does not sufficiently explain the high rate of use of antipsychotic medication. Although these factors are clearly related to the use of antipsychotic medication in nursing homes, multiple regression analysis using behavioral and mental health variables explained less than 15 percent of the variance in the use of antipsychotic medication. Other unmeasured factors must be more important in determining who receives antipsychotic medication. It is perhaps reasonable to speculate that the medical nature of nursing home environments and the training and background of staff in such environments predispose them to seek medical solutions for problems that are addressed behaviorally elsewhere. However, these are merely speculations.
Of the various antipsychotic medications, thioridazine was the most popular, accounting for almost half of the antipsychotic prescriptions. This finding was in agreement with data collected in group homes by Aman and colleagues (9), but the overall use of antipsychotic medication in nursing homes was markedly higher than reported in the group homes. Aggregate data from Oklahoma also indicate a lower rate of antipsychotic use among persons who are retarded and living in group homes rather than in nursing homes (6).
Nevertheless, concern must be expressed about the high rate of use of antipsychotic medication. Given the potential risks of these medications (10), the fact that 91 of the subjects in our sample had screened positive for symptoms of tardive dyskinesia, as well as the evidence that persons with mental retardation are able to live in other environments with less reliance on medication, one must question the appropriateness of placing persons with mental retardation in nursing homes as they are currently conceived.
Considerable medical monitoring of medication use in nursing homes is evident. In the retrospective part of the analysis, we found that 83.7 percent of the subjects in the sample were receiving either different medications or different amounts of medication than in 1991. Despite these various changes, the net effect was that the amount of antipsychotic medication was largely unchanged in 1995. This finding is somewhat similar to that of a previous report (11) in which planned medication reduction trials in an intermediate care facility for mentally retarded persons resulted in essentially no change. Some people received less antipsychotic medication, and some received more, but the net effect was that of no change.
Dr. Spreat is administrator of clinical services at Woods Services, Inc., Langhorne, Pennsylvania 19047. Dr. Conroy is president of the Center for Outcome Analysis in Bryn Mawr, Pennsylvania.
Psychotropic medications received by 1,056 persons with mental retardation living in nursing homes in Oklahoma in 1995
Stepwise multiple regression of variables predicting receipt of an antipsychotic medication by persons with mental retardation living in nursing homes in Oklahoma