Journal of Gerontological Nursing


Julie Johnson, PhD, RN


Pharmacologic treatment for sleep disturbances in the elderly may or may not produce the desired results. Find out what factors contribute to successful management of this problem.


Pharmacologic treatment for sleep disturbances in the elderly may or may not produce the desired results. Find out what factors contribute to successful management of this problem.

Among the problems faced by the older adult are changes in sleep patterns. When the individual enters a long-term care facility, the nursing response to these changes is often based on the assumption that sleep time must be concentrated during the night hours. If the patient experiences difficulty in falling asleep, frequent nocturnal awakenings, or early morning arousals, he may be labeled as an "insomniac." Consequently, he is treated pharmacologically with sedative-hypnotics and anxiolytics rather than with nursing measures. The benzodiazepines are among the more frequently used medications of this type.

However, drug-induced sleep is the antithesis of treatment.' Due to increased body fat, altered rates of gastrointestinal absorption, and a decreased ability to metabolize and excrete drugs, the benzodiazepines have a prolonged half-life in the elderly.2 Among the detrimental consequences are an alteration of REM (rapid eye movement) sleep, a reversal of normal sleep patterns, incontinence, decreased body movements, the daytime continuation of drowsiness, and drug dependency. Furthermore, drug-induced restlessness is often seen as an indication of insufficient medication, with a subsequent increase in dosage.1

After receiving the benzodiazepines for several weeks, tolerance to their sleep-inducing properties develops, and the actual quality and quantity of sleep is disturbed.3 Because sleep is not an isolated phenomenon, this disruption influences daytime activity and behavior.4

Whether or not an individual is satisfied witfi his sleep experience and is able to function adequately during the day may be of greater importance than the objective assessment of a "good" or "poor" night's sleep. This subjective perception of sleep may be more influenced by the act of taking a pill, ie, the "placebo effect," than by the actual sleep-inducing properties of that medication. Thus, the benzodiazepines, or an analgesic may affect the elderly individual's perception of me quality and quantity of his sleep.

Although there are many harmful physical and psychological consequences of using the benzodiazepine sedative-hypnotics and anxiolytics on a routine basis, the institutionalized elderly are often given these medications nightly in an effort to improve their sleep. Because nurses have not studied the effects of these drugs on either the nocturnal sleep patterns or daytime behaviors of older institutionalized adults, there is little knowledge available on which to base the comprehensive and individualized nursing care that is essential to these clients. Therefore, this study sought answers to four research questions:

1 . What is the frequency of the selfreported nocturnal sleep patterns and daytime behaviors of aged, ambulatory, institutionalized adults who take a nightly benzodiazepine sedati ve-hypnotic/anxioly tic, a nightly analgesic, and no nightly medication?

2. What is the relationship between selected demographic variables (age, sex, length of time on medication) and the self-reported nocturnal sleep patterns and daytime behaviors of aged, ambulatory, institutionalized adults?

3. Is there a difference among the selfreported nocturnal sleep patterns and daytime behaviors of those who take a nightly benzodiazepine sedative-hypnotic or anxiolytic, those who take a nightly analgesic, and those who take neither a nightly benzodiazepine sedative-hypnotic/ anxiolytic nor an analgesic?

4. Is there a difference among the selfreported nocturnal sleep patterns and daytime behaviors of those who take a nightly benzodiazepine sedative-hypnotic/anxiolytic , analgesic , and no bedtime medication according to selected demographic variables (age, sex, and length of time on medication)?


Aged subjects were divided into two age categories: those between 65 and 84 years old and those 85 years old and over.

Ambulatory is operationally defined as walking to the dining room at least twice a day as recorded in the medical records, or as confirmed verbally by a licensed member of the nursing staff.

Nocturnal sleep patterns are those behaviors that occur after the patient has folien asleep for the night and before he finally awakens in the morning. They include time to sleep onset; number of awakenings; and physical activities such as going to the bathroom; getting food, drink, medicine; reading; and watching TV. They also include psychosocial activities such as talking, praying, feeling sad or worried, and complaining as well as movement during sleep, soundness of sleep, time of morning awakening, and total sleep time.

Daytime behaviors are those behaviors that occur after the patient has awakened in the morning and before he falls asleep for the night. They include number of daytime naps; feelings of daytime sleepiness; participation in individual, social, or religious activities; side effects related to the benzodiazepine sedative-hypnotics and tranquilizers; reports of fatigue; and state of mind immediately before retiring for the night.

Benzodiazepine sedative-hypnotics and anxiolytics include flurazepam (Dalmane®), temazepam (Restoril®), triazolam (Halcion®), diazepam (Valium®), chlordiazepoxide HCL (Librium®), clorazepate dipotassium (Tranxene®), oxazepam (Serax®), and lorazepam (Ativan®).

Analgesic. Acetaminophen was used as the analgesic.

Length of time on medication refers to those who had taken the drug for less than 30 days and those who had taken it for more than 30 days.

The Study

Data were collected from a convenience sample of 75 ambulatory subjects (18 men and 57 women) who resided in long-term care facilities. They ranged in age from 65 to 98 years, with a mean age of 82.7 years. Twenty-four subjects took acetaminophen nightly, 26 took a benzodiazepine nightly, and 25 took no nightly medication. Of those taking benzodiazepines, 15 took an active metabolite drug (Dalmane®, Valium®, Librium®, Tranxene®) and 11 took an inactive metabolite drug (Restoril®, Halcion®, Serax®, Ativan®).

Subjects with known Alzheimer's or other organic brain disease, Parkinson's disease, thyroid disease, alcoholism, and depression were excluded from the study, because these conditions are known to affect sleep. Patients requiring concurrent treatment with antihistamines, anticholinergics, antidepressants, and anxiolytics (except for sleep) were also excluded, because these medications either alter sleep patterns and/or enhance the sedative-hypnotic effect of the benzodiazepines. Because the elderly commonly experience pain associated with arthritis and other degenerative diseases, those taking analgesics and anti-inflammatory agents were included in the sample in order to obtain more realistic data. Those with sleep apnea and nocturnal myoclonus were also included in the study, because these sleep disorders are common among the elderly.

Two instruments were used to collect the data for this study: the Sleep Pattern and Daytime Behavior Questionnaire and a demographic data sheet.

The Sleep Pattern and Daytime Behavior Questionnaire is a modified version of the Sleep Pattern Questionnaire by Baekeland and Hoy.5 It consists of two parts. Part I contains four questions to be completed immediately before going to bed at night, and Part II contains 12 questions to be completed immediately upon arising in the morning. Each subject completed Part I of the questionnaire for three consecutive nights before retiring. Part II was completed after arising for three consecutive mornings following the completion of Part I.

Demographic data relevant to this study were obtained using each subject's medical records. This information consisted of demographic variables concerning age, sex, type of medication taken, and the length of time the subject was on the medication.

The obtained initial frequencies were submitted to a Pearson ProductMoment Correlation in order to determine variable stability across the three data collection times.

The results showed significant correlations for all variables across time (P < .001). Thus, they were considered to be stable, and average initial frequency scores were obtained. All data were analyzed with descriptive statistics, analysis of variance, factorial analysis of variance, and correlation procedures using these averaged scores.

The .05 level was set as the acceptable level of significance for this study.

The analysis concerning Research Questions #1 and #2 included data from all 75 subjects. However, because of the small number of males in the study and the significant relationships found for gender in Question #2, only data from the 57 women were used in the analysis of Research Questions #3 and #4.


The findings suggest that the individuals in this study perceived some disruption in their nocturnal sleep patterns, regardless of whether or not they had taken medication at bedtime (Research Question #1). Persons who had taken benzodiazepines, acetaminophen, and no medication went to bed and awakened early. While the majority of them spent over eight hours in bed, they experienced prolonged sleep onset latencies (over 15 minutes), frequent nocturnal awakenings, and a light-tomedium level of sleep.

These patterns were reported more frequently by those who had taken a benzodiazepine sedative-hypnotic/anxiolytic than by those who had taken acetaminophen or no medication at bedtime. As could be expected, individuals in that group were also less refreshed on awakening and less satisfied with their sleep than those in the other two groups.

When the relationship between gender and nocturnal sleep patterns was considered (Research Question #2), it was found that females tended to have a higher incidence of subjective sleep complaints than males. They tended to have longer sleep onset latencies, experience more frequent awakenings, and engage in more activities while awake. They also felt more tired on awakening and were more dissatisfied with thensleep.

The findings also indicate that there was no relationship between age and nocturnal sleep patterns (Research Questions #2 and #4). This information suggests that persons between 65 and 84 years old have sleep patterns that are similar to those of individuals over 85 years of age.

When the differences among the selfreported nocturnal sleep patterns of females who had taken benzodiazepines, acetaminophen, and no medication at bedtime were considered, it was found that sleep patterns were least disturbed in those taking no medication and acetaminophen (Research Question #3). Although these patterns were more disturbed in those taking benzodiazepines, there was no difference between those taking drugs with active and inactive metabolites. Sleep was disrupted to the same degree by both categories of benzodiazepines. Individuals in these groups awakened more often, moved more while asleep, and slept less soundly than those in the acetaminophen and no medication groups. They also felt less refreshed on awakening and were less satisfied with their sleep.

Further analysis concerning the differences among the sleep patterns of those who take medication at bedtime suggests that the length of time on medication is an important consideration (Research Questions #2 and #4). Regardless of the type of medication taken, sleep was less disturbed in those who had taken it for less than 30 days. Although statistically significant differences among the medications were not found when the less than 30 day group was looked at, a visual inspection of these results suggests that benzodiazepines with active metabolites may be less disruptive to the sleep patterns of the elderly when taken for a limited time. Acetaminophen and benzodiazepines with inactive metabolites may be more disruptive.

When medication is taken for more than 30 days, statistically significant differences are found between acetaminophen and the benzodiazepines. Individuals who take acetaminophen experience fewer sleep disturbances. However, there is no difference between the sleep patterns of those who take benzodiazepines with active and inactive metabolites. Thus, these drugs may be equally disruptive to the sleep of the elderly when taken for more than 30 days.

The findings concerning the daytime behaviors of those who took benzodiazepines, acetaminophen, and no bedtime medication were as expected (Research Question #1). Regardless of whether or not a bedtime medication was taken, most people napped at least once a day and felt fairly refreshed after napping. Feelings of sleepiness without napping were not pronounced among individuals in this study. Although persons in all three groups reported skipping activities due to tiredness and experiencing problems, such as difficulty walking alone, these behaviors were more frequent in the benzodiazepine group.

When age and daytime behaviors were considered, no relationships were evident (Research Question #2). Individuals between the ages of 65 and 84 years engaged in behaviors that were similar to those of individuals over 85 years old. However, when gender relationships were investigated, it was found that men and women differed. Males tended to take more naps and participate in fewer recreational activities than females. Women tended to feel tired more often without napping than men.

When the differences among the selfreported daytime behaviors of women who had taken benzodiazepines, acetaminophen and no bedtime medication were considered, it was found that the majority of these behaviors were not significantly disturbed (Research Question #3). Individuals who took acetaminophen or no medication at bedtime skipped fewer daytime activities and had fewer problems than those who took benzodiazepines. However, there was no difference between the inactive and active metabolite drug groups for either of these behaviors. Individuals in both of the benzodiazepine groups skipped approximately the same number of activities and experienced the same number of problems. Regardless of the type of medication taken, subjects tended to feel calm and normally tired at bedtime, nap during the day, and participate in approximately the same number of activities.

Length of time on medication was not a significant factor for most of the daytime behaviors in this study (Research Question #4). Subjects who had taken a medication for less than 30 days tended to skip fewer activities due to tiredness than those who had taken one for more than 30 days. There were no differences in the remainder of the daytime behaviors. This suggests that, regardless of the length of time on medication, individuals tended to take as many naps, experience as many problems, participate in as many activities, and experience approximately the same states of mind and fatigue at bedtime.

Finally, it is interesting to note that all of the elderly adults (N = 24) who took acetaminophen at bedtime believed they were taking a sedativehypnotic. Of the 51 subjects taking either acetaminophen or a benzodiazepine at bedtime, 34 took it because it was given to them on a routine basis. Only 17 individuals had requested a bedtime medication.

Nursing Implications

There are several implications of these findings for gerontological nurses. First, nurses must recognize the normal changes in sleep that occur with age in order to plan effective care for the elderly institutionalized adult. Rather than expecting the patient to sleep through the night, the nurse should anticipate disturbed sleep patterns. Patients should be informed that changes in sleep are a normal part of the aging process. Non-pharmacological interventions to induce and maintain sleep should be a planned part of the individual's evening care, i.e., progressive relaxation, back massage, relaxing music, or a non-stimulating activity.

Secondly, nurses must recognize the disruptive influence that the prolonged administration of the benzodiazepines has on the sleep patterns and daytime activities of older adults in long-term care facilities. Unless specifically ordered by the physician and/or requested by the patient, these medications should not be given on a routine basis. If the patient does request a bedtime medication and several alternative drugs are ordered, (ie, acetaminophen versus a sedative-hypnotic) the nurse should encourage the patient to take the medication that is least disruptive to sleep and daytime functioning.

Finally, nurses must continue to conduct research concerning the needs for sleep and rest presented by the older adult who resides in a long-term care facility. More knowledge is needed relative to the basic need for sleep, the identification of individuals who experience sleep disturbances, the effect of various medications on sleep patterns, and the most effective nonpharmacological sleep promoting interventions for use by health professionals.


  • 1 . Lemer R: Sleep loss in the aged: Implications for nursing practice. J Gerontol Nurs 1982; 8:323-326.
  • 2. Mendelson WB: The Use and Misuse of Sleeping Pills. New York, Plenum Medical Book Co, 1980.
  • 3. Goldson RL: Management of sleep disorders in the elderly. Drugs 1981; 21:390-3%.
  • 4. Greenberg R: Insomnia and sleep disturbances in the aged: Introduction. J Geriatr Psychiatry 1980; 13:131-134.
  • 5. Baekeland F, Hoy P: Reported vs. recorded sleep characteristics. Arch Gen Psychiatry 1971;24:548-551.


Sign up to receive

Journal E-contents