BRITISH MEDICAL JOURNAL
27 FEBRUARY 1988
PAPERS AND SHORT REPORTS
Prevalence, frequency, and duration of hypnotic drug use
among the elderly living at home
SHAH EBRAHIM, TOM ARIE,
PETER H FENTEM
Details of consumption of hypnotic drugs derived from a
nationally representative sample ofelderly people were analysed
in terms ofthe prevalence, duration, and likely frequency ofuse.
Of 1020 randomly selected subjects aged 65 and over 16% (166)
reported using (mainly benzodiazepine) hypnotic drugs, and of
these 89% reported having taken such a drug the night before the
interview. Most of these users (73%) had been taking hypnotic
drugs for more than one year, with 25% reporting drug use for
more than 10 years.
These results suggest that for most elderly users of hypnotic
drugs, patterns of consumption encourage the development of
cumulative effects and benzodiazepine dependence.
Most ofthe disadvantages associated with the use ofhypnotic drugs
in the elderly are directly related to the continuity and duration
of drug use. Thus confusional states,' impaired psychomotor
performance,2 increased daytime anxiety,3 rebound insomnia,4 and
dependence5 are all more likely if the hypnotic drug is taken
regularly over long periods. The type of hypnotic drug taken also
influences unwanted side effects, with long acting drugs being more
likely to accumulate and disrupt daytime activities,2 while very
short acting drugs produce an earlier and more severe rebound on
withdrawal.6 Surveys ofhypnotic drug use, however, have tended to
focus mainly on overall use, paying less (if any) attention to the
The Activity and Ageing Research Group, Departments ofHealth Care ofthe
Elderly and of Physiology and Pharmacology, University of Nottingham
Medical School, Queen's Medical Centre, Nottingham NG7 2UH
KEVIN MORGAN, BSC, PHD, lecturer
HELEN DALLOSSO, MSC, PHD, research fellow
SHAH EBRAHIM, DM, MRCP, lecturer
TOM ARIE, FRCP, FRCPSYCH, professor, department of health care of the
PETER H FENTEM, MSc, MRCP, professor, department of physiology and
Correspondence to: Dr Kevin Morgan.
particular drugs prescribed or the frequency and duration with
which such drugs are taken. If surveys of the use ofhypnotic drugs
among the elderly are to serve as useful feedback this additional
information is essential. Here we describe the characteristics and
implications of hypnotic drug use in a large random sample of
elderly people living at home.
Patients and methods
Data were derived from the activity and aging survey, full details ofwhich
are presented elsewhere.78 Briefly, the study was designed to investigate
relations between mental and physical health, psychological wellbeing, and
customary physical activity among the old (aged 65-74) and the very old
(aged 75 or over) living at home. Quality of sleep was identified as an
important index of both physical and psychological wellbeing. Of 1599
elderly patients randomly selected from the Nottinghamshire Family
Practitioner Committee's records (747 aged 65-74; 852 aged 75 or over) an
effective sample of 1299 were available for interview. Causes ofloss from the
original target sample included death, admission to hospital or residential
care, and change of address. Of the sample of 1299, 1042 agreed to
participate in the survey (a reponse rate of 80%). Within this stratified
sample the age group
-75 was intentionally overrepresented to admit
sufficient numbers for both present and prospective analyses. Thus while
the ratio of "old" to "very old" people in the general population is roughly
1 -62: l,9 a ratio of 1:1 was aimed at in this survey.
Interviews were conducted between May and September 1985 by
10 women aged 27-56. Each interviewer received two weeks of intensive
classroom based training and was closely monitored for a further two weeks
of practical experience. The structured questionnaire covered aspects of
health and lifestyle and included questions about subjective characteristics
of sleep, the existence and nature of sleep problems, and currently
prescribed hypnotic drugs. Reported use of drugs was verified by the
interviewer, who was instructed to ask for and examine tangible evidence of
prescribed medication (bottles, cartons, prescriptions, etc). Hypnotics were
defined as prescription drugs with sedative properties, the administration of
which was intended to promote sleep.
PREVALENCE AND FREQUENCY OF USE
Information on use of hypnotic drugs was available from 1020 re-
spondents. The distribution of responses to the question "Do you ever
BRITISH MEDICAL JOURNAL
27 FEBRUARY 1988
take tablets to help you get to sleep?" is shown in table I. A total of
166 respondents (16%) reported using prescription hypnotics at least
"sometimes," equivalent to a weighted prevalence (correcting for the over
sampled >75 age group) of 15%. Levels of use of hypnotic drugs increased
significantly with age, showing a prevalence of 13% among those aged 65-74,
and 20% among those aged >75 (X2=902, df=1, p<O005). When the
results for the sexes were analysed separately, however, this age related
increase was significant only for women, who, overall, were more likely to
report using drugs to help them sleep than were men (table II).
TABLE I-Distribution of responses to the question
"Doyou ever take tablets to helpyou get to sleep?"
No (%) ofpatients answering
All the time
TABLE iI-Reported hypnotic drug use by sex
No (%) ofmen No (%) ofwomen
Hypnotics used at least sometimes
Hypnotics used seldom or never
x2= 19-31,df=1, p<0-001.
TABLE III-Distribution ofhypnotic drugs prescribed
Generic name or
No (%) ofpatients
Generic name or
No (%) ofpatients
Respondents were also asked when they had last taken a sleeping tablet.
Of those reporting use at least "sometimes," 89% had taken a hypnotic the
night before the interview, while a further 8% reported taking such drugs in
the preceding week. The time since last taking a hypnotic was not
significantly related to age or sex.
DURATION OF USE
Of those reporting the use of hypnotics at least "sometimes," 27%
reported taking these drugs for less than one year. Durations of 1-5 years,
5-10 years, and more than 10 years were reported by the remaining 30%,
19%, and 25%, respectively. The duration of hypnotic drug use was
significantly related to age, with the younger age group (65-74) tending to
predominate among the longer term users (X2=8-28) df=3, p<OO5).
Duration ofuse showed no significant relation with sex.
Table III shows the distribution ofhypnotic drugs prescribed. The most
frequently prescribed drug was nitrazepam (accounting for 43% of all
prescriptions). To some extent, the reported duration of drug use reflected
the length of time a particular preparation had been on the market. Thus
63% of those taking diazepam and 55% of those taking nitrazepam reported
using drugs for longer than five years. A similar duration ofuse was reported
by only 32% of those using temazepam.
In 1980 the Committee on the Review of Medicines recom-
mended that benzodiazepine hypnotics among the elderly should be
prescribed "for short periods of time, and only after careful
consideration."'° Nevertheless, the fairly high levels of hypnotic
drug use found in this survey are very similar to those reported
throughout 1961-82." Thus from our data and those reported
earlier it seems that roughly 10-15% ofthe elderly population take a
hypnotic drug each night. In Britain this translates into between
0-8 million and 1 million people each night. In addition to fairly
high levels of use, protracted use, often over five years, continues
to typify consumption of hypnotic drugs among the elderly.
Furthermore, judging by the high proportion of the users in our
study who reported taking a hypnotic on the night before the
interview regular long term use is not uncommon.
The likelihood ofhaving ahypnotic drug prescribed seemed, as in
other surveys, to increase with age, particularly among women." It
is particularly interesting, however, to note that while sex seems to
influence the overall level of use of hypnotic drugs significantly, it
was not associated with the reported duration or likely frequency of
The most frequently prescribed drug was nitrazepam, a par-
ticularly long acting hypnotic associated with confusional states,'
impaired psychomotor performance,2 and residual sedation'2 in
elderly patients. Nevertheless, there was evidence in our study of a
transition from longer to shorter acting hypnotic agents. Given the
relation already described between duration of drug use and the
actual drug prescribed, it should be noted that many ofthose using
nitrazepam were probably the survivors of earlier prescribing
practices. In contrast, among those who had been taking hypnotics
for only a fairly short time a preference for shorter halflife products
Estimates of the effective "lifespan" of hypnotic drugs vary,
with the Committee on the Review of Medicines supporting the
suggestion that "most hypnotics tend to lose their sleep promoting
properties within three to 14 days of continuous use."'0 From our
data, therefore, it seems that many, perhaps most, elderly users of
hypnotic drugs are unnecessarily exposed to the risks of drug
accumulation or daytime withdrawal effects. Furthermore, the
overall pattern ofuse ofhypnotic drugs in this representative sample
was consistent with the development ofbenzodiazepine dependence
in a substantial number ofelderly patients.
This study was conducted by the Activity and Ageing Research Group,
supported by a grant from the Grand Charity.
I Evans JG, Jarvis EH. Nitrazepam and the elderly. BrMedJ 1972;iv:487.
2 Morgan K. Effects ofrepeated dose nitrazepam and lormetazepam on psychomotor performance
in the elderly. Psychopharmacology 1985;86:209-1 1.
3 Morgan K, Oswald I. Anxiety caused by a short life hypnotic. BrMedJ 1982;284:942.
4 Oswald 1, Adam K, Borrow S, Idzikowski C. The effects of two hypnotics on sleep, subjective
feelings and skilled performance. In: Passouant P, Oswald I, eds. Pharmacology ofthe states of
alertness. Oxford: Pergamon Press, 1979:51-63.
5 Higgit AC, Lader MH, Fonagy P. Clinical management ofbenzodiazepine dependence.BrMed_J
6 Adam K, Oswald I, Shapiro C. Effects of loprazolam and of triazolam on sleep and overnight
urinary cortisol. Psychopharmacology 1984;82:389-94.
7 Ebrahim S, Morgan K, Dallosso H, Bassey J, Harries U, Terry A. Interviewing the elderly about
their health: validity and effects on family doctor contacts. AgeAgeing 1987;16:52-7.
8 Morgan K, Dallosso HM, Arie T, Byrne EJ, Jones R, Waite J. Mental health and psychological
wellbeing among the old and the very old living at home. BrJ Psychiatr 1987;150:801-7.
9 Office of Population Censuses and Surveys. Census 1981: sex, age, and marital status. London:
10 Committee on the Review of Medicines. Systematic review of the benzodiazepines. Br Med J
11 Morgan K. Sedative-hypnotic drug use and ageing. Arch Gerontol Geriatr 1983;2:181-99.
12 Greenblatt DJ, Allen M. Toxicity of nitrazepam in the elderly: a report from the Boston
collaborative drug surveillance program. BrJ Clin Pharmacol 1978;5:407-13.
(Accepted 3 November 1987)