It is important to emphasize two related points. Hypnotic usage becomes more extensive and also more chronic with increasing age. The elderly and particularly the aged are very susceptible to depressant effects because of both pharmacokinetic and pharmacodynamic differences from younger individuals. Secondly, insomniac individuals tend to wake during the night even after taking a sleeping draught. Accordingly, it is important to evaluate the effects of these medications at various times after ingestion and not just residual effects the next day. This applies with greatest force to the elderly.

Systematic research into the psychological effects of hypnotic medication dates back a little over 30 years. Awareness of the problem was stimulated by clinical reports of the effects of first barbiturates and then benzodiazepines persisting into the next day. This was a consequence of these compounds having durations of action which extended beyond 6 - 8 hours. The earliest benzodiazepines marketed for insomnia, such as nitrazepam and flurazepam, were long-lasting. As well as residual effects the next day, these drugs could accumulate and in the elderly in particular could result in marked sedation and even confusion and pseudo-dementia.

Since the 1970s, the trend has been towards shorter-acting medications. The advent of lorazepam, temazepam, lormetazepam and others gave rise to a spate of studies showing fewer and sometimes minimal residual effects. However, dose-effect studies established quite clearly that the magnitude of these effects was dose-dependent. Furthermore, predictions from the pharmacokinetics of the drugs, and any known active metabolites, were generally borne out.

The next step was the introduction of even shorter-acting compounds such as the benzodiazepines, triazolam and midazolam, and the non-benzodiazepines, zopiclone and zolpidem (and zaleplon under development). Residual effects were now minimal or even absent at clinically recommended doses. The newer compounds also appeared less potent at inducing these effects, perhaps reflecting their more selective binding profile.

As well as developments in the benzodiazepines, there has been a trend in some other countries to exploit the sedative actions of other classes of psychotropic drugs in order to facilitate sleep. For example, the sedative antidepressant Amitriptyline has been given as a large single dose at night. The sedative effects are due to a blockade of histamine receptors in the brain. Antipsychotic drugs have also been used in this way.

Antihistamines themselves such as Promethazine are available in several over-the-counter remedies. Unfortunately, many are long-acting and persist the next morning. Herbal remedies such as Valerian are also popular as are folk remedies such as hop-filled pillows. On the negative side, drugs such as sympatho-mimetic bronchiodilators can impede the onset of sleep. So can caffeine, and alcohol results in short periods of sleep with subsequent periods of rebound.

Populations

Much of the early work was carried out in normal volunteers, and often after single-dose exposures only. These studies enabled potential problems to be identified but only went part way in delineating drawbacks in insomniac patients. Such patients already complain of poor day time functioning because of the insomnia and objective testing doses suggest some impairment. Consequently, hypnotic drugs could be postulated to have two opposing effects:

1) an indirect effect in improving sleep and next-day psychological functioning, and;

2) a direct effect with residual depressant effects the next day.

Both effects are dose-related but may not necessarily follow the same dose-effect curve. Some studies have involved elderly subjects, both healthy and insomniac. As expected more marked and prolonged effects are seen than in younger adults. The elderly brain seems more sensitive in general to depressant drug effects perhaps because there are fewer receptors for the drug to bind to, and hence a higher receptor occupancy rate.

A recent development has been the study of early drug effects, for example, from 1-hour onwards, thus capturing maximum effects. Because newer hypnotics are shorter acting, their efficacy resides in curtailing the Time to Sleep Onset (TSO). To do this, they must have a rapid onset of action and this may result in a high Cmax, with the danger of marked depressant effects or even neuropsychiatric reactions such as amnesia or hostility.

All these potential effects need careful evaluation in the target populations.

Techniques

A wide range of techniques has been used ranging from apparently simple laboratory tasks such as reaction-time to complex real-life situations such as car-driving, simulated or real (see table). Earlier studies were largely empirical with tests being chosen for their ability to detect effects. The interpretation of those effects was often difficult. For example, an apparently simple task such as reaction time, involves perception, control processing, perhaps decision-making and an executive output. More recently, attempts have been made to use tests known to be sensitive to benzodiazepine effects and more easily interpretable. An example is semantic memory.

The other development has been to attempt to evaluate hypnotic drug effects on complex learned skills such as driving a car or piloting a plane. The problem is that over-learned skills of this sort may show little significant drug effect until the unexpected happens and overload occurs. Epidemiological data, for example, relating to road traffic accidents have been adduced in evaluating hypnotic drug effects but need the institution of standard and valid data-gathering techniques coupled with assessments of drug ingestion or bodily concentrations, e.g. plasma, urine or saliva.

2. Long term use needs monitoring. Does tolerance occur to the extent that unwanted effects become nugatory or do some functions retain their sensitivity to these drugs indefinitely?

3. The elderly must continue to be the focus of special attention. Usage is greater in terms of prevalence and duration, and side effects more frequent, severe and potentially serious than in younger individuals.

2. The onset and offset of depressant effects needs further detailed study as effects may follow change in bodily concentrations rather than absolute levels. Receptor occupancy studies using PET or SPECT should be encouraged.

3. Real-life tests should be further elaborated but correlations to laboratory tests should be incorporated into this program of research.

4. Advantage should be taken of any long term withdrawal studies to detect long term depressant effects.

5. Newer compounds need thorough evaluation as it must not be assumed that all hypnotics have the same profile of action.

6. Effects may not be identical in men and women so studies should be designed and analysed to preserve potential sex differences.

7. Non-pharmacological techniques should be evaluated in a parallel fashion to drugs.

8. The public need education on the possible risk of hypnotic medication, particularly interactions with alcohol.