Below is the uncorrected machine-read text of this chapter, intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text of each book. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.
INTRODUCTION AND SUMMARY Many drugs have been marketed in the past several decades for the alleviation of sleep problems. Most of them have been regarded initially as safer than barbiturates and later were found to pose hazards as great as those attributed to barbiturates. The benzodiaze- pines, first marketed in the early 1960's, possibly may be exceptions to this pattern of disappointments. It is unusual, for example, for an overdose of a benzodiazepine alone to be fatal, and this particular safety aspect has been widely recognized In view of the availability of these newer, apparently safer drugs, the Institute of Medicine was asked by the White House and the National Institute on Drug Abuse to review the available evidence and make recommendations concerning the continuing clinical need for the short-to medium-acting barbiturates. The request was based, in part, on the President's expressed concern that needless deaths from accidental overdose or suicide were occurring among individuals using certain barbiturate drugs. 1/ The drugs of particular concern to policymakers were the commonly prescribed sleeping pills (hypnotics) containing secobarbital, pentobarbital, and amobarbital. From the point of view of public health policy, little or no concern was expressed about phenobarbital or butabarbital, which are prescribed mostly to provide daytime sedation or to control epileptic seizures; unlike the hypnotics, these barbiturates are only infrequently used in suicide and rarely in drug abuse. The Institute of Medicine decided that the public health issues associated not only with the barbiturate hypnotics, but also with other hypnotic drugs, justified a study of the uses of all these substances in medical practice. The steering committee appointed to guide the study included specialists in consumer affairs, food and drug law, internal medicine, nursing, health professional educa- tion, pharmacology, psychiatry, and sleep disorders. The study was begun in October 1977. This study report has its major emphasis on the drugs prescribed as sleeping pills. Some effort also is made to assess the extent to which anti-anxiety drugs such as diazepam (Valium(R)) and anti-depressant drugs such as amitriptyline (Elavil(R)) are prescribed to help promote sleep, and to examine the rationale for this particular use; however, the report does not explore the broad range of questions that are
often raised about their roles in modern health care. Hospital practice is not discussed in detail, although the principles for prescribing hypnotic medication should be similar both in hospital and ambulatory care settings. Nine-tenths of the prescribing of hypnotic drugs occurs in the context of ambulatory care 2/ and insofar as these drugs are implicated in suicide or drug abuse, this usually occurs outside the framework of hospital practice. Over-the-counter sleep aids, available without prescription, are not considered in detail. The Food and Drug Administration, having reviewed safety and efficacy studies of those preparations, recently initiated proceedings to withdraw most of them from the market. Alcohol, which often is used to induce sleep, is not discussed in depth except to document its inefficiency as a hypnotic agent and the additive nature of its undesirable effects when consumed with hypnotic drugs. The committee recognizes that alcohol is a drug, available without a prescription, that has similarities in its behavioral effects to barbiturates and other hypnotics. The health problems associated with the use of potentially hazardous hypnotic drugs result both from drugs obtained through legitimate prescription and from those obtained through illegitimate or unlawful means. For purposes of defining the scope of this report, the committee identified the following prototypical circumstances of acquisition and use of hypnotic drugs: (1) appropriate prescribing for valid medical indica- tions, within the confines of a legitimate physician- patient relationship; (2) inappropriate prescribing/inadequate medical surveillance -- pursuant to some health complaint - and within a valid physician-patient relationship, but inappropriate for any of several reasons, for example, prescribing more dosage units than are medically indicated over a given time period, prescribing for improper or questionable indica- tions, long-term prescribing without monitoring the patient's progress or condition through periodic reassessment (including any tendency to increase dosage or otherwise misuse the drug), failure to ascertain other medications that a patient is taking simultaneously, or failure to alert a patient to the particular hazards associa- ted with the use of a drug; 2
(3) manipulation of the prescriber -- obtaining prescrip- tions from a physician by subterfuge, for example, by posing as a patient or visiting several different physicians to obtain multiple prescriptions for the same drug; (4) illicit prescribing -- prescribing where there is no physician-patient relationship and/or legitimate health complaint by the person receiving the prescription; (5) nonprescription channels -- diversion in the form of thefts at retail or wholesale levels, or prescrip- tion forgeries, unrecorded direct sales without prescriptions by physicians or pharmacists, smuggling, illegal manufacture and street sales. The committee's mandate and expertise lay in identifying impli- cations of hypnotic prescribing and use for medical practice; there- fore its deliberations centered on the first two categories. The committee recognizes that the demarcation among the categories is not absolute, and that even if the categories were more clearly de- fined it is not known what portion of the hypnotics involved in the various public health problems originates from each of the categories. The adverse effects and public health problems of hypnotic drugs that are of principal concern in this report are those that arise in the context of conscientious health care and fairly typical prescribing practices. These include suicide attempts, interactions between the drugs and alcohol, and residual effects on daytime performance. The only type of "drug dependence" discussed here in detail is nightly reliance on drugs for sleep. A number of inadequacies in available data have prevented the committee from reaching definitive conclusions on some issues. There are no data available, for example, on the circumstances in which a drug was being used by persons treated in emergency rooms for adverse reactions. Additionally, the reporting of deaths attributed to drug overdose is recognized to be quite variable and inaccurate, and only a relatively few such deaths are subjected to adequate toxicological confirmation. As part of the same problem, the current reporting code system for deaths is inadequate for purposes of tracing trends in the involvement of individual drugs in overdose deaths. There also are no data available on the determinants of physicians' choice of therapy for complaints of unsatisfactory sleep. Although the time frame for this study did not permit the under- taking of the kind of national survey of physicians that would 3
provide such data, a sample instrument was designed and tested. Forty- nine members of the Institute of Medicine who are in medical practice were asked to complete the questionnaire; 35 responded. Ten of those members met with steering committee members in August 1978 to discuss their responses to the questionnaire and to present their viewpoints and experiences in dealing with sleep complaints in their practices. In October 1978 the survey instrument was submitted to the study's governmental sponsors for their possible future use. A summary of each section of the report follows, together with the committee's conclusions in brief. The committee's major findings and full conclusions and recommendations will be found in Chapter 6. Sleep and Medication - An Overview Sleep consists of two distinct stages: REM sleep, a period in which (R)apid (E)ye (M)ovements may be detected in which dreaming occurs, and (N)onREM sleep. NREM sleep is classified in four stages on the basis of relatively distinguishable brain wave patterns by electroencephalography. Stage 1 is a brief transitional stage between wakefulness and sleep, and makes up about 5 percent of the total night' sleep. Stage 2 usually constitutes 50 to 60 percent of total sleep in the young adult. Stages 3 and 4 are characterized respectively by moderate and large numbers of delta waves. Most stage 3 and 4 sleep occurs during the first one to three hours of sleep in young adults. Sleep patterns typically vary with age. In the newborn, for example, sleep stages are not well-defined, total sleep time averages 14 to 16 hours, and about 50 percent of that time is spent in REM sleep. In middle and old age, stage and 3 and 4 sleep decreases markedly and sleep tends to become progressively more fragmented, with brief arousals and longer periods of wakefulness. Neurological mechanisms that underlie the different sleep states may significantly alter control of vital regulatory functions, such as breathing, temperature, and heart beat, with the result that some individuals who are physically normal while awake can develop serious respiratory or cardiac problems with asleep. Recognition of these physiological alterations during sleep is important to the prac- ticing physician; for instance, hypnotic drugs can cause central respir- atory depression and generally should not be administered to patients whose breathing already is compromised by sleep. The mayor classes of hypnotics are the barbiturates, which have been in use since the early 1900s; the benzodiazepines, only one of which -- flurazepam (Dalmane(R)) -- has been marketed exclusively 4
as a hypnotic in the United States since its introduction in 1970; chloral hydrate, which has been in clinical use since the nineteenth century; and glutethimide (Doriden(R)), introduced in the early 1950s. The barbiturates used primarily as hypnotics are pentobarbital, secobarbital, and amobarbital. These are usually classified as short- to-intermediate acting based on their rate of elimination from the central nervous system. The plasma half-life of an initial dose of a barbiturate varies from 14 to 42 hours. With consecutive doses, the half-lives become shorter because of the tendency of these drugs to stimulate the liver enzymes which metabolize them. The longer-acting barbiturates, butabarbital and phenobarbital, are most commonly used as anxiolytics, and phenobarbital is especially valued as an anticonvulsant. These two drugs, together with the ultra- short-acting barbiturates used only for anesthesia, are not included in the deliberations of this report because the available data indicates that they are very rarely part of the problems which were the concern of the study committee. The benzodiazepines were introduced into medical practice in 1961, and were marketed initially as anxiolytics and muscle relaxants. Flurazepam, today the most widely-prescribed hypnotic was introduced into clinical practice in 1970. Most benzodiazepines have psychoactive metabolites which have a substantially longer duration in the body than the barbiturates. Thus, the major metabolite of flurazepam has a plasma half-life of SO to 100 hours, leading to accumulation when the drug is taken on consecutive nights. This build-up is of some concern because of demonstrated residual daytime effects on tasks requiring visual motor coordination. The most widely prescribed non-barbiturate, non-benzodiazepines are glutethimide (Doriden (R)), methaqualone (Quaalude(R)), chloral hydrate, methyprylon (Noludar(R)), and ethchlorvynol (Placidyl CRIB. Except for methaqualone, these hypnotics have half-lives varying from 4 to 8 hours; that of methequalone, previously believed to be 18-42 hours, has recently been shown to be as long as 70 hours. All of these hypnotics exhibit increased toxic effects when combined with alcohol, and many deaths, either accidental or suicidal, are attributed to the mixture. Even with the benzodiazepines, where there are few reported deaths resulting from ingestion of the drug alone, mixing them with alcohol or other depressant drugs creates a lethal risk.
Epidemiology of Sleep Problems and Prescribing The major indication for prescribing hypnotics is insomnia, or sleep disturbance. Surveys intended to measure the prevalence of sleep complaints, however, are difficult to compare because of the ambiguity of the word "insomnia." Nonetheless, a review of such studies shows that between 29 to 38 percent of individuals over 18 in the United States (or 45 to 60 million persons) perceive themselves as having had "trouble" sleeping within a given year, but only about 2 percent of persons would characterize the problem as "insomnia." About one-sixth of persons with a sleeping problem take the complaint to a physician, and about half of those going to the doctor -- 4 to 6 million persons -- receive a prescription for sleeping medication. About half of all patients admitted to short-term general hospitals (in 1977 there were 33 million admissions) are given sleeping medication, apparently more as a hospital policy than in response to specific sleep complaints. Because physicians receive most of their clinical training in the hospital setting, it is reasonable to assume that exposure to casual prescribing of hypnotics would influence their future conduct in ambulatory practice. Physician propensity for writing hypnotic prescriptions in ambula- tory care has declined appreciably over the past seven years. In 1971, 41.7 million prescriptions were written, 47 percent of which were for barbiturates. By contrast, in 1977 only 25.6 million prescriptions were written for hypnotics, a decrease of 39 percent; 17 percent of these were for barbiturates and 53 percent (13.6 million prescriptions) were for the single drug flurazepam (Dalmane(R)~. In 1977, two to three percent of anti-anxiety drugs, antidepressants and antihistamines -- an additional 5 million prescriptions -- also were specifically prescribed to induce sleep. An unknown additional quantity of these drugs was used by patients to induce sleep, even though not prescribed specifically for that purpose. In 1977, about one-third of hypnotic prescribing was by family and general practitioners, who also dealt with one-third of physician office visits. Psychiatrists accounted for 23 percent of hypnotic prescribing, and internists another 15 percent. These two specialties provide 3 and 11 percent of all physician office visits, respectively. General and family practitioners accounted for 41 percent of instances of new therapy with barbiturates, whereas psychiatrists appear to have virtually abandoned the barbiturates in favor of flurazepam. Thirty-nine percent of hypnotic prescribing in 1977 was for persons over age 60, who constitute 15 percent of the population and account for approximately 11 percent of all office visits. More than half of the barbiturate prescribing was for this age group, most of it being a continuation of previous therapy. 6
~ Surveys intended to measure the prevalence of hypnotic use in the non-hospitalized population are difficult to compare because of defini- tional problems, sample design, and question format. Estimating from two major national studies, it appears that about 5.7 percent of the United States population over 18 about 8.5 million people - used a prescription hypnotic at some time in 1977. The proportions of persons within each age group who had used a hypnotic increased with age and women were somewhat more likely to have used a hypnotic than men. About a quarter of the people who use hypnotics during a year will use them on consecutive nights for two months or longer. Public Health Problems Associated with Hypnotics The public health problems associated with the wide availability of hypnotic medication include increased mortality from its use in suicide or accidental overdose; costs of treatment for non-fatal over- dose emergencies; morbidity, mortality, increased crime and lost production due to addiction to these drugs; and potential hazards to public and personal safety from persons driving automobiles or operating other machinery while under the influence of these drugs. National statistics show that there was a 52 percent decrease between 1970 and 1976 in the number of suicides using barbiturates. During the same period there was a partially offsetting increase in the numbers of suicides using other drugs, combinations of drugs, or combinations of drugs and alcohol. The result was an overall decrease in the drug-related suicide rate of 13 percent. But the total number of drug deaths from accidental overdose increased slightly, so the net result was little change in the total number of drug-related deaths. The decline in barbiturate deaths has paralleled a decline in the number of prescriptions written for them. A similar pattern was found in Australia after the imposition of controls on the prescrip- tion of barbiturates. Laboratory experiments have shown that hypnotic drugs can cause impairment of psychomotor skills the following day, which has impli- cations for traffic safety. In studies of the driving population and accident victims, however, the involvement of alcohol often makes it difficult to measure the strength of the relationships between drugs and traffic accidents. Insomnia Research, Diagnosis, and Therapy Insomnia is a subjective complaint, and as with other types of subjective distress, such as lower back pain, neither the nature of the problem nor the value of the relief commonly provided is well
understood. It also is not clear why some people complain of inadequate or unsatisfactory sleep when, according to sleep laboratory studies, the objective measures of sleep for many insomniacs do not really differ from those of non-complaining normal subjects. At the same time, there are well-documented cases of individuals who sleep very little each night but do not f ind reas on f or comp faint . The physician faced with a sleep complaint, then, must bear in mind that the discomf ort may be exaggerated out of all proportion to anything that can be measured obj actively. The patient may fall as leep more quickly than is stated, and use of a hypnotic would Perhaps shorten the time to fall asleep by only 10 to 20 minutes. These possibilities must be weighed against the dangers of releasing a potent drug to the control of the patient who, for example, may be depressed and therefore at risk to commit suicide, or who may be suffering from a physiological sleep disorder such as sleep apnea. Individuals with sleep apnea are unable to breathe properly while asleep. After falling asleep, they will experience a series of events in which breathing stops for a few seconds to more than a minute. These periods terminate with partial arousal and a gasping for air. This sequence of events may occur hundreds of times during the night without any awareness on the part of the pat lent . The prescription of a hypnotic medication to such a patient could prove fatal. Although the classification and management of insomnia are still in the early stages of development, much has been learned in recent years that is of importance to the practicing physician. Perhaps the two most important principles are: · consider the differential diagnosis · treat the underlying disorder whenever possible. If hypnotic drug therapy is selected, the physician ' s prudent pre- scribing of medication should be mainly f or short-term management of transient insomnia in patients whose use of the hypnotics can be monitored. Each class of hypnotic drug and related psychotropic medication has its advantages and disadvantages, which should be weighed each time a prescription is written. Until recently, findings about the suppression of REM sleep by most hypnotic drugs were considered important in selecting hypnotics. When benzodiazepine hypnotics were f ound not to greatly suppres s REM s leep, this was thought to be an advance toward a hypnotic that produced "natural" sleep. In fact, benzodiazepines have been shown to powerfully suppress Stage 3 and 4 sleep, the clinical importance of which could be as great (or as insignificant ~ as is REM suppression by other types of hypnotics. 8
Various non-pharmacological methods of dealing with insomnia have been proposed. These include behavioral therapies, office counseling, and psychotherapy. The results of studies of behavioral therapy with chronic insomniacs are inconclusive, and office counseling and psycho- therapy have not been evaluated specifically for relief of insomnia. Efficacy and Hazards of Hypnotics The efficacy of hypnotic agents is difficult to judge. If a medical treatment is stated to be effective, the usual implication Is that it cures or ameliorates some aspect of a specific disorder in a given proportion of patients. A statement of efficacy also implies that the disorder or syndrome can be defined, that even if etiologies are unknown the changes due to the treatment can be measured, and that appropriate controls are available with which to compare treatment. These criteria have never been met in studies of the treatment of insomnia with hypnotic medication. Most of the currently available research has been conducted with young, adult, normal, male volunteers, and even when sleep-disturbed patients have been studied, the focus usually has been on short-term studies of a few patients in whom neither the cause nor magnitude of the sleep disturbance has been objectively established. On the other hand, studies of hypnotic drugs' effects on highly selected severe insomniacs are not necessarily in- formative about their appropriate use with more typical patients. Although it has been estimated that in a given year more than one percent of the adult population will use sleeping pills on consecutive nights for a period of more than two months, little research has been conducted concerning the efficacy of long-term use of hypnotic medica- tion for sleep. Of drugs marketed in the United States, only flurazepam has been shown to affect sleep for as long as 28 days of continuous use, and then only in 10 insomniac subjects. Most other hypnotics that have been studied in the sleep laboratory appear to lose their sleep- promoting properties within three to 14 days of continuous use. There have been no published reports of a hypnotic having been clinically evaluated by sleep laboratory methods for more than four weeks or with more than ten insomniac subjects. Nor have any hypnotics been fully assessed with respect to such daytime measures as patient mood, occupa- tional performance, or psychomotor skills. There is some evidence to suggest that chronic use of hypnotics may exacerbate rather than ameliorate the difficulties of the insomniac patient. But little systematic investigation has been made of this aspect of hypnotics, and it is not known whether patients who use hypnotics regularly on a long-term basis are being helped or whether the regular use is due to physical or psychological dependence. _9_
It is possible that a powerful placebo effect is present. However, these drugs are certainly not inert placebos, even in low doses. "Drug withdrawal insomnia' or "rebound insomnia'' has been documented following the abrupt cessation of nightly use of both benzodiazepine and non-benzo- diazepine hypnotics, except for flurazepam, which does not appear to have this effect. Other hazards of long-term use and availability include the risk of increased mortality, risk of abuse or misuse by persons other than the patient, and risk of deliberate or accidental overdose. The commonly used hypnotics remain in the body much longer than the few hours needed to aid sleep. If certain drugs are taken on several consecutive nights, there will be an accumulation leading to a potentially increased risk of undesired effects during the day; this is especially characteristic of long-acting benzodiazepines like flurazepam and nitraze- pam. Laboratory tests of visual motor skills related to driving show functional impairment of performance on the morning following ingestion of these two drugs which is greatly exacerbated if alcohol is also in- gested. Somewhat similar findings have been reported for diazepam, but not for the shorter-acting benzodiazepine, oxazepam (Serax(R)~. The barbiturates that have been tested after consecutive nightly use have a lesser effect on performance than flurazepam. The practical signifi- cance of such impairment is as yet unclear, but some added risk of having or causing accidents appears likely. Experience with hypnotics in clinical use has shown that there are certain patients who are especially at risk for adverse effects: patients with respiratory difficulties or with impaired renal function, patients who are pregnant, who are alcoholics or who are elderly. The effects of hypnotics on sleep-related respiratory disturbancehave never been directly tested, but the central respiratory depressant effects of the barbiturates are well known, and similar effects have been found with the other hypnotics. Renal impairment is of particular concern with the use of benzodiazepines because of their long half-lives and psychoactive metabolizes. Adverse clinical effects with the benzo- diazepines and chloral hydrate have been found to be significantly correlated with renal impairment. The evidence is inconclusive, but there are preliminary indi- cations of an association between exposure to the benzodiazepines early in pregnancy and birth defects, especially cleft lip. Cardiovascular anomalies at birth and brain tumors later in childhood have been puta- tively associated with the use of barbiturates during pregnancy. Inadequacy of research makes it impossible to judge the effective- ness of hypnotic treatment of insomnia or to clearly delineate risk- benefit ratios of these drugs in patients. Criteria for measuring -10-
outcomes are not standardized to allow comparison of findings from one investigation to another. Neither is there consistency in the selection of variables reported in published articles. Sleep Disturbance in the Elderly There are changes in the sleep/wake cycles with age that lead to many complaints of insomnia among the elderly. Of particular concern in this report are the 95 percent of non-institutionalized elderly persons who monitor their own health care. The elderly are more vulnerable to the hazards of hypnotic use because they are more likely already to have disorders aggravated by hypnotics, such as impaired respiratory, hepatic, or renal function. The elderly also are more likely to be taking other medication, leaving them at increased risk for toxic interactions from the consumption of several different drugs at the same time. There is, however, a lack of data on the efficacy and suitable dosage of hypnotics in the elderly, and the few studies available show no particular agent to be more effective than another. Prudent choice of a hypnotic for an elderly patient might be from among those drugs that are converted to inactive metabolites or have relatively short half-lives, so as to minimize the chance of accumulation and adverse effects. Before prescribing a hypnotic for an elderly person, the physician would be wise to suggest non-pharmacological measures, such as establish- ing a ritual for retiring to bed, avoiding daytime naps, and increasing the daily amount of physical activity. CONCLUSIONS Clinical Practice, Patient Information, and Public Health The committee concludes that prudent clinical practice in the matter of sleep disorders should entail: -- a thorough medical and psychosocial appraisal of the insomniac patient before any decision to prescribe drugs for relief of the sleep complaint -- choosing a hypnotic drug on the basis of anticipated risks to individual patients - close monitoring by the physician of all patients receiving hypnotic drugs 11
-- limiting, in most cases, the number of sleeping pills prescribed to that required for short-term management of the complaint, and refraining from providing automatic or "as needed" refills -- exercise of particular care in the choice of medication for elderly patients, for patients who may be pregnant, for those who may have to operate machinery, or for those with certain impairments such as kidney or liver disease. Patients should receive the clearest possible directions and warnings about the use of hypnotic drugs. Oral instructions from health care personnel could be augmented by informational brochures on the proper use of the medication. Information also should be available to the patient about non-pharmacological measures. To reduce the ease of suicidal impulsive pill ingestion, the com- mittee urges the evaluation of measures like individual pill packaging and enlarged pill size. To provide more accurate mortality and highway safety statistics, the committee recommends that modern toxicological screening for the benzodiazepines be added to the traditional tests for alcohol, opiates, and barbiturates. Additionally, since the current method of coding causes of death on death certificates does not permit the tracing of trends in non-medical use of particular individual drugs, the committee recommends that a new code system which amplifies the system currently in international use be devised for use within the United States. Benzodiazepines vs. Barbiturates as Hypnotics Largely because of the benzodiazepines' well-known safety in overdose, physicians now prescribe them much more frequently than the barbiturate hypnotics. The committee finds that insufficient recognition has been given to undesirable or hazardous effects of long-acting benzodiazepines that are not found with the barbiturates. These include the accumulation of long-acting metabolites, increased probability of adverse reactions with age, and an increased probability of adverse reactions in patients with diminished kidney function. Benzodiazepines also share a number of hazards with the barbiturates. The current state of knowledge about hypnotics leads the committee to conclude that the barbiturates and benzodiazepines are probably equally effective in short-term use. Neither the barbiturates nor the currently available benzodiazepines can be considered "ideal" -12-
hypnotics and the committee finds there is a need for continuing evalu- ation of alternatives. In the meantime there is evidence to suggest that there may be rather little difference in effect on sleep between the 15 mg and 30 mg doses of flurazepam although there is a strong relationship between adverse effects and the higher dose. In view of these studies, the committee recommends that physicians who prescribe flurazepam be advised to initially prescribe the lower dose for all patients, rather than, as is now the case, only for elderly or debilitated patients. Professional Education Although hypnotics are widely prescribed, the committee finds that physicians receive little specific education or clinical training in the physiology of sleep and use of hypnotic medication. The committee urges that major professional education efforts be developed in this area, and that the efforts should not be sponsored solely by the pharmaceutical industry. Programs in insomnia should become part of the continuing medical education that has become mandatory in many specialty societies. Advertising and Labeling Studies have shown that drug company detail men, advertising in medical journals, and information in the Physician's Desk Reference continue to be important sources of information to physicians on dosage and adverse reactions to drugs. The committee finds the information from these sources tends to be incomplete and of questionable value to the physician. In the judgment of the committee, consideration should be given to requiring that complete prescribing information appear in advertisements. Research Needs The committee finds a need for basic research into the physiology of sleep, clinical investigations of various types of hypnotic drug use, and epidemiological studies to delineate the hazards of hypnotic drugs. Existing data leave the committee unable to recommend one thera- peutic approach to insomnia over another, or one drug over another. The committee therefore recommends that new, more sophisticated multi- disciplinary research be initiated to test the efficacy of the different -13-
modes of treatment for the different types of insomnia. At the same time, the committee recommends that every effort be made to standardize and upgrade the quality of hypnotic drug efficacy studies. The committee finds that there is a need for research and evaluation of hypnotics that is independent of drug manufacturers. The committee also believes that independent clinical pharmacology should be strength- ened through support and networks of peer review. In the committee's judgment, ways need to be found to supplement, independently, the information provided to regulatory officials by pharmaceutical companies in New Drug Applications and in post-marketing reports. One approach urged by the committee would be to increase the number of qualified expert consultants available to and used by the Food and Drug Administration in evaluating new and marketed drugs. Another approach would be the establishment of a standing advisory committee to deal specifically with hypnotic drugs. Initiation and Coordination of Federal Efforts The committee urges that a central authority within the Public Health Service be formed promptly to review the current scientific and educational issues associated with sleep disorders and their treatment. Working with specialty societies, medical associations and academic health centers, this authority would begin the process of improving research and education. The committee suggests that an ongoing mechanism within the Public Health Service could usefully identify and fund quality research on the practical questions raised in this study, as well as collecting and disseminating related information. -14-
REFERENCES 1 2 Special Message to Congress on Drug Abuse, August 2, 1977. President Jimmy Carter. Cooper, J. (ed.), Sedative-Hypnotic Drugs: Risks and Benefits, Dept. of REW, National Institute on Drug Abuse, 1977. -15-