Click for next page ( 80


The National Academies | 500 Fifth St. N.W. | Washington, D.C. 20001
Copyright © National Academy of Sciences. All rights reserved.
Terms of Use and Privacy Statement



Below are the first 10 and last 10 pages of uncorrected machine-read text (when available) of this chapter, followed by the top 30 algorithmically extracted key phrases from the chapter as a whole.
Intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text on the opening pages of each chapter. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.

Do not use for reproduction, copying, pasting, or reading; exclusively for search engines.

OCR for page 79
SShaping Drinking Practices Directly Chapter 4 examined the potential efficacy of taxes, restrictions on avail- ability, and the detailed regulation of stores and taverns to palliate Ringing practices primarily with respect to consumption, but also with respect to contexts. While such approaches appear to hold significant potential for controlling the alcohol problem, their influence is indirect: they produce effects on drinking practices and the shape of alcohol problems only as drinkers react in personalized, somewhat unpredictable ways to the new terms of commercial availability. Arguably, a more precise and direct approach to moderating drinking practices would be to specify "safe" and "appropriate" drinking practices based on solid facts, then use the prestige, resources, and authority of government to facilitate widespread adoption of the preferred practice. Since much drinking now occurs under private auspices, and since most on-premise drinking is heavily influenced by one's companions as well as the mar- keting efforts of servers, policy instruments focusing directly on private drinking practices can complement instruments regulating commercial availablility. Governments can seek to influence collective drinking practices through instruments that operate with varying degrees of force. At one extreme, governmental authorities can embody conceptions of appro- priate or tolerable drinking norms in statutes and regulations (e.g., laws prohibiting drunken driving, being drunk in public, being intoxicated while flying a plane, and drinking by minors). Clearly, these legal modes draw heavily on the moral, economic, and political resources of gov- 79

OCR for page 79
80 REPORT OF THE PANEL ernments: prestige is staked on conceptions of drinking that may or may not correspond closely to commonly shared views of drinking, consid- erable money may be spent to enforce such laws, and the authority of the government is invoked against individual conduct. Much less extreme are government-sponsored programs to "educate" citizens. Sometimes the programs do nothing more than provide infor- mation about drinking through labeling or mass media advertising. Other times, the programs take advantage of organizations such as schools, churches, and health maintenance organizations to disseminate the messages in more localized ways. In some cases, the programs go beyond merely providing information and seek to motivate people to moderate their drinking or actually train them in techniques for doing so. To the extent that such programs are sponsored by government and embody a specific conception of drinking, they claim the prestige and credibility of the government. The volume of resources claimed, how- ever, can vary significantly, depending on how intensive the program is and how much of the costs can be defrayed by taking advantage of existing institutions as vehicles for the messages, motivation, or training. At the other extreme, the government could seek to alter drinking practices through the symbolic impact of its own independent actions or statements about what constitutes safer drinking practices. In areas in which federal governmental power is largely uncontested (e.g., man- aging its own employees, regulating interstate transportation, and man- aging national parks), it could make its operations reflect a coherent and consistent idea of safe drinking practices. Compared with statutes and regulations, shaping practices through symbolic action is light and unburdensome: it operates only through the force of government's pres- tige and leaves individuals free to draw their own conclusions from observed policies. But these characteristics may be liabilities as well as assets: citizens may simply fail to note government's exemplary actions, draw the wrong inference, or ignore the example as unpersuasive. The purpose of this chapter is to explore the potential of using these sorts of instruments (law; education, information, and training; and symbolic action) to shape collective drinking practices. We try to give some specific content to the general concepts of "safe" or "appropriate" drinking practices. Specifically, we look for restrictions on drinking that could command widespread support and, if more widely and enthusi- astically adopted, could change the shape of alcohol problems in im- portant areas of concern.

OCR for page 79
Shaping Drinking Practices Directly CONCEPTS OF SAFER AND MORE APPROPRIATE DRINKING 81 It is relatively easy to state and find consensus on prevention goals, so long as one keeps away from specifics. There is little controversy, sci- entific or political, over the goal of reducing alcohol abuse and alco- holism. When one looks more specifically at the behaviors entailed and the range of individuals potentially involved, however, the easy con- sensus evaporates along with the simplicity of the terms. A list of drink- ing practice targets that have at various times been pursued by govern- ment and/or public interest groups includes: eating while drinking drinking more diluted forms of alcohol drinking nonalcoholic beverages in preference to alcoholic ones drinking in association with recreational opportunities substituting sports for drinking not driving while drinking or after drinking not drinking at all by people of certain ages or ethnicities not drinking more than a specified amount per day not drinking or appearing drunk in public. To discuss ways of achieving any such behavioral results without being specific about which one is in view cannot lead to effective results. We have identified three aspects of drinking behavior as suitable targets of prescriptive efforts: driving and performing other hazardous mechanical tasks while intoxicated, drinking so much over a lifetime that one begins to suffer from organic damage such as cirrhosis of the liver, and using drunkenness as an excuse for violent or aggressive behavior. In our view, these aspects of drinking behavior are suitable targets in part because the behavior is tied to some of the worst consequences of current drinking practices and in part because current attitudes and practices are supportive enough to allow governmental efforts but not so widely honored that such efforts would be redundant. Some statistical and causal elaboration, focusing for illustrative pur- poses on the relation of alcohol to mortality, indicates what is at stake in these areas. At least 12,000, and perhaps twice this number, of the more than 100,000 accidental deaths that occur each year in the United

OCR for page 79
82 REPORT OF THE PANEL States may be directly attributed to the effects of drinking. In the most extensively studied area, 12,000 annual deaths are attributable to drun- ken driving alone (Reed, in this volume), and the majority of these deaths involve people under 30 years old.i If the same proportion of other accidental deaths were attributable to alcohol as were auto fatalities, it would yield another 12,000 or so each year. This seems the maximum possible figure and may be con- siderably smaller (Wingard and Room 1977~. The tradition of research on non-auto crashes, falls, burns, drownings, and other accidents is less rigorous and voluminous than the driving literature and cannot provide any firmer estimates. Of the more than 30,000 annual liver cirrhosis deaths in the United States, between 20,000 and 25,000 can be attributed directly to the effects of alcohol (Ouellet et al. 1977, 1978~. Nearly all of these deaths are people over 35, and most are under 65. Only about half the alcohol- caused cirrhosis deaths are drinkers who would qualify as alcoholic under clinical critena. Some degree of risk of liver injury is actuarially present in people who sustain an average of only three to four drinks daily over a long period (Popham et al. 1976~. Finally, in regard to violence, about 10,000 murders occur each year in situations involving alcohol. The question of attribution here is quite open. It has often been taken for granted that alcohol makes people violent and that the connection is pharmacological the textbook ex- planation being that alcohol "disinhibits" violent impulses. Recently, evidence has accumulated from a variety of fields suggesting that the connection between alcohol and violence is cultural rather than phar- macological. MacAndrew and Edgerton (1969) marshalled anthropo- logical evidence of the enormous variations between cultures in behav- iors while drunk, including manifestations of aggressiveness. Levine's (1977) historical study of American attitudes toward the role of alcohol in casualties and crime shows that, on the whole, colonists in this country believed alcohol made people clumsy; by and large, they did not believe it made people violent. The latter belief was an innovation of the tem- perance era. Stevenson's tale of Dr. Jekyll and Mr. Hyde captures the essence of how the 19th century came to view alcohol, as a magical substance that turned a man into a beast. Social psychological experi- ' It is illuminating to note that this figure is about one-quarter of all auto crash fatalities; compare the "half of all highway deaths" conventionally cited "involving" alcohol. The difference is due largely to the surprisingly high prevalence of drunken drivers at the times (late nights) when fatal crash risk is at its highest anyway because of driver fatigue, poor visibility, and lack of routine traffic stimuli. The number of crashes that would be expected even in the absence of alcohol substantially reduces the conventional figure. in a causal analysis, but it is still very large.

OCR for page 79
Shaping Drinking Practices Directly meets with college students using a "balanced placebo" design (four conditions: believing that one is drinking alcohol and actually drinking alcohol, believing but not drinking, drinking but not believing, neither drinking nor believing) have shown that the belief that one is drinking alcohol, not the fact of drinking alcohol, is associated with aggression. There is also the negative evidence of an extensive tradition in research of looking for physiological connections between alcohol and violence that has not produced positive findings; indeed, the disinhibition theory shelters a wide variety of essentially incompatible as well as unsubstan- tiated explanations (Pernanen 1976~. Nevertheless, even without a pharmacological connection, there is a strong causal connection between alcohol and violence: the culture tells people this is so, and in acting on this belief they make it come true. Research designed to sort out the specific quantitative contribution of cultural belief to alcohol-involved violence is in its infancy, but this connection requires consideration even in the absence of good epide- miological estimates. Of course, all of the alcohol-related deaths analyzed above are simply the visible tip of the iceberg: not only tragic in themselves, but also important as signs locating the larger quantities of harmphysical, psy- chological, and socialthat we wish to reduce. In the remainder of this chapter we look at the three modes of government influence (law, ed- ucation and symbolic action) with this set of practices and corresponding problems in mind. 83 DRINKING PRACTICES AND THE LAW In two major areas, governments have sought to shape drinking practices through laws backed by criminal penalties. The two areas are drunken driving and public drunkenness. These are by far the most common criminal offenses reported to be engaged in by young men (O'Donnell et al. 1976) and are the two largest categories of arrests reported to the Federal Bureau of Investigation (1979), each approximating 1.25 million arrests annually of the roughly 10 million reported. These involvements of law in the direct control of drinking practices dwarf all other efforts of government. DRUNKEN DRIVING AND THE LAW The law in the United States (and throughout the world) establishes clear prescriptions for drunkenness: one should not drive while drunk. The legal codes usually stipulate a specific blood alcohol content (BAC)? conventionally 0.08 or 0.10, that defines drunkenness for judicial pur-

OCR for page 79
84 REPORT OF THE PANEL poses. These laws reinforce public attitudes, for the great majority of the population believes that driving while drunk is wrong and dangerous (NIAAA 1978~. The question of whether the existence of these laws actually influences the rate of drunken driving and, if so, through what mechanisms is still unsettled. Some useful arguments and evidence bearing on these issues are available, however. At the outset, one might reasonably be skeptical about the value of such laws. Even though we now spend significant sums enforcing these laws and have more arrests nationwide for drunken driving than for any other offense, the probability of being arrested for drunken driving is now estimated to be roughly 1 in 2,000. Compared with the perceived risks of injury or economic losses from drunken driving (which must also operate to deter drunken driving), the risks and consequences of arrest must appear small. Moreover, one can rea- sonably ask if intoxicated people are generally able to calibrate risks, whether created by enforcement agencies or the laws of physics. Given this initial skepticism about the potential value of legal sanctions in controlling drunken driving, it is somewhat startling to discover that the few careful studies in this area indicate that drunken driving is responsive to the perceived risk of arrest. Two bodies of evidence are especially relevant, and both focus not on drunk driving per se but on its most serious risk: fatal injury in an auto crash. The first instance is the 1967 British Road Safety Act, carefully ana- lyzed by Ross (1973~. The Road Safety Act (RSA) defined driving with a BAC of 0.08 or higher, as determined by breath or laboratory analysis, to be an offense. Drivers were required to submit to BAC tests if asked to by police upon reasonable cause (a road accident, moving violation, erratic driving). Refusal to be tested would create the presumption of illegal intoxication. Judges were denied discretion in sentencing: a first BAC offense was to result in a mandatory 1-year suspension of the license to drive (the same statutory maximum penalty as before). The 1967 RSA thus differed from earlier custom and law (the 1962 Road Traffic Act) in three principal respects: (1) the use of BAC measurement as definitive rather than supplementary evidence of drunkenness, (2) mandatory submission to testing, and (3) mandatory sentencing linked specifically to this evidence. In addition, unprecedented publicity was generated by the government in the months before the RSA took effect, especially regarding the prospective use by police of the roadside "Al- cotest" breathalyzer device, 1 million of which were purchased by the Home Office from a German manufacturer. The device was a novelty in England. The immediate impact of the RSA was positive and dramatic. For the

OCR for page 79
Shaping Drinking Practices Directly 85 three months immediately following its effective date, total auto fatal- ities were reduced 23 percent, and the proportion of fatally injured drivers with BAC levels of 0.08 or higher (postmortem blood deter- mination) dropped from 27 percent before passage to 17 percent the following year. However, this dramatic effect gradually diminished. The relevant time series data, adjusted by Ross for seasonal and other ex- ternal variations, show a gradual decline in auto injuries from 1966 to 1967, rapidly falling when the RSA went into effect; the curve then flattened, and began to rise in 1970. As Ross concludes: "In the case at hand it appears that the benefits initially produced by the legislation had been largely canceled by the end of 1970, and a return to normal would be predicted for subsequent years" (p. 77~. The explanation proposed by Ross for this sequence of events is that the well-publicized passage of the RSA convinced many drivers that the risk of arrest and punishment when driving drunk would become much higher than it had been. They were hence deterred from drinking before driving and/or from driving after drinking. But law enforcement did not markedly increase the certainty of either detection or punishment, and, as time passed, drivers observed this and the immediate deterring impact of the RSA thus evaporated. While this general explanation seems quite plausible, Ross's data support a richer interpretation with respect both to how drinking prac- tices changed and to how the deterrent effect was produced. With respect to the change in behavior, Ross clearly rules out a significant overall decline in either drinking or driving. Instead, he presents evidence that the law focused its deterrent effect on a fairly narrow slice of behavior: the custom of driving to and from pubs especially on weekend nights. In response to the new law, many habitues took to walking or drinking closer to home. This aroused a considerable outcry from several thou- sand pub owners; a number of less convenient establishments closed. Similarly, a detailed analysis of the government's "massive publicity effort" reveals that the focus of the campaign was on the novel roadside breathalyzer test. While about 40 percent of adults learned the legal penalty or actual BAC limit, 99 percent could identify the breath test. It appeared that this new technology would bring about drastic changes in the activities of police and the courts, replacing the earlier workings of patrol and trial with a revolutionary scientific mechanism. In fact, there was no such result. As Ross documents, well-publicized cases soon established narrow limits to the authority of the Alcotest as evidence. It took several years to establish reliable standards for its use. Even though testing increased steadily from 1967 onward, the British police did not employ it frequently, compared with police in other countries.

OCR for page 79
86 REPORT OF THE PANEL If the initial dramatic effect on pub-going customs is thus attributable to anticipation that a technical innovation would transform the system of traffic surveillance and justice, that anticipation was due largely, it appears, to a somewhat superstitious belief, shared and promoted by the government, in the power of the new technology. This added information enriches our understanding of how the law operated to produce effects, beyond the simple notion that perception of an increased risk of arrest generally deters a given behavior. It sug- gests that the actual changes in behavior produced may be quite narrow as citizens react to the new contingencies in ways that preserve as much of their customary activity as possible. It also suggests that deterrence works best when its threat is widely advertised and becomes the focus of much public conversation, and perhaps when the enlarged threat is tied to a technological novelty. The question of whether and how the dramatic initial results of the British Road Safety Act might have been sustained thus remains somewhat obscure. A series of cases potentially as significant as the RSA are the Alcohol Safety Action Programs (ASAP). Between 1970 and 1977, the U.S. Department of Transportation funded 35 local sites, which varied in size but in the aggregate covered millions of drivers, in a demonstration effort to control the problem of drunken drivers. The programs cost more than $88 million. These "comprehensive, multifaceted" programs employed four categories of countermeasures: enforcement, adjudica- tion, rehabiliation, and public information and education (National Highway Traffic Safety Administration 1979b). For our purposes, the studies documenting and evaluating this ex- perience have serious flaws (Zador 1976, Zimring 1977, Cameron 1979, Reed, in this volume). Serious methodological problems in the design of the program and the collection of data make it difficult to consider these program innovations as experiments. Equally important is the difficulty of isolating the separate effects of each of the program com- ponents. Since our concern is largely with the effects of enforcement and adjudication on drunken driving (including those portions of re- habilitation and education efforts that were tied to enforcement efforts), such isolation is clearly desirable. Despite these difficulties, some useful information was developed by ASAP evaluators in their final report (Levy et al. 1978; National High- way Traffic Safety Administration 1979a,b,c,d); they attempted to meet a number of the criticisms leveled at the earlier evaluation (National Highway Traffic Safety Administration 1974~. The final evaluation found that 12 of the 35 ASAP programs had produced a statistically discernible effect on nighttime auto fatalitiesa good indicator of drun- ken driving. These 12 programs averaged a 30-percent reduction over

OCR for page 79
Shaping Drinking Practices Directly 87 about 3 years. This result appears broadly comparable to Ross's finding of a 23-percent reduction in this index over a similar time frame a coincidence that lends some increased credibility to the ASAP findings. On the other hand, it should be noted that when previous optimistic evaluations by the National Highway Traffic Safety Administration (NHTSA) of the ASAP and other traffic safety programs have been reanalyzed or redone by independent investigators, much smaller or no effects were reported (Cameron 1979~. At any rate, the reduction and the particular characteristics of those programs that produced it, led NHTSA to conclude that the program had been successful and, some- what more surprisingly, that "raising the enforcement level ti.e., alcohol- related arrests, by a factor of two to three] and public awareness of that level in order to deter the social drinker" were central to producing this result (Levy et al. 1978, p. 174~. Thus, some moderately persuasive evidence exists suggesting that effectively enforced drunken driving laws will deter drunken driving and reduce accidents and fatalities associated with it. The main problem with trying to take advantage of this opportunity is that it is likely to be expensive; exactly how expensive is unclear. Using NHTSA's figures on lives saved (563) and the total costs of ASAP ($88 million) yields an estimate of roughly $156,000 per life saved (but cf. Reed, in this volume). But this simple calculation ignores many of the costs of enforcement borne by local and state agencies, the expenses of treatment programs borne by arrestees themselves, and the "cost" to innocent drivers of increased police surveillance of driving practices. Moreover, we cannot hope to save much in the way of resources by punishing more severely those whom we catch. Current research (summarized by Reed, in this volume) indicates that the prevalence of drunken driving is much more effectively influenced by changing the rate of arrests and convictions than by trying to alter the severity of punishment delivered. Our jury system, wherein many jurors identify with the defendant accused of drunken driving, prohibits the imposition of penalties out of proportion to those the public feels are appropriate. Nonetheless, drunken driving laws are probably important. At a minimum, they help sustain a widely shared disapprobation of drunken driving. They also provide an op- portunity to attack a given drinking practice more aggressively if the society is willing to commit the resources, publicity, and attention nec- essary to make deterrence a social phenomenon rather than an abstract concept. PU B L I C D RU N K ENN E S S The second aspect of drinking practices controlled by criminal statutes is that of being drunk in public. In principle, these statutes could be

OCR for page 79
88 REPORT OF THE PANEL justified in several different ways. Since those who become publicly drunk make themselves vulnerable to criminal attacks and are often the people experiencing the worst consequences of drinking, the laws could be understood as paternalistic efforts to discourage dangerous conduct or to allow the state to take "protective custody" of people who are in danger. Alternatively, the laws could be seen as devices for preventing drunks from harming others. To the extent that public drunkenness is related to violence, vandalism, and other forms of disorder, the laws against public drunkenness could be understood as devices to allow the police to intervene early and moderately in situations that could become dangerous much as laws against drunken driving or the illegal carrying of weapons are assumed to operate. The laws could also be understood as devices to protect the sensibilities of people shocked by breaches in public decorum created by drunkenness. In recent years, the laws against public drunkenness have been under- stood primarily in these last terms, i.e., as devices that fixed a preference about public decorum in criminal statutes and were used to stigmatize and control a minority of the population. Since this use of state power seemed suspect, a social movement developed that urged the "decri- minalization" of public drunkenness. The central argument of this move- ment is that the real problems created by public drunkenness can be handled more effectively through extensions of civil authority and the provision of social and medical services than through the application of criminal law. To the extent that criminal laws served the purposes of protecting occasional drunks from victimization and chronic drunks from exposure and starvation, the purpose could be better served by a form of civil "protective custody" allied with social and medical services. To the extent that the laws guarded against public violence and disorder, the police could rely on statutes regulating conduct more reliably related to the worrisome offenses (e.g., assault, vandalism, etc.~. About half the states and cities of the United States have been per- suaded by these arguments to "decriminalize" public drunkenness (U.S. Senate 1976~. Typically, the decriminalization statutes not only eliminate criminal sanctions but also mandate that drunks be taken into protective custody and supplied with a variety of social and medical services. While these new laws have succeeded in changing the legal status of public drunkenness and much of the rhetoric about the problem, the actual handling of public drunks has changed only modestly. The police usually remain involved because they are the only agency trained and deployed in ways that bring them into contact with public drunks. Moreover, they retain explicit authority to hold a drunk in protective custody for a fixed period of time before release or transfer to a treatment facility. The social and medical services remain largely absent because the legal man-

OCR for page 79
Shaping Drinking Practices Directly 89 date for increased services was in few cases backed up by suitable ap- propriations and programmatic activities. The effects of this liberalizing trend are unclear. There are virtually no data on how the laws regulating public drunkenness have affected the frequency and character of public violence. Some data exist (Oki et al. 1977) that suggest that neither the old laws nor the new laws have had much effect on the health and welfare of chronic public inebriates- though this may be the result of having failed to make investments in the medical and social services mandated by the reforms. In fact, the only clear effect of decriminalization seems to be the most immediate and most obvious: chronic public inebriates have, according to surveys of public impressions, become more visible on city streets (Aaronson et al. 1978, Giffen and Lambert 1978~. The increase in visibility might be taken as evidence for increased frequency of public drunkenness, but this need not be so. It may be that public inebriates are now diffused more broadly throughout cities, giving the impression of greater fre- quency. Alternatively, it may be that chronic public drunks spend less of their time in institutions (a possibility that is consistent with the idea that episodes of drunkenness on city streets have increased but incon- sistent with the idea that the number of public drunks has increased). Such effects are the natural result of increased official tolerance of public drunkenness. There remains at least the possibility that public drunkenness has increased in frequency and that some increase in violence, vandalism, and disorder is associated with this change. We wish the research in this area were more complete and illuminating, but we suspect that the laws against public drunkenness were not operating to hold drunken violence within tight bounds all those years. It is more likely that the role of the law with regard to violence linked to alcohol involves the acceptance, by judges, juries, or other participants in the legal process, of the cultural presumption that alcohol causes violence and therefore serves to excuse it. In the absence of a body of studies on the treatment of drunkenness in this connection, we are unable to extend this analysis into the question of what might be done in this area. (Chapter 6 includes a discussion of public drunkenness from an environmental perspective). EDUCATION, INFORMATION, AND TRAINING PROGRAMS A second way to directly influence the drinking practices of individuals is through education, information, and training programs. To a degree, of course, such efforts must be seen as important complements to legal approaches. Laws are effective in shaping conduct only if they are pro-

OCR for page 79
9o REPORT OF THE PANEL mulgated, a function that is at least in part one of information and education. Moreover, the language of a law has educational value: it instructs people about what the society considers desirable as well as threatens them with official action if they fail to comply. Finally, inten- sive education and training programs are sometimes used as alternative sanctions for violations of the legal requirements. Thus, education, in- formation, and training are often intertwined with laws. They can also operate as instruments separate from the law. They can seek to persuade and advise people in areas in which society is reluctant to establish legal restrictions. In fact, in a democratic society, educational programs operating apart from the law enjoy great prestige and legiti- macy, despite the scanty evidence of their effectiveness. In any case, it is this advisory form of educational effort that is the focus of this section. Specifically, we look at three kinds of programs: educational programs directed at youths through educational institutions (primarily public schools); mass media information campaigns; and community-based health information and training programs. SCHOOL EDUCATION PROGRAMS Alcohol education in the public schools was originally a creature of the temperance movement. By the late 19th century, state authorization and implementation of alcohol education programs was almost univer- sal. The character and tone of this education were consistent with pre- vailing views at the time and were strongly antialcohol. While some attention was paid to the adverse health consequences of drinking, the dominant focus was on alcohol's disastrous effects on character, family life, and the fabric of society. Ruin was predicted for the vast majority who drank at all. With repeal and the early beginnings of the alcoholism movement in the 1930s and the 1940s, the approach shifted to a more health-oriented, factual presentation, but the emphasis was still on the negative consequences of drinking. By the late l950s and the early 1960s, alcohol education had become less judgmental regarding alcohol use and more colorless, scanty, and limited. During the 1960s the emphasis shifted gradually from alcohol education toward alcoholism education. This carefully neutral approach presented alcoholism as a disease while avoiding alarmist interpretations of the small but quite real risks of alcoholism for the individual drinker. In the 1970s, attention to alcohol education as a device for controlling the alcohol problem increased markedly, as a result of at least three factors: (1) the establishment of NIAAA with a broad mandate for controlling the alcohol problem; (2) the heavy reliance on drug education

OCR for page 79
Shaping Drinking Practices Directly 91 tor coping with the drug problem; and (3) the "discovery" of significant levels of teenage drinking. Using such approaches as "values clarifica- tion," "enhancement of self-concept," "training in coping skills," and "peer counseling," the approach focused on teaching individuals strat- egies to cope with a variety of life problems not just alcohol and drink- ing. This emphasis appears in the "responsible drinking" campaign pro- moted by Morris Chafetz as founding director of the NIAAA. The alliance of the National Clearinghouse for Alcohol Information in NIAAA and the network of prevention coordinators within state al- coholism authorities resulted in a proliferation of new educational ma- terials as well as a marked increase in attention to evaluation and re- search. Whereas 15 years ago alcohol education was a desultory part of the high school health curriculum, there are now numerous alcohol- centered programs sponsored by NIAAA's Prevention Division, the National Institute on Drug Abuse, the U.S. Department of Education, and a number of private organizations. Replication and evaluation of projects in anticipation of national dissemination are now under way (NIAAA 1979~. The ultimate effects of these educational programs remain unclear. The data that would allow evaluation of the programs have become more voluminous and sophisticated, but remain inconclusive (Staulcup et al. 1979, Wittman 1980~. Still, it is possible to reflect a bit on the strategic orientation of these programs in the light of what we do know about drinking practices. In general, we think that the ambitions of school-based alcohol ed- ucation programs are too grand. They are seeking to produce broad and durable changes in the orientations of students toward drinking: in ef- fect, to innoculate the student population against future drinking prob- lems. Moreover, they are relying on a minor part of the curriculum in the schools to produce this effect. Given what we know about drinking practices, this strategy seems overly ambitious. It is known, for example, that drinking practices change dramatically over a person's life. Con- sequently, the idea that we might fix lifetime drinking practices within certain parameters by educating people early evokes skepticism. Simi- larly, it is known (Blanc 1976) that most drinking behavior is learned at home, and that students listen with only half an ear to teachers when they raise questions about life-style. Consequently, the public schools may not be the most powerful institutional base from which to launch an alcohol-prevention education program. For these reasons, we believe that it is inappropriate to define the objectives of school-based education programs as the innoculation of students against future alcohol prob- lems.

OCR for page 79
92 REPORT OF THE PANEL A different objective might be effectively pursued through school- based programs. Significant problems are associated with the significant amount of drinking done by young people of high school and college age (e.g., traffic accidents, fights, and diminished classroom perform- ance due to drunkenness on school nights, in class, or instead of class). It would be valuable and feasible for education programs to focus on the current problems of youthful drinkers without reference to the ques- tion of whether the training efforts shaped long-term practices. It could conceivably be useful to teach youths that they are responsible for violent and vicious behavior whether drunk or sober that it is a matter of responsible behavior, not just responsible drinking. In effect, high schools and universities should adopt the objective of shaping drinking practices for the student populations while they are of school age and should register successes if they succeed in reducing drinking and prob- lems associated with drinking for their current populations. This re- orientation may not require much change in current approaches and materials, but it would certainly require replacing the prevailing vague objectives of school-based alcohol education programs with clear be- havioral goals. In any case, it would make a big difference in how we evaluate these programs and in the urgency that the schools and students may feel in coping with the problems. MASS MEDIA INFORMATION CAMPAIGNS Typically, education for adults is viewed primarily in terms of mass media campaigns. The publicists of the temperance movement were masters of media campaigns. After repeal, media efforts along temper- ance lines for the most part disappeared. The alcoholism movement at this time invested little effort in the mass media to ain acceptance of the disease concept. Since the mid-1960s, however, interest has greatly increased, as has investment in media campaigns. The consensus of those who have surveyed the potential of using mass media for reducing alcohol problems parallels our judgment on school- based education programs. There is very little hard evidence pertaining to the impact of these programs, with some recent, notable exceptions. The criticisms are numerous: poorly designed evaluation, no evaluation, outcomes limited to changes in attitudes or information with little at- tempt to connect these outcomes with behavior. Blane and Hewitt (1977b, p. 32) speak for many reviewers in concluding that: "the effects of public education are largely limited to increasing knowledge and reinforcing established attitudes or behavior patterns." Blane, in an- other article (1976, p. 357), notes that "The effects of public information

OCR for page 79
Shaping Drinking Practices Directly 93 and education programs on alteration of drinking patterns is indeter- minant, although it is generally felt to be slight. Research evidence is almost totally lacking; all that exists are bits and pieces of information that are no more than suggestive." Moreover, on the basis of a com- bination of general empirical findings, common sense, and logic, Wal- lack (1980) makes a persuasive case that noticeable behavioral changes would be unlikely as a result of mass media campaigns alone.2 The future of mass media campaigns in controlling the alcohol prob- lem is not likely to be as an independent instrument but as a program undertaken in conjunction with something else: as a device for increasing the salience of a new legal restriction as in the British Road Safety Act, as reinforcement for ideas presented in school-based programs, or as a way of triggering interest in programs that can provide training to individuals who want to change these current practices. It is to this last idea that we next turn. COMMUNITY-BASED HEALTH INFORMATION AND TRAINING Perhaps the most promising new approach in the area of education and training is a program that combines carefully designed mass media cam- paigns with personalized behavioral training provided through schools, volunteer associations, or health maintenance organizations. The most widely known (but by no means the only) example of such a program is the Three Community Study (TCS) of the Stanford Heart Disease Prevention Project. In the area of education and training the TCS is an example parallel to the study of the Road Safety Act in drunken-driving deterrence; we present a detailed review in order to give readers a more concrete idea of the genre. The TCS was intended to determine whether state-of-the- art programming of mass media in a community and intensive training of a segment of a community would be effective in modifying behaviors known to contribute to the risk of heart disease (Hochheimer, in this volume; Maccoby 1980; Farquhar et al. 1977~. Two towns of about 2 Despite the enormous expenditure for advertising different brands and types of bever- ages, "there are few reports of empirical studies of the effects of alcohol advertising and no clear evidence that advertising affects consumption" (Ogburne and Smart 1980. p. 293~. It is generally thought that the main effect of commercial advertising is to alert the public to new brands, in competition with older ones, and conversely to protect or expand the market shares of established brands. The available scientific evidence is too sparse to permit us any extended discussion of the effects of advertising policies. Nevertheless, important issues of principle are involved in such policies, which may serve as barometers of the influence of different governing ideas.

OCR for page 79
94 REPORT OF THE PANEL 13,000 in central California were selected for public health education, and a third town served as a control. A random probability sample of adults (age 35-59) in all three communities completed interviews about their knowledge of heart disease and about behaviors and life-styles affecting their risk of heart attack. The participants also provided meas- urements of serum cholesterol, triglyceride, weight, and blood pressure. After the initial survey, educational campaigns via mass media were started in the two towns. Specifically, the programs advocated dietary changes (a reduction in animal fats, cholesterol, sugar, alcohol, and salt and an increase in fiber), giving up cigarettes, a return to ideal weight, and a program of regular exercise. Most of the behavior changes that were advocated involved not only motivation to change (almost all adult cigarette smokers know that smoking is dangerous to their health and the great majority have tried to stop smoking) but also the learning of new information? the acquisition of new skills, and the practice of tech- niques for maintaining these skills. A multimedia program, developed in the Department of Communi- cation at Stanford University, was carefully orchestrated to achieve a series of objectives. With the cooperation of local broadcasting and newspaper outlets, messages were delivered through a variety of media. The sponsors of TCS reasoned that televised and radio spots of 10 to 60 seconds' duration, if widely viewed many times, could get messages attended to very well; mass media might not succeed in telling people what to think, but they could do an excellent job of telling people what to think about (Cohen 19634. There were some 50 TV spots and more than 100 radio spots. In addition, more than three hours of television programming, many hours of radio programming, weekly newspaper columns, newspaper articles and advertisements, and direct mail were involved. In one town, a group of people found to be at high risk of cardiovascular disease were selected at random to receive an additional 14-week program of intensive instruction. They received home coun- seling or took part in group classes led by university physicians, health educators, and graduate students trained in health education. Surveys and medical examinations were undertaken after one, two, and three years of the campaign. As might be expected, those receiving intensive training in addition to the media campaign showed the sharpest initial reduction in risk. By the end of two years, however, the town receiving health messages through the media only had caught up with the community including the intensive instruction group. When overall risk of heart disease was calculated, participants showed reductions of between 16 and 18 percent after two years. In the town that received no educational campaign, the average risk had increased by 6.5 percent.

OCR for page 79
Shaping Drinking Practices Directly 95 There was some retrogression during the third year in the mass-media- only town when educational programming was sharply curtailed, but not in the mass-media town that included intensive instruction. Appar- ently the supplemental value of face-to-face instruction had more staying power. Certain detailed findings are of particular interest. Smoking cessation was not achieved to any great extent in the mass-media-only town, but 50 percent of those receiving intensive instructions (versus 0 to 15 per- cent of the high-risk controls) had quit smoking after three years. Dietary cholesterol levels decreased by more than 30 percent in both experi- mental communities (versus 10 percent in the control). Changes in weight were not achieved by either method (see Stunkard and Penick 1979~; in contrast, however, the control sample gained weight. The TCS is only a single study of community-based health training practices. Extensions and replications now under way will improve the breadth and reliability of the data base. One extension of the intensive instruction method using peer-group trainers has been put into effect in one of two matched California junior high schools (McAlister et al. t980~. The students who were trained to resist social pressures toward substance use were reporting cigarette, marijuana, and alcohol con- sumption rates less than half those in the control school, nearly two years after training was initiated. The authors note that (p. 721~: "The exact processes through which this apparent effect has been produced are uncertain. Our impression is that the program created generally negative attitudes about smoking at least as much as it actually taught skills for resisting pressures to smoke." The attitude change, it should be emphasized, was an effect of the training, not its focus. In conclusion, there is now some accumulating evidence, from TCS and other efforts? that health information and training programs can be effective in reducing health-damaging activities involving overconsump- tion or poorly managed consumption. There are, however, many ques- tions still to be resolved. Efforts are still in an experimental stage, and there is a substantial difference between experimental innovative pro- grams (and closely associated replications), on one hand, and widely diffused "mass production" equivalents, on the other. The step from research-based implementation by highly committed originators to rou- tine dissemination by initially indifferent or skeptical organizations and operatives is a long one whose outcome cannot be ensured. Neverthe- less, the experiments are promising in an area in which there has been a dearth of empirically based encouragement for many years. Moreover, an institutional base that could be exploited in implementing such pro- grams is now being expanded. That is the network of health maintenance

OCR for page 79
96 REPORT OF THE PANEL organizations springing up around the country. Such organizations are likely to be interested in and capble of providing suitable training pro- grams at relatively low cost. CONCLUSIONS ABOUT EDUCATION, INFORMATION, AND TRAINING PROGRAMS Reflecting on this accumulated experience with education, information, and training programs, and considering what distinguishes relatively promising efforts from less successful ones, we can draw several tentative conclusions. First, it is important to define the behavior that is the target of the educational effort as concretely as possible. Otherwise, it is impossible to design or evaluate a program effectively. Second, the emphasis should be on teaching specific new knowledge, acquiring new skills, and practicing techniques for maintaining the new skills. Education, information, and training are intended as tools of informed choice. They should be viewed as ways of expanding individual and community capacities to achieve desired ends, rather than as ways of shaping, changing, or clarifying these ends. Education may have this latter effect, but it is not the sole or main focus of attention. Third, the programs that have been successful have adopted an ex- perimental approach and have drawn heavily on the professional knowl- edge and skills of people trained in behavioral and communication sci- ences. This professional and experimental orientation has not been characteristic of past education efforts, which have relied on the ex- pertise of commercial advertisers and the enthusiasm of a wide variety of local volunteer groups. Such operations have generally failed in the detailed planning, precise execution, and frequent evaluations that seem essential to fielding an effective training program. Thus, we think there is potential in these areas, but it does not lie where we have commonly looked. It is not exclusively in the schools or in mass media advertising. It may be in information and training pro- grams sponsored by universities and health maintenance organizations focusing on the health risks of some drinking practices and teaching techniques for modifying personal drinking habits. SETTING A GOOD EXAMPLE: A NOTE ON SYMB OLIC EFFECTS OF GOVERNMENT ACTIONS As Mosher and Mottl document (in this volume), there are a surprising number of jurisdictions, populations, and spheres of work and leisure

OCR for page 79
Shaping Drinking Practices Directly 97 in which the federal government retains formal control over the supply of alcohol, drinking practices, or products or facilities directly relevant to drinking problems. In certain areas the government has chosen to clearly recognize alcohol as an intoxicating, controlled substance that is, a drug. Two examples are the prohibition on the presence or sale of alcoholic beverages in virtually all federal work sites and the establish- ment of the National Institute on Alcohol Abuse and Alcoholism with responsibility both to treat alcoholism and to work at the prevention of alcohol abuse and alcoholism. In other areas, the federal government has treated alcoholic beverages as commercial goods that provide le- gitimate business opportunities for entrepreneurs (e.g., in the granting of small business loans to fledgling package stores). In still other areas, such as its management of most national parks, the federal government has chosen to fit its alcohol policies to local views. It can be argued that a reasonably consistent approach to alcohol throughout the federal government's jurisdiction would be a desirable component of any government-sponsored program to shape drinking practices. This argument is appealing in institutional terms because it seems easier to establish a coherent line within one level of government than to operate through the myriad state and local agencies that write laws and regulations, enforce them, mount preventive training pro- grams, and so on. Not only are there fewer agencies at the federal level, but one can rely on NIAAA to provide a natural and accepted insti- tutional focus for such an effort. Moreover, it seems conceivable that there would be substantial benefits from striking a consistent position within the federal government. The terrain covered by the federal gov- ernment is large enough to suggest that the direct effects of federal efforts would not necessarily be small. In addition, the symbolic weight of the federal government's position might play an important role in reinforcing individual decisions about drinking practices or mobilizing institutions at other levels of government and in the private sector to move in similar directions. In essence, if the weight of the federal gov- ernment could swing in a consistent, coherent direction, some broad social momentum could be developed even though many of the relevant agencies lie outside the direct control of the federal government. This argument seems plausible, but one can be skeptical about how easy it would be to develop a coherent line throughout the federal domain and how much this would matter to the positions of private individuals and other agencies concerned with the problem. Probably the strongest argument for undertaking the arduous bureaucratic task of coordinating alcohol policy throughout the federal government is not for the symbolic effects, but for the material effects. Enough drinking

OCR for page 79
98 REPORT OF THE PANEL behavior is within the direct reach of the federal government to make it worthwhile to manage this drinking, quite apart from any symbolic effect. CONCLUSION We began this chapter by rehearsing the substantial normative and prag- matic objections to the general idea that government might be able to shape drinking practices directly. We conclude with some guarded op- timism about the potential for some carefully designed efforts to ac- complish this purpose in selected areas. In general, we recognize that the law is a clumsy and expensive in- strument to wield in shaping private drinking practices, but in one area we think it is feasible and desirable to do so: the regulation of drunken driving. The problems associated with drunken driving are large enough and involve enough innocent third parties to warrant legal intervention. Moreover, institutions exist that can enforce such laws more aggressively and are backed by a consensus against drunken driving. Major barriers to more aggressive action in this area seem to be some confusion about the nature of the offense and ambiguity about institutional leadership. Ironically, the confusion about the offense has been created partly by mass media campaigns against drunken driving. Ads suggest that any level of drinking is dangerous in combination with driving. This causes many people to assume that the criminal offense is simply having a drink and then driving something that many people have done. Hence, they are inclined to be lenient with accused offenders. Actually, the offense is drunken driving, and the standard for drunkenness is usually set far above the levels of intoxication that most people reach. If people (par- ticularly jurors) knew just how drunk one had to be to be charged with the offense, it is conceivable that they would be more willing to convict, because the legal level of drunkenness really does establish a deviant level of drinking. The confusion about institutional leadership exists because NHTSA and NIAAA share federal jurisdiction in this area, and virtually all of the enforcement capacity is at state and local levels. Education programs seem attractive because they generally appear to be less intrusive, more flexible, and more precise, but there is little direct evidence indicating that they can succeed, particularly on a large scale. We are especially skeptical of the idea that adults or youths in high school and college can be told things about drinking that will then keep their drinking within safe limits for the whole of their lives. We do think that educational institutions might be able to shape current drinking practices in safer directions, by talking about current dangers

OCR for page 79
Shaping Drinking Practices Directly 99 and risks and providing some techniques for improving practices if peo- ple are so inclined. More generally, we think that the key to successful education programs is to combine (1) well-designed local mass media campaigns that increase public salience and private concern about drink- ing practices with (2) low-keyed programs offering simple behavioral training techniques to those who are motivated to change their drinking practices. In particular, we think programs focusing on the health con- sequences of too much or too reckless drinking mounted by local hos- pitals and health maintenance organizations might succeed in reducing some of the adverse consequences of drinking. In undertaking such programs it is very important to understand that the objective is to change specific bits of behavior and also that we are fundamentally dependent on the preexisting motivation of people to produce these changes. The task is thus to make it relatively easy and unembarassing for people to learn how better to understand and control their drinking. Finally, we think it is desirable for the federal government to review its programs in a variety of areas and ensure that its actions are consistent with its position with respect to drinking. This is important primarily because the cumulative impact of the various areas of federal influence might be substantial and also because the symbolic effect of doing so might stimulate other institutions to move in similar directions. We note that actions of the type suggested here always carry with them significant social risks. In a society as large and heterogeneous with respect to drinking practices as ours is, it is difficult to establish any kind of boundary around appropriate drinking practices. Those whose practices are now well within the boundary may feel outraged that the government could be so irresponsible as to license a kind of drinking that seems very reckless to them. Those whose practices are outside the boundary may feel indignant that the government is dis- couraging conduct that seems quite safe to them and may complain that the government is interfering. Moreover, to make matters worse, when- ever the government draws lines with respect to drinking practices, many third parties are mobilized and feel entitled to disapprove of and com- ment on practices that are outside those lines. This can exacerbate social conflict over drinking and produce important consequences even if the level and character of drinking does not change. It is for these reasons that we considered the substantive changes in drinking practices simul- taneously with the instruments. We believe there is sufficient consensus about drunken driving and heavy cumulative consumption to make these suitable subjects for the law and for community-based health training intervention, respectively; we remain uncertain about how we might proceed to discourage drunken violence.