Promoting Health: Intervention Strategies from Social and Behavioral Research (2000)

Lawrence O.Gostin, J.D., LL.D (Hon.)¹

Government has a number of “levers” at its disposal to try to prevent injury and disease and promote the population's health.² Laws and regulations, like other prevention strategies, can intervene at a variety of levels, each designed to secure safer behavior among the population—individual behavior, agents of behavior change, and the environment (Haddon et al., 1964; Institute of Medicine, 1999).

First, government interventions are aimed at individual behavior—through education, incentives, or deterrence. For example, health communications campaigns educate and persuade people to make healthier choices; taxing and spending powers discourage risk behaviors or encourage healthy activities; and police powers deter risk behaviors by imposing civil and criminal penalties. Second, the law regulates the agents of behavior change by requiring safer product design. For example, government regulates unsafe products directly by requiring safety features or indirectly through the tort system. Finally, the law

alters the informational, physical, social, or economic environment to facilitate safer behavior. For example, government demands accurate labeling and instructions or restricts commercial advertising of hazardous products; enacts housing and building codes to prevent injury and disease; and regulates the environment in diverse ways.

Despite the importance of legislation and regulation to promote the public's health, the effectiveness of government interventions is poorly understood. In particular, policy makers often cannot answer important empirical questions—do legal interventions work and at what economic and social cost? Policy makers need to know if legal interventions achieve their intended goals (e.g., reducing risk behavior). If so, do legal interventions unintentionally increase other risks (“risk-risk” trade-offs)?³ Similarly, cost is important because, if legal interventions are very expensive, they can reduce opportunities to spend those resources in more effective ways.

Policy makers need to understand not only the cost or effectiveness of an intervention, but also the burdens it imposes on individuals and groups. Public health regulations can intrude on personal interests, economic interests, and the rights of minorities. Some public health regulations affect personal interests in autonomy, privacy, expression, or liberty. Other regulations affect economic interests in property and, more generally, dampen economic development. Still other regulations are unjust because they disproportionately burden vulnerable groups such as the poor and racial minorities. The laws themselves may be unfair or they may be enforced selectively and in a discriminatory manner. These intrusive regulations should concern us not only because freedom has intrinsic value in our democracy, but because denial of human rights can undermine public health efforts (Mann et al., 1994). Even detriments to professional opportunities and economic prosperity can have health effects. Researchers, for example, have long understood the positive correlation between socioeconomic status and health (DHHS, 1991).

In this paper, I first examine legal interventions as a tool in achieving the public's health and safety. Here, I present five models of public health regulation: (1) economic incentives and disincentives (taxing and spending powers); (2) the informational environment (education, labeling, and commercial speech regulation); (3) direct regulation (penalties for engaging in risk behavior); (4) indirect regulation (the tort system); and (5) deregulation (dismantling legal barriers to desired public health behaviors). In relation to each of these five models, I explore the trade-offs entailed between advancing the population's health and invading private rights and interests. Second, I examine the empirical question, do legal interventions work? Here, I present data from two of the most studied areas of intervention—one involving disease control (cigarettes) and the other

injury control (automobile crashes). I also examine some of the methodological problems that thwart many evaluations of law reform, as well as the striking absence of systematic research evaluating the large majority of legal interventions. Finally, I draw some conclusions about the use of the law as a tool for achieving public health goals. How important is the law, what risks are entailed in legal strategies, and how can legal interventions best be used in combination with other public health strategies?

Elsewhere, I define public health law as the study of the legal powers and duties of the state to assure the conditions for people to be healthy and the limits on that power to constrain the liberty or property of individuals for protection or promotion of community health (Gostin, 2000). Through this definition, I suggest five essential characteristics of public health law: (1) government has the responsibility to advance the public's health; (2) populations are the main focus of public health interventions; (3) relationships are primarily between the state and the population; (4) services meet the needs of populations and are grounded in scientific methodologies; and (5) coercion of individuals and businesses may be necessary for the protection of the community.

If government has an obligation to assure the conditions for the population's health, what tools are available to accomplish that task? Here, I present five regulatory models intended to demonstrate the breadth and importance of legal and public policy interventions to advance the public's health. Public health interventions, however, are not an inherent good. More often than not, when government acts to promote the population's health it undermines an important private right or interest. In this section, I also explore the trade-offs between the common good and private interests in liberty and property.

The government has a number of powers at its disposal to create potent incentives or disincentives for changing the behavior of individuals and businesses—the power to tax and spend. By Article I, Section 8 of the Constitution, “Congress shall have Power To lay and collect Taxes…and provide for the common Defence and general Welfare of the United States.” State constitutions provide similar taxing and spending powers. On its face, the power to tax and spend has a single, overriding purpose—to raise revenue and allocate resources to provide for the good of the community. Absent the ability to generate sufficient revenue, the legislature could not provide services such as education, medical services to the poor, sanitation, and environmental protection.

The taxing power, while affording government the financial resources to provide public health services, has another, equally important, purpose. The power to tax is also the power to regulate risk behavior and influence health-promoting activities. Virtually all taxes achieve ancillary regulatory effects by imposing an economic burden on the taxed activity or providing economic relief for certain kinds of private spending. Consequently, the tax code provides incentives and disincentives to perform, or to refrain from performing, certain acts.

Through various forms of tax relief (e.g., excluding benefits from taxable income, deducting spending from gross income, and providing credits against tax owed), government provides incentives for private activities that it views as advantageous to community health. Employer-sponsored health plans, for example, receive generous tax incentives that “deeply [affect] how health care is provided in the United States, to whom it is provided, and who provides it.” (Fox and Schaffer, 1987, p. 610). The tax code influences private health-related spending in many other ways: encouraging child care to enable parents to enter the work force; inducing investment in low-income housing; and stimulating charitable spending for research and care.

Public health taxation also regulates private behavior by economically penalizing risk-taking activities. Tax policy discourages a number of activities that government regards as unhealthy or dangerous. Consider excise or manufacturing taxes on tobacco and alcoholic beverages. Tax policy also influences individual and business decisions that adversely affect human health or the environment, such as taxes on gasoline or on ozone-depleting chemicals that contribute to environmental degradation. It is difficult to imagine a public health threat caused by human behavior or business activity that cannot be influenced by the taxing power.

While many Americans resent the use of government taxing power, they generally understand its importance in a good society. Still, high taxation of inherently dangerous products such as cigarettes and alcoholic beverages is not universally supported. On one level, businesses and consumers denounce increased taxation of hazardous products as they would any tax—liberal government overreaching and economically burdening the population. To some, excessive taxation of lawful conduct represents government intrusion into the private sphere, where individuals should be permitted to make free, autonomous choices. To others, the taxation itself appears regressive because the poorer and working classes disproportionately engage in the taxed behavior.

A high tax on cigarettes, for example, is widely supported by public health professionals because it provides a disincentive to smoking. Taxation may reduce smoking particularly among young people whose demand for the product is elastic. The cigarette tax, however, is burdensome. It may compel individuals to pay a “health tax” on a behavior that is addictive and, to a certain extent, beyond the person's control. More importantly, the tax is highly regressive because most adults who smoke are in lower socioeconomic classes.

The power to spend is closely aligned with the power to tax. Economists regard legislative decisions to provide tax relief for certain activities as indirect

expenditures because government is, in fact, subsidizing the activity from the national treasury. Economists project, for example, that favorable tax treatment afforded to employer-sponsored health care plans will cost the federal government $438 billion between 1999 and 2003 (OMB, 1998).

The spending power does not simply grant legislatures the authority to allocate resources; it is also an indirect regulatory device. Congress may prescribe the terms upon which it disburses federal money to the states. The conditional spending power is akin to a contract: in return for federal funds, the states agree to comply with federally imposed conditions.⁴ For example, in the mid-1980s, Congress appropriated highway funds on the condition that states adopt a 21-year-old minimum drinking age. Since a major purpose of highway funds is traffic safety, the Supreme Court upheld this public health measure.⁵

Congress' power to set the terms upon which state appropriations shall be distributed is an effective regulatory device. States and localities can seldom afford to decline public health grants. Congress and the federal agencies use conditional appropriations to induce states to conform to federal standards in numerous public health contexts. Federal funding programs for AIDS/HIV, for example, require involuntary postconviction testing of sex offenders, adoption of Centers for Disease Control and Prevention (CDC) guidelines for HIV counseling and testing of pregnant women, and compliance with specific community planning and program priorities. Conditional spending induces states to conform to federal regulatory requirements in other areas as well: eligibility and quality standards relating to Medicare and Medicaid, prohibition on recipients of fam-ily-planning funds from engaging in abortion counseling, and state and local planning for land use and solid waste management.

The federal government and the states also award grants and contracts to individuals and groups in the private and charitable sectors. These awards frequently contain conditions that require individuals to refrain from, or conform with, specific behavior. For example, the Supreme Court upheld a federal regulation that prohibited recipients of federal family-planning funds from providing abortion counseling or referrals.⁶

It is obvious that the power to tax and spend is not value neutral, but rather laden with political overtones. Collection of revenues and allocation of resources go to the very heart of the political process. Legislators, as influenced by the public and interest groups, purport to promote the public's health, safety, and security. Many of their economic decisions do promote the common welfare, such as taxes on cigarettes and expenditures for antismoking campaigns. But their vision is also influenced by moral, social, and cultural values so that government's economic power may be used to discourage fetal research, sex educa-

tion, or needle exchange. The power to tax and spend, then, may be used to impede, as well as to promote, legitimate public health goals.

The field of public health is deeply concerned with the communication of ideas. Human behavior is a powerful contributor to injury and disease, so government strives to influence behavioral choices. Public health authorities understand that many factors influence behavior, but information is a prerequisite for change. The population must at least be aware of the health consequences of risk behaviors to make informed decisions. The citizenry is bombarded with behavioral messages that affect its health—by the media and entertainment, trade associations and corporations, religious and civic organizations, and family and peers. Public health authorities strive to be heard above the din of conflicting and confusing communications. Consequently, the field of public health is a virtual battleground of ideas. Consider the controversy surrounding the Office of Drug Policy's approving scripts in lieu of public service announcements on prime-time programming. While many people would applaud the government's attempts to promote antidrug messages, they might object to the deceptive manner in which those messages are conveyed.

Public health agencies deliver messages to promote healthy behavior and restrict messages that encourage risktaking. Health authorities exert influence on behavior at various levels of intervention: at the individual level, they counsel individuals at risk for injury and disease (e.g., HIV prevention, genetic disease, or reproductive choice); at the group level, they inform and promote behavior change (e.g., hospital or health maintenance organizations [HMO] newsletters, support groups for stress reduction, or smoking cessation); at the population level, they educate the public and market healthy life-styles (e.g., health communication campaigns or school-based health education).

We like to think that public health's concern with “healthy messages” is an inherent good, and we are prepared to give government considerable leeway in constructing a “favorable” informational environment. But the concern of public health authorities with health, communication, and behavior conflicts with other important social values. When public health authorities counsel individuals or educate groups or populations about the government's view concerning unsafe sex, abortion, smoking, a high-fat diet, or a sedentary life-style, there is no formal coercion. Yet education conveys more than information; it is also a process of inculcation and acculturation intended to change behavior. Even though they are not always particularly adept or successful, public health authorities routinely construct social norms or meanings to achieve healthier populations. Not everyone believes that public funds should be expended, or the veneer of government legitimacy used, to prescribe particular social orthodoxies (Lessig, 1995).

No matter how much government educates the public, it is bound to have difficulty changing risk behaviors. The reason is that the private sector has a

strong economic interest in influencing consumer preferences. Advertising commands ever-increasing resources and engenders ever more creative approaches to stimulate consumer demands. As a result, government regulation of commercial speech is an important strategy to safeguard consumer health and safety. First, government is concerned with advertising that increases use of hazardous products and services such as cigarettes, alcoholic beverages, firearms, and gambling. For example, advertisements extol the virtues of handguns for home defense and safety, despite scientific evidence of increased risk of suicide, domestic violence, and unintended death (Vernick et al., 1997). Second, government is concerned with marketing “age-restricted” products to children and adolescents. For example, the Distilled Spirits Council lifted its 48-year-old voluntary ban on liquor advertising on radio and television, prompting concern about the targeting of young people (Elliott, 1996). Third, government is concerned with industry health claims that mislead the public. For example, government actively reviews food and dietary supplement labels⁷ and direct-to-consumer pharmaceutical marketing for unsubstantiated health claims. Regulation of commercial speech is becoming even more challenging as government monitors Internet marketing of the sale of hazardous products or the expression of deceptive health messages.

Government regulation of commercial speech raises profound social and constitutional questions. Substantial public health benefits can be achieved by reducing risk behavior, but regulation of communications can stifle freedom of expression. Nominally, the Supreme Court affords commercial speech “lower-value” constitutional protection than social or political discourse.⁸ But in reality, the level of scrutiny for commercial speech has changed over the years and the modern Court steadfastly defends truthful, nonmisleading advertisements. Most public health advocates argue that advertisements that associate hazardous products with glamorous, adventuresome, sexy life-styles are, in fact, misleading. But the Court has not viewed mere “puffery” as inherently misleading and appears to prefer to have consumers make the decisions for themselves.

Perhaps more importantly, the Court is insisting that the government must demonstrate, most probably with credible evidence, that the regulation will, in fact, achieve the asserted public health goal and not be an “ineffective” or “remote” method.⁹ This standard demands an empirical response. Absent careful research into the effectiveness of commercial speech regulation, this form of

public health intervention may be found unconstitutional by the Supreme Court.¹⁰

Public health authorities have the power to set health and safety standards as well as to monitor compliance and punish noncompliance. Monitoring activities include injury and disease surveillance, screening individuals for disease and reporting positive cases to health authorities, and inspecting commercial premises. Government can directly prohibit behavior (e.g., exceeding highway speed limits) or require individual behavior (e.g., wearing seatbelts); mandate businesses to alter product designs to promote safety (e.g., air bags in automobiles, trigger locks on firearms, safety caps on medicines); or modify sales practices (e.g., sale of cigarettes or alcoholic beverages to children). In the area of infectious diseases, government can compel vaccination, treatment, or isolation.

The state's authority to monitor, prohibit, or mandate behavior is derived from the “police power” —the most famous expression of the natural authority of sovereign governments. The police power is the state's inherent authority to regulate private interests to protect, preserve, and promote the health, safety, morals, and general welfare of the people. To achieve these communal benefits, the state retains the power to restrict —within federal and state constitutional limits—private interests.

In the foundational case of Jacobson v. Massachusetts, the Supreme Court explained its early twentieth century community-oriented philosophy: “There are manifold restraints to which every person is necessarily subject for the common good. On any other basis organized society could not exist with safety to its members… A fundamental principle of the social compact is that the people covenant that each citizen shall be governed by certain laws for the common good, and that government is instituted for the common good, for the protection, safety, prosperity and happiness of the people.”¹¹

Despite the undoubted police power authority of government to act for the welfare of society, there are constitutional limits beyond which it cannot go. Every measure that sets health and safety standards, monitors compliance and punishes noncompliance, creates individual burdens. Surveillance (i.e., government collection of personal health data) affects privacy; compulsory screening affects autonomy; mandatory treatment affects bodily integrity; isolation affects liberty. Commercial regulation also affects proprietary interests. Licenses affect professional opportunities, while inspections and nuisance abatements affect

property rights. These trade-offs “between public goods and private interests” are highly complex and politically contested in a constitutional democracy.

Perhaps the most controversial area of police power regulation is the control of so-called self-regarding behavior—that is, behavior that risks primarily the health or safety of the individual concerned. Classical regulation of self-regarding behavior includes mandatory motorcycle helmet and seatbelt use, as well as gambling prohibitions. From the public health perspective, these kinds of laws are justified because they reduce behavioral risks, thereby improving the population 's health and longevity. At the same time, many argue that individuals should have the right to make their own decisions about behaviors that primarily affect themselves. After all, a person declines to wear a motorcycle helmet not because he or she is oblivious to the risk, but presumably because that person places one value (freedom) above another (physical security). There is a special resistance to regulation that involves government controlling how people behave when they pose little public risk.

Court decisions usually support regulation of self-regarding behavior, but often reduce the justification to a strained conception of social harms rather than recognizing certain public health interventions as justified paternalism. Consider one state court's justification for motorcycle helmet regulation: “From the mo-ment of the [motorcycle] injury, society picks the person up off the highway, delivers him to a municipal hospital, and provides him with unemployment compensation … We do not understand a state of mind that permits plaintiff to think that only he himself is concerned.”¹²

If public health authorities continue to press for regulation of self-regarding behavior, it will certainly be important to be able to point to empirical evidence of effectiveness in reducing harms within the population.

The foregoing methods of regulation are comprised principally of actions taken by legislatures and administrative agencies to prevent injury or disease or to promote public health. The creation of private rights of action in the courts can also be an effective means of public health regulation (Jacobson and Warner, 1999). Perhaps the most important form of civil litigation for public health purposes can be found in tort law to redress harms to people and the environment they inhabit.

A tort is a civil, noncontractual wrong for which an injured person or group of persons seeks a remedy in the form of monetary damages. Tort law, then, characteristically is a private, rather than a public, right of action, a civil rather than a criminal proceeding; it remedies a wrong by awarding monetary damages rather than an injunction or civil or criminal penalties. Tort law is comprised of

a series of related doctrines that impose civil liability upon persons or businesses whose (usually) substandard conduct causes injury or disease.

The functions, or goals, of tort law—although highly controversial and imperfectly achieved—are the (1) assignment of responsibility to individuals or businesses that impose unreasonable risks, causing injury or disease; (2) compensation of persons for loss caused by the conduct of individuals or businesses; (3) deterrence of unreasonably hazardous conduct; and (4) encouragement of innovation in product design, packaging, labeling, and advertising to reduce the risk of injury or disease. In thinking about tort law as a tool of public health, it is important to emphasize the role of litigation in preventing risk behavior and providing incentives for safer product design.

There exists a vast potential for using tort litigation as an effective tool to reduce the burden of injury and disease. Attorneys general, public health authorities, and private citizens resort to civil litigation to redress many different kinds of public health harms: environmental damage or exposure to toxic substances; unsafe pharmaceuticals or vaccines; and hazardous products such as tobacco, firearms, or alcoholic beverages.

Private citizens, as well as government officials, have adopted a civil litigation strategy in two highly publicized public health contexts—cigarettes and firearms. In the context of cigarettes, lawsuits have been filed by private litigants, populations of smokers (class actions), and government entities (medical reimbursement suits) (Gostin et al., 1991; Daynard and Kelder, 1998). For example, in November 1998, a Settlement Agreement between tobacco companies and 46 states required industry to compensate states in perpetuity, with payments totaling $206 billion through the year 2025. The agreement also restricted outdoor advertising, the use of cartoon characters, tobacco merchandising, and sponsorship of sporting events. In exchange, the industry received civil immunity for future state claims, but not for individual or class action lawsuits.¹³

Thus far, litigants have not been as successful in firearm litigation. Most firearm design defect cases fail because courts rule that consumers have a “reasonable expectation” that firearms will be dangerous.¹⁴ Firearm risks, for the most part, are open and obvious, and consumers do not require expert knowledge to assess the risks.¹⁵ Firearms, moreover, are in “common use.” If they are functioning properly, they may be exempt from strict liability based on the the-

ory that they are “good” (well-made) guns.¹⁶ Nevertheless, recent litigation has followed the model used against cigarette manufacturers. Many cities (e.g., New Orleans and Chicago) have filed, or intend to file, tort actions against the gun industry (Kairys, 1998). The city lawsuits proceed on different theories of recovery including product liability theory¹⁷ and public nuisance theory claiming that industry marketing practices encourage movement of firearms into the possession of criminals. ¹⁸

Tobacco and firearm litigation strategies are premised on the theory that these industries have made conscious choices that increase disease, injury, and death, particularly among children and adolescents. Tort litigation is seen as one strategy, among many, to encourage safer design, greater disclosure, and reduced consumption of potentially hazardous products. Whether tort law can achieve all of these public health goals is uncertain. Litigation is still languishing in the terrain of personal responsibility—the idea that it is not the manufacturer of the product that creates harm, but the user. It is in this regard that the emerging tort strategy of government lawsuits has advantages over individual tort actions. Governments have the resources to engage in mass litigation and seek damages incurred collectively by taxpayers. Furthermore, a manufacturer cannot attribute blame for harm to government entities the way it can on those who purchase and use hazardous products. Finally, discovery powers in civil litigation may provide needed evidence about the industries' intentions in design, promotion, and sale.

Like other forms of regulation, the tort system does not achieve only socially desirable results. Tort costs (liability and litigation expenses) have potentially detrimental effects that warrant consideration. Tort costs are absorbed by the enterprise, which will often pass the costs onto its employees or consumers through loss of employment (or benefits of employment such as wages or health insurance) and/or higher prices. Since socioeconomic status has important health effects, lower employment or wages and higher consumer prices are important from a public health perspective. Tort costs may be, or may be perceived to be, so high that businesses leave the market or curtail research and development. Society, it may be argued, would not be any the poorer if tort costs made it difficult for dangerous enterprises (e.g., tobacco and firearms) to operate within the market. Tort costs, however, appear to be just as high for socially advantageous goods and services such as drugs, vaccines, and medical devices. While it is in

society's interest to encourage safety in the production of health-related products and services, overdeterrence may deter investment and innovation.

Statutes and regulations may actually impede the achievement of public health objectives, and the necessary social response is to deregulate. Sometimes policy makers enact laws that criminalize or otherwise penalize provision of, or access to, the means for achieving behavior change. Consider government regulation designed to impede access to sterile syringes and needles even though public health authorities recommend their use to reduce transmission of blood-borne diseases such as HIV and HBV (hepatitis B virus). At least three kinds of statutes and regulations impede access to sterile drug injection equipment (Gostin and Lazzarini, 1997; Gostin, 1998).

Virtually all states have drug paraphernalia statutes that ban the manufacture, sale, distribution, possession, or advertisement of a wide range of devices, including syringes, if it is known that they may be used to introduce illicit substances into the body. It is not an offense under these statutes to sell or distribute syringes without knowledge that they will be used to inject illicit drugs. Thus, a pharmacist who sells syringes over the counter, believing they will be used for a lawful purpose such as to treat diabetes with insulin, commits no offense under drug paraphernalia laws. Most state statutes are based on the Model Drug Paraphernalia Act (DEA, 1991); the federal government also has enacted the Federal Mail Order Drug Paraphernalia Control Act that prohibits the sale and transportation of drug paraphernalia, including syringes, in interstate commerce. ¹⁹

A minority of states (mostly in the Northeast corridor) have syringe prescription statutes that prohibit the dispensing or possession of syringes without a valid medical prescription. Prescription statutes differ from drug paraphernalia laws in that criminal intent is not required for a violation. The pharmacist dispensing a syringe without a prescription in a jurisdiction with a prescription law does not have to be aware that the buyer intends to use it to administer illegal drugs; the act of selling or dispensing the syringe without a prescription is itself a violation. In this sense, prescription laws have the potential to encompass many more transactions than paraphernalia laws.

The government also uses its spending powers to restrict access to sterile injection equipment. Since 1988, Congress has enacted statutes that contain provisions prohibiting or restricting the use of federal funds for syringe exchange programs (SEPs). The U.S. General Accounting Office concluded that the Department of Health and Human Services (DHSS) is authorized to conduct demonstration and research projects involving the provision of syringes, but is prohibited from using certain funds to support SEPs directly, which may prevent

DHSS from funding ancillary support services provided by SEPs. Accordingly, SEPs must rely on state, local, or philanthropic funds for their operational activities, which is problematic given their uncertain legal status (GAO, 1993).

Under current law, before the federal government can fund SEPs, the Secretary of Health and Human Services must find that the programs effectively prevent the spread of HIV and do not encourage the use of illegal drugs. The National Research Council (1995) recommended that the Surgeon General make such a determination, and a National Institutes of Health (1997) consensus panel concurred. Nevertheless, the Secretary, while acknowledging that SEPs are effective, did not permit the use of federal funding.

Taken together, these laws and funding restrictions, make it difficult (but not impossible) for public health agencies and community groups to operate needle exchange programs in many states (Burris et al., 1996; Paone et al., 1999). They also make it difficult for injection drug users to reliably purchase sterile equipment in pharmacies (Case et al., 1998). There have been numerous research studies evaluating the effectiveness of deregulation of drug injection equipment. While the evidence for reduction of hepatitis B and hepatitis—is mixed (Hagan et al., 1999), most studies show that SEPs (Heimer, 1998; Heimer et al., 1998; Vlahov and Junge, 1998) and pharmacy access to syringes (Singer et al., 1998) reduce the incidence of HIV. In October 1999, the American Medical Association and many other national public health organizations sent a joint letter to the states urging the removal of legal penalties for possession of syringes that discourage injection drug users from carrying syringes and contribute to increased risk behaviors (AMA et al., 1999).

It is important to emphasize that the position of elected officials in these areas is not irrational. Rather, these officials, sensitive to the desires of many constituents, prefer one value (discouraging drug use) over another value (public health). Empirical evidence may help inform public policy in this controversy as well. It ought to be a “testable” proposition to inquire whether support for SEPs actually facilitates drug use. Most public health authorities believe that there is no relationship between access to syringes and greater drug use (AMA et al., 1999).

The five models of public health regulation just described are all highly complex, involving trade-offs between competing public goods as well as conflicting public and private interests. As the experience with drug injection equipment shows, politicians do not uniformly follow evidence of effectiveness because there exist many different values involved in policy making. Nevertheless, policy makers cannot even begin to assess the importance of public health regulation unless they have the benefit of impartial and rigorous research assessments. If policy makers are to support regulation they need to justify that decision by the probable achievement of an important public health objective.

Despite its importance, legal interventions often are not subjected to rigorous research evaluation. Part of the reason for the absence of systematic evaluative research is that universities do not always train investigators to evaluate legal interventions. Further, research in this area is not adequately funded. Finally, policy makers do not always understand the need for research. Many regulatory approaches have been in place for so long that they are simply presumed to be effective.

Just as important is the fact that evaluations of legal interventions face daunting methodological challenges. There exist so many variables that can affect behaviors (e.g., changes in the informational, physical, social, and cultural environment) that it can be extraordinarily difficult to demonstrate scientifically a causal relationship between the intervention and the perceived health effect (Flay and Cook, 1990). Consider the methodological constraints in identifying the causal effects of specific laws designed to reduce driving while under the influence of alcoholic beverages. Several different kinds of laws may be enacted within a short period of time, making it hard to isolate the effect of each law; publicity about the problem and the legal response may cross state borders, making state comparisons more difficult; people who drive under the influence may also engage in other risky driving behaviors (e.g., speeding, failing to wear safety belts, and running red lights), so that researchers need to control for changes in other highway safety laws and traffic law enforcement; and treatment and control communities may differ in unanticipated ways (Hingson, 1996; DeJong and Hingson, 1998).

Despite these challenges (relating to training, funding, and methodology), social science researchers have studied a number of discrete legal interventions, often with encouraging results. The following review of the social and behavioral science literature is not intended to be systematic, but does illustrate two important points. Firstly, legal interventions can be an essential part of a comprehensive strategy to influence behaviors desirable for the public health; and secondly, social and behavioral science research is pivotally important in evaluating the effectiveness and costs of legal interventions.

Of course, not only is research important in evaluating existing legislation; it is also important in helping policy makers to decide whether to enact legislation. Without adequate data, policy makers cannot make informed judgments about the need for, or importance of, legislation. For example, surveillance data demonstrated that increasing the minimum age for alcohol purchases from 18 to 21 resulted in fewer teenage traffic fatalities (Cook and Tauchen, 1984; GAO, 1987). These research findings were instrumental in influencing the Congress to enact legislation requiring states to adopt a 21-year-old drinking age as a condition for the receipt of federal highway funds (Wagenaar, 1993). All states currently have a 21-year-old drinking age, which is credited with saving many lives (NHTSA, 1996).

The social science, medical, and behavioral literature contains evaluations of interventions in a number of subject areas, particularly in relation to injury prevention (Rivara, 1997a,b; IOM, 1999). For example, studies have evaluated the

effectiveness of regulation to prevent head injuries (bicycle helmets: Lund et al., 1991; Dannenberg et al., 1993; Kraus et al., 1994; Thompson et al., 1996a,b; Ni et al., 1997; Rivara et al., 1998), choking and suffocation (refrigerator disposal and warning labels on thin plastic bags: Kraus, 1985), child poisoning (childproof packaging: Rogers, 1996), and burns (tap water: Erdmann et al., 1991). The regulatory measures that have received a great deal of attention include reduction in cigarettes and motor vehicle crashes. Funding agencies and researchers should give much greater attention to systematic evaluation in other areas of governmental intervention that have not been subjected to systematic evaluation.

Since tobacco is the primary preventable cause of morbidity and mortality, government has adopted multiple strategies to reduce smoking, particularly among children and adolescents. These interventions are directed mostly toward individual-level behavioral changes: education (e.g., antismoking campaigns), deterrence (e.g., bans on retail sales to minors), and disincentives (e.g., tobacco taxes). Government has also directed interventions at the level of the manufacturer (e.g., varied tobacco litigation strategies) and at the level of the informational (e.g., advertising restrictions) and physical environment (e.g., bans on smoking in the workplace and other public places). While there has been no systematic evaluation of all of these interventions, researchers have sought to analyze several governmental antitobacco strategies.

Perhaps the most encompassing studies have evaluated broad, multidimensional antitobacco campaigns that have been initiated in several states. While these state antitobacco initiatives differ in certain respects, they all involve taxation of cigarettes and use of the revenues for a multipronged approach to reducing smoking. In California, for example, the state mandated funding for health education campaigns, local health agencies to provide technical support and monitor adherence to antismoking laws, community-based interventions, and enhancement of school-based prevention programs (Bal et al., 1990). Evaluation studies in Massachusetts (Abt, 1997), California (Pierce et al., 1998), and Oregon (CDC, 1999a) all reported per capita reductions in cigarette consumption. An Institute of Medicine (IOM) analysis of these and other studies concluded that a regulatory approach toward cigarette use can be effective. The IOM found that states with “ intense” tobacco control initiatives (as measured by laws, law enforcement, and funding) showed greater decreases in tobacco use than comparable states (IOM, 2000).

Despite the promising results of these studies, there are certain limitations—both empirical and methodological. The California study reported a sharp decline in smoking directly following the intervention, but the effect dissipated over time and, ultimately, failed to significantly affect smoking behavior (Pierce et al., 1998). There are several possible explanations for the fact that the program's initial effects did not persist. One possible reason is that the effect demonstrated

was due as much to the political and social environment as it was to the interventions themselves. Undoubtedly, changes in the law, which are often associated with political debate and media coverage, have a “declarative effect” that influences behavior indirectly by changing attitudes; legal norms reinforce, stimulate, accelerate, or symbolize changes in public attitudes about socially desirable behaviors (Bonnie, 1986). Other possible explanations are temporary reductions in state funding for antitobacco programs, political interference with antitobacco messages, and industry advertising and pricing policy.

These studies also face difficult methodological challenges. Most importantly, public or private funding for evaluation research, including population surveys of smoking behavior, becomes available only after the intervention has occurred. Research funding after the fact thwarts a comparison of specific behaviors before and after the intervention. Even if this important methodological problem could be overcome, however, these studies will find it hard to identify precisely the intervention that is having the desired effect. The Massachusetts, California, and Oregon studies, by definition, examined multiple interventions without an ability to separate the effect of each element of the intervention.

The most distressing finding is that in spite of all the efforts, tobacco use among adolescents has recently increased. The data show, moreover, that current cigarette smoking prevalence among African Americans is similar to rates among whites and Hispanics (CDC, 2000).

There are, of course, numerous studies of discrete tobacco regulation strategies. Two of the major interventions that have been studied are programs to prevent youth access to cigarettes and bans on smoking in the workplace and other public places. In the absence of enforcement, banning cigarette sales to minors does not significantly reduce teenage tobacco use. Compliance with the law by retailers is low; youth access studies have demonstrated that the majority of retailers sell to children. The Tobacco Institute's own “It's the Law” campaign, perhaps predictably, has not been effective (DiFranza and Brown, 1992). Efforts to educate retailers to reduce cigarette sales to minors have shown an effect, but a small one (Wildey et al., 1995). The most successful legal strategy to reduce youth access imposes civil penalties against the store owner (not just the clerk), has progressive increases in fees culminating in suspension or revocation of the tobacco retailer' s license and has regular enforcement using minors in unannounced purchase attempts to monitor compliance. This strategy, implemented by legislation in Woodbridge, Illinois, has been demonstrated to be successful over time (Jason et al., 1999). Other states, such as Minnesota, have also demonstrated the effectiveness of well-enforced antitobacco ordinances restricting youth access to tobacco (Forster et al., 1998). Additionally, locking devices on cigarette vending machines do not appear to be as effective as an outright ban on these machines (Forster et al., 1992). Finally, some states have reported modest success with antitobacco education campaigns directed toward youth (Noland et al., 1998).

Legal restrictions on smoking in the workplace and other public places have demonstrated high conformance with the rule and some reduction in cigarette use. A study of compliance with a clear indoor air act in Brookline, Massachu-

setts, showed that the law was popular and the incidence of smoking restrictions was high (Rigotti et al., 1992). Further, a summary of 19 studies that evaluated the effects of smoke-free workplaces on smoking habits showed that both smoking rates (cigarettes consumed in a 24-hour period) and smoking prevalence (proportion of workers who smoke) decreased as a result of the indoor smoking bans. The authors estimate an annual reduction of 9.7 billion cigarettes (2%) in the United States as a result of smoke-free workplaces (Chapman et al., 1999).

What is notably absent from the evaluative research are rigorous studies of the effects of tobacco and other forms of litigation. Although tort strategies are much used and publicized by private parties, classes, and national and state governments themselves, there remains considerable debate about the effects. Public health advocates strongly favor this approach, while some law and economic scholars are skeptical about the tort system as a useful and cost-effective intervention (Rose-Ackerman, 1991; Vicusi, 1992).

The experience with government strategies to reduce cigarette smoking shows that both the intervention and the behavior are highly complex. Many interventions have not been carefully evaluated and methodological difficulties have thwarted some of the studies that have been undertaken. A comprehensive, multipronged approach to cigarette reduction appears to work best, but problems still exist in the long term. This may suggest that government interventions may also have to restrict manufacturers, particularly in their advertisements and promotions. This, of course, is the current strategy of the federal Food and Drug Administration, but its proposed regulations are being challenged by the industry on jurisdictional grounds; the case is being heard in the Supreme Court's 1999–2000 term. At the same time, there are substantial societal issues surrounding the First Amendment protection to be afforded to commercial speech (Gostin et al., 1997). Once these legal issues are resolved, of course, it will be essential to study the effects of stricter regulation of the tobacco industry.

Unlike the case of tobacco, government has had relatively little difficulty regulating the automobile industry. Indeed, government regulation of motor vehicles has received such popular support and has been so successful that the CDC (1999b) called motor vehicle safety one of the 10 great public health achievements of the twentieth century.

The twentieth century may be known more for the proliferation of automobiles and other motorized vehicles than for any other technological change. Six times as many people drive today as in 1925, and the number of motor vehicles has increased elevenfold since then to approximately 215 million. The number of miles traveled is also 10 times higher than in the mid-1920s (CDC, 1999b).

Systematic safety efforts began during the mid-1960s when 41% of the nearly 94,000 unintentional injury deaths were caused by motor vehicle crashes. In 1966, the Highway Safety Act and the National Traffic and Motor Vehicle Safety Act authorized the federal government to set and monitor standards for motor vehicles and highways. The Highway Safety Act also created the National Highway Safety Bureau, which later became the National Highway Traffic Safety Administration (NHTSA).

The pervasive regulation of motor vehicle safety over the last three decades resulted in a steep decline in the mileage-based motor vehicle death rate (Fingerhut and Warner, 1997). A broad and creative array of regulatory strategies have been used to reduce the vast toll of motor vehicle injuries and deaths. As the IOM Committee on Injury Prevention and Control observed: “Dramatic progress has been made in reducing motor vehicle injuries by understanding the factors that increase the risk of injury, designing interventions to reduce these risks, implementing and evaluating a wide array of interventions and assessing their benefits and costs, and providing a scientific foundation for individual and business choices and public policy judgments” (IOM, 1999, p. 115). In short, government has been successful largely because it has adopted a sophisticated and comprehensive approach that addresses safety issues for the host (driver and passenger), agent (vehicle), and physical environment (highway system) (Haddon, 1972, 1980).

Individual-level interventions have been carefully studied, including general and specific deterrence, alcohol control policies, mass communications campaigns (including advertising restrictions), and community traffic safety programs (DeJong and Hingson, 1998). Two of the most important individual-level interventions are legislative policies to prevent alcohol-impaired driving and to increase the use of seatbelts and child restraints.

In 1996, 38.6% of fatal traffic crashes involved a driver or pedestrian who had been drinking (CDC, 1998; NHTSA, 1998a). Still, fatalities involving alcoholic beverages have decreased 39% since 1982 (CDC, 1999b). Government has sought to reduce drinking-impaired driving by aiming its interventions at those who have been convicted for driving under the influence (DUI) of alcoholic beverages (specific deterrence) and at the general driving population (general deterrence). A great deal of evaluative research has studied the differential impact of various rehabilitation and punishment schemes for those convicted of driving while under the influence of alcoholic beverages (Jacobs, 1989; Ross, 1993). One meta-analysis found that rehabilitation reduced alcohol-related driving recidivism and crashes by 7–9% when averaged across all kinds of offenders and rehabilitation techniques. Strategies that combined punishments with treatment were more effective than either strategy alone. Treatment did not substitute for sanctions, and sanctions did not substitute for treatment (Wells-Parker et al., 1995).

Various punishment strategies have been tried, with mixed results (Hingson, 1996):(1) mandatory license suspension, which has been found to be effective in reducing crashes and violations (e.g., administrative “per se” laws that automatically suspend driver's licenses as a penalty for driving under the influence of alcohol; Me Arthur and Kraus, 1999); (2) jail terms, which generally have not been found to effect postincarceration deterrence; (3) probation, which has a slight deterrent effect on drivers at low risk for DUI recidivism, but has no effect on those at high risk; (4) lower blood alcohol concentration (BAC) limits to 0.04% for previously convicted drivers, which resulted in a 38% decline in fatal nighttime crashes compared with a 50% increase in a neighboring state with no such law; and (5) license plate impoundment, which was effective when managed administratively, but not when managed through the courts.

The enactment of laws to impose firm and consistent punishment against convicted drunk drivers, with the hope of decreasing repeat offenses, worked only to a limited extent. Policy makers began to realize that successful prevention would require more than punishing the relatively small number of drunk drivers who are arrested and convicted each year. There emerged a policy of general deterrence targeted at the general population. DeJong and Hingson (1998) reviewed the extant research of various general deterrence policies: (1) administrative license revocation (where police remove driver's licenses if they exceed the legal BAC limit) probably reduces alcohol-related traffic fatalities, but studies have had difficulty isolating the effect from other policies (Hingson, 1996); (2) sobriety checkpoints (where police roadblocks check for drivers who have been drinking) resulted in reductions in alcohol-related fatalities compared with contiguous states; (3) criminal per se laws (which lower BAC limits from 0.10 to 0.08%) reduce fatalities (Hingson et al., 1996); (4) reduced availability of alcoholic beverages had positive effects if servers were trained (DeJong and Hingson, 1998), but not with respect to other server interventions (McKnight, 1996); and (5) increased alcohol excise taxes, which reduced fatality rates in the 15- to 24-year-old age group.

General deterrence strategies have particularly targeted young drivers who tend to take greater risks. Researchers have identified three successful strategies to reduce injuries and deaths among young drivers. One successful strategy has been to increase the minimum legal drinking age from 18 to 21 years of age. This strategy reduced traffic fatalities in the target ages, compared with states that did not adopt such laws (Hingson et al., 1983; Cook and Tauchen, 1984; GAO, 1987; Decker et al., 1988; O'Malley and Wagenaar, 1991; Jones et al., 1992; Hingson, 1996). A second strategy, which has been enacted in 49 states, is to establish lower legal blood alcohol limits for younger drivers than for adult drivers. Researchers in six studies reported that this approach reduced injuries: “Despite the methodologic difficulties of ecologic studies, the six studies reviewed represent accumulating evidence in support of the effectiveness of these laws” (Zwerling and Jones, 1999). A third strategy aimed at young persons is to place restrictions on the licenses of beginning drivers, such as proscribing nighttime driving or prohibiting beginning drivers from having other young passengers in the vehicle

(Foss and Evenson, 1999). Nighttime driving restrictions, for example, have been shown to reduce injuries and deaths (Williams and Preusser, 1997).

During the years 1985 to 1995, drivers younger than 21 years experienced nearly a 50% drop in fatal crashes involving alcoholic beverages (NHTSA, 1997). While much of this effect can probably be attributed to the declarative and actual effects of regulation, the reasons for the decrease remain unclear. For example, a recent study demonstrated that while drivers below the age of 21 were still drinking heavily, a smaller percentage of younger drivers than of older drivers had BACs of 0.01 or higher. The study concluded that younger drivers are more likely than drivers older than 21 years to separate drinking from driving (Roeper and Voas, 1999).

Another strategy aimed at individual-level interventions involves laws mandating child restraint and safety belt use (Graham, 1993). Since 1979, all states have had legislation requiring infants and young children to be restrained when traveling in passenger cars. While differences in penalties, enforcement, and public education make it difficult to compare states, the extant data show that these laws significantly reduced child injuries and deaths (Guerin and MacKinnon, 1985; Muller, 1986; Margolis et al., 1988; Goldstein and Spurlock, 1998). For example, following an Orange County statute that mandated restraints for children under 4, restraint use doubled, there were fewer injuries overall, and fewer head injuries, yet emergency room utilization did not decrease (Agran et al., 1987).

Child restraint laws, however, appear to have an insignificant effect on reducing the high-risk behavior of children traveling in the front seat of a car. Interventions aimed at promoting the use of rear seats by children traveling in motor vehicles continue to be very important (Segui-Gomez, 1998).

Since New York became the first state to enact a mandatory seat belt law in December 1984 (Preusser et al., 1988), studies have consistently demonstrated the beneficial safety effects of such laws (Thyer and Landis, 1988; Bernstein et al., 1989; Williams et al., 1996). After the enactment of seat belt laws, there are fewer crashes, the crashes are less severe, and the occupants require less medical treatment for head and face injuries (Lestina et al., 1991).

In 1993, California became the first state to modify existing safety belt laws from a secondary to a primary enforcement provision (Ulmer et al., 1994). A primary law gives police the authority to stop a vehicle solely on the basis of their observation of noncompliance with the safety belt law; a secondary law permits police officers to cite unbelted occupants only when the vehicle is stopped for another violation. After enactment of California's primary law, rates of safety belt use rose from 73 to 95.6% (Lange and Voas, 1998). Rivara and colleagues (1999) reviewed 48 studies of primary and secondary enforcement of seat belt laws, finding that primary enforcement laws are likely to be more effective. Nevertheless, these researchers observed that few studies are of good

quality, and quantitative estimates of the relative effect of primary compared with secondary laws are limited.

Individual-level approaches, while often effective, do raise important constitutional and ethical issues in a constitutional democracy. As explained above, compulsory seat belt and motorcycle helmet laws have been attacked as unwarranted paternalism and challenged on the basis of personal freedom (Cole, 1995). Yet courts and philosophers have justified these minimally coercive measures on consequentialist grounds because they save lives.²⁰ Commentators also draw attention to the social and economic burdens caused by vehicular injuries and deaths (Dworkin, 1986). Various restrictions on drivers' freedoms are criticized because they reduce mobility, which is seen by the public as an important value. Indeed, mobility interests have prevailed over safety interests as speed limits have been increased in many states. As a result, vehicular fatalities have correspondingly increased (Farmer et al., 1997; NHTSA,1998b). Concerns about freedom and mobility have also been expressed about other effective strategies such as license restrictions on young persons (e.g., nighttime driving restrictions). Finally, and importantly, some personal-level interventions raise issues of fundamental fairness. For example, minority communities point to selective and discriminatory enforcement of primary enforcement laws. The communities' concerns are that police will use these laws to intrude on civil liberties, particularly of young persons and minorities.

Regulatory strategies have gone well beyond individual-level interventions. Legislation enacted in 1966 enabled the federal government to set safety standards for new vehicles²¹ and to develop a coordinated national highway safety program.²² The Highway Safety Act of 1970 charged NHTSA with the responsibility of reducing deaths, injuries, and economic losses resulting from motor vehicle crashes. At the state level, government has established programs for motor vehicle inspections, speed limits, road design, and highway maintenance. Federal and state agencies, moreover, fund research, which has demonstrated the effectiveness of multiple interventions on the vehicular and environmental level—for example, center high-mounted rear brake lights, improved tire and brake performance, breakaway sign and light poles, and protective guardrails (TRB, 1990; IOM, 1999).

Legal and regulatory interventions represent critically important components of societal strategies to prevent injury and disease and promote the public's health. Yet, regulation remains substantially underevaluated. There are many reasons for the absence of systematic evaluation, including lack of training

and funding, difficulty in timing research strategies, and most importantly, the many methodological difficulties and confounding variables that are inherent in such research. Despite the challenges, rigorous research evaluations are critically important. Since society spends so much of its resources on regulation, policy makers need to better understand what works and at what cost. Arguably, the increasingly close working relationship among policy makers, industry, and researchers to improve traffic safety is a good model to emulate. Traffic safety interventions, despite concerns about mobility and civil liberties, have been broadly supported by consumers and the public. Careful and systematic research, moreover, has bolstered political and public support.

There have been many areas affecting the public health that have not had the same research base, industry cooperation, and public and political support. Many public health experts believe that regulation should be included within a broad, multidimensional strategy to reduce injuries and diseases. Yet we live in a society that is apathetic, and sometimes hostile, to government regulation. In order to assure political and public acceptance of regulation in America's future, interventions should be carefully and systematically evaluated; and the results should be made widely available to policy makers and the public. The best antidote to lack of political support for government regulation is to show that it works, at a reasonable cost, and with socially and constitutionally acceptable burdens on liberty and property.

Abt. 1997. Independent Evaluation of the Massachusetts Tobacco Control Program . Cambridge, MA: Abt Associates.

Agran P.P., Dunkle D.E., Winn D.G. 1987. Effects of legislation on motor vehicle injuries to children. American Journal of Diseases of Children 141:959–964.

AMA (American Medical Association), American Pharmaceutical Association, National Alliance of State and Territorial AIDS Directors, et al. 1999. HIV Prevention and Access to Sterile Syringes. Chicago. AMA.

Bal D.G., Kizer KW, Felton PG, et al. 1990. Reducing tobacco consumption in California: Development of a statewide antitobacco use campaign. Journal of the American Medical Association 264:1570–1574.

Bernstein E., Pathak D, Rutledge L, Demarest G. 1989. New Mexico safely restraint law: Changing patterns of motor vehicle injury, severity, and cost. American Journal of Emergency Medicine 7(3):271–277.

Bonnie R.. 1986. The efficacy of law as a paternalistic instrument. In: Melton G, ed. Nebraska Symposium on Human Motivation, 1985: The Law as a Behavioral Instrument. Lincoln, NE: University of Nebraska Press. Pp. 131–211.

Burris S. et al. 1996. The legal strategies used in operating syringe exchange programs in the United States. American Journal of Public Health 86:1161–1165.

Case P., Meehan T, Jones TS. 1998. Arrests and incarceration of injection drug users forsyringe possession in Massachusetts: Implications for HIV prevention. Journal of Acquired Immune Deficiency Syndromes and Retrovirology 18(Suppl. 1):S71–S75.

CDC (Centers for Disease Control and Prevention). 1998. Alcohol involvement in fatal motor-vehicle crashes—United States, 1996–1997. Morbidity and Mortality Weekly Report 47:1055–1056, 1063.

CDC (Centers for Disease Control and Prevention). 1999a. Decline in cigarette consumption following implementation of a comprehensive tobacco prevention and education program—Oregon, 1996–1998. Morbidity and Mortality Weekly Report 48:140–143.

CDC (Centers for Disease Control and Prevention). 1999b. Motor vehicle safety: A 20th century public health achievement. Morbidity and Mortality Weekly Report 48(18): 369–374.

CDC (Centers for Disease Control and Prevention). 2000. Tobacco use among middle and high school students—United States, 1999. Morbidity and Mortality Weekly Report 49:49–53.

Chapman S, Borland R, Scollo M, et al. 1999. The impact of smoke-free workplaces on declining cigarette consumption in Australia and the United States. American Journal of Public Health 89(7):1018–1023.

Cole P. 1995. The moral bases for public health interventions. Epidemiology 6(1):78–83.

Cook P.J., Tauchen G. 1984. The effect of minimum drinking age legislation on youthful auto fatalities . Journal of Legal Studies 13:169–190.

Dannenberg A.L., Gielen AC, Beilenson PL, et al. 1993. Bicycle helmet laws and educational campaigns: An evaluation of strategies to increase children's helmet use. American Journal of Public Health 83(5):667–674.

Daynard R., Kelder G., Jr., 1998. The many virtues of tobacco litigation. Trial 34:36–48.

DEA (U.S. Department of Justice, Drug Enforcement Administration) . 1991. Controlled Substances Act: As Amended to July 1, 1991. Washington DC: Department of Justice.

Decker, M.D., Graitcer P.L., Schaffher W. 1988. Reduction in motor vehicle fatalities associated with an increase in the minimum drinking age. Journal of the American Medical Association 260(24):3604–3610.

DeJong W, Hingson R. 1998. Strategies to reduce driving under the influence of alcohol. Annual Review of Public Health 19:359–378.

DHHS (Department of Health and Human Services), 1991. Health Status of Minorities and Low Income Groups. Washington DC: U.S. Government Printing Office.

DiFranza J.R., Brown L.J. 1992. The Tobacco Institute's “It's the Law” campaign: Has it halted illegal sales of tobacco to children? American Journal of Public Health 82(9): 1271–1273.

Dworkin G. 1986. Paternalism. In: Feinberg J, Gross H, eds. Philosophy of Law. Belmont, CA: Wadsworth Publishing Co., 3rd edition.

Elliott S. 1996. Liquor industry ends its ad ban in broadcasting. N.Y. Times, November 8, A1.

Erdmann T.C., Feldman KW, Rivara FP, et al. 1991. Tap water burn prevention: The effect of legislation. Pediatrics 88(3):572–577.

Farmer C.M., Rettig R.A., Lund A.K. 1997. Effect of 1996 Speed Limit Changes on Motor Vehicle Occupant Fatalities. Arlington, VA: Insurance Institute for Highway Safety.

Fingerhut L.A., Warner M. 1997. Injury Chartbook. Health, United States, 1996–97. Hyattsville, MD: National Center for Health Statistics.

Flay B.R., Cook T..D. 1990. Evaluation of mass media prevention campaigns. In: Rice R.E., Atkin C.K., Eds. Public Communication Campaigns. Newbury park, CA: Sage Publications.

Forster J.L., Hourigan M.E., Kelder S. 1992. Locking devices on cigarette vending machines: Evaluation of a city ordinance. American Journal of Public Health 82(9): 1217–1219.

Forster J.L., Murray D.M., Wolfson M. 1998. The effects of community policies to reduce youth access to tobacco . American Journal of Public Health 88(8):1193–1198.

Foss R,D, Evenson K.R. 1999. Effectiveness of graduated driver licensing in reducing motor vehicle crashes. American Journal of Preventive Medicine 16(1 Suppl.):47–56.

Fox D.M., Schaffer D.C. 1987. Tax policy as social policy: Cafeteria plans, 1978A–1985. Journal of Health Politics Policy and Law 12:609–621.

GAO (General Accounting Office). 1987. Drinking-Age Laws: An Evaluation Synthesis of Their Impact on Highway Safety. Washington, DC:GAO.PEMD-87–10.

GAO (General Accounting Office). 1993. Needle Exchange Programs: Research Suggests Promise as an AIDS Prevention Strategy. Washington, DC:GAO/HRD-93–60.

Goldstein LA, Spurlock CW. 1998. Kentucky's child restraint law has saved lives: A 20-year review of fatalities among children (aged 0–4) as motor vehicle occupants. Kentucky Medical Journal 96:97–100.

Gostin L.O. 1998. The legal environment impeding access to sterile syringes and needles: The conflict between law enforcement and public health. Journal of Acquired Immune Deficiency Syndromes and Human Retrovirology 18 (Suppl. 1):S60–S70.

Gostin L.O. 2000. Public Health Law: Power, Duty, Restraint,. Berkeley, CA: University of California Press.

Gostin L.O., Lazzarini Z. 1997. Prevention of HIV/AIDS among injection drug users: The theory and science of public health and criminal justice approaches to disease prevention. Emory Law Journal 46:587–696.

Gostin L.O., Arno P.S., Brandt A.M. 1997. FDA regulation of tobacco advertising and youth smoking: Historical, social, and constitutional perspectives. Journal of the American Medical Association 277(5):410–418.

Gostin L.O., Brant A.M., Cleary P.D.. 1991. Tobacco liability and public health policy. Journal of the American Medical Association 266:3178–3183.

Graham J.D., 1993. Injuries from traffic crashes: Meeting the challenge. Annual Review of Public Health 14:515–543.

Guerin D., MacKinnon D.P. 1985. An assessment of the California child passenger restraint requirement . American Journal of Public Health 75:142–144.

Haddon W. Jr. 1972. A logical framework for categorizing highway safety phenomena and activity. Journal of Trauma 12(3):193–207.

Haddon W. Jr. 1980. Options for the prevention of motor vehicle crash injury. Israel Journal of Medical Science 16(1):45–65.

Haddon W. Jr., Sussman EA, Klein D. 1964. Accident Research: Methods and Approaches. New York: Harper and Row.

Hagan H, McGough J.P. Thiede H. et al. 1999. Syringe exchange and risk of infection with hepatitis B and viruses . American Journal of Epidemiology 149(3):203–213.

Heimer R. 1998. Syringe exchange programs: Lowering the transmission of syringeborne diseases and beyond. Public Health Reports 113(Suppl. 1):67–74.

Heimer R, Khoshnood K, Bigg D. 1998. Syringe use and reuse: Effects of syringe exchange programs in four cities. Journal of Acquired Immune Deficiency Syndrome and Human Retroviruses 18(Suppl. 1):S37–S44.

Hingson R. 1996. Prevention of drinking and driving. Alcohol Health and Research World 20(4):219–226.

Hingson R, Scotch N, Mangione T, et al. 1983. Impact of legislation raising the legal drinking age in Massachusetts from 18 to 20. American Journal of Public Health 73(2):163–170.

Hingson R, Heeren T, Winter M. 1996. Lowering state legal blood alcohol limits to 0.09%: The effect on fatal motor vehicle crashes. American Journal of Public Health 86(9):1297–1299.

IOM (Institute of Medicine). 1988. The Future of Public Health. Washington DC: National Academy Press.

IOM (Institute of Medicine). 1999. Reducing the Burden of Injury: Advancing Prevention and Treatment. Washington DC: National Academy Press.

IOM (Institute of Medicine). 2000. State Programs Can Reduce Tobacco Use. Washington DC: National Academy Press.

Jacobs J.B.. 1989. Drunk Driving: An American Dilemma. Chicago: University of Chicago Press.

Jacobson P, Warner K.E.. 1999. Litigation and public health policy making: The case of tobacco control . Journal of Health Politics, Policy and Law. 24:769–802.

Jason L.A.,, Berk M., Schnopp-Wyatt, Talbot B. 1999. Effects of enforcement of youth access laws on smoking prevalence . American Journal of Community Psychology 21(2):143–160.

Jones N.E., Pieper C.F., Robertson L.S. 1992. The effect of legal drinking age on fatal injuries of adolescents and young adults. American Journal of Public Health 82(1): 112–115.

Kairys D. 1998. Legal claims of cities against the manufacturers of handguns. Temple Law Review 71:1–32.

Kraus J.F., Peek C., McArthur D.L., Williams A. 1994. The effect of the 1992 California motorcycle helmet use law on motorcycle crash fatalities and injuries. Journal of the American Medical Association 272(19):1506–1511.

Kraus J.F. 1985. Effectiveness of measures to prevent unintentional deaths of infants and children from suffocation and strangulation. Public Health Report 100(2):231–240.

Lange J.E., Voas R.B. 1998. Nighttime observations of safety belt use: An evaluation of California 's primary law. American Journal of Public Health 88(11):1718–1720.

Lessig L. 1995. The regulation of social meaning. Columbia Law Review 92:943B–1045.

Lestina D.C., Williams A.F., Lund A.K. et al. 1991. Motor vehicle crash injury patterns and the Virginia seat belt law . Journal of the American Medical Association 265(11):1409–1413.

Lund A.K., Williams A.F., Womack K.N. 1991. Motorcycle helmet use in Texas. Public Health Reports 106(5):576–578.

Mann J., Gostin L.O., Gruskin S. et al. 1994. Health and Human Rights, 1. Journal of Health and Human Rights 1:6–22.

Margolis L.H., Wagenaar A.C., Liu W. 1988. The effects of a mandatory child restraint law on injuries requiring hospitalization. American Journal of Diseases of Children 142:1099–1103.

McArthur D.L., Kraus J.F. 1999. The specific deterrence of administrative per se laws in reducing drunk driving recidivism. American Journal of Preventive Medicine 16(1S):68–75.

McKnight A.J. 1996. Server intervention to reduce alcohol-involved traffic crashes. Alcohol Health and Research World 20(4):227–229.

Muller A. 1986. Is the Oklahoma child restraint law effective? [Letter]. American Journal of Public Health 76(10):1251–1252.

National Institutes of Health. 1997. Consensus Development Statement. Interventions to Prevent HIV Risk Behaviors. February 11–13. http://odp.od.nih.gov/consensus.

National Research Council. 1995. Preventing HIV Transmission: The Role of Sterile Needles and Bleach. Normand J, Vlahov D, Moses LE, eds. Washington DC: National Academy Press.

NHTSA (National Highway Traffic Safety Administration). 1996. Traffic Safety Facts, 1996. Alcohol. Washington, DC: NHTSA.

NHTSA (National Highway Traffic Safety Administration). 1998a. Traffic Safety Facts, 1997. Washington, DC: NHTSA.

NHTSA (National Highway Traffic Safety Administration). 1998b. The Effect of Increased Speed Limits in the Post-NMSL Era. Report to Congress. Washington, DC: NHTSA.

NHTSA (National Highway Traffic Safety Administration). 1999. 1995 Youth Fatal Crash and Alcohol Facts. Washington, DC: NHTSA.

Ni H., Sacks J.J., Curtis L., Cieslak P.R., Hedberg K. 1997. Evaluation of a statewidebicycle helmet law via multiple measures of helmet use. Archives of Pediatric and Adolescent Medicine 151:59–65.

Noland M.P., Kryscio R.J., Riggs R.S. et al. 1998. The effectiveness of a tobacco prevention program with adolescents living in a tobacco-producing region. American Journal of Public Health 88(12):1862–1865.

O'Malley P.M., Wagenaar A.C. 1991. Effects of minimum drinking laws on alcohol use, related behaviors, and traffic crash involvement among American youth: 1976–1987. Journal of the Study of Alcohol 52:278–491.

OMB (Office of Management and Budget). 1998. The Budget of the United States Government: Fiscal Year 1999, p. 218. Washington, DC: OMB.

Paone D., Clark J., Shi Q. et al. 1999. Syringe exchange in the United States, 1996: A national profile. American Journal of Public Health 89(1):43–46.

Pierce J.P., Gilpin E.A., Emery S.L. 1998. Has the California tobacco control program reduced smoking? Journal of the American Medical Association 280(1): 893–899.

Preusser D.F., Lund A.K., Williams A.F. et al. 1988. Belt use by high-risk drivers before and after New York's seat belt use law. Accident Analysis and Prevention 20(4): 245–250.

Rigotti N.A., Bourne D., Rosen A. et al. 1992. Workplace compliance with a no-smoking law: A randomized community intervention trial. American Journal of Public Health 82(2):229–235.

Rivara F.P., Grossman D.C., Cummings P. 1997a. Injury prevention. First of two parts. New England Journal of Medicine 337(8):543–548.

Rivara F.P., Grossman D.C., Cummings P. 1997b. Injury prevention. Second of two parts. New England Journal of Medicine 337(9):613–618.

Rivara F.P., Thompson D.C., Cummings P. Effectiveness of primary and secondary enforced seat belt laws. 1999. American Journal of Preventive Medicine 16(1 Suppl.): 30–39.

Rivara F.P., Thompson D.C., Patterson M.Q. Thompson R.S. 1998. Prevention of bicycle-related injuries: Helmets, education, and legislation . Annual Review of Public Health 19:293–318.

Rogers G.B. 1996. The safety effects of child-resistant packaging for oral prescription drugs. Two decades of experience. Journal of the American Medical Association 275(21):1661–1665.

Robinson D.L. 1996. Head injuries and bicycle helmet laws. Accidental Analysis and Prevention 28(4):463–475.

Roeper P.J.Voas. 1999. Underage drivers are separating drinking from driving. American Journal of Public Health 89(5):755–757.

Rose-Ackerman S. 1991. Tort law in the regulatory state. In: Schuck, ed. Tort Law and the Public Interest. New York: Norton. Pp. 105–126.

Ross H.L. 1993. Punishment as a factor is preventing alcohol-related accidents. Addiction 88:997–1002. Segui-Gomez M. 1998. Evaluating interventions that promote the use of rear seats for children . American Journal of Preventive Medicine 16(1S):23–29. Singer M., Baer H.A., Scott G. et al. 1998. Pharmacy access to syringes among injecting drug users: Follow-up findings from Hartford, Connecticut. Public Health Reports 113(Suppl. 1):81–89.

Thompson D.C., Nunn M.E., Thompson R.S., Rivara F.P., 1996a. Effectiveness of bicycle safety helmets in preventing serious facial injury. Journal of the American Medical Association 276(24):1974–1975. Thompson D.C., Rivara F.P., Thompson R.S. 1996b. Effectiveness of bicycle safety helmets in preventing head injuries: A case-control study. Journal of the American Medical Association 276(24):1968–1973. Thyer BA, Landis J. 1988. Is the Florida safety belt law effective? [letter]. American Journal of Public Health 78(3):323.

TRB (Transportation Research Board, National Research Council) 1990. Safety Research for a Changing Highway Environment. Washington, DC:TRB.

Ulmer R., Preusser C., Preusser D.F. 1994. Final Report (December 1992–December 1993): Evaluation of California's Safety Belt Law Change to Primary Enforcement. Washington, DC: NHTSA.

Vernick J.S., Teret S.P., Webster D.W. 1997. Regulating firearm advertisements that promise home protection. Journal of the American Medical Association 277:1391–1396.

Vicusi W.K. 1992. Fatal Tradeoffs. New York: Oxford University Press.

Vlahov D., Junge B. 1998. The role of needle exchange programs in HIV prevention. Public Health Reports 113(Suppl. 1):75–80.

Wagenaar A.C. 1993. Research affects public policy: The case of legal drinking age in the United States. Addiction 88(Suppl.):758–818.

Wells-Parker E., Bangert-Drowns R., McMillen R., Williams M. 1995. Final results from a meta-analysis of remedial interventions with drink/drive offenders. Addiction 90: 907–926.

Wildey M.B., Woodruff S.I., Agro A. 1995. Sustained effects of educating retailers to reduce cigarette sales to minors. Public Health Reports 110:625–629.

Williams A.F., Preusser DF. 1997. Night driving restrictions for youthful drivers: A literature review and commentary. Journal of Public Health Policy 18(3):334–345.

Williams A.F., Reinfurt D. Wells J.K. 1996. Increasing seat belt use in North Carolina. Journal of Safety Research 27(1):33–41.

Zwerling C., Jones M.P. 1999. Evaluation of the effectiveness of low blood alcohol concentration laws for younger drivers. American Journal of Preventive Medicine 16(1S):76–80.