The public health research literature has largely ignored the role of alcohol service laws in reducing problems related to alcohol intoxication. What little research is available strongly suggests that: (a) there is an association between serving practices and the over-consumption of alcohol, and (b) intervention in support of improved serving practices and the enforcement of laws governing these practices is associated with a decrease in alcohol-related harm.

Research of Happy Hour and Other Drink Specials Practices:
Previous research demonstrates that alcohol consumption, intoxication, and drinking/driving rates are sensitive to the price of alcoholic beverages (Chaloupka, et al., 2002). Underage people and young adults are particularly affected by the cost of alcohol. Studies show that increases in the price of alcohol significantly reduce the number of drinks consumed by this population (Grossman, et al., 1998; Chaloupka, et al., 2002). Happy hours, drinking contests, "all you can drink" specials, and the like encourage over-consumption by reducing prices, a potent inducement to drinking large amounts of alcohol in short time periods. The research offers strong evidence for the negative health outcomes of happy hour and other drink specials practices, thereby suggesting that policies restricting these practices could have a positive impact on public health.

In the 1970s, an experimental study was conducted in which a small group of subjects was tested in a clinical setting (Babor, et al., 1978). The subjects were divided into experimental and control groups. The experimental group was given a 50 percent price reduction for alcoholic beverages during a daily three-hour period in the afternoon, and the control group was offered drinks at full price. A significant increase in consumption was observed among both casual and heavy drinkers in the experimental group, with consumption returning to normal when happy hour price reductions were discontinued. Casual and heavy drinkers in the happy hour group drank about twice as much as those in the non-happy hour group.

Using data collected in 2001 by the Harvard School of Public Health College Alcohol Study, researchers examined the relationship between binge-drinking rates on college campuses and the availability of large volumes of alcohol, low sales prices, and frequent promotions and advertisements in the vicinity of campus (Kuo, et al., 2003). Binge-drinking rates for 119 colleges were determined using college students’ self-reports on alcohol consumption. An assessment study of the alcohol environment surrounding each campus was conducted, which included the monitoring of on-site premises for serving sizes, prices, promotions, and so forth. The results demonstrated a significant correlation between lower drink prices and higher binge-drinking rates. The presence of weekend beer specials and alcohol promotions was also highly correlated with a higher binge-drinking rate. This same study demonstrated a reduction in self-reported drinking-and-driving rates when laws limited underage access to alcohol and high volume sales of alcohol (Wechsler, et al., 2003). The presence of these laws was associated with lower rates of drinking-and-driving among college students, a group at risk for both binge drinking and alcohol-related traffic fatalities (Wechsler, et al., 2003; NHTSA, 2002; NHTSA, 2004). This effect was enhanced when there was active enforcement of laws limiting underage access to alcohol and high-volume sales of alcohol.

Only one study has attempted to directly evaluate the efficacy of happy hour laws in lowering alcohol consumption. The banning of happy hour practices in Ontario, Canada, was studied by observation of drinking habits before and after the ban, supplemented with analyses of total per capita consumption in the city (Smart and Adlaf, 1986; Smart, 1996). No significant decline in alcohol consumption was observed following the ban. Given that there was little time (two days) allotted to observing pre-ban drinking habits, and given that aggregate consumption figures may not be that sensitive to changes in happy hour practices, the results were inconclusive as far as the overall effect on alcohol consumption of the presence or absence of happy hour practices.

Research on the Prohibition of Sales and Service of Alcohol to Intoxicated People:
Nearly every State prohibits sales and service of alcohol to obviously intoxicated people. Little research is available to determine how these laws are enforced, the extent with which they are complied with, and the impact enforcement and compliance might have on public health outcomes. One study that examined compliance rates found that 79 percent of alcohol establishments will serve alcohol to patrons who appear obviously intoxicated (Toomey, et al., 1999; Toomey, et al., 2004). Actors simulating intoxication attempted to make alcohol purchases at both on- and off-premise establishments in 11 communities in a large Midwestern metropolitan area. Seventy-six percent of the on-premise sites sold alcohol to the pseudo-intoxicated patrons, as did 83 percent of the off-premise establishments. The high non-compliance rates highlight the importance of further research into the effectiveness of enforcement of laws prohibiting sales to intoxicated people or other interventions designed to reduce over-consumption and subsequent driving.

A study that directly examined enforcement of these laws was conducted in Washtenaw County, Michigan. Compliance was observed before and after the implementation of a publicized campaign to enforce laws directed at sales to intoxicated individuals (Edwards, et al., 1994). Enforcement included the issuing of warnings to businesses that violated the law, followed by enforcement visits and citations. These actions were conducted in conjunction with education and training of bar and restaurant staff. Compliance with the law was measured before and during the enforcement program by the rate at which patrons simulating intoxication were refused service. Refusals of service rose from 17.5 percent before the enforcement program to a peak of 54.3 percent after the first three months of the enforcement intervention. Significantly, the percentage of impaired drivers arrested after leaving bars and restaurants declined from 31.7 percent to 23.3 percent during the same period. While refusals of service to pseudo-intoxicated people declined from the initial peak of 54.3 percent to 47.4 percent after six months, and 41.0 percent after one year of the program, these later refusal rates remained significantly higher than the baseline, indicating that the intervention had an enduring effect on server compliance with no-sale-to-intoxicated laws (McKnight and Streff, 1994).

A few studies have examined dram shop liability laws, which hold alcohol servers responsible for harm caused by intoxicated or underage patrons, another avenue toward curbing over-intoxication². Studies indicate that enforcement and prosecution of dram shop laws (and resulting case decisions) are associated with a substantial reduction in alcohol-related harm. The initiation of a dram shop liability lawsuit in Texas in 1983 resulted in 6.5 percent fewer single-vehicle nighttime injury crashes (which are associated with high percentages of alcohol involvement). After a second suit was filed the following year, an additional 5.3 percent decrease in such crashes resulted (Wagenaar and Holder, 1991). One study found that in States where servers have a relatively high level of exposure to liability, there are fewer low-price drink promotions and more servers check identifications for underage purchases. Both of these changes in serving practices can prevent alcohol-related traffic crashes (Holder, et al., 1993).

A study evaluating the effects of the Alcohol Risk Management (ARM) program highlights the benefits of promoting voluntary compliance with over-service laws. The ARM program is a five-session one-on-one consultation program for owners and managers of on-site alcohol outlets. The purpose of the program is to help owners and managers develop policies and practices that increase compliance with State alcohol laws and reduce the risk of alcohol-related problems (Toomey, et al., 2001). Pseudo-intoxicated and underage patrons were sent to test sites and control sites before and after the training sessions to determine the efficacy of the educational efforts at the intervention outlets. Underage sales declined by 11.5 percent after the intervention at the test sites, while sales to the pseudo-intoxicated patrons declined by 46 percent.

Another recent finding of a strong relationship between enforcement efforts and reduced alcohol-related harm comes from New South Wales, Australia (Wiggers, et al., 2001). The Linking Project was a collaborative effort between researchers and law enforcement officers. Law enforcement officers in selected districts of New South Wales identified licensed establishments that were listed as “last place of drink” by people apprehended for alcohol-related incidents (including drinking and driving, assault, domestic violence, and other criminal activities). A random sample of these establishments was chosen for intervention, which included giving the licensees feedback on incidents associated with their establishments, conducting audits of responsible alcohol service practices, and the offer of resources and assistance to the licensees for improvement in their service practices. Following these interventions, a follow-up workshop on responsible alcohol service was conducted. The number of alcohol-related incidents associated with the intervention sites decreased by 36 percent following the intervention actions, compared to a 21 percent decline for a control group of establishments. The results were a clear demonstration that enforcement efforts focused on alcohol serving practices could have a much larger impact on reducing crime, and therefore benefit the public and reduce the burden on law enforcement. These impressive findings have convinced political leaders and law enforcement agencies to expand the Linking Project to the entire enforcement system of New South Wales.

Enforcement and Adjudication of State Laws Restricting Over-consumption of Alcohol
Methodology:
As the research reviewed in the previous section suggests, improving the rates of compliance with alcohol service laws is a promising strategy for reducing alcohol-related harm, including drinking and driving. However, there is little information regarding the nature of these laws or current enforcement practices. To address this gap in the research literature and to promote additional studies of the topic, NHTSA funded this project to collect detailed information about the statutes governing sales to intoxicated people and happy hour practices, and their enforcement and adjudication in the United States. The following methods were employed to collect this data.

• Legal research: Attorneys and staff working on the project reviewed State alcohol beverage control statutes and regulations to determine those statutes and regulations governing sales to intoxicated people and happy hour practices in 51 jurisdictions, including all 50 States and the District of Columbia. This legal research, current as of January 1, 2003, was completed using national legal databases and a variety of secondary sources. The research focused exclusively on existing State statutes and did not include local regulations.
  
  
• Interviews with alcohol enforcement representatives³: To confirm the data collected in the legal research phase, Alcohol Beverage Control representatives from 45 States and the District of Columbia were interviewed about 12 key alcohol policies, including sales to intoxicated and happy hour regulations⁴. Each interview also included questions about penalties, licensing systems, enforcement resources and strategies, data collection processes, and the adjudication of alcohol violations.
  
  
• Additional interviews and research about enforcement strategies: In 2003, additional research was conducted about promising State and local strategies employed for the enforcement of sales to intoxicated and happy hour laws. Representatives from the appropriate agencies (identified during the initial interview process) were interviewed and any available reports or documentation of the programs were collected.

Current Statutes and Regulations:
Nearly every State and the District of Columbia has a provision prohibiting sales to intoxicated people, and over one-half (27) of the States have laws that specifically prohibit happy hours, drink specials and other practices that encourage drinking to intoxication.

Sales to Intoxicated Statutes: A review of the statutes prohibiting sales and service of alcohol to intoxicated people revealed that 47 States and the District of Columbia have such laws as of January 1, 2003 (PIRE, 2003). Florida, Nevada, and Wyoming do not have comprehensive laws prohibiting sales to intoxicated people. (See Appendix A.) State provisions vary in terms of language used to describe the state of intoxication (e.g., obviously intoxicated; visibly intoxicated; appears to be intoxicated; noticeably intoxicated; reason to believe is intoxicated; apparently under the influence of liquor), as well as that used to describe the provision of alcohol (e.g., serving, selling, furnishing, giving, bartering, exchanging, providing, delivering, and procuring).

Interpretation of these statutes in court may vary, most notably regarding the level of proof required for a finding that the law has been violated. Most statutes state or imply that a violation occurs if the server acted negligently—the server failed to act in a manner expected of a reasonable person in like circumstances. Some statutes use language that could be interpreted to require proof that the server knew the person being served was intoxicated or was reckless rather than merely negligent in his or her actions. These are higher standards of proof, making findings of violation much more difficult. Requiring proof that the server knew that the patron was intoxicated is particularly difficult to establish since it requires evidence of the server’s state of mind. This requirement is rare in cases determining whether a violation has occurred. Courts have tended to interpret statutes as requiring only the negligent standard even when the language of the statute suggests a higher level of proof. The reckless standard in particular is more common in dram shop liability lawsuits, which involve monetary compensation to those injured by the intoxicated patron. In the case of some older alcohol statutes, strict liability for the sale or service of alcohol to an intoxicated person is imposed; that is, no defense is allowed that encompasses knowledge or recognition of the signs of intoxication. If the person served is intoxicated, the establishment selling the alcohol to that person is liable, regardless of whether or not they were aware of the person’s intoxication (Moore, 2003). Case law must be carefully reviewed to determine the level of proof issue and even a detailed analysis may result in inconclusive findings. Because of the ambiguities in the law and the difficulty of conducting comprehensive case law research, we have not attempted to report level of proof requirements in our State law review.

Statutes may also vary in terms of who can be held in violation. Most State statutes apply to both commercial and noncommercial servers, although in some cases the statutory language is vague and may be subject to an interpretation that it only applies to service in commercial establishments. Finally, statutes will vary in terms of the types of penalties that may be imposed on violators. For commercial servers, violations may be either civil or criminal. Civil offenses are handled administratively by the agency responsible for adjudicating violations of the Alcoholic Beverage Control (ABC) laws. As discussed below, penalties can include suspensions or revocations of licenses and/or fines. Civil offenses are more easily prosecuted because they are administrative in nature, requiring only that the preponderance of the evidence supports the finding of violation. Both commercial and noncommercial servers can be found criminally liable. Criminal liability suggests moral approbation, is adjudicated in courts of law, and can involve both fines and imprisonment. Because of the more serious consequences, a conviction must be proven by the prosecutor beyond a reasonable doubt, the highest legal standard of proof, and defendants are given the right to a jury trial.

Happy Hour and Drink Specials Statutes: As of January 1, 2003, 27 States had provisions expressly prohibiting excessive drinking practices, or “happy hour” types of promotions. (See Appendix B.) In addition, many communities have passed local ordinances prohibiting these practices. The following information refers to State legislation only.

Drinking practices referred to in these statutes include:

• Free beverages—10 States have happy hour provisions that contain specific prohibitions against the distribution of free alcoholic beverages.
  
  
• Additional servings—16 States prohibit an establishment from providing additional servings of alcoholic beverages until previous servings have been consumed.
  
  
• Reduced price – specified day or time—18 States prohibit the sale of alcoholic beverages at reduced prices during specified days or times.
  
  
• Unlimited beverages – fixed price, fixed time—23 States prohibit the sale of alcoholic beverages during a fixed period of time for a fixed price.
  
  
• Increased volume—12 States prohibit increasing the volume of alcoholic beverages in a drink without increasing the price.
  
  
• Prizes—15 States have happy hour provisions that contain specific prohibitions against giving alcoholic beverages as prizes.

Most of the States prohibiting happy hour practices specifically proscribe two or more of the practices listed above. For example, the Texas State statute specifies 11outlawed practices:

• “Two for one” or other discounted multiple alcoholic beverage sales;
  
  
• Increasing the volume of alcohol in a drink without increasing the price;
  
  
• Serving more than one free alcoholic beverage to any identifiable segment of the population;
  
  
• Fixed-price or “all you can drink” sales;
  
  
• Selling alcoholic beverages at a reduced price for a fixed “buy in” price;
  
  
• Selling alcoholic beverages at a price contingent on the amount consumed by an individual;
  
  
• Reduced drink prices after 11:00 p.m.;
  
  
• Selling more than two drinks to a single consumer at one time;
  
  
• Imposing an entry fee for the purpose of recovering financial losses incurred because of reduced drink prices;
  
  
• Drinking contests or awarding of alcoholic beverages as prizes;
  
  
• Any practice that is reasonably calculated to induce consumers to drink to excess, or that would impair the ability of the licensee to monitor or control the consumption of alcohol by their customers⁵.

Enforcement and Adjudication of Sales to Intoxicated and Happy Hour Statutes:
Limited information about the enforcement and adjudication of laws restricting the over-consumption of alcohol was collected during interviews with State alcohol enforcement officials. Reductions in budgets, decreasing available personnel, the absence of public and governmental support, and difficulties coordinating efforts with local law enforcement are some of the problems that affect enforcement of over-consumption policies. The representatives also reported that enforcement is hampered by the difficulties of proving that the patron being served was obviously intoxicated. Gathering such evidence usually involves undercover operations, which are both costly and time intensive.

The most commonly reported enforcement strategy (reported by 12 States) was the use of surveillance and undercover agents to identify violations of sales to intoxicated and drink specials laws. In many States, investigations are primarily complaint-driven. A few State agencies identified walk-through inspections as their primary method of identifying violations and enforcing these laws.

A promising strategy that is being implemented by some States (e.g., Massachusetts, Oregon, Utah, and Washington) involves identifying the place of last drink for those arrested on driving under the influence/driving while intoxicated (DUI/DWI) charges. The collection of this data allows States to identify and target problem outlets that may be in violation of laws prohibiting sales to intoxicated people or drink specials that encourage over-consumption of alcohol.

The imposition of penalties for violations of the law is an integral part of the enforcement process and can play an important role in deterring future violations. As discussed above, laws addressing sales to intoxicated people and happy hour and drink specials can be adjudicated through administrative proceedings and can lead to fines and license suspensions and revocations.

Researchers examined the penalties specific to sales to intoxicated laws. The States vary widely in the range and severity of administrative penalties imposed for violations of sales to intoxicated laws. Most States increase the severity of the punishment as the number of offenses committed increase. There may be an increase in number of days of suspension, revocation may become more likely, and fines may increase. The suspension of a license is included as a potential maximum penalty for a first-time offense in most States. At least 36 States and the District of Columbia allow for the revocation of a liquor license as a potential maximum penalty for a first-time offense. However, interviews with enforcement officials reveal that revocations rarely occur, and are highly unlikely after a first offense. In a smaller number of States (seven, plus the District of Columbia), license revocation is the only allowable penalty for a licensee convicted of a fourth violation of sales to intoxicated laws. In at least 10 States, fines can be paid in lieu of license suspension, primarily for first offenses only. For an expanded chart of penalties imposed for sales to intoxicated violations by State, please see Appendix C.

Promising Enforcement Strategies:
Interviews were conducted with representatives from State and local law enforcement agencies regarding innovative programs to enforce sales to intoxicated people and happy hour statutes. The following examples illustrate some of the enforcement strategies being employed by these agencies in their efforts to reduce alcohol-related harm.

Enforcing Sales to Intoxicated Laws by Identifying “Place of Last Drink”: As indicated above, both Washington and Utah are using the “place of last drink” strategy, sometimes in conjunction with other enforcement efforts, to reduce sales to intoxicated people.

Washington: In 2002, the Washington State Liquor Control Board (WSLCB) launched an enforcement program with the goals of reducing the number of DUI arrests, and reducing the average blood alcohol concentration (BAC) level of those arrested. Despite the fact that Washington lowered the maximum allowable BAC level to 0.08 in 1999, high BAC levels continued to be involved in fatal traffic crashes. In 79 percent of fatal crashes, the BAC level of the driver at fault was over 0.15; 52 percent of these exceeded 0.20 (WSLCB, 2003).

In response, the WSLCB is conducting a monthly analysis of DUI arrest reports supplied by the Washington State Patrol, which include “last drink” locations as well as the BAC levels of arrested drivers. The Washington State Patrol administers the BAC Datamaster database, which contains every breath test administered statewide by every law enforcement agency. Law enforcement officers are required by Washington law⁶ to administer a breath test to individuals arrested for driving or being in physical control of a vehicle while under the influence. The location of the individual’s last drink is entered into a BAC Datamaster machine while the breath test is administered. All BAC test data and accompanying information (such as place of last drink) are uploaded to the State Patrol’s database⁷.

This comprehensive database provides the necessary information for the WSLCB to create a “worst offenders” list of establishments. These establishments are associated with the highest number of DUIs or highest BAC readings recorded among DUI arrestees. The board then executes a plan that begins with notifying the establishments that they have a high number of DUI arrestees who identified their establishment as the place where they had their last drink. The corrective plan includes educating the licensee and their employees in training sessions about responsible beverage service, signs of intoxication, and laws governing sales to intoxicated patrons. Routine premise checks and undercover operations are increased to monitor the establishment’s progress and to maintain an enforcement presence. If necessary, corrective actions, ranging from notification of infraction to administrative or criminal actions, are taken. The progress of each targeted location is monitored and evaluated. If their DUI count increases or remains the same, the enforcement plan is continued. If the count has decreased, the enforcement focus shifts to the next worst offender, but routine premise checks continue.

Data collected to date has not only identified problem establishments, but has also provided valuable information about the relationships between type of licensee, DUI arrests, and average BAC levels. In addition to identifying specific problem licensees, the data provides enforcement agencies with an overview from which to plan the allocation of their resources and develop enforcement strategies. Analysis of the data is now in progress.

Utah: The Liquor Enforcement Section of the Utah Department of Public Safety conducts a statewide program called SIP (Serving Intoxicated Patrons) to enforce laws prohibiting sales to intoxicated people. SIP has targeted those establishments either identified by local law enforcement as problem locations, or those listed as place of last drink for individuals involved in traffic fatalities. Employing covert agents who observe the establishment’s serving practices, SIP operations result in referrals to the State Alcohol Beverage Control agency when violations are observed. Licensees referred for disciplinary action are offered the opportunity to attend training sessions that review relevant laws and teach attendees how to identify signs of intoxication.

SIP operations will soon be enhanced by a statewide program to identify “place of last drink” for all DUI cases, not just those involving fatalities. Funded by a grant from the Utah Office of Highway Safety, DUI data gathered from drivers’ license data will be used to identify problem outlets throughout the State. A pilot study recently completed in Salt Lake County collected and analyzed place of last drink data from DUI arrestees, demonstrating the viability of this strategy for identifying problem outlets. The SIP program will use the statewide data to target establishments for SIP interventions, and will track DUI data before and after SIP interventions to evaluate the program’s effectiveness (Michaud, 2003).

Enforcing Happy Hour and Drink Specials Laws Through Observation, Surveillance, and Undercover Operations: A number of strategies may be employed to reduce the over-consumption of alcohol by enforcing happy hour and drink specials laws. In the following examples, emphasis was placed on identifying violations through observation, surveillance, and other undercover operations, in coordination with other concentrated enforcement efforts.

Champaign, Illinois: The Champaign, Illinois, Alcohol Enforcement Unit has conducted a successful campaign against over-service and happy hour practices as part of a larger effort to reduce underage drinking and alcohol-related harm (CPD, 2002; Friedlein, 2003). Champaign and its twin city, Urbana, share a large student population from the University of Illinois (attended by some 38,000 students) and a community college (UIUC, 2003). In conjunction with its focus on underage drinking, the unit has given special emphasis to those bars engaging in such practices as drinking contests, reducing the price of drinks at certain times of the day, and other promotions that encourage excessive drinking, particularly among younger patrons. Bar advertisements are reviewed on a daily basis and the unit conducts follow-up, observational visits to identify possible violations. If a violation is observed, the unit may notify the owners of the bar of a need to correct the problem or proceed to other actions, depending on the severity of the violation and past history of the establishment. If the problem is not corrected after notification, the unit initiates an undercover investigation that can conclude with enforcement actions if violations are observed. The enforcement program is part of a broader, community policing strategy that includes making regular visits to drinking establishments and building a cooperative relationship between law enforcement and the licensed alcohol establishment community.

To enhance its current program, the Unit is considering the implementation of a new local policy to reduce the allowable alcoholic beverage serving size. Some establishments serve mixed drinks in 48-ounce personal containers, allowing individuals to purchase a large volume of alcohol in one serving. This reduces the server’s ability to gauge a patron’s intoxication level and regulate the number of drinks served. Should this policy be put into practice, training for licensees and their employees would be integrated into the existing beverage service training offered by the Unit. This represents one more method in Champaign’s multiple strategy approach, which combines observation, undercover work, direct enforcement, community policing, training, and altering serving practices.

Texas: The Texas Alcoholic Beverage Commission (TABC) Enforcement Division collaborated with researchers from the Pacific Institute for Research and Evaluation to field-test a model enforcement program, funded by a contract from NHTSA⁸. The project was designed to assess the impact of proactive enforcement of laws restricting serving practices that encourage intoxication on compliance rates among commercial alcohol servers. As noted above, Texas has a comprehensive set of regulations restricting such serving practices, including prohibitions against serving pitchers to individuals and discounting the cost of drinks when served in double portions (prohibited by the statute against increasing the volume of alcohol in a drink without proportionately increasing the price).

The TABC identified 50 high-volume sales-on-premises establishments in two counties (a total of 100 establishments). TABC enforcement officers conducted a sales test (purchase survey) at all 100 sites to collect information on current serving practices. Officers entered the establishments and attempted to purchase a pitcher of beer, or a single and a double shot of spirits to determine if the server was complying with the relevant regulations regarding these serving practices. Focused and concentrated serving practice enforcement (including sending letters to all on-premise establishments in that county to notify them of increased enforcement actions) was implemented in both counties during separate time intervals. A final sales test was conducted at all 100 outlets to determine whether any changes in serving practices continued over time.

Preliminary analysis of the data indicates that this relatively modest intervention resulted in significant reductions in violation rates in both counties. In one county the number of violations fell by 100 percent at the end of four months of enforcement intervention. Even after a two-month interval of no intervention, the final sales test revealed that violations were still reduced by 100 percent. In the second county, the number of serving practice violations dropped 68 percent between the first and final sales tests. Since some on-premise alcohol outlets received multiple violations during one visit, the reduction in the number of outlets found in violation was also examined. Not surprisingly, the number of outlets found in violation dropped significantly in both counties—by 100 percent and 63 percent, respectively.

²A "dram" refers to a unit of liquid measure used during colonial times in the United States. "Dram shops" refer to the establishments that served alcohol by the dram (Holder, et al., 1993).

³These interviews were conducted between September 2001 and December 2002.

⁴In the remaining five States, the researchers were either not able to locate a State agency that held the primary responsibility for enforcing alcohol laws, or the agency identified declined to participate in the interview. In addition, the enforcement of alcohol laws in Hawaii and Maryland is primarily conducted at the county level. In those States, interviews were conducted with representatives from one county, and therefore, the results are not applicable to the entire State.

⁵Texas statute §45.103. On-Premises Promotions

⁶Revised Code of Washington (RCW) 46.20.308. Note that arrested people have the right to refuse the breath test, but they face revocation of their driver’s licenses if they do so.

⁷Data is stored in each BAC Datamaster machine until the machine is polled or automatically sends its stored information to the central database. Note that place of last drink data is recorded and stored even if the individual refuses the breath test.

⁸NHTSA contract DTNH22-03-H-05134.