Abstract
Overservice of alcohol, defined as commercial provision of alcohol to an individual who is obviously intoxicated, is illegal in most states and contributes to motor vehicle crashes and violence. Law enforcement agencies use various strategies that aim to reduce overservice at licensed alcohol establishments (e.g., bars, restaurants). Place of Last Drink (POLD) data collection is an emerging overservice enforcement strategy. POLD identifies patterns of overservice, which can provide support for targeted interventions to prevent overservice at offending establishments. We describe the prevalence of POLD and other overservice enforcement strategies and associations with agency characteristics, which has important implications for public health and safety. We conducted a national survey of 1024 municipal (e.g., town, city) and county law enforcement agencies in 2019 (response rate = 73%). We assessed the use of overservice enforcement strategies conducted by the agencies over the past year. We examined associations of each type of overservice enforcement strategy with agency and jurisdiction characteristics using regression models. Twenty-seven percent of responding agencies reported conducting overservice enforcement and 7% conducted POLD data collection specifically. Municipal (vs. county) agencies and agencies with an officer assigned primarily to alcohol enforcement activities were significantly more likely to conduct overservice enforcement generally but not POLD data collection specifically. Overservice enforcement in general, and POLD data collection specifically, are not widely conducted. Prevention of overservice has the potential to reduce harms related to excessive alcohol consumption. Increased evaluation of overservice enforcement strategies should be prioritized.
Keywords: Alcohol, Overservice, Prevention, Enforcement
INTRODUCTION
The provision of alcohol to an obviously intoxicated person (i.e., overservice) contributes to various social and health problems. Overservice facilitates excessive alcohol consumption at on-premise alcohol establishments (such as bars, clubs, and restaurants), and leads to motor vehicle crashes, falls, and violence [1–3]. Overservice is illegal in 48 of 50 U.S. states [4], yet these laws are seldom enforced [5] and overservice of alcohol is common. A study using pseudo-intoxicated patrons found that 82% successfully purchased alcohol from licensed establishments [6].
Alcohol overservice has been addressed primarily through responsible beverage service (RBS) training of servers and bartenders to recognize signs of intoxication and employ strategies to stop service to obviously intoxicated customers [7–8]. The success of RBS training in reducing overservice is mixed [9]. Another strategy to reduce overservice is dram shop liability laws, which provide a mechanism by which alcohol servers and establishments can be held liable for harms inflicted by a customer that was overserved in their establishment. However, a high burden of proof and the complexity of the litigation process make dram shop laws challenging to enforce [10]. Additional law enforcement strategies to address overservice include random inspections, walk-throughs, observations at alcohol establishments to identify and prevent overservice, and pseudo-intoxicated purchase attempts to determine the likelihood of overservice in their communities [6]. The effectiveness of these strategies has not been rigorously evaluated.
Place of Last Drink (POLD) is a relatively new overservice enforcement strategy. When a law enforcement officer encounters an incident where alcohol consumption is suspected, the officer inquires where the persons involved were last drinking and systematically records this information. Systematically collected POLD information can be analyzed to identify licensed alcohol establishments that demonstrate a pattern of overservice. By identifying establishments that are frequent places of last drink through systematic data collection, law enforcement agencies can concentrate their resources with specific enforcement strategies at the establishments that are less likely to comply with responsible service practices. Data-driven enforcement can involve compliance-based approaches (e.g., requiring server training, reducing drink specials), alcohol license suspension or revocation, or targeted enforcement such as retailer observations [11].
Evaluation of POLD as an enforcement strategy is limited and has shown mixed results regarding its effectiveness. A process evaluation of a POLD initiative in Auckland, New Zealand suggests that POLD data can persuade establishment managers to change their serving practices, inform action on liquor license decisions, and motivate targeted intervention [12]. An outcome evaluation of a POLD initiative in Washington State showed a decrease in driving under the influence (DUI) arrests and a lower average blood alcohol content (BAC) among DUI arrestees in intervention communities post-intervention compared to pre-intervention [13]. However, the same study found an increase in the likelihood of service to pseudo-intoxicated patrons in intervention communities post-intervention compared to pre-intervention. A study conducted by Fell and colleagues [7] evaluated a multi-faceted intervention of RBS training, POLD data collection, and additional targeted overservice enforcement (e.g., observations by plainclothes officers). This intervention was associated with a post-intervention increase (compared to pre-intervention levels) in refusal of service to pseudo-intoxicated patrons at one set of intervention sites but not the other. The proportion of customers with a BAC ≥ 0.08 and number of calls for law enforcement response near the establishments showed similarly mixed results among intervention vs. comparison sites.
Although more research is needed to assess the effectiveness of overservice enforcement strategies, more understanding is also needed about how these strategies are currently used as well as attitudes about overservice among law enforcement agencies [10]. In this study we describe the prevalence of strategies used to reduce alcohol overservice using data from a national survey of law enforcement agencies in the United States. Our findings can guide future research and evaluation and ultimately help determine evidence-based strategies to reduce overservice and related harms.
MATERIALS AND METHODS
We conducted a survey among a national sample of municipal-level (e.g., town, city) and county-level law enforcement agencies from March 2019 to January 2020 to assess alcohol-related law enforcement practices. The survey sample was obtained by expanding on a national survey of municipal and county-level law enforcement agencies that we conducted in 2010. For the 2010 survey, the unit of analysis was the state and the sampling frame of agencies was constructed based on: (1) the number of agencies per state, (2) the proportion of large vs. small agencies (based on number of officers), and (3) stratified by local police vs. county sheriff in each state (see [5] for full details on sampling). The sample for the current survey started with all agencies that responded to the 2010 survey. However, to adjust the sample to better represent county-level populations, we added municipal agencies within counties as needed (beginning with the most populous areas) to cover at least 50% of the county population. In this process, we added 487 agencies, for a total of 1452. Of these, 41 were deemed ineligible because they were closed or had merged with another agency or did not conduct primary enforcement. Our final sample was 1,411.
The survey was conducted via email and telephone by the Office of Measurement Services (OMS) at the University of Minnesota. OMS sent an email invitation to all agencies in the sample with a valid email address. OMS sent up to four email reminders to the non-responding agencies for which an email address was available. If an agency did not respond to email reminders or an email address was not available, OMS attempted up to ten calls to complete the survey over the phone. A paper survey was sent to agencies upon request. When contacting agencies by phone or email, the data collector asked to speak with the individual most knowledgeable about enforcement of alcohol laws at the agency. We received a total of 1,024 responses to the survey from law enforcement agencies for a response rate of 73% (1,024/1411). Fifty-four percent completed the survey online. Agencies that didn’t respond to our survey were not significantly different from agencies that did respond in terms of proportion of residents living in poverty (<11% vs. ≥11%) and proportion of Hispanic residents living in the jurisdiction (<18% vs. ≥18%); however, agencies with a higher proportion of Black residents in the jurisdiction (≥ 18% vs. <18%) and sheriff agencies (vs. police) were slightly less likely to respond (p=0.04 for both).
Measures
Measures from the survey include those specific to overservice as well as general agency characteristics. The survey asked all agencies about perceptions of how common alcohol overservice was in their jurisdiction (not common, somewhat common, very common), how large of a priority overservice enforcement was for their agency (very low/low/average, fairly/very high, and “do not have the authority to enforce laws prohibiting sales to obviously intoxicated patrons”) and whether they conducted overservice enforcement (yes/no). Agencies that reported conducting overservice enforcement were asked how often they conducted four types of overservice enforcement at alcohol establishments: (1) random inspections, (2) walk-throughs, (3) observations for overservice, and (4) pseudo-intoxicated purchase attempts, and also whether they collected place of last drink (POLD) data. The response options for the first four types were consolidated to never, once/a few times per year, and weekly/monthly. The response options for POLD were yes vs. no, with those responding yes asked to indicate the type(s) of alcohol-related incidents that prompted recording of POLD data: DUI, domestic violence, disorderly conduct, physical assault, underage consumption, medical, and other. They were also asked whether they used POLD data to focus enforcement efforts and whether they shared POLD data with other jurisdictions (yes/no).
Measures for agency characteristics included type of agency (municipal police vs. county police or sheriff), number of full-time sworn officers per capita, whether the agency had any officers assigned primarily to alcohol enforcement (dichotomized to none vs. 1 or more), and whether the agency had a division dedicated specifically to alcohol enforcement. The number of officers per capita was dichotomized to < 2 full-time sworn officers per 1000, and ≥ 2 full-time sworn officers per 1000.
Statistical Analysis
We calculated descriptive statistics for all measures including agency and jurisdiction characteristics and overservice enforcement items. We estimated generalized linear mixed models (accounting for agencies nested within states) to evaluate associations between each overservice measure and agency characteristics. Analyses were conducted using SAS statistical software version 9.4 (SAS/STAT, Cary NC) and Stata statistical software release 15.1 (StataCorp LLC, College Station, TX), with a significance threshold of p ≤ 0.05.
RESULTS
Of the 1,024 agencies surveyed, 752 (73%) were municipal agencies and 272 (27%) were county agencies. Fifty-nine percent of responding agencies were small (fewer than two officers per 1,000 population). Thirty-one percent of agencies reported having an officer assigned primarily to alcohol enforcement and 14% reported having an alcohol-specific division.
Descriptive statistics for the overservice measures are in Table 1. Of the 1,024 agencies that responded to the survey, about half reported that overservice was at least somewhat common in their jurisdiction, one in eight said overservice was a fairly or very high priority for their agency, and less than one in four reported conducting overservice enforcement. Among agencies conducting overservice enforcement, the majority conducted random inspections (68%), walk-throughs (82%) and observations (65%) at alcohol establishments at least once per year; 30% reported collecting POLD data.
Table 1.
Overservice perceptions/enforcement (n=1024)
| Percentage (n) of agencies |
|
|---|---|
| How common is overservice in your jurisdiction? | |
| Not common | 49 (484) |
| Somewhat common | 42 (411) |
| Very common | 9 (87) |
| How big a priority is overservice enforcement at your agency? | |
| Fairly low/Very low/Average | 88 (873) |
| Fairly high/Very high | 12 (120) |
| Conduct overservice enforcement | |
| Yes | 23 (238) |
| No | 64 (650) |
| Do not have authority | 5 (55) |
| Don’t know | 7 (76) |
| Of those that have authority: | |
| Yes | 27 (238) |
| No | 73 (650) |
| Among 238 agencies reporting overservice enforcement: | |
| Types of overservice enforcement | |
| Random inspections | |
| Never | 32 (77) |
| Once/a few times | 50 (118) |
| Monthly/weekly | 18 (43) |
| Walk-throughs | |
| Never | 18 (42) |
| Once/a few times | 43 (102) |
| Monthly/weekly | 39 (93) |
| Observations for overservice | |
| Never | 35 (84) |
| Once/a few times | 41 (97) |
| Monthly/weekly | 24 (57) |
| Pseudo-intoxicated purchase attempts | |
| Never | 92 (218) |
| Once/a few times | 7 (16) |
| Monthly/weekly | 1 (3) |
| Place of last drink (POLD) | |
| Yes | 30 (71) |
| No | 70 (167) |
Table 2 shows associations between agency characteristics and perceptions/enforcement of overservice among all 1024 agencies. Agencies with an officer assigned to alcohol enforcement, agencies with an alcohol-specific division (compared to other agencies), small (compared to large) agencies, and county agencies (compared to municipal agencies) were more likely to report that overservice is common in their jurisdiction. Large agencies were more likely than small agencies to report that overservice enforcement is a fairly or very high priority in their agency. Municipal (compared to county) agencies, those with an officer assigned primarily to alcohol enforcement and agencies with an alcohol division were more likely to conduct overservice enforcement.
Table 2.
Bivariate associations: Overservice perceptions/enforcement by agency characteristics (n=1024)a
| Agency characteristics |
||||||||
|---|---|---|---|---|---|---|---|---|
| Overservice perceptions/enforcement | Type of agency | Size of agencyb | Officer assigned to alcohol enforcement | Alcohol-specific division | ||||
|
| ||||||||
| Municipal | County | Small | Large | No | Yes | No | Yes | |
|
|
||||||||
| Percentage of agencies | ||||||||
| How common is overservice | ||||||||
| Not common | 52 | 42 | 45 | 55 | 52 | 42 | 52 | 33 |
| Somewhat common | 41 | 45 | 44 | 38 | 41 | 44 | 40 | 51 |
| Very common | 8 | 13 | 11 | 7 | 7 | 14 | 8 | 16 |
| Priority of overservice enforcement | ||||||||
| Fairly high/Very high | 13 | 10 | 10 | 15 | 12 | 12 | 12 | 14 |
| Fairly low/Very low/Average | 87 | 90 | 90 | 85 | 88 | 88 | 88 | 86 |
| Conduct overservice enforcement (if have authority) | ||||||||
| Yes | 30 | 18 | 24 | 29 | 23 | 34 | 24 | 43 |
| No | 70 | 82 | 76 | 71 | 77 | 66 | 76 | 57 |
Separate regression model for each perception/enforcement measure with each agency characteristic (bold indicates p < 0.05)
Small = < 2 officers/1000 population; large = ≥ 2 officers/1000 population
Note: Due to rounding, some columns may not add to exactly 100 percent
Table 3 shows associations between agency characteristics and types of overservice enforcement conducted among the 238 agencies that reported conducting overservice enforcement. There were no significant associations between use of overservice enforcement strategies (random inspections, walk-throughs, observations, pseudo-intoxicated purchase attempts, and POLD) and agency characteristics.
Table 3.
Bivariate associations: Types of overservice enforcement by agency characteristics (among 238 agencies reporting overservice enforcement)a
| Agency characteristics | ||||||||
|---|---|---|---|---|---|---|---|---|
|
| ||||||||
| Types of overservice enforcement | Type of agency | Size of agencyb | Officer assigned to alcohol enforcement | Alcohol-specific division | ||||
|
| ||||||||
| Municipal | County | Small | Large | No | Yes | No | Yes | |
|
| ||||||||
| Percentage of agencies | ||||||||
|
|
||||||||
| Random inspections | ||||||||
| Never | 34 | 26 | 35 | 30 | 36 | 27 | 36 | 20 |
| Once/a few times | 48 | 59 | 49 | 51 | 48 | 51 | 48 | 58 |
| Monthly/weekly | 18 | 15 | 16 | 19 | 15 | 22 | 17 | 21 |
| Walk-throughs | ||||||||
| Never | 17 | 22 | 16 | 19 | 20 | 12 | 19 | 14 |
| Once/a few times | 43 | 40 | 46 | 39 | 41 | 47 | 41 | 49 |
| Monthly/weekly | 40 | 37 | 38 | 42 | 39 | 40 | 40 | 37 |
| Observations | ||||||||
| Never | 35 | 38 | 36 | 34 | 39 | 27 | 38 | 27 |
| Once/a few times | 41 | 43 | 41 | 42 | 37 | 49 | 40 | 45 |
| Monthly/weekly | 24 | 20 | 23 | 24 | 24 | 24 | 22 | 29 |
| Pseudo-intoxicated purchase attempts | ||||||||
| Never | 91 | 98 | 94 | 91 | 94 | 89 | 94 | 86 |
| Once or more | 9 | 2 | 6 | 9 | 6 | 11 | 6 | 14 |
| Place of last drink (POLD) | ||||||||
| Yes | 30 | 24 | 29 | 28 | 25 | 35 | 27 | 36 |
| No | 70 | 76 | 71 | 72 | 75 | 65 | 73 | 64 |
Separate regression model for each enforcement measure with each agency characteristic (bold indicates p < 0.05)
Small = < 2 officers/1000 population; large = ≥ 2 officers/1000 population
Note: Due to rounding, some columns may not add to exactly 100 percent
Seventy-one agencies in our sample reported systematically collecting POLD data. The most common incident types collected by agencies included DUI (97%) and underage consumption (63%). Approximately one in three agencies collected POLD data in cases of disorderly conduct, physical assault, domestic violence, and medical incidents, and 21% reported collecting POLD data for other types of incidents (e.g., public intoxication). More than three in four agencies reported using POLD information to focus enforcement efforts and two in three shared POLD data with other jurisdictions. Table 4 shows the relationship between agency characteristics and types of incidents included in POLD data collection. Among agencies collecting POLD data, larger agencies were significantly more likely than smaller agencies to collect POLD data in disorderly conduct incidents. Agencies with an alcohol-specific division were less likely to collect POLD data in cases of domestic violence and physical assault.
Table 4.
Bivariate associations: POLD data collection by agency characteristics (n=71)a
| Agency Characteristics | ||||||||
|---|---|---|---|---|---|---|---|---|
|
| ||||||||
| POLD data collection | Type of agency | Size of agencyb | Officer assigned to alcohol enforcement | Alcohol-specific division | ||||
|
| ||||||||
| Municipal | County | Small | Large | No | Yes | No | Yes | |
|
| ||||||||
| Percentage of agencies | ||||||||
|
|
||||||||
| Types of incidents where POLD data is collected | ||||||||
| DUI | 97 | 100c | 97 | 97 | 97 | 97 | 98 | 94 |
| Domestic Violence | 34 | 56 | 32 | 41 | 41 | 32 | 50 | 13 |
| Disorderly Conduct | 41 | 31 | 25 | 51 | 43 | 32 | 43 | 29 |
| Physical assault | 39 | 43 | 28 | 48 | 40 | 36 | 46 | 25 |
| Underage consumption | 63 | 77 | 67 | 62 | 69 | 59 | 70 | 52 |
| Medical | 30 | 44 | 32 | 29 | 28 | 34 | 33 | 28 |
| Other (e.g. public intoxication) | 18 | 44 | 26 | 16 | 8 | 34 | 13 | 44 |
| Share POLD info with other jurisdictions | 66 | 78 | 58 | 77 | 69 | 66 | 67 | 67 |
| Use POLD to focus efforts | 77 | 78 | 82 | 71 | 72 | 82 | 75 | 83 |
Separate regression model for each type of incident/action with each agency characteristic (bold indicates p ≤ 0.05)
Small = < 2 officers/1000 population; large = ≥ 2 officers/1000 population
regression model did not converge due to zero cell; Chi square test used to test significance
DISCUSSION
Enforcement of state laws prohibiting service of alcohol to intoxicated patrons by local law enforcement agencies is not common in the U.S. Less than a quarter of agencies that responded to our national survey reported conducting overservice enforcement. This finding is consistent with a previous survey we conducted in 2010 [5]. POLD data collection is a strategy to reduce overservice that has received more attention over the past decade. Approximately 7% of agencies nationally, and 30% of those that conducted any overservice enforcement, collected POLD data in 2019.
As we found in our previous study [14], agencies with an officer assigned primarily to alcohol enforcement or that had a specific alcohol division (compared to agencies that did not) were more likely to conduct overservice enforcement in general. Though not statistically significant, agencies with an alcohol-specific officer and/or division were also more likely to conduct POLD enforcement specifically—a relationship that was not previously examined. The presence of an alcohol division or an officer assigned to alcohol enforcement may indicate that the agency places a priority on enforcing alcohol laws in general or has dedicated resources to alcohol enforcement. Agencies with at least one officer assigned to alcohol enforcement may have the time, resources, and clear delineation of responsibility necessary to develop such a program. Similarly, we found that municipal agencies (compared with county agencies) were more likely to conduct overservice enforcement in general and POLD specifically but the association was not statistically significant for POLD. Municipalities may have a greater number of licensed alcohol establishments in a smaller geographic area compared with counties which may facilitate overservice enforcement.
Perceptions of how common overservice was in the community and prioritization of overservice enforcement also varied by agency characteristics. Agencies that had an officer assigned primarily to alcohol enforcement or an alcohol division were more likely to perceive overservice to be common within their jurisdiction. It is possible that agencies with an officer assigned primarily to alcohol enforcement or agencies with an alcohol division conduct more alcohol-related enforcement in general and are therefore more likely to encounter alcohol-related offenses. It is also possible that agencies create alcohol divisions or assign an officer primarily to alcohol enforcement when they perceive alcohol-related incidents to be common. Given the cross-sectional nature of this study, it is not possible to determine whether agencies’ perceptions precede the creation of an alcohol division or hiring of an alcohol officer, or vice-versa. Perceptions of how common certain alcohol-related offenses are is likely based on the extent of enforcement in addition to the true number of offenses committed within a jurisdiction. However, municipal agencies were less likely to perceive that overservice is common within their jurisdiction despite being more likely to conduct overservice enforcement. As mentioned previously, it may be easier for municipal-level agencies to conduct overservice enforcement in their jurisdictions, as establishments with liquor licenses issued by the municipality are likely to be concentrated in a smaller geographic area than establishments with liquor licenses issued by a county and there are likely to be more licenses overall in a municipality. Small agencies (< 2 officers/1000 population) were more likely than large agencies (≥2 officers/1000 population) to perceive overservice as very common within their jurisdiction, but were less likely to rate overservice enforcement as a fairly or very high priority for their agency. Smaller agencies, compared to larger agencies, may simply have a shorter list of issues they can prioritize. Despite this difference in prioritization between small and large agencies, there was not a significant difference between these groups in the proportion that conducted overservice enforcement. No other agency characteristics were associated with prioritization of overservice enforcement. It is not clear why, but prioritization of overservice enforcement was predominantly low for responding agencies, suggesting a need for increased attention to this issue.
Agencies more commonly reported conducting random inspections, walk-throughs, and observations to address overservice than conducting pseudo-intoxicated purchase attempts or collecting POLD data. Of these strategies, only POLD has been evaluated in terms of its ability to reduce overservice but with mixed results regarding its effectiveness. More research on the effectiveness of POLD is needed in various contexts. Random inspections, walk-throughs, and observations may be preferred by agencies because they fit within existing departmental infrastructure and practices. In contrast, POLD or pseudo-intoxicated purchase attempts may require establishing new systems for managing data and/or hiring additional personnel. Evaluating the effectiveness of these various types of overservice enforcement is an area for future research, especially in light of their current use by law enforcement agencies.
DUI offenses were the most likely type of offense to be recorded by collecting POLD data. The emphasis on DUIs that we observed in POLD programs across the country is consistent with reports that describe how some law enforcement agencies in Wisconsin train officers to collect place of last drink information only for driving while intoxicated incidents [16–17]. In contrast, a report from Minnesota shows that agencies implementing POLD train officers to ask about the place of last drink for any incident that involves alcohol, capturing a broader range of incident types [11]. Data from POLD efforts conducted in Minnesota indicated that DUI offenses accounted for roughly half of all POLD incidents [18], but this sample included agencies that only conducted POLD data collection for DUI offenses as well as those that conducted POLD across multiple types of incidents. Nevertheless, it is clear agencies that conduct POLD only for DUI offenses miss a substantial proportion of cases in which overservice has occurred.
Limitations
This study has several limitations. First, while we surveyed a large national sample of agencies, the sample was not designed to be nationally representative. Our enforcement survey expanded on a similar survey conducted in 2010 that was focused on state-level enforcement. We started with the 2010 sample and added agencies to improve generalizability. The large sample size and broad geographic distribution of our sample give us confidence that our results provide a reasonable picture of U.S. enforcement practices.
Second, this analysis is focused on the municipal and county levels. Overservice enforcement strategies (including POLD) can be implemented by statewide agencies, which could affect their use in specific jurisdictions. For example, a state agency conducting POLD may change the likelihood of conducting POLD at the local level due to encouragement or pressure from larger agencies, or reduce the likelihood of conducting POLD so as not to duplicate efforts. An analysis incorporating the state level would be able to paint a fuller picture of the prevalence of overservice enforcement strategies in the U.S.
Third, this survey was cross-sectional, leaving us unable to determine causal relationships between overservice enforcement and characteristics of agencies and jurisdictions. Another limitation involves the self-report nature of the data, and we collected responses from just one representative of each agency. While we took steps to ensure this representative was the person most knowledgeable about the agency’s alcohol enforcement efforts, it is possible that this individual was unable to respond accurately to all of our questions. An additional limitation is that while we asked agencies whether they systematically record POLD data and whether they use the data to focus enforcement efforts, we did not ask about how agencies used the data to inform enforcement. It is likely that some enforcement strategies used by agencies are more effective than others in reducing overservice. Unintended negative consequences of enforcement may also vary depending on the strategies employed by agencies. Agencies that employ discriminatory practices targeting certain communities or businesses, for example, may have an overall negative effect on public health and safety compared with agencies that seek to enforce overservice laws in an equitable way. Proponents of POLD should seek to identify ways of collecting and using POLD data to inform enforcement in a way that minimizes potential for discriminatory practices or unfairly targeting certain communities or commercial establishments.
Conclusions
Excessive alcohol consumption remains a major public health issue and overservice of alcohol contributes to alcohol-related problems. Multiple strategies are currently used by law enforcement to prevent overservice. The strategies used by agencies vary based on the priority an agency places on alcohol enforcement and the resources they dedicate to this enforcement. Identifying the factors associated with adoption of overservice enforcement strategies can advance our understanding how these strategies work and may help us predict the agencies that are likely to adopt them in the future. Most overservice enforcement strategies have not been evaluated for their effectiveness; our research team is currently undertaking some of this evaluation. Increasing our understanding of overservice enforcement strategies may also help identify other agencies that could be involved in or lead the implementation of these strategies, including Alcohol Beverage Control (ABC) agencies, public safety departments, transportation departments, and health departments. This is particularly relevant in light of recent calls to reform and restructure law enforcement and public safety in the U.S. The enforcement structure that maximizes potential public health and safety benefits while minimizing harms should be determined based on the context in which the strategy is implemented.
Funding:
This work was supported by a grant from National Institute on Alcohol Abuse and Alcoholism under Grant R01 AA026610-01.
Footnotes
Conflicts of interest: The authors have no conflicts of interest to disclose.
Ethics approval: Our proposed research was reviewed by the University of Minnesota Human Research Protection Program Institutional Review Board and was determined that the study was not research involving human subjects as defined by federal regulations (Study ID 00001798).
Consent to participate: Not applicable
Consent for publication: Not applicable
Code availability: The code used to analyze the data is available from the corresponding author upon reasonable request.
Availability of data and material:
The data upon which the presented findings are based are available from the corresponding author upon reasonable request.
References
- 1.Naimi TS, Nelson DE, & Brewer RD (2009). Driving after binge drinking. American Journal of Preventive Medicine, 37(4), 314–320. [DOI] [PubMed] [Google Scholar]
- 2.Savola O, Niemelä O, & Hillbom M (2005). Alcohol intake and the pattern of trauma in young adults and working aged people admitted after trauma. Alcohol and Alcoholism, 40(4), 269–273. [DOI] [PubMed] [Google Scholar]
- 3.Gruenewald PJ, Stockwell T, Beel A, & Dyskin EV (1999). Beverage sales and drinking and driving: The role of on-premise drinking places. Journal of Studies on Alcohol, 60(1), 47–53. [DOI] [PubMed] [Google Scholar]
- 4.Mosher JF, Hauck A, Carmona M, et al. (2009) Laws Prohibiting Alcohol Sales to Intoxicated Persons. National Highway Traffic Safety Administration. https://www.nhtsa.gov/sites/nhtsa.gov/files/811142.pdf [Google Scholar]
- 5.Lenk KM, Toomey TL, Nelson TF, Jones-Webb R, & Erickson DJ (2014). State and local law enforcement agency efforts to prevent sales to obviously intoxicated patrons. Journal of Community Health, 39(2), 339–348. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Toomey TL, Lenk KM, Nederhoff DM, et al. (2016). Can obviously intoxicated patrons still easily buy alcohol at on-premise establishments? Alcoholism: Clinical and Experimental Research, 40(3), 616–622. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Fell JC, Fisher DA, Yao J, & McKnight AS (2017). Evaluation of a responsible beverage service and enforcement program: Effects on bar patron intoxication and potential impaired driving by young adults. Traffic Injury Prevention, 18(6), 557–565. [DOI] [PubMed] [Google Scholar]
- 8.George MD, Bodiford A, Humphries C, Stoneburner KA, & Holder HD (2018). Media and education effect on impaired driving associated with alcohol service. Journal of Drug Education, 48(3–4), 86–102. [DOI] [PubMed] [Google Scholar]
- 9.Jones L, Hughes K, Atkinson AM, & Bellis MA (2011). Reducing harm in drinking environments: A systematic review of effective approaches. Health & Place, 17(2), 508–518. [DOI] [PubMed] [Google Scholar]
- 10.Rammohan V, Hahn RA, Elder R, et al. (2011). Effects of dram shop liability and enhanced overservice law enforcement initiatives on excessive alcohol consumption and related harms. American Journal of Preventive Medicine, 41(3), 334–343. [DOI] [PubMed] [Google Scholar]
- 11.Mikkelson S (2020). Putting a cork in over service. Minnesota Chiefs of Police Association. https://www.mnchiefs.org/place-of-last-drink [Google Scholar]
- 12.Conway K, McTaggart S (2002). Evaluation of Alcohol Healthwatch Last Drink Survey Programme. Alcohol & Public Health Research Unit, University of Auckland. https://popcenter.asu.edu/sites/default/files/evalua2.pdf [Google Scholar]
- 13.Ramirez R, Nguyen D, Cannon C, Carmona M, & Freisthler B (2008). A campaign to reduce impaired driving through retail-oriented enforcement in Washington State. National Highway Traffic Safety Administration. https://rosap.ntl.bts.gov/view/dot/16482 [Google Scholar]
- 14.Erickson DJ, Farbakhsh K, Toomey TL, Lenk KM, Jones-Webb R, & Nelson TF (2015). Enforcement of alcohol-impaired driving laws in the United States: A national survey of state and local agencies. Traffic Injury Prevention, 16(6), 533–539. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Wisconsin Alcohol Policy Project. (2018). Place of Last Drink Projects: What, Why and How. In University of Wisconsin-Madison Law School. [Google Scholar]
- 16.Bosma L, Schriemer D, Delehanty E, Lenk K, & Toomey T (2022). Summary and Lessons Learned from POLD Case Studies [Unpublished manuscript]. School of Public Health, University of Minnesota. [Google Scholar]
- 17.Mikkelson S, Smith A (2016). Addressing over service of alcohol: Place of last drink [PowerPoint slides]. 2016 Minnesota Toward Zero Deaths Conference. http://tzd.state.mn.us/events/conference/2016/documents/Mikkelson_Smith_Kringen.pdf [Google Scholar]