5th October 2018, Volume 131 Number 1483

Kate Ford, James Foulds, Oliver Coleman, Michael Ardagh, Scott Pearson, Nicolas Droste, Giles Newton-Howes, J Douglas Sellman

The harm arising from acute alcohol intoxication is a significant economic and public health problem in New Zealand1 and worldwide.2 A marker for this burden of morbidity is alcohol-related emergency department (ED) presentations, especially those due to injuries incurred after drinking.3

From a public health perspective, prevention of harms arising from acute alcohol intoxication is difficult. One reason for this is that while the density, location and type of alcohol outlets influence alcohol availability and therefore alcohol-related harm,4 efforts to regulate availability often encounter opposition from organisations who profit from the sale of alcohol.5

In 2010, a New Zealand Law Commission report into alcohol-related harm was published.6 Shortly after its publication, The Sale and Supply of Alcohol Act 2012 (“the Act”) came into effect. The Act was intended to provide a new legal framework for regulating the availability of alcohol in New Zealand. Its explicit objective, as stated in Section 4 of the Act, is that “the harm caused by the excessive or inappropriate consumption of alcohol should be minimised”.7 Although the Justice Minister at the time noted that the Act adopted 126 of the 153 recommendations of the Law Commission report,8 many of these were changes were minor. The changes included setting default trading hours restricting on-licence premises from trading after 4am and stopping off-licenses from opening before 7am. More substantive recommendations of the Law Commission’s work, including greater regulation of alcohol marketing and pricing,9 were not implemented. Provision for raising the alcohol purchase age was included in the original Bill, but this was defeated in a conscience vote and therefore was not incorporated into the Act.10

A key part of the Act was a provision to allow territorial authorities to develop their own Local Alcohol Policies (LAPs). The Act was implemented in stages from December 2012 onwards and subsequently, many local authorities attempted to introduce LAPs in their jurisdictions. In many cases these efforts met legal obstacles, particularly from the owners of supermarket chains and liquor stores.11 For example, a provisional LAP in Christchurch was developed in 2013 but by late 2017 it still had not been introduced.12 This provisional LAP was finally put on hold in 2018.13 Similar problems have been encountered in other regions, and it has recently been argued that the maximum trading hours provisions have been the only element of the Act that has significantly altered the alcohol environment to date, because “the strength of LAPs has been muted by alcohol industry appeals”.14

On this background, in a previous study conducted in late 2013 (prior to the provisions of the Act coming into full effect) we reported on patterns of alcohol-related ED attendances in Christchurch as a proxy for harm related to acute intoxication.15,16 In the present study, we aimed to see whether patterns of alcohol-related ED attendances had changed since the Act was implemented.


Study design

Cross-sectional observational study of hospital ED attendances during three-week waves of data collection in 2013 and 2017.


Christchurch Hospital is a large teaching hospital and tertiary referral centre in the South Island of New Zealand. The hospital is the only major acute referral centre in a region with a population of approximately 550,000. Christchurch Hospital has one of the busiest EDs in Australasia, with over 100,000 attendances annually.

Participants and data collection

Interviewers were four University of Otago students funded by summer studentships (two in 2013 and two in 2017). The first wave of data collection spanned a 23-day period from 16 November to 8 December 2013. The equivalent of two full weeks (42 x 8-hour shifts) was sampled during that period. Shifts were non-randomly selected in order to mitigate effects of fatigue on the interviewers. However, shifts were scheduled to ensure equal numbers of shifts on each day of the week and at each time of day (midnight to 0800; 0800 to 1600; and 1600 to midnight) were sampled. In the second wave, a further 42 x 8-hour shifts were sampled from 17 November to 9 December 2017 using the same non-random sampling method, with the days and times sampled designed to correspond as closely as possible to those in the first wave. For patients with two or more attendances within the study period, only the first attendance was considered. One public holiday weekend fell in each sampling period. At least two major public events associated with alcohol consumption occurred during both sampling periods, New Zealand Cup Day (a horse racing event) and “Crate Day” (an informal event promoted by a radio station, in which people are encouraged to socialise at private locations).

The interviewers identified all potentially eligible participants among patients who attended the Christchurch Hospital ED within the sampling frame. To determine who was potentially eligible, interviewers attempted to talk to all ED attendees. For those patients who were not screened by the interviewers (for example because they were uncooperative, too unwell or had left the department), eligibility was determined by reviewing routinely-collected ED triage data. These data record alcohol consumption within four hours and the presence of an alcohol-related problem causing the presentation, for all attendances. Uncooperative patients, those lacking capacity to consent for any other reason (eg, altered level of consciousness) and those who were too unwell to be interviewed (eg, requiring immediate surgery) were not considered for inclusion. Parental consent was obtained for subjects under age 17.

Participants were required to fulfil at least one of the following criteria:

  1. Reported consuming any alcohol in the four hours prior to attending ED.
  2. Patient stated that a specific drinking episode contributed to the attendance.
  3. ED clinical staff or the interviewer judged that acute alcohol consumption causally contributed to the condition which prompted the ED attendance.
  4. Patient exhibited outward signs of alcohol intoxication, using a standardised intoxication scale developed by the New Zealand Health Promotion Agency (see Supplementary material).

The Southern Health and Disability Ethics Committee approved the study (13/STH/147/AM02).


Age, gender and ethnicity were taken from district health board electronic records. Ethnicity data were recorded according to standard New Zealand Ministry of Health ethnicity data protocols.17 To facilitate analyses in the present study, and consistent with the ethnicity data protocols, ethnicity was prioritised in the following order: New Zealand Māori; Pacific Peoples; other; European.

The reason for the attendance was taken from clinical notes and was categorised as being either due to an injury or other causes. Alcohol use data in relation to the index drinking episode were obtained by self-report while participants were in the ED. The index drinking episode was defined as the episode in which the injury or acute physical symptoms prompting the ED attendance had occurred, or the current drinking episode for patients who met only criterion 1, as above.

Participants were asked the number of standard drinks they had consumed in the index episode, the source(s) of alcohol consumed and the place the last alcoholic drink was consumed.

The Alcohol Use Disorders Identification Test (AUDIT)18 was administered to participants in the 2017 wave but was not administered in 2013. The AUDIT is a widely used and well-validated 10-item self-report tool used to measure patients’ usual drinking patterns and identify the presence of hazardous or harmful drinking patterns. Scores of 8 or over suggest the presence of a hazardous drinking pattern while scores of 20 or more are indicative of a more severe alcohol problem requiring specialised assessment and treatment.19

Assessment of bias and statistical analysis

Sampling bias was assessed by comparing demographic characteristics of those who were included in the study with those who were eligible but not included in the study.

Statistical comparisons were performed in SPSS version 25 using chi-square tests for categorical outcomes and Mann-Whitney U tests or t-tests for continuous outcomes.

As participants who had consumed alcohol within four hours of the attendance but met no other inclusion criteria may be a population with different characteristics to the remainder of the sample, a secondary analysis was then conducted in which these participants were excluded. The participants included in these analyses were classed as “alcohol-affected” while the remainder of the participants were labelled “non alcohol-affected”.


Participant selection

Figure 1 shows the process for selection of participants in the two waves of the study. As shown in the Figure, 3,400 people attended the ED at least once during the 2013 sampling frame and 3,721 attended in the corresponding period in 2017. The percentage of all ED attendees whose attendance occurred within four hours of drinking or as a direct short-term result of alcohol consumption (and were therefore eligible for the study) was 7.4% (n=253) in 2013 and 6.9% (n=258) in 2017 (2013 vs 2017: χ2= 0.69, p=.41). The gender and ethnicity distributions did not differ between those included and those not included, but potentially eligible patients who were not included were on average 3.3 years younger (t=2.21, p=.03) than the participants.

Figure 1: Participant selection flowchart.


Comparing the two waves of data collection, those who were potentially eligible but not included did not differ by gender (χ2= 0.68, p=.41), age (t=0.64, p=.52), ethnicity (χ2=3.4, p=.33) or length of stay in ED (U=6,266, p=.22) between 2013 and 2017.

Characteristics of participants

In all, 163 participants were included in 2013 and 139 in 2017.

As shown in Table 1, participants were predominantly male. They ranged in age from 14 to 87. There were no statistically significant differences in gender, age, ethnicity or ED arrival times between 2013 and 2017 but the median length of stay in the ED was significantly longer in 2017 than 2013.

Table 1: Characteristics of sample.



Data collection wave






Statistical test (2013 vs 2017)

Number of participants





Male gender: n (%)


105 (64.4)

96 (69.1)

χ2=0.73, p=.39

Age categories: n (%)


9 (5.5)

5 (3.6)

χ2=2.00, p=.74



46 (28.2)

36 (25.9)



49 (30.1)

47 (33.8)



41 (25.2)

31 (22.3)



18 (11.0)

20 (14.4)

Prioritised ethnicity: n (%)

NZ Māori

21 (12.9)

20 (14.4)

χ2=4.46, p=.22


Pacific Island

3 (1.8)

6 (4.3)


Other ethnicity

3 (1.8)

7 (5.0)



136 (83.4)

106 (76.3)






Emergency department arrival time: n (%)

Day (0800–1600)

31 (19.0)

34 (24.5)

χ2=1.39, p=.50


Evening (1600–2400)

82 (50.3)

67 (48.2)


Night (0000–0800)

50 (30.7)

38 (27.3)

Alcohol-affected:a n (%)


114 (69.9)

96 (69.1)

χ2=0.03, p=.87

Length of stay in ED in minutes: median (interquartile range)


169 (123–231)

204 (150–284)

Mann-Whitney U=14,083, p<.001

a. “Alcohol-affected” patients were those in whom either the patient indicated they thought alcohol contributed to the attendance, or staff observed intoxication signs or judged that alcohol consumption in the most recent drinking episode contributed to the attendance.

Comparison of alcohol measures between 2013 and 2017: whole sample


Table 2 shows a comparison of alcohol-related measures between the 2013 and 2017 waves of data collection, including alcohol consumption in the index episode; source of alcohol; and place of the last drink for the whole sample in 2017 compared to 2013.

Table 2: Comparison of alcohol-related measures between 2013 and 2017 waves of data: all patients.




Statistical test: 2013 vs 2017

Total number of participants




Standard drinks consumed in index drinking episode: n (%)


53 (41.1)

41 (29.5)


U=10,364, p=.56d


31 (24.0)

39 (28.1)


22 (17.1)

19 (13.7)


23 (17.8)

10 (7.2)


32 (19.9)

25 (18.0)


2 (1.2)

5 (3.6)

Injury-related attendance: n (%)

79 (48.4)

74 (53.2)

χ2=0.69 p=.41

Source of alcohol:a (n; %)

Liquor store

68 (41.7)

78 (56.1)

χ2=6.22 p=.01


49 (30.0)

32 (23.0)

χ2=1.89 p=.17

Unknown or other off-licence source

19 (11.7)

15 (10.8)

χ2=0.06, p=.81

On-licence venueb

49 (30.0)

29 (20.9)

χ2=3.31 p=.07

Home brew

4 (2.5)

3 (2.2)

χ2=0.03, p=.86

Pre-drinking:c n (%)

17 (10.4)

13 (9.4)

χ2=0.09, p=.79

Place last drink consumed: n (%)

Private location

114 (69.9)

97 (69.8)

χ2=3.10, p=.38

On-licence venueb

38 (23.3)

27 (19.4)

Unlicensed public location

6 (3.7)

5 (3.6)

Other or unknown

5 (3.1)

10 (7.2)

AUDIT score

median (range)

Not administered

11 (1–38)


% scoring 8 or more



% scoring 20 or more



a. Note that some patients obtained alcohol from more than one location. See text for breakdown of those purchasing from only off-licence; only on-licence; and both types of sources. Source of methylated spirits was categorised under unknown or other off-licence source.
b. Includes bars, restaurants, licensed clubs and private functions at licensed venues. 
c. Consumption from off-licence sources prior to drinking at licensed premises.
d. Non-parametric comparison of number of standard drinks (as a continuous outcome measure) in 2013 vs 2017.

As shown in the table, the number of standard drinks consumed; the percentage of participants with an injury-related attendance; pre-drinking (consuming from off-licence sources prior to engaging in consumption at licensed premises); and the place the last drink was consumed did not change significantly between 2013 and 2017. There was a significant (p<.01) increase in the percentage of participants who had purchased from liquor stores and non-significant (p>.05) reductions in supermarket and on-licence purchasing from 2013 to 2017. Comparing patterns of on-licence versus off-licence purchasing between 2013 and 2017, the percentage of participants who purchased solely from off-licence sources was 67.5% in 2013 and 79.1% in 2017; solely from on-licence sources was 17.2% in 2013 and 10.8% in 2017; and from both on and off-licence sources was 12.3% in 2013 and 9.4% in 2017. The remaining participants were methylated spirits drinkers (n=2) or those who consumed solely home brew (n=4).

Findings for alcohol-affected participants

Table 3 shows alcohol consumption in the index episode; source of alcohol; and place of the last drink for alcohol-affected participants (that is, those participants who had consumed alcohol within four hours of the attendance but met no other inclusion criteria) in 2017 compared to 2013.

Table 3: Comparison of alcohol-related measures between 2013 and 2017 waves of data: alcohol-affecteda patients only.




Statistical test: 2013 vs 2017

Total number of participants




Standard drinks consumed in index drinking episode: n (%)


15 (13.2)

3 (3.1)

 Mann-Whitney U=4,880, p=.61e


22 (19.3)

36 (37.5)


21 (18.4)

18 (18.8)


22 (19.3)

10 (25.0)


32 (28.1)

24 (25.0)


2 (1.7)

5 (5.2)


Injury-related attendance: n (%)

61 (53.5)

61 (63.5)

χ2=2.16, p=.14

Source of alcohol:b n (%)

Liquor store

57 (50.0)

65 (67.7)

χ2=6.71, p=.01


29 (25.4)

15 (15.6)

χ2=3.03, p=.08

Unknown or other off-licence source

13 (11.4)

7 (7.3)

χ2=1.11, p=.29

On-licence venuec

36 (31.6)

21 (22.9)

χ2=2.74, p=.10

Home brew

3 (2.6)

2 (2.1)

χ2=0.08, p=.78

Pre-drinking:d n (%)

16 (14.0)

10 (10.4)

χ2=0.71, p=.40

Place last drink consumed: n (%)

Private location

79 (69.3)

65 (67.7)

χ2=2.04, p=.57

On-licence venue

26 (22.8)

20 (19.8)

Unlicensed public location

4 (3.5)

5 (5.2)

Other or unknown

3 (2.6)

6 (6.2)

AUDIT score

median (range)

Not administered

15 (4–38)


% scoring 8 or more



% scoring 20 or more



a. “Alcohol-affected” patients were those in whom either the patient indicated they thought alcohol contributed to the attendance, or staff observed intoxication signs or judged that alcohol consumption in the most recent drinking episode contributed to the attendance.
b. Note that some patients obtained alcohol from more than one location. 
c. Includes restaurants, licensed clubs and private functions at licensed venues. 
d. Consumption from off-licence sources before drinking at licensed premises.
e. Non-parametric comparison of number of standard drinks (as a continuous outcome measure) in 2013 vs 2017.

As shown in Table 3, findings for the comparisons in alcohol measures between 2013 and 2017 were similar for alcohol-affected participants and the whole sample. In particular there was no change in most of alcohol measures from 2013 to 2017, but the apparent increase in liquor store purchasing seen in the whole sample was also noted in the alcohol-affected subgroup.

Combining both waves of data, alcohol-affected participants consumed significantly more (U=17,371, p<.001) in the index drinking episode (median 13.0 standard drinks) than non alcohol-affected participants (median 2.3 standard drinks). Similarly, the AUDIT score was significantly higher (U=3,082, p<.001) in alcohol-affected (median 15) than in non alcohol-affected participants (median 6).The percentage of participants scoring 8 or more and 20 or more on the AUDIT was far higher in the alcohol-affected group than the remainder of the sample. Alcohol-affected participants were also significantly more likely (χ2 =18.0, p<.001) to have presented because of an injury (59.2%) than non alcohol-affected participants (32.6%).


The proportion of ED attendances that occurred immediately after alcohol consumption or as a direct short-term result of alcohol did not change significantly from 2013 to 2017, and was about 1 in 14 ED attendances overall. This figure is likely to underestimate the total impact of alcohol-related harm in the ED, since it does not account for ED presentations associated with chronic alcohol use rather than acute intoxication.

Participants had mainly purchased alcohol from off-licence sources and had consumed it at private locations. The 2017 findings were consistent with our first wave of data collection15,16 and with recent Australian data.20 Our data also suggest there was an increase in purchasing from liquor stores from 2013 to 2017. Conversely there was a modest decline in supermarket purchasing but this change was not statistically significant. However the rebuild of Christchurch after the 2010–2011 earthquakes (which destroyed the central bar district) provides a unique local context to these findings. Caution is therefore needed in generalising these findings beyond Christchurch. Nonetheless, media reports suggest bar districts in other parts of New Zealand are struggling for survival as their customers desert them,21,22 and the predominance of packaged liquor consumption reported by ED attendees is also reflected by Australian research.20 One core driver of this behaviour is that packaged alcohol from off-licence venues is much cheaper per unit of alcohol than alcohol bought from bars and restaurants. This economic fact has given rise to the strategy known as “pre-drinking”, whereby drinkers consume packaged liquor before going to licensed venues. Pre-drinkers have regularly cited price differentials between on-licence and off-licence drinks as the most common motivator of this behaviour.23–25 In the present study, pre-drinking was reported by only about 10% of the sample overall, although it was reported by almost half the people whose last drink was at an on-licence venue. This pre-drinking figure is lower than that reported in a recent survey of drinkers at Australian licensed venues.25

This study was designed to measure changes following the introduction of the Sale and Supply of Alcohol Act 2012, many provisions of which were not expected to occur until after the 2013 wave of data collection. In particular, the study was designed to investigate the impact of a new Local Alcohol Policy in Christchurch, which was scheduled to be introduced soon after our first wave. However, as outlined in the introduction, many of the changes implemented in the Act were relatively minor, while Local Alcohol Policies have proven difficult and costly for territorial authorities to implement, primarily because of legal opposition from the alcohol industry.11 This problem has been very apparent in Christchurch, where an LAP is still not in place.13

As expected, most alcohol-affected participants’ AUDIT score suggested a hazardous drinking pattern beyond the index drinking episode and more than one quarter of alcohol-affected patients had an AUDIT score in the range suggesting they needed specialist alcohol assessment.19 The potential opportunity to offer alcohol interventions to this group of patients has been well-recognised.26 However, the literature on alcohol screening and brief interventions has methodological limitations, and findings overall have been mixed.27 In particular, brief alcohol interventions delivered in the ED appear to have very modest and short-lived effects.28,29 Nonetheless, the collection of individual-level alcohol consumption and purchasing data among ED attendees can help to inform effective, intelligence-based public health interventions by identifying hotspots of alcohol harm at a community level.29,30

Limitations of this study include the relatively small sample size and short sampling timeframes; lack of objective measures of alcohol exposure (ie, blood or breath alcohol); and reliance on self-report data regarding source of alcohol and location of consumption. People who agreed to participate in the study may not be fully representative of the whole pool of eligible attendees. In particular, more heavily intoxicated and uncooperative individuals, those with reduced level of consciousness, and those who left the ED before being seen were less likely to be included.

In conclusion, the number and pattern of alcohol-related ED attendances did not change significantly from 2013 to 2017 despite new legislation in 2012 aimed at reducing alcohol-related harm. Most people with an alcohol-related ED attendance had purchased from off-licence outlets, particularly (and increasingly) liquor stores, and most had consumed their last drink at a private location. Policy measures designed to reduce the harm from acute alcohol intoxication therefore need to target all sources of alcohol, but particularly off-licence venues. This could be achieved via Local Alcohol Policies placing controls on outlet density and opening hours for these venues.


Medical problems arising after alcohol use are a major cause of people needing to attend their local emergency department. A new law introduced in New Zealand (The Sale and Supply of Alcohol Act 2012) was intended to reduce the amount of harm due to alcohol. We found that there had been little change in the patterns of alcohol-related emergency department attendances in Christchurch after the new law came into effect. Most people who came to the emergency department because of a medical problem after drinking had bought their alcohol at a supermarket or liquor store, and had consumed their last drink at a private location.



To measure changes in alcohol-related emergency department (ED) attendances after introduction of the Sale and Supply of Alcohol Act 2012.


Cross-sectional survey of Christchurch ED attendees in three-week sampling periods in 2013 and 2017. Participants had consumed alcohol within four hours, or their drinking had directly contributed to the attendance. The quantity of alcohol consumed and places of purchase and consumption for the index drinking episode were recorded.


From 2013 to 2017 there was a non-significant (p=.41) reduction in the proportion of ED attendees eligible for the study, from 253/3400 (7.4%) to 258/3721 (6.9%). Among participants (n=169 in 2013, n=139 in 2017), liquor store purchasing increased from 41.7% in 2013 to 56.1% in 2017 (p<.01) but there was no significant change in quantity consumed in the index episode; last drink location; percentage of participants with an injury-related attendance; or pre-drinking. In both waves, most participants had purchased alcohol from off-licence venues and consumed their last drink at a private location.


Alcohol-related ED attendances remained common after the Sale and Supply of Alcohol Act 2012 was introduced, and they mainly occurred in people who sourced alcohol from off-licence outlets and had their last drink at private locations. 

Author Information

Kate Ford, Department of Psychological Medicine, University of Otago Christchurch;
James Foulds, Department of Psychological Medicine, University of Otago Christchurch;
Oliver Coleman, Department of Psychological Medicine, University of Otago Christchurch;
Michael Ardagh, Emergency Department, Christchurch Hospital, Christchurch;
Scott Pearson, Emergency Department, Christchurch Hospital, Christchurch;
Nicolas Droste, School of Psychology, Deakin University, Geelong, Victoria, Australia;
Giles Newton-Howes, Department of Psychological Medicine, University of Otago, Christchurch;
J Douglas Sellman, Department of Psychological Medicine, University of Otago Christchurch.


James Foulds, Senior Lecturer, Department of Psychological Medicine, University of Otago, PO Box 4345, Christchurch 8140.

Correspondence Email


Competing Interests

Mr Coleman, Ms Ford and Dr Foulds report grants from New Zealand Health Promotion Agency during the conduct of the study.


  1. Connor J, Broad J, Rehm J, Vander Hoorn S, Jackson R. The burden of death, disease, and disability due to alcohol in New Zealand. N Z Med J 2005; 118:1–12.
  2. Rehm J, Mathers C, Popova S, Thavorncharoensap M, Teerawattananon Y, Patra J. Global burden of disease and injury and economic cost attributable to alcohol use and alcohol-use disorders. The Lancet 2009; 373:2223–33.
  3. Rehm J. The risks associated with alcohol use and alcoholism. Alcohol Research: Current Reviews 2011; 34:135–43.
  4. Campbell CA, Hahn RA, Elder R, et al. The Effectiveness of Limiting Alcohol Outlet Density As a Means of Reducing Excessive Alcohol Consumption and Alcohol-Related Harms. Am J Prev Med 2009; 37:556–69.
  5. Babor TF, Robaina K. Public Health, Academic Medicine, and the Alcohol Industry’s Corporate Social Responsibility Activities. Am J Public Health 2012; 103:206–14.
  6. New Zealand Law Commission. Alcohol in our lives : curbing the harm. A report on the review of the regulatory framework for the sale and supply of liquor (NZLC R114). Wellington: New Zealand Law Commission; 2010.
  7. New Zealand Sale and Supply of Alcohol Act. New Zealand 2012.
  8. Government outlines balanced plan for alcohol reform. 2010. at http://www.beehive.govt.nz/release/government-outlines-balanced-plan-alcohol-reform
  9. Babor TF, Caetano R, Casswell S, et al. Alcohol: No ordinary commodity: research and public policy. Oxford: Oxford University Press; 2010.
  10. Sellman D, Connor J, Robinson G, McBride S. Alcohol reform – New Zealand style: Reflections on the process from 1984 to 2012. Psychotherapy and Politics International 2017; 15:e1398.
  11. New Zealand Drug Foundation. Local Alcohol Policies: Has the new Act delivered? Matters of Substance. Wellington, 2017.
  12. Local Alcohol Policy (or LAP). 2018. at http://www.ccc.govt.nz/the-council/plans-strategies-policies-and-bylaws/policies/alcohol-and-psychoactive-products-policies/provisional-local-alcohol-policy-or-lap
  13. Truebridge N. Liquor policy for city on ice. The Christchurch Press 2018 April 6 
  14. Randerson S, Casswell S, Huckle T. Changes in New Zealand’s alcohol environment following implementation of the Sale and Supply of Alcohol Act (2012). New Zealand Medical Journal 2018; 131:14–23.
  15. Das M, Stewart R, Ardagh M, et al. Patterns and sources of alcohol consumption preceding alcohol-affected attendances to a New Zealand hospital emergency department. The New Zealand Medical Journal (Online) 2014; 127:40–55.
  16. Stewart R, Das M, Ardagh M, et al. The impact of alcohol-related presentations on a New Zealand hospital emergency department. N Z Med J 2014; 127:23–39.
  17. New Zealand Ministry of Health. HISO 10001:2017 Ethnicity Data Protocols. Wellington, NZ. 2017.
  18. Saunders JB, Aasland OF, Babor TF, De La Fuente JR, Grant M. Development of the Alcohol Use Disorders Identification Test (AUDIT): WHO Collaborative Project on Early Detection of Persons with Harmful Alcohol Consumption-II. Addiction 1993; 88:791–804.
  19. Babor TF, Higgins-Biddle JC, Saunders JB, Monteiro MG. The Alcohol Use Disorders Identification Test: Guidelines for use in primary care, second edition. Geneva: World Health Organisation; 2001.
  20. Miller P, Droste N, Baker T, Gervis C. Last drinks: A study of rural emergency department data collection to identify and target community alcohol-related violence. Emergency Medicine Australasia 2015; 27:225–31.
  21. Bars struggle to survive as New Zealand’s drinking culture changes. Fairfax Digital, 2016. 2018, at http://www.stuff.co.nz/business/industries/83000678/is-a-shift-in-new-zealands-drinking-culture-changing-the-hospitality-industry
  22. Morris C. Risks seen in bar regulation. Otago Daily Times 2016 3 October 2016.
  23. MacLean S, Callinan S. “Fourteen Dollars for One Beer!” Pre-drinking is associated with high-risk drinking among Victorian young adults. Australian and New Zealand Journal of Public Health 2013; 37:579–85.
  24. Wells S, Graham K, Purcell J. Policy implications of the widespread practice of ‘pre-drinking’ or ‘pre-gaming’ before going to public drinking establishments—are current prevention strategies backfiring? Addiction 2009; 104:4–9.
  25. Miller P, Droste N, Groot F, et al. Correlates and motives of pre-drinking with intoxication and harm around licensed venues in two cities. Drug and Alcohol Review 2016; 35:177–86.
  26. Havard A, Shakeshaft A, Sanson-Fisher R. Systematic review and meta-analyses of strategies targeting alcohol problems in emergency departments: interventions reduce alcohol-related injuries. Addiction 2008; 103:368–76.
  27. Kypri K. Methodological Issues in Alcohol Screening and Brief Intervention Research. Substance Abuse 2007; 28:31–42.
  28. McGinnes RA, Hutton JE, Weiland TJ, Fatovich DM, Egerton-Warburton D. Review article: Effectiveness of ultra-brief interventions in the emergency department to reduce alcohol consumption: A systematic review. Emergency Medicine Australasia 2016; 28:629–40.
  29. Florence C, Shepherd J, Brennan I, Simon T. Effectiveness of anonymised information sharing and use in health service, police, and local government partnership for preventing violence related injury: experimental study and time series analysis. BMJ 2011; 342.
  30. Droste N, Miller P, Baker T. Emergency department data sharing to reduce alcohol-related violence: A systematic review of the feasibility and effectiveness of community-level interventions. Emergency Medicine Australasia 2014; 26:326–35.