A Need for Tailored Programs and Policies to Reduce Rates of Alcohol-related Crimes for Vulnerable Communities and Young People: An Analysis of Routinely Collected Police Data

A Need for Tailored Programs and Policies to Reduce Rates of Alcohol-related Crimes for... Abstract Background and aims Given ongoing community concern about high rates of alcohol-related crimes (ARCs) experienced by disadvantaged populations, a more specific and nuanced understanding of factors associated with ARCs would help inform the development of more sophisticated programs and policies aimed at reducing ARCs. This study estimates rates of ARCs across all communities in New South Wales (NSW), Australia, using routinely collected police data; investigates whether there are differences between communities; and identifies individual and community characteristics that are significantly associated with higher rates of ARCs. Short summary This study analysed routinely collected police data in New South Wales, Australia, to identify individual and community characteristics associated with alcohol-related crimes. Young people, Aboriginal Australians, socio-economically disadvantaged communities, remote and regional communities and communities with higher per capita rate of on-venue liquor licenses are at risk of alcohol-related crimes. Methods Age standardized rates of ARCs were calculated. A multi-level Poisson regression analysis was conducted to investigate the individual and community factors that were statistically significantly associated with higher rates of ARC, separately for Aboriginal and non-Aboriginal Australians. Results Rates of ARCs were statistically significantly higher for Aboriginal Australians, young people (aged 13–37 years) and on weekends. ARCs varied significantly across communities, and were significantly higher in remote or regional communities, in communities with a higher per capita rate of on-venue licences, and for socio-economically disadvantaged communities for non-Aboriginal Australians, but not for Aboriginal females. Conclusion This analysis shows that the impact of national-level and jurisdictional-level legislation and policies is uneven across communities and defined populations, leaving young people, socio-economically disadvantaged communities and Aboriginal Australians at increased risk of ARCs. To more equitably reduce the exposure of all Australians to ARC, mechanisms that effectively engage vulnerable communities and defined populations, need to be developed in consultation with them, implemented and evaluated. alcohol-related crimes, administrative data, Indigenous, community, socio-economic status, risk factors INTRODUCTION Internationally, a strong relationship exists between single occasion risky drinking and crime (Rehm et al., 2009). This relationship imposes an avoidable economic burden, as it costs the criminal justice system ~25 billion USD per year in the United States, 2.4 billion USD per year in Australia and 7 billion USD per year in the United Kingdom (Manning et al., 2013; Sacks et al., 2010; Lister et al., 2008). Individual characteristics significantly associated with higher rates of alcohol-related crime (ARC) include being young, male, unmarried, and unemployed (Teece and Williams, 2000; Williams, 2001; Palk et al., 2007; Collins, 2016). Community characteristics significantly associated with higher rates of ARC include socio-economic disadvantage, income inequality, increased density of alcohol outlets, and increased remoteness (Breen et al., 2011; Collins, 2016; Gmel et al., 2016). While these characteristics have been identified using data from a select number of communities, there has been no published analysis of whether those characteristics differ between all communities in whole jurisdictions, or between specific populations, such as comparing Indigenous and non-Indigenous Peoples. Indigenous Peoples are known to experience disproportionately higher rates of ARCs than non-Indigenous People, despite efforts to address this (Wundersitz, 2010; Perry, 2004; d’Abbs, 2015). A better understanding of the generalizability of individual and community characteristics that are significantly associated with higher rates of ARC can contribute to more effective efforts aimed at reducing ARCs as they can target the specific characteristics of different communities and defined populations. In order to improve the generalizability of existing data this study has three aims. Firstly, to estimate and specify rates of ARCs by individual, community, and offence characteristics for all communities in New South Wales (NSW), Australia, separately for Aboriginal and non-Aboriginal Australians. Secondly, to investigate whether differences in rates of ARCs between Aboriginal and non-Aboriginal Australians vary across geographical communities. Thirdly, to identify individual and community characteristics significantly associated with rates of ARCs for all communities in NSW separately for Aboriginal and non-Aboriginal Australians. METHODS Ethics The NSW Aboriginal Health and Medical Research Council Ethics Committee (approval no. 987/13) and the NSW Population & Health Services Research Ethics Committee (approval no. 2014/02/516) approved this study. Study design and setting This observational study used NSW Police data. NSW is the most populous state of Australia. It has an estimated population of 7 million, of whom an estimated 2.4% identify as Aboriginal (the term Aboriginal is used in this paper following recommendations from the Aboriginal Health and Medical Research Council for New South Wales (2013)). The majority (72%) of the NSW population resides in major cities, which are defined as having a high accessibility to services and 250,000 or more inhabitants (Australian Bureau of Statistics, 2011). Data sources and definitions Police recorded criminal incidents The NSW Bureau of Crime Statistics and Research (BOCSAR) collects data on all NSW Police recorded criminal incidents. BOCSAR defines a criminal incident as ‘an activity detected by or reported to police which involved the same offender(s) and the same victim(s), occurred at the one location, during one uninterrupted period of time, falls into one offence category (e.g. assault, offensive conduct, and theft) and falls into one incident type (e.g. actual, attempted, conspiracy)’ (NSW BOCSAR, 2016, p. 49). For this study, BOCSAR provided de-identified, unit-level criminal incident data and person of interest (POI) data. A POI is a suspected offender recorded by police in connection with a criminal incident. The POI is not necessarily legally proceeded against. The POI dataset is not composed solely of unique offenders; the same POI can be linked to more than one criminal incident, and one criminal incident can involve multiple POIs. Data comprised all criminal incidents in NSW that occurred from 1 January 2005 to 31 December 2014 and involved POIs aged 13 years and older (NSW BOCSAR, 2015). Criminal incident data included incident number, date and time of incident, postcode of incident location, offence category and subcategory, and whether a criminal incident was flagged by the police as alcohol-related. POI data included incident number, age, gender, postcode of POI, offence category and subcategory, and whether the POI is recorded as non-Aboriginal, Aboriginal and/or Torres Strait Islander, or unknown. Aboriginal status is based on whether the POI has been identified as Aboriginal and/or Torres Strait Islander in any previous contact with the police, either as POI or as victim. POIs with an unknown Aboriginal status were excluded from the analyses (n = 5,725,974; 58%). Criminal incident and POI datasets were combined by the researchers using incident numbers. All incidents had associated POI data. The extent to which NSW Police accurately identify criminal incidents as alcohol-related is unclear, but it is likely that the reported numbers underestimate the true incidence (Wiggers et al., 2016). The intra-rater reliability (the consistency with which a police officer identifies different incidents as alcohol-related) and the inter-rater reliability (the consistency with which different police officers identify incidents as alcohol-related) is also unknown. Given this uncertainty, this study used a proxy measure to identify ARCs (Matthews et al., 2002). Although proxy measures are problematic for quantifying valid estimates of the incidence of ARCs, they are more reliable than police judgements for comparing rates of ARCs between jurisdictions, between communities, or over time (Matthews et al., 2002; Breen et al., 2011). The Australian-derived proxy measure used in this study comprises two components: incident types and time of occurrence (Matthews et al., 2002; Breen et al., 2011). Incident types included are those that have been shown to correlate strongly with alcohol use, specifically: assault (domestic, non-domestic and police assaults), sexual assault (sexual and indecent assaults), disorderly conduct (offensive language and conduct), and malicious damage to property. Times of occurrence are the time periods in which a disproportionately high number of ARCs occur (Sunday 10 p.m.–Monday 6 a.m., Monday 10 p.m.–Tuesday 2 a.m., Wednesday 10 p.m.–Thursday 2 a.m., Friday 10 p.m.–Saturday 6 a.m., and Saturday 6 p.m.–Sunday 6 a.m.). Incidents satisfying both components were coded as ARCs. Individual characteristics The individual characteristics included in the analyses were the Aboriginal status, gender, and age of the POI, as identified in the dataset. Aboriginal status was categorized as Aboriginal or non-Aboriginal. Gender was categorized as male or female. Age was categorized in 5-year categories. The youngest age category started at 13 years, because this was the youngest age available in the data provided by BOCSAR. Community characteristics Community characteristics were obtained for each Postcode Area (POA) in which a criminal incident occurred. Although a variety of geographical units can be used to identify a ‘community’ in Australia, POA was used because postcodes were included in the criminal incident dataset and the community characteristics of interest, obtained from various resources as specified below, were more readily available by POA, compared to other geographical units. Ten postcodes in the criminal incident dataset (n = 53 ARCs) were excluded because no data on community characteristics were available for those postcodes. Community characteristics were selected based on a previous study, investigating community characteristics associated with ARCs in the general population of NSW: Socio-economic status (SES), remoteness, per capita rate of liquor licences, and population characteristics (Breen et al., 2011). SES Each community’s SES was determined using the Index of Relative Socio-economic Disadvantage from the Socio-Economic Indexes for Areas. This index assesses variables that reflect the relative disadvantage of an area compared to the rest of the state (Pink, 2011). Quintile scores were obtained from the Australian Bureau of Statistics using 2011 POA census data: A higher quintile equals more disadvantage (Australian Bureau of Statistics, 2013a). Remoteness Remoteness of each community was determined using the Australian Standard Geographical Classification remoteness structure, which defines remoteness based on road distance to service centres. Researchers grouped all NSW communities into three categories: major city, regional and remote (Australian Bureau of Statistics, 2007). Liquor licences Data on the per capita rate of liquor licences per 1000 population in communities were provided by Liquor and Gaming NSW (2015). Given that different types of liquor licences are related to different types of ARCs (Gmel et al., 2016), distinct licences were grouped into three categories: on-venue licences (e.g. bars, clubs and restaurants), packaged licences (e.g. liquor stores), and hotel licences (which can have both on-venue and packaged licences). Population characteristics For each community, the proportion of the population that identified as Aboriginal and the proportion of young (20–29 years old) male residents were obtained from the Australian Bureau of Statistics using 2011 POA Census data (Australian Bureau of Statistics, 2011). The age range for young males was based on the minimum drinking age (18 years), however, as POA Census data were only available in 5- or 10-year age categories, the minimum age was set at 20 rather than 18 years. Additionally, findings from previous research indicate that young males’ involvement in ARCs declines after 30 years of age (Teece and Williams, 2000; Palk et al., 2007). Offence characteristics The offence categories were limited to those included in the proxy definition: assaults, sexual assaults, disorderly conduct and malicious damage to property. Time of occurrence of the criminal incident identified whether the incident took place during the week or weekend. Statistical analyses Rates of ARCs for Aboriginal and non-Aboriginal Australians in NSW were standardized using the Standard Population for Use in Age-Standardization Table provided by the Australian Bureau of Statistics, which uses 2001 Census data as the most recent standard population available in Australia (2013b). Age-Standardized Rates (ASRs) were estimated separately for males and females for the total number of ARCs and separately for all individual characteristics (Aboriginal identity, gender and age), for the categorical community characteristics (SES, remoteness), and for the offence characteristics (offence category and time of occurrence). A preliminary analysis was conducted using a single-level Poisson model with ARCs as outcome variable that was composed of age, gender, Aboriginal status, and an interaction term of gender and Aboriginal status. This preliminary analysis identified that the effect of Aboriginal status on the rates of ARCs was modified by gender (RR 0.42 (99% CI 0.40; 0.43) P ≤ 0.05). Therefore, all models in this study were stratified by gender. The statistical significance of differences in ASRs for Aboriginal and non-Aboriginal POIs were estimated by calculating rate ratios using single-level Poisson models for individual characteristics (each age category) and categorical community characteristics (SES and remoteness) with ARC as the outcome. To investigate whether differences in the ASRs of ARCs between Aboriginal and non-Aboriginal Australians were statistically significantly different across communities, a random intercept for POA was added to all Poisson models. To identify individual and community characteristics that are significantly associated with the rates of ARCs, separate multi-level Poisson models with rates of ARCs as outcome variable for Aboriginal and non-Aboriginal Australians were estimated. These models comprised individual characteristics (age), community characteristics (SES, remoteness, per capita rate of liquor licences, proportion of young males, and proportion of Aboriginal Australians), and the random intercept for POA. All analyses were conducted using Stata 14 (StataCorp, 2015). Confidence intervals were calculated at the 99% level. RESULTS Rates of ARCs A total of 330,063 ARCs were identified between 2005 and 2014 involving POIs aged 13 and over in NSW (43,590 [13%] involving Aboriginal POIs and 286,473 [87%] involving non-Aboriginal POIs). For individual characteristics, Table 1 shows that the ASRs of ARCs per 1000 were 48.7 (CI 47.8–49.5) for Aboriginal males and 21.9 (CI 20.5–23.2) for Aboriginal females, compared to 10.0 (CI 10.0–10.1) for non-Aboriginal males and 1.8 (CI 1.7–1.8) for non-Aboriginal females. ASRs were highest for 18–22-year olds and second highest for 23–27 year olds among both Aboriginal and non-Aboriginal groups. ARCs progressively reduced as age categories increased. The only exception to these patterns was for non-Aboriginal females, where the second highest ASRs were for 13–17-year olds. Table 1. Age-standardized rates of ARCs in NSW between 2005 and 2014, by Aboriginal status and gender ARCs Aboriginal Non-Aboriginal Male Female Male Female N ASRa (99% CI) N ASRa (99% CI) N ASRa (99% CI) N ASRa (99% CI) Total ARCs 32,032 48.7 (47.8–49.5) 11,558 21.9 (20.5–23.2) 244,738 10.0 (10.0–10.1) 41,735 1.8 (1.7–1.8) Individual characteristics Age of POI  13–17 6464 62.9 (60.9–65.0) 2523 29.7 (29.3–31.3) 32,323 15.7 (15.5–15.9) 7694 4.1 (3.9–4.2)  18–22 8846 107.6 (104.7–110.6) 2920 45.7 (43.5–47.9) 74,930 36.7 (36.4–37.1) 11,031 5.7 (5.6–59)  23–27 5424 69.7 (66.8–72.6) 1822 37.7 (35.5–40.0) 44,875 21.8 (21.5–22.0) 6369 3.1 (3.0–3.2)  28–32 3853 24.7 (22.8–26.6) 1316 34.6 (32.2–37.1) 29,055 13.6 (13.4–13.8) 4502 2.1 (2.0–2.2)  33–37 3023 72.9 (69.6–76.4) 1174 31.1 (28.9–33.5) 22,034 10.4 (10.2–10.6) 4189 2.0 (1.9–2.1)  38–42 2122 46.0 (43.5–48.7) 902 21.4 (19.6–23.3) 16,616 7.4 (7.2–7.5) 3436 1.5 (1.4–1.6)  43–47 1230 31.7 (29.5–34.2) 558 15.4 (13.8–17.2) 11,151 5.3 (5.2–5.4) 2254 1.1 (1.1–1.2)  48–52 639 23.8 (21.4–26.2) 225 9.9 (8.3–11.8) 6602 3.0 (2.9–3.1) 1209 0.6 (0.6–0.7)  53–57 274 15.7 (13.3–18.3) 81 8.2 (6.0–10.8) 3651 1.9 (1.8–2.0) 560 0.4 (0.3–0.4)  58–62 104 10.9 (8.3–14.0) 22 6.3 (3.4–10.7) 1821 1.1 (1.1–1.2) 268 0.3 (0.2–0.3)  63–67 40 10.5 (6.7–15.6) 11 6.0 (2.4–12.5) 887 0.7 (0.6–0.8) 116 0.3 (0.2–0.4)  68+ 13 4.2 (1.8–8.3) 4 7.3 (1.2–22.9) 793 0.3 (0.3–0.3) 107 0.1 (0.0–0.1) Community characteristics SES quintile  1 (least disadvantaged) 835 26.0 (22.3–29.8) 243 15.0 (12.0–18.1) 36,947 6.4 (6.3–6.5) 5269 1.0 (0.9–1.0)  2 2330 33.0 (27.1–38.8) 857 18.3 (16.0–20.6) 45,305 8.5 (5.4–8.6) 7093 1.4 (1.4–1.5)  3 8222 44.8 (43.2–46.4) 2776 18.6 (17.0–20.3) 74,542 13.4 (12.9–13.2) 12,330 2.3 (2.2–2.3)  4 8735 53.1 (46.5–59.7) 3219 23.7 (17.2–30.2) 48,842 13.6 (13.4–13.7) 9186 2.6 (2.5–.27)  5 (most disadvantaged) 11,910 62.4 (60.8–64.0) 4463 25.7 (24.6–26.8) 39,102 10.4 (10.3–10.6) 7857 2.2 (2.1–2.3) Remoteness  Major city 7213 30.6 (29.2–32.0) 2541 13.5 (12.6–14.4) 143,798 8.2 (8.1–8.3) 23,967 1.4 (1.4–1.5)  Regional 18,871 51.5 (50.4–52.7) 6667 22.5 (21.1–24.0) 96,181 15.0 (14.9–15.1) 16,800 2.7 (2.7–2.8)  Remote 5948 119.9 (116.1–123.7) 2350 44.4 (42.0–46.8) 4755 24.4 (23.5–25.3) 968 5.8 (5.2–6.4) Offence characteristics Offence categories  Assaults 17,435 32.4 (31.4–33.4) 6718 14.9 (13.5–16.3) 114,632 4.8 (4.7–4.8) 26,348 1.2 (1.2–1.2)  Sexual assaults 1096 6.0 (5.2–6.8) 28 2.4 (0.6–4.3) 14,794 0.7 (0.7–0.7) 433 0.2 (0.2–0.2)  Disorderly conduct 5903 15.0 (13.5–16.5) 2809 10.4 (8.0–12.8) 58,664 2.7 (2.7–2.7) 6630 0.5 (0.5–0.5)  Malicious damage 7598 14.3 (13.5–15.2) 2003 7.3 (4.8–9.7) 56,648 2.4 (2.4–2.4) 8324 0.5 (0.5–0.5) Time of occurrence  Weekend 22,237 36.2 (35.2–37.2) 8164 17.0 (15.8–18.3) 176,379 7.3 (7.3–7.4) 29,630 1.3 (1.3–1.4)  Week day 9795 19.1 (18.3–19.8) 3394 11.4 (0.5–17.9) 68,359 2.9 (2.9–2.9) 12,105 0.6 (0.6–0.6) ARCs Aboriginal Non-Aboriginal Male Female Male Female N ASRa (99% CI) N ASRa (99% CI) N ASRa (99% CI) N ASRa (99% CI) Total ARCs 32,032 48.7 (47.8–49.5) 11,558 21.9 (20.5–23.2) 244,738 10.0 (10.0–10.1) 41,735 1.8 (1.7–1.8) Individual characteristics Age of POI  13–17 6464 62.9 (60.9–65.0) 2523 29.7 (29.3–31.3) 32,323 15.7 (15.5–15.9) 7694 4.1 (3.9–4.2)  18–22 8846 107.6 (104.7–110.6) 2920 45.7 (43.5–47.9) 74,930 36.7 (36.4–37.1) 11,031 5.7 (5.6–59)  23–27 5424 69.7 (66.8–72.6) 1822 37.7 (35.5–40.0) 44,875 21.8 (21.5–22.0) 6369 3.1 (3.0–3.2)  28–32 3853 24.7 (22.8–26.6) 1316 34.6 (32.2–37.1) 29,055 13.6 (13.4–13.8) 4502 2.1 (2.0–2.2)  33–37 3023 72.9 (69.6–76.4) 1174 31.1 (28.9–33.5) 22,034 10.4 (10.2–10.6) 4189 2.0 (1.9–2.1)  38–42 2122 46.0 (43.5–48.7) 902 21.4 (19.6–23.3) 16,616 7.4 (7.2–7.5) 3436 1.5 (1.4–1.6)  43–47 1230 31.7 (29.5–34.2) 558 15.4 (13.8–17.2) 11,151 5.3 (5.2–5.4) 2254 1.1 (1.1–1.2)  48–52 639 23.8 (21.4–26.2) 225 9.9 (8.3–11.8) 6602 3.0 (2.9–3.1) 1209 0.6 (0.6–0.7)  53–57 274 15.7 (13.3–18.3) 81 8.2 (6.0–10.8) 3651 1.9 (1.8–2.0) 560 0.4 (0.3–0.4)  58–62 104 10.9 (8.3–14.0) 22 6.3 (3.4–10.7) 1821 1.1 (1.1–1.2) 268 0.3 (0.2–0.3)  63–67 40 10.5 (6.7–15.6) 11 6.0 (2.4–12.5) 887 0.7 (0.6–0.8) 116 0.3 (0.2–0.4)  68+ 13 4.2 (1.8–8.3) 4 7.3 (1.2–22.9) 793 0.3 (0.3–0.3) 107 0.1 (0.0–0.1) Community characteristics SES quintile  1 (least disadvantaged) 835 26.0 (22.3–29.8) 243 15.0 (12.0–18.1) 36,947 6.4 (6.3–6.5) 5269 1.0 (0.9–1.0)  2 2330 33.0 (27.1–38.8) 857 18.3 (16.0–20.6) 45,305 8.5 (5.4–8.6) 7093 1.4 (1.4–1.5)  3 8222 44.8 (43.2–46.4) 2776 18.6 (17.0–20.3) 74,542 13.4 (12.9–13.2) 12,330 2.3 (2.2–2.3)  4 8735 53.1 (46.5–59.7) 3219 23.7 (17.2–30.2) 48,842 13.6 (13.4–13.7) 9186 2.6 (2.5–.27)  5 (most disadvantaged) 11,910 62.4 (60.8–64.0) 4463 25.7 (24.6–26.8) 39,102 10.4 (10.3–10.6) 7857 2.2 (2.1–2.3) Remoteness  Major city 7213 30.6 (29.2–32.0) 2541 13.5 (12.6–14.4) 143,798 8.2 (8.1–8.3) 23,967 1.4 (1.4–1.5)  Regional 18,871 51.5 (50.4–52.7) 6667 22.5 (21.1–24.0) 96,181 15.0 (14.9–15.1) 16,800 2.7 (2.7–2.8)  Remote 5948 119.9 (116.1–123.7) 2350 44.4 (42.0–46.8) 4755 24.4 (23.5–25.3) 968 5.8 (5.2–6.4) Offence characteristics Offence categories  Assaults 17,435 32.4 (31.4–33.4) 6718 14.9 (13.5–16.3) 114,632 4.8 (4.7–4.8) 26,348 1.2 (1.2–1.2)  Sexual assaults 1096 6.0 (5.2–6.8) 28 2.4 (0.6–4.3) 14,794 0.7 (0.7–0.7) 433 0.2 (0.2–0.2)  Disorderly conduct 5903 15.0 (13.5–16.5) 2809 10.4 (8.0–12.8) 58,664 2.7 (2.7–2.7) 6630 0.5 (0.5–0.5)  Malicious damage 7598 14.3 (13.5–15.2) 2003 7.3 (4.8–9.7) 56,648 2.4 (2.4–2.4) 8324 0.5 (0.5–0.5) Time of occurrence  Weekend 22,237 36.2 (35.2–37.2) 8164 17.0 (15.8–18.3) 176,379 7.3 (7.3–7.4) 29,630 1.3 (1.3–1.4)  Week day 9795 19.1 (18.3–19.8) 3394 11.4 (0.5–17.9) 68,359 2.9 (2.9–2.9) 12,105 0.6 (0.6–0.6) CI = confidence interval; POI = person of interest; SES = socio-economic status. aAge standardized rates per 1000 population. Table 1. Age-standardized rates of ARCs in NSW between 2005 and 2014, by Aboriginal status and gender ARCs Aboriginal Non-Aboriginal Male Female Male Female N ASRa (99% CI) N ASRa (99% CI) N ASRa (99% CI) N ASRa (99% CI) Total ARCs 32,032 48.7 (47.8–49.5) 11,558 21.9 (20.5–23.2) 244,738 10.0 (10.0–10.1) 41,735 1.8 (1.7–1.8) Individual characteristics Age of POI  13–17 6464 62.9 (60.9–65.0) 2523 29.7 (29.3–31.3) 32,323 15.7 (15.5–15.9) 7694 4.1 (3.9–4.2)  18–22 8846 107.6 (104.7–110.6) 2920 45.7 (43.5–47.9) 74,930 36.7 (36.4–37.1) 11,031 5.7 (5.6–59)  23–27 5424 69.7 (66.8–72.6) 1822 37.7 (35.5–40.0) 44,875 21.8 (21.5–22.0) 6369 3.1 (3.0–3.2)  28–32 3853 24.7 (22.8–26.6) 1316 34.6 (32.2–37.1) 29,055 13.6 (13.4–13.8) 4502 2.1 (2.0–2.2)  33–37 3023 72.9 (69.6–76.4) 1174 31.1 (28.9–33.5) 22,034 10.4 (10.2–10.6) 4189 2.0 (1.9–2.1)  38–42 2122 46.0 (43.5–48.7) 902 21.4 (19.6–23.3) 16,616 7.4 (7.2–7.5) 3436 1.5 (1.4–1.6)  43–47 1230 31.7 (29.5–34.2) 558 15.4 (13.8–17.2) 11,151 5.3 (5.2–5.4) 2254 1.1 (1.1–1.2)  48–52 639 23.8 (21.4–26.2) 225 9.9 (8.3–11.8) 6602 3.0 (2.9–3.1) 1209 0.6 (0.6–0.7)  53–57 274 15.7 (13.3–18.3) 81 8.2 (6.0–10.8) 3651 1.9 (1.8–2.0) 560 0.4 (0.3–0.4)  58–62 104 10.9 (8.3–14.0) 22 6.3 (3.4–10.7) 1821 1.1 (1.1–1.2) 268 0.3 (0.2–0.3)  63–67 40 10.5 (6.7–15.6) 11 6.0 (2.4–12.5) 887 0.7 (0.6–0.8) 116 0.3 (0.2–0.4)  68+ 13 4.2 (1.8–8.3) 4 7.3 (1.2–22.9) 793 0.3 (0.3–0.3) 107 0.1 (0.0–0.1) Community characteristics SES quintile  1 (least disadvantaged) 835 26.0 (22.3–29.8) 243 15.0 (12.0–18.1) 36,947 6.4 (6.3–6.5) 5269 1.0 (0.9–1.0)  2 2330 33.0 (27.1–38.8) 857 18.3 (16.0–20.6) 45,305 8.5 (5.4–8.6) 7093 1.4 (1.4–1.5)  3 8222 44.8 (43.2–46.4) 2776 18.6 (17.0–20.3) 74,542 13.4 (12.9–13.2) 12,330 2.3 (2.2–2.3)  4 8735 53.1 (46.5–59.7) 3219 23.7 (17.2–30.2) 48,842 13.6 (13.4–13.7) 9186 2.6 (2.5–.27)  5 (most disadvantaged) 11,910 62.4 (60.8–64.0) 4463 25.7 (24.6–26.8) 39,102 10.4 (10.3–10.6) 7857 2.2 (2.1–2.3) Remoteness  Major city 7213 30.6 (29.2–32.0) 2541 13.5 (12.6–14.4) 143,798 8.2 (8.1–8.3) 23,967 1.4 (1.4–1.5)  Regional 18,871 51.5 (50.4–52.7) 6667 22.5 (21.1–24.0) 96,181 15.0 (14.9–15.1) 16,800 2.7 (2.7–2.8)  Remote 5948 119.9 (116.1–123.7) 2350 44.4 (42.0–46.8) 4755 24.4 (23.5–25.3) 968 5.8 (5.2–6.4) Offence characteristics Offence categories  Assaults 17,435 32.4 (31.4–33.4) 6718 14.9 (13.5–16.3) 114,632 4.8 (4.7–4.8) 26,348 1.2 (1.2–1.2)  Sexual assaults 1096 6.0 (5.2–6.8) 28 2.4 (0.6–4.3) 14,794 0.7 (0.7–0.7) 433 0.2 (0.2–0.2)  Disorderly conduct 5903 15.0 (13.5–16.5) 2809 10.4 (8.0–12.8) 58,664 2.7 (2.7–2.7) 6630 0.5 (0.5–0.5)  Malicious damage 7598 14.3 (13.5–15.2) 2003 7.3 (4.8–9.7) 56,648 2.4 (2.4–2.4) 8324 0.5 (0.5–0.5) Time of occurrence  Weekend 22,237 36.2 (35.2–37.2) 8164 17.0 (15.8–18.3) 176,379 7.3 (7.3–7.4) 29,630 1.3 (1.3–1.4)  Week day 9795 19.1 (18.3–19.8) 3394 11.4 (0.5–17.9) 68,359 2.9 (2.9–2.9) 12,105 0.6 (0.6–0.6) ARCs Aboriginal Non-Aboriginal Male Female Male Female N ASRa (99% CI) N ASRa (99% CI) N ASRa (99% CI) N ASRa (99% CI) Total ARCs 32,032 48.7 (47.8–49.5) 11,558 21.9 (20.5–23.2) 244,738 10.0 (10.0–10.1) 41,735 1.8 (1.7–1.8) Individual characteristics Age of POI  13–17 6464 62.9 (60.9–65.0) 2523 29.7 (29.3–31.3) 32,323 15.7 (15.5–15.9) 7694 4.1 (3.9–4.2)  18–22 8846 107.6 (104.7–110.6) 2920 45.7 (43.5–47.9) 74,930 36.7 (36.4–37.1) 11,031 5.7 (5.6–59)  23–27 5424 69.7 (66.8–72.6) 1822 37.7 (35.5–40.0) 44,875 21.8 (21.5–22.0) 6369 3.1 (3.0–3.2)  28–32 3853 24.7 (22.8–26.6) 1316 34.6 (32.2–37.1) 29,055 13.6 (13.4–13.8) 4502 2.1 (2.0–2.2)  33–37 3023 72.9 (69.6–76.4) 1174 31.1 (28.9–33.5) 22,034 10.4 (10.2–10.6) 4189 2.0 (1.9–2.1)  38–42 2122 46.0 (43.5–48.7) 902 21.4 (19.6–23.3) 16,616 7.4 (7.2–7.5) 3436 1.5 (1.4–1.6)  43–47 1230 31.7 (29.5–34.2) 558 15.4 (13.8–17.2) 11,151 5.3 (5.2–5.4) 2254 1.1 (1.1–1.2)  48–52 639 23.8 (21.4–26.2) 225 9.9 (8.3–11.8) 6602 3.0 (2.9–3.1) 1209 0.6 (0.6–0.7)  53–57 274 15.7 (13.3–18.3) 81 8.2 (6.0–10.8) 3651 1.9 (1.8–2.0) 560 0.4 (0.3–0.4)  58–62 104 10.9 (8.3–14.0) 22 6.3 (3.4–10.7) 1821 1.1 (1.1–1.2) 268 0.3 (0.2–0.3)  63–67 40 10.5 (6.7–15.6) 11 6.0 (2.4–12.5) 887 0.7 (0.6–0.8) 116 0.3 (0.2–0.4)  68+ 13 4.2 (1.8–8.3) 4 7.3 (1.2–22.9) 793 0.3 (0.3–0.3) 107 0.1 (0.0–0.1) Community characteristics SES quintile  1 (least disadvantaged) 835 26.0 (22.3–29.8) 243 15.0 (12.0–18.1) 36,947 6.4 (6.3–6.5) 5269 1.0 (0.9–1.0)  2 2330 33.0 (27.1–38.8) 857 18.3 (16.0–20.6) 45,305 8.5 (5.4–8.6) 7093 1.4 (1.4–1.5)  3 8222 44.8 (43.2–46.4) 2776 18.6 (17.0–20.3) 74,542 13.4 (12.9–13.2) 12,330 2.3 (2.2–2.3)  4 8735 53.1 (46.5–59.7) 3219 23.7 (17.2–30.2) 48,842 13.6 (13.4–13.7) 9186 2.6 (2.5–.27)  5 (most disadvantaged) 11,910 62.4 (60.8–64.0) 4463 25.7 (24.6–26.8) 39,102 10.4 (10.3–10.6) 7857 2.2 (2.1–2.3) Remoteness  Major city 7213 30.6 (29.2–32.0) 2541 13.5 (12.6–14.4) 143,798 8.2 (8.1–8.3) 23,967 1.4 (1.4–1.5)  Regional 18,871 51.5 (50.4–52.7) 6667 22.5 (21.1–24.0) 96,181 15.0 (14.9–15.1) 16,800 2.7 (2.7–2.8)  Remote 5948 119.9 (116.1–123.7) 2350 44.4 (42.0–46.8) 4755 24.4 (23.5–25.3) 968 5.8 (5.2–6.4) Offence characteristics Offence categories  Assaults 17,435 32.4 (31.4–33.4) 6718 14.9 (13.5–16.3) 114,632 4.8 (4.7–4.8) 26,348 1.2 (1.2–1.2)  Sexual assaults 1096 6.0 (5.2–6.8) 28 2.4 (0.6–4.3) 14,794 0.7 (0.7–0.7) 433 0.2 (0.2–0.2)  Disorderly conduct 5903 15.0 (13.5–16.5) 2809 10.4 (8.0–12.8) 58,664 2.7 (2.7–2.7) 6630 0.5 (0.5–0.5)  Malicious damage 7598 14.3 (13.5–15.2) 2003 7.3 (4.8–9.7) 56,648 2.4 (2.4–2.4) 8324 0.5 (0.5–0.5) Time of occurrence  Weekend 22,237 36.2 (35.2–37.2) 8164 17.0 (15.8–18.3) 176,379 7.3 (7.3–7.4) 29,630 1.3 (1.3–1.4)  Week day 9795 19.1 (18.3–19.8) 3394 11.4 (0.5–17.9) 68,359 2.9 (2.9–2.9) 12,105 0.6 (0.6–0.6) CI = confidence interval; POI = person of interest; SES = socio-economic status. aAge standardized rates per 1000 population. For community characteristics, ASRs of ARC generally increased as communities became more disadvantaged, regardless of Aboriginal status or gender, although they were lower in the most disadvantaged areas (5) than to the second most disadvantaged areas (4) for non-Aboriginal POIs. Furthermore, ASRs increased as communities became more remote. For offence characteristics, assaults were the most commonly recorded ARCs, regardless of Aboriginal status or gender, followed by disorderly conduct and malicious damage, with a relatively small number of sexual assaults. ASRs of ARCs were at least twice as high on weekends than on weekdays, Aboriginal status and gender notwithstanding. Despite similar patterns in the ASRs of ARCs among Aboriginal and non-Aboriginal POIs, and for males and females, the rate ratios presented in Table 2 indicate that ARCs were statistically significantly higher for Aboriginal than for non-Aboriginal POIs. Table 2. Rate ratios of ARCs for Aboriginal compared to non-Aboriginal POIs, stratified by gender resulting from single-level Poisson models Male Female Rate ratioa (99% CI) Rate ratioa (99% CI) Individual characteristics Age of POI  13–17 2.78 (2.67–2.89) 4.63 (4.33–4.95)  18–22 1.97 (1.91–2.04) 4.40 (4.14–4.68)  23–27 2.95 (2.83–3.07) 6.66 (6.15–7.22)  28–32 3.82 (3.64–4.01) 9.18 (8.32–10.12)  33–37 4.33 (4.10–4.58) 8.42 (7.61–9.33)  38–42 3.84 (3.59–4.09) 9.27 (8.27–10.39)  43–47 4.10 (3.76–4.46) 9.77 (8.47–11.26)  48–52 5.90 (5.25–6.63) 14.46 (11.63–17.98)  53–57 6.78 (5.67–8.11) 22.62 (16.20–31.59)  58–62 7.65 (5.69–10.29) 22.77 (12.86–40.32)  63–67 14.22 (8.97–22.91) 21.59 (9.58–48.65)  68+ 12.81 (5.66–28.96) 85.28 (22.97–316.60) Community characteristics SES quintile  1 (least disadvantaged) 5.37 (4.90–5.88) 16.89 (14.18–20.12)  2 4.48 (4.24–4.73) 13.74 (12.48–15.12)  3 4.90 (4.75–5.06) 9.66 (9.11–10.24)  4 4.97 (4.81–5.12) 8.21 (7.75–8.70)  5 (most disadvantaged) 5.38 (5.22–5.55) 7.67 (7.22–8.14) Remoteness  Major city 4.11 (3.98–4.25) 9.14 (8.64–9.67)  Regional 5.30 (5.19–5.41) 9.20 (8.83–9.59)  Remote 6.83 (6.47–7.22) 7.32 (6.53–8.19) Male Female Rate ratioa (99% CI) Rate ratioa (99% CI) Individual characteristics Age of POI  13–17 2.78 (2.67–2.89) 4.63 (4.33–4.95)  18–22 1.97 (1.91–2.04) 4.40 (4.14–4.68)  23–27 2.95 (2.83–3.07) 6.66 (6.15–7.22)  28–32 3.82 (3.64–4.01) 9.18 (8.32–10.12)  33–37 4.33 (4.10–4.58) 8.42 (7.61–9.33)  38–42 3.84 (3.59–4.09) 9.27 (8.27–10.39)  43–47 4.10 (3.76–4.46) 9.77 (8.47–11.26)  48–52 5.90 (5.25–6.63) 14.46 (11.63–17.98)  53–57 6.78 (5.67–8.11) 22.62 (16.20–31.59)  58–62 7.65 (5.69–10.29) 22.77 (12.86–40.32)  63–67 14.22 (8.97–22.91) 21.59 (9.58–48.65)  68+ 12.81 (5.66–28.96) 85.28 (22.97–316.60) Community characteristics SES quintile  1 (least disadvantaged) 5.37 (4.90–5.88) 16.89 (14.18–20.12)  2 4.48 (4.24–4.73) 13.74 (12.48–15.12)  3 4.90 (4.75–5.06) 9.66 (9.11–10.24)  4 4.97 (4.81–5.12) 8.21 (7.75–8.70)  5 (most disadvantaged) 5.38 (5.22–5.55) 7.67 (7.22–8.14) Remoteness  Major city 4.11 (3.98–4.25) 9.14 (8.64–9.67)  Regional 5.30 (5.19–5.41) 9.20 (8.83–9.59)  Remote 6.83 (6.47–7.22) 7.32 (6.53–8.19) CI = confidence interval; POI = person of interest; SES = socio-economic status. aAll presented rate ratios are statistically significant (P ≤ 0.01). Table 2. Rate ratios of ARCs for Aboriginal compared to non-Aboriginal POIs, stratified by gender resulting from single-level Poisson models Male Female Rate ratioa (99% CI) Rate ratioa (99% CI) Individual characteristics Age of POI  13–17 2.78 (2.67–2.89) 4.63 (4.33–4.95)  18–22 1.97 (1.91–2.04) 4.40 (4.14–4.68)  23–27 2.95 (2.83–3.07) 6.66 (6.15–7.22)  28–32 3.82 (3.64–4.01) 9.18 (8.32–10.12)  33–37 4.33 (4.10–4.58) 8.42 (7.61–9.33)  38–42 3.84 (3.59–4.09) 9.27 (8.27–10.39)  43–47 4.10 (3.76–4.46) 9.77 (8.47–11.26)  48–52 5.90 (5.25–6.63) 14.46 (11.63–17.98)  53–57 6.78 (5.67–8.11) 22.62 (16.20–31.59)  58–62 7.65 (5.69–10.29) 22.77 (12.86–40.32)  63–67 14.22 (8.97–22.91) 21.59 (9.58–48.65)  68+ 12.81 (5.66–28.96) 85.28 (22.97–316.60) Community characteristics SES quintile  1 (least disadvantaged) 5.37 (4.90–5.88) 16.89 (14.18–20.12)  2 4.48 (4.24–4.73) 13.74 (12.48–15.12)  3 4.90 (4.75–5.06) 9.66 (9.11–10.24)  4 4.97 (4.81–5.12) 8.21 (7.75–8.70)  5 (most disadvantaged) 5.38 (5.22–5.55) 7.67 (7.22–8.14) Remoteness  Major city 4.11 (3.98–4.25) 9.14 (8.64–9.67)  Regional 5.30 (5.19–5.41) 9.20 (8.83–9.59)  Remote 6.83 (6.47–7.22) 7.32 (6.53–8.19) Male Female Rate ratioa (99% CI) Rate ratioa (99% CI) Individual characteristics Age of POI  13–17 2.78 (2.67–2.89) 4.63 (4.33–4.95)  18–22 1.97 (1.91–2.04) 4.40 (4.14–4.68)  23–27 2.95 (2.83–3.07) 6.66 (6.15–7.22)  28–32 3.82 (3.64–4.01) 9.18 (8.32–10.12)  33–37 4.33 (4.10–4.58) 8.42 (7.61–9.33)  38–42 3.84 (3.59–4.09) 9.27 (8.27–10.39)  43–47 4.10 (3.76–4.46) 9.77 (8.47–11.26)  48–52 5.90 (5.25–6.63) 14.46 (11.63–17.98)  53–57 6.78 (5.67–8.11) 22.62 (16.20–31.59)  58–62 7.65 (5.69–10.29) 22.77 (12.86–40.32)  63–67 14.22 (8.97–22.91) 21.59 (9.58–48.65)  68+ 12.81 (5.66–28.96) 85.28 (22.97–316.60) Community characteristics SES quintile  1 (least disadvantaged) 5.37 (4.90–5.88) 16.89 (14.18–20.12)  2 4.48 (4.24–4.73) 13.74 (12.48–15.12)  3 4.90 (4.75–5.06) 9.66 (9.11–10.24)  4 4.97 (4.81–5.12) 8.21 (7.75–8.70)  5 (most disadvantaged) 5.38 (5.22–5.55) 7.67 (7.22–8.14) Remoteness  Major city 4.11 (3.98–4.25) 9.14 (8.64–9.67)  Regional 5.30 (5.19–5.41) 9.20 (8.83–9.59)  Remote 6.83 (6.47–7.22) 7.32 (6.53–8.19) CI = confidence interval; POI = person of interest; SES = socio-economic status. aAll presented rate ratios are statistically significant (P ≤ 0.01). Community variation in the differences between Aboriginal and non-Aboriginal Australians in rates of ARCs Single-level Poisson models identified that rates of ARCs for Aboriginal POIs were 6.75 (CI 6.67–6.83) and 14.58 (CI 14.29–14.89) times higher than rates for non-Aboriginal POIs, for males and females, respectively. However, these estimates reduced to 4.65 (CI 4.59–4.71) times higher for Aboriginal males and 8.42 (CI 8.22–8.63) times higher for Aboriginal females after adjusting for community variation in the rate ratios of ARCs of Aboriginal and non-Aboriginal POIs. This change in rate ratio provides evidence that differences in rates of ARCs between Aboriginal and non-Aboriginal POIs significantly varied among communities for both males (P ≤ 0.05) and females (P ≤ 0.05). Individual and community characteristics associated with ARCs In terms of individual characteristics, the same pattern of results was observed for Aboriginal and non-Aboriginal POIs: rates of ARCs were statistically significantly higher in age categories younger than the reference group (38–42-year olds) and statistically significantly lower in age categories older than the reference group (Table 3). Table 3. Adjusted rate ratios of ARCs among Aboriginal and non-Aboriginal POIs in NSW, resulting from the multi-level Poisson regression model Aboriginal (N = 43,590) Non-Aboriginal (N = 286,473) Males Females Males Females RRa 99% CI RRa 99% CI RRa 99% CI RRa 99% CI Individual characteristics (age POI)  13–17 1.63** (1.53–1.74) 1.63** (1.47–1.80) 2.20** (2.15–2.26) 2.74** (2.59–2.89)  18–22 2.97** (2.79–3.17) 2.47** (2.24–2.73) 5.12** (5.01–5.24) 3.79** (3.60–4.00)  23–27 2.46** (2.39–2.63) 1.81** (1.62–2.02) 2.78** (2.71–2.85) 1.91** (1.80–2.02)  28–32 2.04** (1.90–2.19) 1.56** (1.39–1.75) 1.68** (1.64–1.73) 1.30** (1.22–1.38)  33–37 1.73** (1.60–1.86) 1.48** (1.39–1.75) 1.37** (1.34–1.41) 1.30** (1.23–1.39)  38–42 (ref) 1.00 1.00 1.00 1.00  43–47 0.69** (0.62–0.75) 0.68** (0.59–0.78) 0.72** (0.69–0.74) 0.73** (0.68–0.78)  48–52 0.44** (0.39–0.50) 0.40** (0.33–0.48) 0.40** (0.39–0.42) 0.39** (0.36–0.43)  53–57 0.26** (0.22–0.31) 0.30** (0.22–0.41) 0.24** (0.23–0.25) 0.22** (0.20–0.25)  58–62 0.19** (0.14–0.24) 0.25** (0.14–0.44) 0.14** (0.13–0.15) 0.15** (0.12–0.17)  63–67 0.16** (0.11–0.25) 0.21** (0.10–0.46) 0.08** (0.07–0.09) 0.15** (0.12–0.19)  68+ 0.05** (0.03–0.11) 0.28* (0.08–1.02) 0.03** (0.03–0.04) 0.05** (0.04–0.06) Community characteristics SES quintile  1 (least disadvantaged) (ref) 1.00 1.00 1.00 1.00  2 1.10 (0.79–1.54) 0.83 (0.54–1.26) 1.13 (0.91–1.39) 1.23* (0.98–1.53)  3 1.40* (1.01–1.95) 0.83 (0.55–1.24) 1.53** (1.25–1.88) 1.66** (1.34–2.07)  4 1.33* (0.94–1.88) 1.05 (0.69–1.61) 1.88** (1.51–2.34) 2.09** (1.66–2.64)  5 (most disadvantaged) 1.32* (0.93–1.88) 0.91 (0.59–1.39) 1.81** (1.45–2.26) 2.08** (1.65–2.63) Remoteness  Major City (ref) 1.00 1.00 1.00  Regional 1.38* (1.05–1.82) 1.51* (1.10–2.06) 1.39** (1.16–1.67) 1.40** (1.14–1.68)  Remote 2.41** (1.41–4.11) 2.10* (1.20–3.67) 1.47* (1.00–2.16) 1.39* (0.91–2.11) Liquor licences  Hotels 0.79** (0.67–0.93) 0.97 (0.78–1.21) 1.04 (0.95–1.45) 1.30** (1.15–1.46)  On-venue 1.18** (1.11–1.25) 1.14* (1.06–1.23) 1.17** (1.13–1.21) 1.12** (1.07–1.17)  Packaged 0.91 (0.81–1.03) 0.92 (0.79–1.07) 0.99 (0.93–1.07) 1.01 (0.93–1.10) Population characteristicsb  Young males 1.03 (0.99–1.08) 1.06* (1.01–1.12) 0.98 (0.95–1.01) 0.97* (0.94–1.00)  Aboriginal 1.03** (1.01–1.05) 1.02* (1.00–1.04) 1.01 (1.00–1.02) 1.03** (1.01–1.04) Aboriginal (N = 43,590) Non-Aboriginal (N = 286,473) Males Females Males Females RRa 99% CI RRa 99% CI RRa 99% CI RRa 99% CI Individual characteristics (age POI)  13–17 1.63** (1.53–1.74) 1.63** (1.47–1.80) 2.20** (2.15–2.26) 2.74** (2.59–2.89)  18–22 2.97** (2.79–3.17) 2.47** (2.24–2.73) 5.12** (5.01–5.24) 3.79** (3.60–4.00)  23–27 2.46** (2.39–2.63) 1.81** (1.62–2.02) 2.78** (2.71–2.85) 1.91** (1.80–2.02)  28–32 2.04** (1.90–2.19) 1.56** (1.39–1.75) 1.68** (1.64–1.73) 1.30** (1.22–1.38)  33–37 1.73** (1.60–1.86) 1.48** (1.39–1.75) 1.37** (1.34–1.41) 1.30** (1.23–1.39)  38–42 (ref) 1.00 1.00 1.00 1.00  43–47 0.69** (0.62–0.75) 0.68** (0.59–0.78) 0.72** (0.69–0.74) 0.73** (0.68–0.78)  48–52 0.44** (0.39–0.50) 0.40** (0.33–0.48) 0.40** (0.39–0.42) 0.39** (0.36–0.43)  53–57 0.26** (0.22–0.31) 0.30** (0.22–0.41) 0.24** (0.23–0.25) 0.22** (0.20–0.25)  58–62 0.19** (0.14–0.24) 0.25** (0.14–0.44) 0.14** (0.13–0.15) 0.15** (0.12–0.17)  63–67 0.16** (0.11–0.25) 0.21** (0.10–0.46) 0.08** (0.07–0.09) 0.15** (0.12–0.19)  68+ 0.05** (0.03–0.11) 0.28* (0.08–1.02) 0.03** (0.03–0.04) 0.05** (0.04–0.06) Community characteristics SES quintile  1 (least disadvantaged) (ref) 1.00 1.00 1.00 1.00  2 1.10 (0.79–1.54) 0.83 (0.54–1.26) 1.13 (0.91–1.39) 1.23* (0.98–1.53)  3 1.40* (1.01–1.95) 0.83 (0.55–1.24) 1.53** (1.25–1.88) 1.66** (1.34–2.07)  4 1.33* (0.94–1.88) 1.05 (0.69–1.61) 1.88** (1.51–2.34) 2.09** (1.66–2.64)  5 (most disadvantaged) 1.32* (0.93–1.88) 0.91 (0.59–1.39) 1.81** (1.45–2.26) 2.08** (1.65–2.63) Remoteness  Major City (ref) 1.00 1.00 1.00  Regional 1.38* (1.05–1.82) 1.51* (1.10–2.06) 1.39** (1.16–1.67) 1.40** (1.14–1.68)  Remote 2.41** (1.41–4.11) 2.10* (1.20–3.67) 1.47* (1.00–2.16) 1.39* (0.91–2.11) Liquor licences  Hotels 0.79** (0.67–0.93) 0.97 (0.78–1.21) 1.04 (0.95–1.45) 1.30** (1.15–1.46)  On-venue 1.18** (1.11–1.25) 1.14* (1.06–1.23) 1.17** (1.13–1.21) 1.12** (1.07–1.17)  Packaged 0.91 (0.81–1.03) 0.92 (0.79–1.07) 0.99 (0.93–1.07) 1.01 (0.93–1.10) Population characteristicsb  Young males 1.03 (0.99–1.08) 1.06* (1.01–1.12) 0.98 (0.95–1.01) 0.97* (0.94–1.00)  Aboriginal 1.03** (1.01–1.05) 1.02* (1.00–1.04) 1.01 (1.00–1.02) 1.03** (1.01–1.04) CI = confidence interval; POI = person of interest; SES = socio-economic status. aAdjusted incidence rate ratios. bPercentage of young males (20–29) and Aboriginal Australians per POA. *P ≤ 0.05; **P ≤ 0.01. Table 3. Adjusted rate ratios of ARCs among Aboriginal and non-Aboriginal POIs in NSW, resulting from the multi-level Poisson regression model Aboriginal (N = 43,590) Non-Aboriginal (N = 286,473) Males Females Males Females RRa 99% CI RRa 99% CI RRa 99% CI RRa 99% CI Individual characteristics (age POI)  13–17 1.63** (1.53–1.74) 1.63** (1.47–1.80) 2.20** (2.15–2.26) 2.74** (2.59–2.89)  18–22 2.97** (2.79–3.17) 2.47** (2.24–2.73) 5.12** (5.01–5.24) 3.79** (3.60–4.00)  23–27 2.46** (2.39–2.63) 1.81** (1.62–2.02) 2.78** (2.71–2.85) 1.91** (1.80–2.02)  28–32 2.04** (1.90–2.19) 1.56** (1.39–1.75) 1.68** (1.64–1.73) 1.30** (1.22–1.38)  33–37 1.73** (1.60–1.86) 1.48** (1.39–1.75) 1.37** (1.34–1.41) 1.30** (1.23–1.39)  38–42 (ref) 1.00 1.00 1.00 1.00  43–47 0.69** (0.62–0.75) 0.68** (0.59–0.78) 0.72** (0.69–0.74) 0.73** (0.68–0.78)  48–52 0.44** (0.39–0.50) 0.40** (0.33–0.48) 0.40** (0.39–0.42) 0.39** (0.36–0.43)  53–57 0.26** (0.22–0.31) 0.30** (0.22–0.41) 0.24** (0.23–0.25) 0.22** (0.20–0.25)  58–62 0.19** (0.14–0.24) 0.25** (0.14–0.44) 0.14** (0.13–0.15) 0.15** (0.12–0.17)  63–67 0.16** (0.11–0.25) 0.21** (0.10–0.46) 0.08** (0.07–0.09) 0.15** (0.12–0.19)  68+ 0.05** (0.03–0.11) 0.28* (0.08–1.02) 0.03** (0.03–0.04) 0.05** (0.04–0.06) Community characteristics SES quintile  1 (least disadvantaged) (ref) 1.00 1.00 1.00 1.00  2 1.10 (0.79–1.54) 0.83 (0.54–1.26) 1.13 (0.91–1.39) 1.23* (0.98–1.53)  3 1.40* (1.01–1.95) 0.83 (0.55–1.24) 1.53** (1.25–1.88) 1.66** (1.34–2.07)  4 1.33* (0.94–1.88) 1.05 (0.69–1.61) 1.88** (1.51–2.34) 2.09** (1.66–2.64)  5 (most disadvantaged) 1.32* (0.93–1.88) 0.91 (0.59–1.39) 1.81** (1.45–2.26) 2.08** (1.65–2.63) Remoteness  Major City (ref) 1.00 1.00 1.00  Regional 1.38* (1.05–1.82) 1.51* (1.10–2.06) 1.39** (1.16–1.67) 1.40** (1.14–1.68)  Remote 2.41** (1.41–4.11) 2.10* (1.20–3.67) 1.47* (1.00–2.16) 1.39* (0.91–2.11) Liquor licences  Hotels 0.79** (0.67–0.93) 0.97 (0.78–1.21) 1.04 (0.95–1.45) 1.30** (1.15–1.46)  On-venue 1.18** (1.11–1.25) 1.14* (1.06–1.23) 1.17** (1.13–1.21) 1.12** (1.07–1.17)  Packaged 0.91 (0.81–1.03) 0.92 (0.79–1.07) 0.99 (0.93–1.07) 1.01 (0.93–1.10) Population characteristicsb  Young males 1.03 (0.99–1.08) 1.06* (1.01–1.12) 0.98 (0.95–1.01) 0.97* (0.94–1.00)  Aboriginal 1.03** (1.01–1.05) 1.02* (1.00–1.04) 1.01 (1.00–1.02) 1.03** (1.01–1.04) Aboriginal (N = 43,590) Non-Aboriginal (N = 286,473) Males Females Males Females RRa 99% CI RRa 99% CI RRa 99% CI RRa 99% CI Individual characteristics (age POI)  13–17 1.63** (1.53–1.74) 1.63** (1.47–1.80) 2.20** (2.15–2.26) 2.74** (2.59–2.89)  18–22 2.97** (2.79–3.17) 2.47** (2.24–2.73) 5.12** (5.01–5.24) 3.79** (3.60–4.00)  23–27 2.46** (2.39–2.63) 1.81** (1.62–2.02) 2.78** (2.71–2.85) 1.91** (1.80–2.02)  28–32 2.04** (1.90–2.19) 1.56** (1.39–1.75) 1.68** (1.64–1.73) 1.30** (1.22–1.38)  33–37 1.73** (1.60–1.86) 1.48** (1.39–1.75) 1.37** (1.34–1.41) 1.30** (1.23–1.39)  38–42 (ref) 1.00 1.00 1.00 1.00  43–47 0.69** (0.62–0.75) 0.68** (0.59–0.78) 0.72** (0.69–0.74) 0.73** (0.68–0.78)  48–52 0.44** (0.39–0.50) 0.40** (0.33–0.48) 0.40** (0.39–0.42) 0.39** (0.36–0.43)  53–57 0.26** (0.22–0.31) 0.30** (0.22–0.41) 0.24** (0.23–0.25) 0.22** (0.20–0.25)  58–62 0.19** (0.14–0.24) 0.25** (0.14–0.44) 0.14** (0.13–0.15) 0.15** (0.12–0.17)  63–67 0.16** (0.11–0.25) 0.21** (0.10–0.46) 0.08** (0.07–0.09) 0.15** (0.12–0.19)  68+ 0.05** (0.03–0.11) 0.28* (0.08–1.02) 0.03** (0.03–0.04) 0.05** (0.04–0.06) Community characteristics SES quintile  1 (least disadvantaged) (ref) 1.00 1.00 1.00 1.00  2 1.10 (0.79–1.54) 0.83 (0.54–1.26) 1.13 (0.91–1.39) 1.23* (0.98–1.53)  3 1.40* (1.01–1.95) 0.83 (0.55–1.24) 1.53** (1.25–1.88) 1.66** (1.34–2.07)  4 1.33* (0.94–1.88) 1.05 (0.69–1.61) 1.88** (1.51–2.34) 2.09** (1.66–2.64)  5 (most disadvantaged) 1.32* (0.93–1.88) 0.91 (0.59–1.39) 1.81** (1.45–2.26) 2.08** (1.65–2.63) Remoteness  Major City (ref) 1.00 1.00 1.00  Regional 1.38* (1.05–1.82) 1.51* (1.10–2.06) 1.39** (1.16–1.67) 1.40** (1.14–1.68)  Remote 2.41** (1.41–4.11) 2.10* (1.20–3.67) 1.47* (1.00–2.16) 1.39* (0.91–2.11) Liquor licences  Hotels 0.79** (0.67–0.93) 0.97 (0.78–1.21) 1.04 (0.95–1.45) 1.30** (1.15–1.46)  On-venue 1.18** (1.11–1.25) 1.14* (1.06–1.23) 1.17** (1.13–1.21) 1.12** (1.07–1.17)  Packaged 0.91 (0.81–1.03) 0.92 (0.79–1.07) 0.99 (0.93–1.07) 1.01 (0.93–1.10) Population characteristicsb  Young males 1.03 (0.99–1.08) 1.06* (1.01–1.12) 0.98 (0.95–1.01) 0.97* (0.94–1.00)  Aboriginal 1.03** (1.01–1.05) 1.02* (1.00–1.04) 1.01 (1.00–1.02) 1.03** (1.01–1.04) CI = confidence interval; POI = person of interest; SES = socio-economic status. aAdjusted incidence rate ratios. bPercentage of young males (20–29) and Aboriginal Australians per POA. *P ≤ 0.05; **P ≤ 0.01. In terms of community characteristics, more socio-economic disadvantage in the community was associated with higher rates of ARCs for non-Aboriginal POIs and Aboriginal males, but not for Aboriginal females. For both Aboriginal and non-Aboriginal POIs, rates of ARCs were higher in remote and regional communities than in major cities. A higher per capita rate of on-venue licences within a community was significantly associated with higher rates of ARCs, regardless of Aboriginal status. For non-Aboriginal females and Aboriginal males, a higher per capita rate of hotel licences in the community was also associated with higher rates of ARCs. For non-Aboriginal females, a higher proportion of young males in the community was significantly associated with lower rates of ARCs. For Aboriginal females a higher proportion of young males was associated with higher rates of ARCs. A small, but statistically significant, association was found between a higher proportion of residents in a community identifying as Aboriginal and a higher rate of ARCs, regardless of the Aboriginal status of the POI, except among non-Aboriginal males, where this was not statistically significant. Nevertheless, the marginal nature of this association (e.g. a 1% increase in Aboriginal Australians in a community lead to a 3% increase in ARCs for Aboriginal males), suggests its statistical significance is a consequence of the large sample size, rather than a practically meaningful association. DISCUSSION This study found that age standardized rates of ARCs were statistically significantly higher for Aboriginal POIs, young people (aged 13–37 years) and on weekends. Differences in the rates of ARCs for Aboriginal and non-Aboriginal POIs varied significantly across communities. ARC rates were significantly higher in remote or regional communities, in communities with a higher per capita rate of on-venue licences, and for more socio-economically disadvantaged communities. For national- or jurisdictional-level policy frameworks, these findings highlight the need to ensure legislation continues to target young people, such as the enforcement of minimum legal drinking age and potentially further consideration of increasing the legal minimum drinking age, especially considering those aged 18–22 had the highest rates of ARCs (Cobiac et al., 2009). Minimum Unit Pricing (MUP) policies are an additional policy strategy that has shown to reduce alcohol consumption and related assaults in countries such as Canada, United Kingdom and Australia (Boniface et al., 2017). However, more evidence is required for the evidence of MUP for sub-groups, such as Indigenous Peoples and young people (Seaman et al., 2013). Similarly, the importance of continuing to target alcohol consumption on weekends is also highlighted by these findings. Currently, policies restricting night time alcohol sales in central entertainment districts have been successfully implemented in cities in Norway, California and Australia (Kypri, 2015). Evaluation of these liquor licensing restrictions in Australia and Norway has shown significant reductions in weekend night time assaults (Menéndez et al., 2017; Rossow and Norström, 2012). Even though all communities included in this study are subject to the same jurisdictional-level framework, there are unique community characteristics that contributed to higher rates of ARCs. First, rates of ARCs were higher in communities with a higher per capita rate of on-venue licences and hotel licenses for Aboriginal males and non-Aboriginal females. Currently, there is no official threshold for the number of liquor licenses allowed in any one community in NSW. Applications for new licenses have to provide a community impact statement and local council feedback, which covers the number of licenses in their community. While these factors are considered in the decision-making process by the NSW government office of Liquor and Gaming, implementing a jurisdiction-wide threshold to cap the density of alcohol licences in any one community may help reduce the variability in rates of ARCs across different communities. Second, rates of ARC were significantly higher in remote and regional communities and in lower SES communities, both of which have been previously associated with higher rates of alcohol harms (Australian Institute for Health and Welfare, 2015; Collins, 2016). Given the impracticality of legislative approaches for communities based on their geographical location or SES, this finding highlights the need to develop a mechanism that effectively engages with vulnerable communities to assist them to effectively reduce their rates of ARC. Previous research has identified multi-component approaches that can effectively help regional communities reduce ARCs, such as increasing licensees’ awareness and police activity (Shakeshaft et al., 2014). In addition to benefiting these high-risk communities, such a process would have jurisdictional-level benefits given the resources expended in responding to ARC, such as policing and the provision of emergency and health care services following serious assaults, are met at the jurisdictional-level rather than by local councils. Despite the finding that the risk factors for ARC were mostly the same for Aboriginal and non-Aboriginal Australians, this study found rates of ARCs were significantly higher for Aboriginal Australians. Two possible explanations for this are: Aboriginal status is being confounded with other factors, and/or there are characteristics not measured in this study that are specific to Aboriginal Australians that increase their risk of experiencing ARC. The findings of this analysis are consistent with the possibility of confounding by remoteness, because both the proportion of Aboriginal Australians and rates of ARCs increase with greater remoteness (Australian Bureau of Statistics and Australian Institute of Health and Welfare, 2008; Gmel et al., 2016). They are also consistent with the possibility of confounding by socio-economic status: although rates of ARC were significantly higher for more socio-economically disadvantaged communities, this was not true for Aboriginal females and less pronounced for Aboriginal males which most likely reflects the high level of disadvantaged experienced by Aboriginal Australians, relative to non-Aboriginal Australians (Australian Bureau of Statistics and Australian Institute of Health and Welfare, 2008). Given this study did not identify any characteristics uniquely associated with ARCs for Aboriginal Australians, it is unlikely that the variables included in this study explain the differences in rates of ARC between Aboriginal and non-Aboriginal Australians. Furthermore, a study by Wundersitz (2010) identified similar factors associated with (non-alcohol-related) violent offending by Aboriginal Australians as our study identified for alcohol-related crimes for Aboriginal Australians, including males, young people and remoteness of community. Additionally she identified childhood abuse and alcohol abuse as predictors of violent offending, with the latter being the number one predictor. Given there are similar factors associated with violent offending and ARCs and similar characteristics for Aboriginal and non-Aboriginal Australians, it is likely that there are other, fundamental drivers to violence among Aboriginal Australians, whether alcohol-related or not. Previous studies have identified that issues regarding race, ethnicity and history can influence rates of ARCs, such as experience of racism or transgenerational trauma resulting from the history of colonization and dispossession (Wundersitz, 2010; Cunningham and Paradies, 2013; Osborne et al., 2013). This highlights the need to develop specific strategies for Aboriginal communities that are devised and implemented by the community themselves and evaluated in partnership with researchers. Particularly strategies that support Aboriginal communities to empower themselves to address ARCs can be promising to reduce ARCs (Tsey, 2008). More work is needed in this space, as was highlighted in a systematic review of Australian Aboriginal community-based studies (Snijder et al., 2015). Only 31 evaluations were published between 1990 and 2015 (one addressed substance use and related harms), with only a moderate level of participation of Indigenous Peoples, especially in the design and evaluation phases of these studies (Snijder et al., 2015). Methodological considerations There are several methodological limitations due to the use of police data that are not collected for research purposes. First, the Aboriginal status for 58% of POIs in the dataset was recorded as unknown, these POIs were excluded from the analyses because it is unclear whether Aboriginal people are more or less likely to be recorded as unknown in police incident data (Thompson et al., 2012). Second, while the proxy measure for ARCs has not been explicitly validated for use with Aboriginal Australians, the proxy has been validated with the NSW population, which included Aboriginal Australians (Breen et al., 2011). Currently, no published research indicates that patterns of ARCs differ between Aboriginal and non-Aboriginal Australians. The findings of this study suggest that these patterns might be very similar and an additional analysis of the times at which ARCs flagged by the police occur also indicated that times are similar for Aboriginal and non-Aboriginal Australians. Regardless, the validation of the proxy measure with Aboriginal populations would help improve future estimates for Aboriginal Australians (Chikritzhs et al., 2004). Third, there are potential biases that influence the police recorded data, such as institutional racism (Cunneen, 2006; Wundersitz, 2010). Previous studies identified that 40% of Aboriginal Australians reported to have experienced racism within the legal system, possibly leading to an artefactual overrepresentation of Aboriginal Australians in police records (Cunneen, 2006; Cunningham and Paradies, 2013). Fourth, the use of POA is not a complete measure of communities; ten POAs in this study were not included in POA classification, either because they covered two or more areas, or only partially covered one area (Australian Bureau of Statistics, 2012). Despite its limitations, POA was considered the most appropriate classification to use, because of the availability of community characteristics data by POA and the availability of postcodes in the datasets containing ARCs. Finally, the possibility of reoffending makes it not possible to identify whether characteristics of POIs contribute to a higher proportion of these groups involved in ARCs, or whether the group has more repeat offenders. The findings of this study can nevertheless identify at risk groups that can benefit from targeted policies or programs. CONCLUSION In 2016, The Lancet published an Indigenous-led paper that called for governments internationally to improve Indigenous health by developing targeted strategies that are grounded in data collected in national surveillance systems (Anderson et al., 2016). The current study operationalized this by utilizing routinely collected police data to identify that there is a need for programs that are tailored to vulnerable people and communities, specifically young people, socio-economically disadvantaged and Aboriginal communities. Current best-evidence suggests such programs are likely to be most effective if they are led by communities themselves, and directly respond to issues of transgenerational trauma, racism and social exclusion, in addition to ARC (Gray and Wilkes, 2011). The lack of published evaluations of such programs highlights the need to develop a practical framework on which partnerships between communities and researchers can be developed (Snijder et al., 2015). ACKNOWLEDGEMENTS The authors thank the NSW Ministry of Health for funding the body of research of which this study is one part. Mieke Snijder thanks the NDARC Education Trustees for funding her doctoral research, which includes this paper. The authors thank BOCSAR for providing the data. FUNDING This study was funded by the NSW Ministry of Health as part of the ‘Aboriginal Injury Prevention and Safety Promotion demonstration Grants Program’. M.S. was funded by a postgraduate scholarship from the NDARC Educationan Trustees. CONFLICT OF INTEREST STATEMENT None declared. REFERENCES Aboriginal Health and Medical Research Council . ( 2013 ) AH&MRC Guidelines For Research Into Aboriginal Health: Key Principles . Sydney : Aboriginal Health and Medical Research Council of New South Wales . Anderson I , Robson B , Connolly M , et al. . ( 2016 ) Indigenous and tribal peoples’ health (The Lancet-Lowitja Institute Global Collaboration): a population study . Lancet 388 : 131 – 57 . Google Scholar CrossRef Search ADS PubMed Australian Bureau of Statistics . ( 2007 ) Australian Standard Geographical Classification (ASGC) Remoteness Area Correspondences, 2006 . Canberra : Australian Bureau of Statistics . Australian Bureau of Statistics . ( 2011 ) Census of Population and Housing . Canberra : Commonwealth of Australia . Australian Bureau of Statistics . ( 2012 ) Postcodes & Postal Areas [Online]. Canberra. Available: http://www.abs.gov.au/websitedbs/censushome.nsf/home/factsheetspoa?opendocument&navpos=450 (16 May 2018 date last accessed). Australian Bureau of Statistics . ( 2013 a) 2033.0.55.001 – Census of Population and Housing: Socio-Economic Indexes for Areas (SEIFA), Australia, 2011 . Canberra : Australian Bureau of Statistics . Australian Bureau of Statistics . ( 2013 b) Standard Population for Use in Age-Standardisation . Canberra : Australian Bureau of Statistics . Australian Bureau of Statistics and Australian Institute of Health and Welfare . ( 2008 ) The Health and Welfare of Australia’s Indigenous and Torres Strait Islander Peoples . Canberra : Commonwealth of Australia . Australian Institute for Health and Welfare . ( 2015 ) The Health and Welfare of Australia’s Aboriginal and Torres Strait Islander Peoples: 2015 . Canberra : AIHW . Boniface S , Scannell JW , Marlow S . ( 2017 ) Evidence for the effectiveness of minimum pricing of alcohol: a systematic review and assessment using the Bradford Hill criteria for causality . BMJ Open 7 : e013497 . Google Scholar CrossRef Search ADS PubMed Breen C , Shakeshaft A , Slade T , et al. . ( 2011 ) Do community characteristics predict alcohol-related crime? Alcohol Alcohol 46 : 464 – 70 . Google Scholar CrossRef Search ADS PubMed Chikritzhs T , Gray D , Stockwell T , et al. . ( 2004 ) Applying National Indicators of Alcohol-related Harms to Indigenous Australians: A Discussion Paper . Perth : National Drug Research Institute, Curtin University of Technology . Cobiac L , Vos T , Doran C , et al. . ( 2009 ) Cost‐effectiveness of interventions to prevent alcohol‐related disease and injury in Australia . Addiction 104 : 1646 – 55 . Google Scholar CrossRef Search ADS PubMed Collins SE . ( 2016 ) Associations between socioeconomic factors and alcohol outcomes . Alcohol Res 38 : 83 – 94 . Google Scholar PubMed Cunneen C . ( 2006 ) Racism, discrimination and the over-representation of Indigenous people in the criminal justice system: some conceptual and explanatory issues . Curr Issues Criminal Justice 17 : 329 – 46 . Cunningham J , Paradies YC . ( 2013 ) Patterns and correlates of self-reported racial discrimination among Australian Aboriginal and Torres Strait Islander adults, 2008–09: analysis of national survey data . Int J Equity Health 12 : 47 . Google Scholar CrossRef Search ADS PubMed d’abbs P . ( 2015 ) Widening the gap: the gulf between policy rhetoric and implementation reality in addressing alcohol problems among Indigenous Australians . Drug Alcohol Rev 34 : 461 – 6 . Google Scholar CrossRef Search ADS Gmel G , Holmes J , Studer J . ( 2016 ) Are alcohol outlet densities strongly associated with alcohol-related outcomes? A critical review of recent evidence . Drug Alcohol Rev 35 : 40 – 54 . Google Scholar CrossRef Search ADS Gray D , Wilkes ET . ( 2011 ) Alcohol restrictions in Indigenous communities: an effective strategy if Indigenous -led . Med J Aus 194 : 508 . Kypri K . ( 2015 ) Evidence of harm from late night alcohol sales continues to strengthen . Addiction 110 : 965 – 6 . Google Scholar CrossRef Search ADS PubMed Lister G , Mcvey D , French J , et al. . ( 2008 ) Measuring the societal impact of behavior choices . Soc Mar Q 14 : 51 – 62 . Google Scholar CrossRef Search ADS Manning M , Smith C , Mazerolle P . ( 2013 ) The societal costs of alcohol misuse in Australia . Trends Issues crime Criminal Justice 545 : 1 – 6 . Matthews S , Chikritzhs T , Catalano P , et al. . ( 2002 ) Trends in Alcohol Related Violence in Australia, 1991/92–1999/00 . Perth : Curtin University of Technology . Menéndez P , Kypri K , Weatherburn D . ( 2017 ) The effect of liquor licensing restrictions on assault: a quasi-experimental study in Sydney, Australia . Addiction 112 : 261 – 8 . Google Scholar CrossRef Search ADS PubMed NSW Bureau of Crime Statistics and Research . ( 2015 ) Dg1513103 . Sydney : BOCSAR . NSW Bureau of Crime Statistics and Research . 2016 . Crime Statistics: Definitions and Explanations: Incidents [Online]. Available: http://www.bocsar.nsw.gov.au/Pages/bocsar_crime_stats/bocsar_glossary.aspx#incidents [16 May 2018, date last accessed]. NSW Office of Liquor Gaming and Racing . ( 2015 ) Liquor Premises List 01/10/2015 . Sydney : NSW Office of Liquor, Gambling and Racing . Osborne K , Baum F , Brown L . ( 2013 ) What Works? A Review of Actions Addressing the Social and Economic Determinants of Indigenous Health . Canberra/Melbourne : Australian Institute for Health and Welfare/Australian Institute for Family Studies . Palk G , Davey J , Freeman J . ( 2007 ) Prevalence and characteristics of alcohol-related incidents requiring police attendance . J Stud Alcohol Drugs 68 : 575 – 81 . Google Scholar CrossRef Search ADS PubMed Perry SW . ( 2004 ) American Indians and Crime: A BJS Statistical Profile, 1992–2002 . Washington, DC : Department of Justice, Bureau of Justice Statistics . Pink B . ( 2011 ) Socio-Economic Indexes for Areas (SEIFA). Technical Paper. Canberra: Australian Bureau of Statistics. Rehm J , Mathers C , Popova S , et al. . ( 2009 ) Global burden of disease and injury and economic cost attributable to alcohol use and alcohol-use disorders . Lancet 373 : 2223 – 33 . Google Scholar CrossRef Search ADS PubMed Rossow I , Norström T . ( 2012 ) The impact of small changes in bar closing hours on violence. The Norwegian experience from 18 cities . Addiction 107 : 530 – 7 . Google Scholar CrossRef Search ADS PubMed Sacks JJ , Gonzales KR , Bouchery EE , et al. . ( 2010 ) National and state costs of excessive alcohol consumption . Am J Prev Med 49 : e73 – 9 . Google Scholar CrossRef Search ADS Seaman P , Edgar F , Ikegwuonu T . ( 2013 ) The role of alcohol price in young adult drinking cultures in Scotland . Drugs (Abingdon Engl) 20 : 278 – 85 . Google Scholar PubMed Shakeshaft A , Doran C , Petrie D , et al. . ( 2014 ) The effectiveness of community action in reducing risky alcohol consumption and harm: a cluster randomised controlled trial . PLoS Med 11 : e1001617 . Google Scholar CrossRef Search ADS PubMed Snijder M , Shakeshaft A , Wagemakers A , et al. . ( 2015 ) A systematic review of studies evaluating Australian indigenous community development projects: the extent of community participation, their methodological quality and their outcomes . BMC Public Health 15 : 1154 . Google Scholar CrossRef Search ADS PubMed StataCorp . ( 2015 ) Stata Statistical Software: Release 14 . College Station, TX : StataCorp LP . Teece M , Williams P . ( 2000 ) Alcohol-related assault: time and place . Trends Issues Crime Criminal Justice 169 : 1 – 6 . Thompson SC , Woods JA , Katzenellenbogen JM . ( 2012 ) The quality of Indigenous identification in administrative health data in Australia: insights from studies using data linkage . BMC Med Inform Decis Mak 12 : 133 – 43 . Google Scholar CrossRef Search ADS PubMed Tsey K . ( 2008 ) The control factor: a neglected social determinant of health . Lancet 372 : 1629 . Google Scholar CrossRef Search ADS PubMed Wiggers JH , Hacker A , Kingsland M , et al. . ( 2016 ) Facilitating police recording of the alcohol-related characteristics of assault incidents: a stepped wedge implementation trial . Drug Alcohol Rev 35 : 30 – 9 . Google Scholar CrossRef Search ADS Williams P . ( 2001 ) Alcohol, Young People and Violence . Canberra : Australian Institute of Criminology . Wundersitz J . ( 2010 ) Indigenous Perpetrators of Violence: Prevalence and Risk Factors for Offending . Canberra : Australian Institute of Criminology . © The Author(s) 2018. Medical Council on Alcohol and Oxford University Press. All rights reserved. This article is published and distributed under the terms of the Oxford University Press, Standard Journals Publication Model (https://academic.oup.com/journals/pages/about_us/legal/notices) http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png Alcohol and Alcoholism Oxford University Press

A Need for Tailored Programs and Policies to Reduce Rates of Alcohol-related Crimes for Vulnerable Communities and Young People: An Analysis of Routinely Collected Police Data

Loading next page...
 
/lp/ou_press/a-need-for-tailored-programs-and-policies-to-reduce-rates-of-alcohol-md00HxIjey
Publisher
Oxford University Press
Copyright
© The Author(s) 2018. Medical Council on Alcohol and Oxford University Press. All rights reserved.
ISSN
0735-0414
eISSN
1464-3502
D.O.I.
10.1093/alcalc/agy034
Publisher site
See Article on Publisher Site

Abstract

Abstract Background and aims Given ongoing community concern about high rates of alcohol-related crimes (ARCs) experienced by disadvantaged populations, a more specific and nuanced understanding of factors associated with ARCs would help inform the development of more sophisticated programs and policies aimed at reducing ARCs. This study estimates rates of ARCs across all communities in New South Wales (NSW), Australia, using routinely collected police data; investigates whether there are differences between communities; and identifies individual and community characteristics that are significantly associated with higher rates of ARCs. Short summary This study analysed routinely collected police data in New South Wales, Australia, to identify individual and community characteristics associated with alcohol-related crimes. Young people, Aboriginal Australians, socio-economically disadvantaged communities, remote and regional communities and communities with higher per capita rate of on-venue liquor licenses are at risk of alcohol-related crimes. Methods Age standardized rates of ARCs were calculated. A multi-level Poisson regression analysis was conducted to investigate the individual and community factors that were statistically significantly associated with higher rates of ARC, separately for Aboriginal and non-Aboriginal Australians. Results Rates of ARCs were statistically significantly higher for Aboriginal Australians, young people (aged 13–37 years) and on weekends. ARCs varied significantly across communities, and were significantly higher in remote or regional communities, in communities with a higher per capita rate of on-venue licences, and for socio-economically disadvantaged communities for non-Aboriginal Australians, but not for Aboriginal females. Conclusion This analysis shows that the impact of national-level and jurisdictional-level legislation and policies is uneven across communities and defined populations, leaving young people, socio-economically disadvantaged communities and Aboriginal Australians at increased risk of ARCs. To more equitably reduce the exposure of all Australians to ARC, mechanisms that effectively engage vulnerable communities and defined populations, need to be developed in consultation with them, implemented and evaluated. alcohol-related crimes, administrative data, Indigenous, community, socio-economic status, risk factors INTRODUCTION Internationally, a strong relationship exists between single occasion risky drinking and crime (Rehm et al., 2009). This relationship imposes an avoidable economic burden, as it costs the criminal justice system ~25 billion USD per year in the United States, 2.4 billion USD per year in Australia and 7 billion USD per year in the United Kingdom (Manning et al., 2013; Sacks et al., 2010; Lister et al., 2008). Individual characteristics significantly associated with higher rates of alcohol-related crime (ARC) include being young, male, unmarried, and unemployed (Teece and Williams, 2000; Williams, 2001; Palk et al., 2007; Collins, 2016). Community characteristics significantly associated with higher rates of ARC include socio-economic disadvantage, income inequality, increased density of alcohol outlets, and increased remoteness (Breen et al., 2011; Collins, 2016; Gmel et al., 2016). While these characteristics have been identified using data from a select number of communities, there has been no published analysis of whether those characteristics differ between all communities in whole jurisdictions, or between specific populations, such as comparing Indigenous and non-Indigenous Peoples. Indigenous Peoples are known to experience disproportionately higher rates of ARCs than non-Indigenous People, despite efforts to address this (Wundersitz, 2010; Perry, 2004; d’Abbs, 2015). A better understanding of the generalizability of individual and community characteristics that are significantly associated with higher rates of ARC can contribute to more effective efforts aimed at reducing ARCs as they can target the specific characteristics of different communities and defined populations. In order to improve the generalizability of existing data this study has three aims. Firstly, to estimate and specify rates of ARCs by individual, community, and offence characteristics for all communities in New South Wales (NSW), Australia, separately for Aboriginal and non-Aboriginal Australians. Secondly, to investigate whether differences in rates of ARCs between Aboriginal and non-Aboriginal Australians vary across geographical communities. Thirdly, to identify individual and community characteristics significantly associated with rates of ARCs for all communities in NSW separately for Aboriginal and non-Aboriginal Australians. METHODS Ethics The NSW Aboriginal Health and Medical Research Council Ethics Committee (approval no. 987/13) and the NSW Population & Health Services Research Ethics Committee (approval no. 2014/02/516) approved this study. Study design and setting This observational study used NSW Police data. NSW is the most populous state of Australia. It has an estimated population of 7 million, of whom an estimated 2.4% identify as Aboriginal (the term Aboriginal is used in this paper following recommendations from the Aboriginal Health and Medical Research Council for New South Wales (2013)). The majority (72%) of the NSW population resides in major cities, which are defined as having a high accessibility to services and 250,000 or more inhabitants (Australian Bureau of Statistics, 2011). Data sources and definitions Police recorded criminal incidents The NSW Bureau of Crime Statistics and Research (BOCSAR) collects data on all NSW Police recorded criminal incidents. BOCSAR defines a criminal incident as ‘an activity detected by or reported to police which involved the same offender(s) and the same victim(s), occurred at the one location, during one uninterrupted period of time, falls into one offence category (e.g. assault, offensive conduct, and theft) and falls into one incident type (e.g. actual, attempted, conspiracy)’ (NSW BOCSAR, 2016, p. 49). For this study, BOCSAR provided de-identified, unit-level criminal incident data and person of interest (POI) data. A POI is a suspected offender recorded by police in connection with a criminal incident. The POI is not necessarily legally proceeded against. The POI dataset is not composed solely of unique offenders; the same POI can be linked to more than one criminal incident, and one criminal incident can involve multiple POIs. Data comprised all criminal incidents in NSW that occurred from 1 January 2005 to 31 December 2014 and involved POIs aged 13 years and older (NSW BOCSAR, 2015). Criminal incident data included incident number, date and time of incident, postcode of incident location, offence category and subcategory, and whether a criminal incident was flagged by the police as alcohol-related. POI data included incident number, age, gender, postcode of POI, offence category and subcategory, and whether the POI is recorded as non-Aboriginal, Aboriginal and/or Torres Strait Islander, or unknown. Aboriginal status is based on whether the POI has been identified as Aboriginal and/or Torres Strait Islander in any previous contact with the police, either as POI or as victim. POIs with an unknown Aboriginal status were excluded from the analyses (n = 5,725,974; 58%). Criminal incident and POI datasets were combined by the researchers using incident numbers. All incidents had associated POI data. The extent to which NSW Police accurately identify criminal incidents as alcohol-related is unclear, but it is likely that the reported numbers underestimate the true incidence (Wiggers et al., 2016). The intra-rater reliability (the consistency with which a police officer identifies different incidents as alcohol-related) and the inter-rater reliability (the consistency with which different police officers identify incidents as alcohol-related) is also unknown. Given this uncertainty, this study used a proxy measure to identify ARCs (Matthews et al., 2002). Although proxy measures are problematic for quantifying valid estimates of the incidence of ARCs, they are more reliable than police judgements for comparing rates of ARCs between jurisdictions, between communities, or over time (Matthews et al., 2002; Breen et al., 2011). The Australian-derived proxy measure used in this study comprises two components: incident types and time of occurrence (Matthews et al., 2002; Breen et al., 2011). Incident types included are those that have been shown to correlate strongly with alcohol use, specifically: assault (domestic, non-domestic and police assaults), sexual assault (sexual and indecent assaults), disorderly conduct (offensive language and conduct), and malicious damage to property. Times of occurrence are the time periods in which a disproportionately high number of ARCs occur (Sunday 10 p.m.–Monday 6 a.m., Monday 10 p.m.–Tuesday 2 a.m., Wednesday 10 p.m.–Thursday 2 a.m., Friday 10 p.m.–Saturday 6 a.m., and Saturday 6 p.m.–Sunday 6 a.m.). Incidents satisfying both components were coded as ARCs. Individual characteristics The individual characteristics included in the analyses were the Aboriginal status, gender, and age of the POI, as identified in the dataset. Aboriginal status was categorized as Aboriginal or non-Aboriginal. Gender was categorized as male or female. Age was categorized in 5-year categories. The youngest age category started at 13 years, because this was the youngest age available in the data provided by BOCSAR. Community characteristics Community characteristics were obtained for each Postcode Area (POA) in which a criminal incident occurred. Although a variety of geographical units can be used to identify a ‘community’ in Australia, POA was used because postcodes were included in the criminal incident dataset and the community characteristics of interest, obtained from various resources as specified below, were more readily available by POA, compared to other geographical units. Ten postcodes in the criminal incident dataset (n = 53 ARCs) were excluded because no data on community characteristics were available for those postcodes. Community characteristics were selected based on a previous study, investigating community characteristics associated with ARCs in the general population of NSW: Socio-economic status (SES), remoteness, per capita rate of liquor licences, and population characteristics (Breen et al., 2011). SES Each community’s SES was determined using the Index of Relative Socio-economic Disadvantage from the Socio-Economic Indexes for Areas. This index assesses variables that reflect the relative disadvantage of an area compared to the rest of the state (Pink, 2011). Quintile scores were obtained from the Australian Bureau of Statistics using 2011 POA census data: A higher quintile equals more disadvantage (Australian Bureau of Statistics, 2013a). Remoteness Remoteness of each community was determined using the Australian Standard Geographical Classification remoteness structure, which defines remoteness based on road distance to service centres. Researchers grouped all NSW communities into three categories: major city, regional and remote (Australian Bureau of Statistics, 2007). Liquor licences Data on the per capita rate of liquor licences per 1000 population in communities were provided by Liquor and Gaming NSW (2015). Given that different types of liquor licences are related to different types of ARCs (Gmel et al., 2016), distinct licences were grouped into three categories: on-venue licences (e.g. bars, clubs and restaurants), packaged licences (e.g. liquor stores), and hotel licences (which can have both on-venue and packaged licences). Population characteristics For each community, the proportion of the population that identified as Aboriginal and the proportion of young (20–29 years old) male residents were obtained from the Australian Bureau of Statistics using 2011 POA Census data (Australian Bureau of Statistics, 2011). The age range for young males was based on the minimum drinking age (18 years), however, as POA Census data were only available in 5- or 10-year age categories, the minimum age was set at 20 rather than 18 years. Additionally, findings from previous research indicate that young males’ involvement in ARCs declines after 30 years of age (Teece and Williams, 2000; Palk et al., 2007). Offence characteristics The offence categories were limited to those included in the proxy definition: assaults, sexual assaults, disorderly conduct and malicious damage to property. Time of occurrence of the criminal incident identified whether the incident took place during the week or weekend. Statistical analyses Rates of ARCs for Aboriginal and non-Aboriginal Australians in NSW were standardized using the Standard Population for Use in Age-Standardization Table provided by the Australian Bureau of Statistics, which uses 2001 Census data as the most recent standard population available in Australia (2013b). Age-Standardized Rates (ASRs) were estimated separately for males and females for the total number of ARCs and separately for all individual characteristics (Aboriginal identity, gender and age), for the categorical community characteristics (SES, remoteness), and for the offence characteristics (offence category and time of occurrence). A preliminary analysis was conducted using a single-level Poisson model with ARCs as outcome variable that was composed of age, gender, Aboriginal status, and an interaction term of gender and Aboriginal status. This preliminary analysis identified that the effect of Aboriginal status on the rates of ARCs was modified by gender (RR 0.42 (99% CI 0.40; 0.43) P ≤ 0.05). Therefore, all models in this study were stratified by gender. The statistical significance of differences in ASRs for Aboriginal and non-Aboriginal POIs were estimated by calculating rate ratios using single-level Poisson models for individual characteristics (each age category) and categorical community characteristics (SES and remoteness) with ARC as the outcome. To investigate whether differences in the ASRs of ARCs between Aboriginal and non-Aboriginal Australians were statistically significantly different across communities, a random intercept for POA was added to all Poisson models. To identify individual and community characteristics that are significantly associated with the rates of ARCs, separate multi-level Poisson models with rates of ARCs as outcome variable for Aboriginal and non-Aboriginal Australians were estimated. These models comprised individual characteristics (age), community characteristics (SES, remoteness, per capita rate of liquor licences, proportion of young males, and proportion of Aboriginal Australians), and the random intercept for POA. All analyses were conducted using Stata 14 (StataCorp, 2015). Confidence intervals were calculated at the 99% level. RESULTS Rates of ARCs A total of 330,063 ARCs were identified between 2005 and 2014 involving POIs aged 13 and over in NSW (43,590 [13%] involving Aboriginal POIs and 286,473 [87%] involving non-Aboriginal POIs). For individual characteristics, Table 1 shows that the ASRs of ARCs per 1000 were 48.7 (CI 47.8–49.5) for Aboriginal males and 21.9 (CI 20.5–23.2) for Aboriginal females, compared to 10.0 (CI 10.0–10.1) for non-Aboriginal males and 1.8 (CI 1.7–1.8) for non-Aboriginal females. ASRs were highest for 18–22-year olds and second highest for 23–27 year olds among both Aboriginal and non-Aboriginal groups. ARCs progressively reduced as age categories increased. The only exception to these patterns was for non-Aboriginal females, where the second highest ASRs were for 13–17-year olds. Table 1. Age-standardized rates of ARCs in NSW between 2005 and 2014, by Aboriginal status and gender ARCs Aboriginal Non-Aboriginal Male Female Male Female N ASRa (99% CI) N ASRa (99% CI) N ASRa (99% CI) N ASRa (99% CI) Total ARCs 32,032 48.7 (47.8–49.5) 11,558 21.9 (20.5–23.2) 244,738 10.0 (10.0–10.1) 41,735 1.8 (1.7–1.8) Individual characteristics Age of POI  13–17 6464 62.9 (60.9–65.0) 2523 29.7 (29.3–31.3) 32,323 15.7 (15.5–15.9) 7694 4.1 (3.9–4.2)  18–22 8846 107.6 (104.7–110.6) 2920 45.7 (43.5–47.9) 74,930 36.7 (36.4–37.1) 11,031 5.7 (5.6–59)  23–27 5424 69.7 (66.8–72.6) 1822 37.7 (35.5–40.0) 44,875 21.8 (21.5–22.0) 6369 3.1 (3.0–3.2)  28–32 3853 24.7 (22.8–26.6) 1316 34.6 (32.2–37.1) 29,055 13.6 (13.4–13.8) 4502 2.1 (2.0–2.2)  33–37 3023 72.9 (69.6–76.4) 1174 31.1 (28.9–33.5) 22,034 10.4 (10.2–10.6) 4189 2.0 (1.9–2.1)  38–42 2122 46.0 (43.5–48.7) 902 21.4 (19.6–23.3) 16,616 7.4 (7.2–7.5) 3436 1.5 (1.4–1.6)  43–47 1230 31.7 (29.5–34.2) 558 15.4 (13.8–17.2) 11,151 5.3 (5.2–5.4) 2254 1.1 (1.1–1.2)  48–52 639 23.8 (21.4–26.2) 225 9.9 (8.3–11.8) 6602 3.0 (2.9–3.1) 1209 0.6 (0.6–0.7)  53–57 274 15.7 (13.3–18.3) 81 8.2 (6.0–10.8) 3651 1.9 (1.8–2.0) 560 0.4 (0.3–0.4)  58–62 104 10.9 (8.3–14.0) 22 6.3 (3.4–10.7) 1821 1.1 (1.1–1.2) 268 0.3 (0.2–0.3)  63–67 40 10.5 (6.7–15.6) 11 6.0 (2.4–12.5) 887 0.7 (0.6–0.8) 116 0.3 (0.2–0.4)  68+ 13 4.2 (1.8–8.3) 4 7.3 (1.2–22.9) 793 0.3 (0.3–0.3) 107 0.1 (0.0–0.1) Community characteristics SES quintile  1 (least disadvantaged) 835 26.0 (22.3–29.8) 243 15.0 (12.0–18.1) 36,947 6.4 (6.3–6.5) 5269 1.0 (0.9–1.0)  2 2330 33.0 (27.1–38.8) 857 18.3 (16.0–20.6) 45,305 8.5 (5.4–8.6) 7093 1.4 (1.4–1.5)  3 8222 44.8 (43.2–46.4) 2776 18.6 (17.0–20.3) 74,542 13.4 (12.9–13.2) 12,330 2.3 (2.2–2.3)  4 8735 53.1 (46.5–59.7) 3219 23.7 (17.2–30.2) 48,842 13.6 (13.4–13.7) 9186 2.6 (2.5–.27)  5 (most disadvantaged) 11,910 62.4 (60.8–64.0) 4463 25.7 (24.6–26.8) 39,102 10.4 (10.3–10.6) 7857 2.2 (2.1–2.3) Remoteness  Major city 7213 30.6 (29.2–32.0) 2541 13.5 (12.6–14.4) 143,798 8.2 (8.1–8.3) 23,967 1.4 (1.4–1.5)  Regional 18,871 51.5 (50.4–52.7) 6667 22.5 (21.1–24.0) 96,181 15.0 (14.9–15.1) 16,800 2.7 (2.7–2.8)  Remote 5948 119.9 (116.1–123.7) 2350 44.4 (42.0–46.8) 4755 24.4 (23.5–25.3) 968 5.8 (5.2–6.4) Offence characteristics Offence categories  Assaults 17,435 32.4 (31.4–33.4) 6718 14.9 (13.5–16.3) 114,632 4.8 (4.7–4.8) 26,348 1.2 (1.2–1.2)  Sexual assaults 1096 6.0 (5.2–6.8) 28 2.4 (0.6–4.3) 14,794 0.7 (0.7–0.7) 433 0.2 (0.2–0.2)  Disorderly conduct 5903 15.0 (13.5–16.5) 2809 10.4 (8.0–12.8) 58,664 2.7 (2.7–2.7) 6630 0.5 (0.5–0.5)  Malicious damage 7598 14.3 (13.5–15.2) 2003 7.3 (4.8–9.7) 56,648 2.4 (2.4–2.4) 8324 0.5 (0.5–0.5) Time of occurrence  Weekend 22,237 36.2 (35.2–37.2) 8164 17.0 (15.8–18.3) 176,379 7.3 (7.3–7.4) 29,630 1.3 (1.3–1.4)  Week day 9795 19.1 (18.3–19.8) 3394 11.4 (0.5–17.9) 68,359 2.9 (2.9–2.9) 12,105 0.6 (0.6–0.6) ARCs Aboriginal Non-Aboriginal Male Female Male Female N ASRa (99% CI) N ASRa (99% CI) N ASRa (99% CI) N ASRa (99% CI) Total ARCs 32,032 48.7 (47.8–49.5) 11,558 21.9 (20.5–23.2) 244,738 10.0 (10.0–10.1) 41,735 1.8 (1.7–1.8) Individual characteristics Age of POI  13–17 6464 62.9 (60.9–65.0) 2523 29.7 (29.3–31.3) 32,323 15.7 (15.5–15.9) 7694 4.1 (3.9–4.2)  18–22 8846 107.6 (104.7–110.6) 2920 45.7 (43.5–47.9) 74,930 36.7 (36.4–37.1) 11,031 5.7 (5.6–59)  23–27 5424 69.7 (66.8–72.6) 1822 37.7 (35.5–40.0) 44,875 21.8 (21.5–22.0) 6369 3.1 (3.0–3.2)  28–32 3853 24.7 (22.8–26.6) 1316 34.6 (32.2–37.1) 29,055 13.6 (13.4–13.8) 4502 2.1 (2.0–2.2)  33–37 3023 72.9 (69.6–76.4) 1174 31.1 (28.9–33.5) 22,034 10.4 (10.2–10.6) 4189 2.0 (1.9–2.1)  38–42 2122 46.0 (43.5–48.7) 902 21.4 (19.6–23.3) 16,616 7.4 (7.2–7.5) 3436 1.5 (1.4–1.6)  43–47 1230 31.7 (29.5–34.2) 558 15.4 (13.8–17.2) 11,151 5.3 (5.2–5.4) 2254 1.1 (1.1–1.2)  48–52 639 23.8 (21.4–26.2) 225 9.9 (8.3–11.8) 6602 3.0 (2.9–3.1) 1209 0.6 (0.6–0.7)  53–57 274 15.7 (13.3–18.3) 81 8.2 (6.0–10.8) 3651 1.9 (1.8–2.0) 560 0.4 (0.3–0.4)  58–62 104 10.9 (8.3–14.0) 22 6.3 (3.4–10.7) 1821 1.1 (1.1–1.2) 268 0.3 (0.2–0.3)  63–67 40 10.5 (6.7–15.6) 11 6.0 (2.4–12.5) 887 0.7 (0.6–0.8) 116 0.3 (0.2–0.4)  68+ 13 4.2 (1.8–8.3) 4 7.3 (1.2–22.9) 793 0.3 (0.3–0.3) 107 0.1 (0.0–0.1) Community characteristics SES quintile  1 (least disadvantaged) 835 26.0 (22.3–29.8) 243 15.0 (12.0–18.1) 36,947 6.4 (6.3–6.5) 5269 1.0 (0.9–1.0)  2 2330 33.0 (27.1–38.8) 857 18.3 (16.0–20.6) 45,305 8.5 (5.4–8.6) 7093 1.4 (1.4–1.5)  3 8222 44.8 (43.2–46.4) 2776 18.6 (17.0–20.3) 74,542 13.4 (12.9–13.2) 12,330 2.3 (2.2–2.3)  4 8735 53.1 (46.5–59.7) 3219 23.7 (17.2–30.2) 48,842 13.6 (13.4–13.7) 9186 2.6 (2.5–.27)  5 (most disadvantaged) 11,910 62.4 (60.8–64.0) 4463 25.7 (24.6–26.8) 39,102 10.4 (10.3–10.6) 7857 2.2 (2.1–2.3) Remoteness  Major city 7213 30.6 (29.2–32.0) 2541 13.5 (12.6–14.4) 143,798 8.2 (8.1–8.3) 23,967 1.4 (1.4–1.5)  Regional 18,871 51.5 (50.4–52.7) 6667 22.5 (21.1–24.0) 96,181 15.0 (14.9–15.1) 16,800 2.7 (2.7–2.8)  Remote 5948 119.9 (116.1–123.7) 2350 44.4 (42.0–46.8) 4755 24.4 (23.5–25.3) 968 5.8 (5.2–6.4) Offence characteristics Offence categories  Assaults 17,435 32.4 (31.4–33.4) 6718 14.9 (13.5–16.3) 114,632 4.8 (4.7–4.8) 26,348 1.2 (1.2–1.2)  Sexual assaults 1096 6.0 (5.2–6.8) 28 2.4 (0.6–4.3) 14,794 0.7 (0.7–0.7) 433 0.2 (0.2–0.2)  Disorderly conduct 5903 15.0 (13.5–16.5) 2809 10.4 (8.0–12.8) 58,664 2.7 (2.7–2.7) 6630 0.5 (0.5–0.5)  Malicious damage 7598 14.3 (13.5–15.2) 2003 7.3 (4.8–9.7) 56,648 2.4 (2.4–2.4) 8324 0.5 (0.5–0.5) Time of occurrence  Weekend 22,237 36.2 (35.2–37.2) 8164 17.0 (15.8–18.3) 176,379 7.3 (7.3–7.4) 29,630 1.3 (1.3–1.4)  Week day 9795 19.1 (18.3–19.8) 3394 11.4 (0.5–17.9) 68,359 2.9 (2.9–2.9) 12,105 0.6 (0.6–0.6) CI = confidence interval; POI = person of interest; SES = socio-economic status. aAge standardized rates per 1000 population. Table 1. Age-standardized rates of ARCs in NSW between 2005 and 2014, by Aboriginal status and gender ARCs Aboriginal Non-Aboriginal Male Female Male Female N ASRa (99% CI) N ASRa (99% CI) N ASRa (99% CI) N ASRa (99% CI) Total ARCs 32,032 48.7 (47.8–49.5) 11,558 21.9 (20.5–23.2) 244,738 10.0 (10.0–10.1) 41,735 1.8 (1.7–1.8) Individual characteristics Age of POI  13–17 6464 62.9 (60.9–65.0) 2523 29.7 (29.3–31.3) 32,323 15.7 (15.5–15.9) 7694 4.1 (3.9–4.2)  18–22 8846 107.6 (104.7–110.6) 2920 45.7 (43.5–47.9) 74,930 36.7 (36.4–37.1) 11,031 5.7 (5.6–59)  23–27 5424 69.7 (66.8–72.6) 1822 37.7 (35.5–40.0) 44,875 21.8 (21.5–22.0) 6369 3.1 (3.0–3.2)  28–32 3853 24.7 (22.8–26.6) 1316 34.6 (32.2–37.1) 29,055 13.6 (13.4–13.8) 4502 2.1 (2.0–2.2)  33–37 3023 72.9 (69.6–76.4) 1174 31.1 (28.9–33.5) 22,034 10.4 (10.2–10.6) 4189 2.0 (1.9–2.1)  38–42 2122 46.0 (43.5–48.7) 902 21.4 (19.6–23.3) 16,616 7.4 (7.2–7.5) 3436 1.5 (1.4–1.6)  43–47 1230 31.7 (29.5–34.2) 558 15.4 (13.8–17.2) 11,151 5.3 (5.2–5.4) 2254 1.1 (1.1–1.2)  48–52 639 23.8 (21.4–26.2) 225 9.9 (8.3–11.8) 6602 3.0 (2.9–3.1) 1209 0.6 (0.6–0.7)  53–57 274 15.7 (13.3–18.3) 81 8.2 (6.0–10.8) 3651 1.9 (1.8–2.0) 560 0.4 (0.3–0.4)  58–62 104 10.9 (8.3–14.0) 22 6.3 (3.4–10.7) 1821 1.1 (1.1–1.2) 268 0.3 (0.2–0.3)  63–67 40 10.5 (6.7–15.6) 11 6.0 (2.4–12.5) 887 0.7 (0.6–0.8) 116 0.3 (0.2–0.4)  68+ 13 4.2 (1.8–8.3) 4 7.3 (1.2–22.9) 793 0.3 (0.3–0.3) 107 0.1 (0.0–0.1) Community characteristics SES quintile  1 (least disadvantaged) 835 26.0 (22.3–29.8) 243 15.0 (12.0–18.1) 36,947 6.4 (6.3–6.5) 5269 1.0 (0.9–1.0)  2 2330 33.0 (27.1–38.8) 857 18.3 (16.0–20.6) 45,305 8.5 (5.4–8.6) 7093 1.4 (1.4–1.5)  3 8222 44.8 (43.2–46.4) 2776 18.6 (17.0–20.3) 74,542 13.4 (12.9–13.2) 12,330 2.3 (2.2–2.3)  4 8735 53.1 (46.5–59.7) 3219 23.7 (17.2–30.2) 48,842 13.6 (13.4–13.7) 9186 2.6 (2.5–.27)  5 (most disadvantaged) 11,910 62.4 (60.8–64.0) 4463 25.7 (24.6–26.8) 39,102 10.4 (10.3–10.6) 7857 2.2 (2.1–2.3) Remoteness  Major city 7213 30.6 (29.2–32.0) 2541 13.5 (12.6–14.4) 143,798 8.2 (8.1–8.3) 23,967 1.4 (1.4–1.5)  Regional 18,871 51.5 (50.4–52.7) 6667 22.5 (21.1–24.0) 96,181 15.0 (14.9–15.1) 16,800 2.7 (2.7–2.8)  Remote 5948 119.9 (116.1–123.7) 2350 44.4 (42.0–46.8) 4755 24.4 (23.5–25.3) 968 5.8 (5.2–6.4) Offence characteristics Offence categories  Assaults 17,435 32.4 (31.4–33.4) 6718 14.9 (13.5–16.3) 114,632 4.8 (4.7–4.8) 26,348 1.2 (1.2–1.2)  Sexual assaults 1096 6.0 (5.2–6.8) 28 2.4 (0.6–4.3) 14,794 0.7 (0.7–0.7) 433 0.2 (0.2–0.2)  Disorderly conduct 5903 15.0 (13.5–16.5) 2809 10.4 (8.0–12.8) 58,664 2.7 (2.7–2.7) 6630 0.5 (0.5–0.5)  Malicious damage 7598 14.3 (13.5–15.2) 2003 7.3 (4.8–9.7) 56,648 2.4 (2.4–2.4) 8324 0.5 (0.5–0.5) Time of occurrence  Weekend 22,237 36.2 (35.2–37.2) 8164 17.0 (15.8–18.3) 176,379 7.3 (7.3–7.4) 29,630 1.3 (1.3–1.4)  Week day 9795 19.1 (18.3–19.8) 3394 11.4 (0.5–17.9) 68,359 2.9 (2.9–2.9) 12,105 0.6 (0.6–0.6) ARCs Aboriginal Non-Aboriginal Male Female Male Female N ASRa (99% CI) N ASRa (99% CI) N ASRa (99% CI) N ASRa (99% CI) Total ARCs 32,032 48.7 (47.8–49.5) 11,558 21.9 (20.5–23.2) 244,738 10.0 (10.0–10.1) 41,735 1.8 (1.7–1.8) Individual characteristics Age of POI  13–17 6464 62.9 (60.9–65.0) 2523 29.7 (29.3–31.3) 32,323 15.7 (15.5–15.9) 7694 4.1 (3.9–4.2)  18–22 8846 107.6 (104.7–110.6) 2920 45.7 (43.5–47.9) 74,930 36.7 (36.4–37.1) 11,031 5.7 (5.6–59)  23–27 5424 69.7 (66.8–72.6) 1822 37.7 (35.5–40.0) 44,875 21.8 (21.5–22.0) 6369 3.1 (3.0–3.2)  28–32 3853 24.7 (22.8–26.6) 1316 34.6 (32.2–37.1) 29,055 13.6 (13.4–13.8) 4502 2.1 (2.0–2.2)  33–37 3023 72.9 (69.6–76.4) 1174 31.1 (28.9–33.5) 22,034 10.4 (10.2–10.6) 4189 2.0 (1.9–2.1)  38–42 2122 46.0 (43.5–48.7) 902 21.4 (19.6–23.3) 16,616 7.4 (7.2–7.5) 3436 1.5 (1.4–1.6)  43–47 1230 31.7 (29.5–34.2) 558 15.4 (13.8–17.2) 11,151 5.3 (5.2–5.4) 2254 1.1 (1.1–1.2)  48–52 639 23.8 (21.4–26.2) 225 9.9 (8.3–11.8) 6602 3.0 (2.9–3.1) 1209 0.6 (0.6–0.7)  53–57 274 15.7 (13.3–18.3) 81 8.2 (6.0–10.8) 3651 1.9 (1.8–2.0) 560 0.4 (0.3–0.4)  58–62 104 10.9 (8.3–14.0) 22 6.3 (3.4–10.7) 1821 1.1 (1.1–1.2) 268 0.3 (0.2–0.3)  63–67 40 10.5 (6.7–15.6) 11 6.0 (2.4–12.5) 887 0.7 (0.6–0.8) 116 0.3 (0.2–0.4)  68+ 13 4.2 (1.8–8.3) 4 7.3 (1.2–22.9) 793 0.3 (0.3–0.3) 107 0.1 (0.0–0.1) Community characteristics SES quintile  1 (least disadvantaged) 835 26.0 (22.3–29.8) 243 15.0 (12.0–18.1) 36,947 6.4 (6.3–6.5) 5269 1.0 (0.9–1.0)  2 2330 33.0 (27.1–38.8) 857 18.3 (16.0–20.6) 45,305 8.5 (5.4–8.6) 7093 1.4 (1.4–1.5)  3 8222 44.8 (43.2–46.4) 2776 18.6 (17.0–20.3) 74,542 13.4 (12.9–13.2) 12,330 2.3 (2.2–2.3)  4 8735 53.1 (46.5–59.7) 3219 23.7 (17.2–30.2) 48,842 13.6 (13.4–13.7) 9186 2.6 (2.5–.27)  5 (most disadvantaged) 11,910 62.4 (60.8–64.0) 4463 25.7 (24.6–26.8) 39,102 10.4 (10.3–10.6) 7857 2.2 (2.1–2.3) Remoteness  Major city 7213 30.6 (29.2–32.0) 2541 13.5 (12.6–14.4) 143,798 8.2 (8.1–8.3) 23,967 1.4 (1.4–1.5)  Regional 18,871 51.5 (50.4–52.7) 6667 22.5 (21.1–24.0) 96,181 15.0 (14.9–15.1) 16,800 2.7 (2.7–2.8)  Remote 5948 119.9 (116.1–123.7) 2350 44.4 (42.0–46.8) 4755 24.4 (23.5–25.3) 968 5.8 (5.2–6.4) Offence characteristics Offence categories  Assaults 17,435 32.4 (31.4–33.4) 6718 14.9 (13.5–16.3) 114,632 4.8 (4.7–4.8) 26,348 1.2 (1.2–1.2)  Sexual assaults 1096 6.0 (5.2–6.8) 28 2.4 (0.6–4.3) 14,794 0.7 (0.7–0.7) 433 0.2 (0.2–0.2)  Disorderly conduct 5903 15.0 (13.5–16.5) 2809 10.4 (8.0–12.8) 58,664 2.7 (2.7–2.7) 6630 0.5 (0.5–0.5)  Malicious damage 7598 14.3 (13.5–15.2) 2003 7.3 (4.8–9.7) 56,648 2.4 (2.4–2.4) 8324 0.5 (0.5–0.5) Time of occurrence  Weekend 22,237 36.2 (35.2–37.2) 8164 17.0 (15.8–18.3) 176,379 7.3 (7.3–7.4) 29,630 1.3 (1.3–1.4)  Week day 9795 19.1 (18.3–19.8) 3394 11.4 (0.5–17.9) 68,359 2.9 (2.9–2.9) 12,105 0.6 (0.6–0.6) CI = confidence interval; POI = person of interest; SES = socio-economic status. aAge standardized rates per 1000 population. For community characteristics, ASRs of ARC generally increased as communities became more disadvantaged, regardless of Aboriginal status or gender, although they were lower in the most disadvantaged areas (5) than to the second most disadvantaged areas (4) for non-Aboriginal POIs. Furthermore, ASRs increased as communities became more remote. For offence characteristics, assaults were the most commonly recorded ARCs, regardless of Aboriginal status or gender, followed by disorderly conduct and malicious damage, with a relatively small number of sexual assaults. ASRs of ARCs were at least twice as high on weekends than on weekdays, Aboriginal status and gender notwithstanding. Despite similar patterns in the ASRs of ARCs among Aboriginal and non-Aboriginal POIs, and for males and females, the rate ratios presented in Table 2 indicate that ARCs were statistically significantly higher for Aboriginal than for non-Aboriginal POIs. Table 2. Rate ratios of ARCs for Aboriginal compared to non-Aboriginal POIs, stratified by gender resulting from single-level Poisson models Male Female Rate ratioa (99% CI) Rate ratioa (99% CI) Individual characteristics Age of POI  13–17 2.78 (2.67–2.89) 4.63 (4.33–4.95)  18–22 1.97 (1.91–2.04) 4.40 (4.14–4.68)  23–27 2.95 (2.83–3.07) 6.66 (6.15–7.22)  28–32 3.82 (3.64–4.01) 9.18 (8.32–10.12)  33–37 4.33 (4.10–4.58) 8.42 (7.61–9.33)  38–42 3.84 (3.59–4.09) 9.27 (8.27–10.39)  43–47 4.10 (3.76–4.46) 9.77 (8.47–11.26)  48–52 5.90 (5.25–6.63) 14.46 (11.63–17.98)  53–57 6.78 (5.67–8.11) 22.62 (16.20–31.59)  58–62 7.65 (5.69–10.29) 22.77 (12.86–40.32)  63–67 14.22 (8.97–22.91) 21.59 (9.58–48.65)  68+ 12.81 (5.66–28.96) 85.28 (22.97–316.60) Community characteristics SES quintile  1 (least disadvantaged) 5.37 (4.90–5.88) 16.89 (14.18–20.12)  2 4.48 (4.24–4.73) 13.74 (12.48–15.12)  3 4.90 (4.75–5.06) 9.66 (9.11–10.24)  4 4.97 (4.81–5.12) 8.21 (7.75–8.70)  5 (most disadvantaged) 5.38 (5.22–5.55) 7.67 (7.22–8.14) Remoteness  Major city 4.11 (3.98–4.25) 9.14 (8.64–9.67)  Regional 5.30 (5.19–5.41) 9.20 (8.83–9.59)  Remote 6.83 (6.47–7.22) 7.32 (6.53–8.19) Male Female Rate ratioa (99% CI) Rate ratioa (99% CI) Individual characteristics Age of POI  13–17 2.78 (2.67–2.89) 4.63 (4.33–4.95)  18–22 1.97 (1.91–2.04) 4.40 (4.14–4.68)  23–27 2.95 (2.83–3.07) 6.66 (6.15–7.22)  28–32 3.82 (3.64–4.01) 9.18 (8.32–10.12)  33–37 4.33 (4.10–4.58) 8.42 (7.61–9.33)  38–42 3.84 (3.59–4.09) 9.27 (8.27–10.39)  43–47 4.10 (3.76–4.46) 9.77 (8.47–11.26)  48–52 5.90 (5.25–6.63) 14.46 (11.63–17.98)  53–57 6.78 (5.67–8.11) 22.62 (16.20–31.59)  58–62 7.65 (5.69–10.29) 22.77 (12.86–40.32)  63–67 14.22 (8.97–22.91) 21.59 (9.58–48.65)  68+ 12.81 (5.66–28.96) 85.28 (22.97–316.60) Community characteristics SES quintile  1 (least disadvantaged) 5.37 (4.90–5.88) 16.89 (14.18–20.12)  2 4.48 (4.24–4.73) 13.74 (12.48–15.12)  3 4.90 (4.75–5.06) 9.66 (9.11–10.24)  4 4.97 (4.81–5.12) 8.21 (7.75–8.70)  5 (most disadvantaged) 5.38 (5.22–5.55) 7.67 (7.22–8.14) Remoteness  Major city 4.11 (3.98–4.25) 9.14 (8.64–9.67)  Regional 5.30 (5.19–5.41) 9.20 (8.83–9.59)  Remote 6.83 (6.47–7.22) 7.32 (6.53–8.19) CI = confidence interval; POI = person of interest; SES = socio-economic status. aAll presented rate ratios are statistically significant (P ≤ 0.01). Table 2. Rate ratios of ARCs for Aboriginal compared to non-Aboriginal POIs, stratified by gender resulting from single-level Poisson models Male Female Rate ratioa (99% CI) Rate ratioa (99% CI) Individual characteristics Age of POI  13–17 2.78 (2.67–2.89) 4.63 (4.33–4.95)  18–22 1.97 (1.91–2.04) 4.40 (4.14–4.68)  23–27 2.95 (2.83–3.07) 6.66 (6.15–7.22)  28–32 3.82 (3.64–4.01) 9.18 (8.32–10.12)  33–37 4.33 (4.10–4.58) 8.42 (7.61–9.33)  38–42 3.84 (3.59–4.09) 9.27 (8.27–10.39)  43–47 4.10 (3.76–4.46) 9.77 (8.47–11.26)  48–52 5.90 (5.25–6.63) 14.46 (11.63–17.98)  53–57 6.78 (5.67–8.11) 22.62 (16.20–31.59)  58–62 7.65 (5.69–10.29) 22.77 (12.86–40.32)  63–67 14.22 (8.97–22.91) 21.59 (9.58–48.65)  68+ 12.81 (5.66–28.96) 85.28 (22.97–316.60) Community characteristics SES quintile  1 (least disadvantaged) 5.37 (4.90–5.88) 16.89 (14.18–20.12)  2 4.48 (4.24–4.73) 13.74 (12.48–15.12)  3 4.90 (4.75–5.06) 9.66 (9.11–10.24)  4 4.97 (4.81–5.12) 8.21 (7.75–8.70)  5 (most disadvantaged) 5.38 (5.22–5.55) 7.67 (7.22–8.14) Remoteness  Major city 4.11 (3.98–4.25) 9.14 (8.64–9.67)  Regional 5.30 (5.19–5.41) 9.20 (8.83–9.59)  Remote 6.83 (6.47–7.22) 7.32 (6.53–8.19) Male Female Rate ratioa (99% CI) Rate ratioa (99% CI) Individual characteristics Age of POI  13–17 2.78 (2.67–2.89) 4.63 (4.33–4.95)  18–22 1.97 (1.91–2.04) 4.40 (4.14–4.68)  23–27 2.95 (2.83–3.07) 6.66 (6.15–7.22)  28–32 3.82 (3.64–4.01) 9.18 (8.32–10.12)  33–37 4.33 (4.10–4.58) 8.42 (7.61–9.33)  38–42 3.84 (3.59–4.09) 9.27 (8.27–10.39)  43–47 4.10 (3.76–4.46) 9.77 (8.47–11.26)  48–52 5.90 (5.25–6.63) 14.46 (11.63–17.98)  53–57 6.78 (5.67–8.11) 22.62 (16.20–31.59)  58–62 7.65 (5.69–10.29) 22.77 (12.86–40.32)  63–67 14.22 (8.97–22.91) 21.59 (9.58–48.65)  68+ 12.81 (5.66–28.96) 85.28 (22.97–316.60) Community characteristics SES quintile  1 (least disadvantaged) 5.37 (4.90–5.88) 16.89 (14.18–20.12)  2 4.48 (4.24–4.73) 13.74 (12.48–15.12)  3 4.90 (4.75–5.06) 9.66 (9.11–10.24)  4 4.97 (4.81–5.12) 8.21 (7.75–8.70)  5 (most disadvantaged) 5.38 (5.22–5.55) 7.67 (7.22–8.14) Remoteness  Major city 4.11 (3.98–4.25) 9.14 (8.64–9.67)  Regional 5.30 (5.19–5.41) 9.20 (8.83–9.59)  Remote 6.83 (6.47–7.22) 7.32 (6.53–8.19) CI = confidence interval; POI = person of interest; SES = socio-economic status. aAll presented rate ratios are statistically significant (P ≤ 0.01). Community variation in the differences between Aboriginal and non-Aboriginal Australians in rates of ARCs Single-level Poisson models identified that rates of ARCs for Aboriginal POIs were 6.75 (CI 6.67–6.83) and 14.58 (CI 14.29–14.89) times higher than rates for non-Aboriginal POIs, for males and females, respectively. However, these estimates reduced to 4.65 (CI 4.59–4.71) times higher for Aboriginal males and 8.42 (CI 8.22–8.63) times higher for Aboriginal females after adjusting for community variation in the rate ratios of ARCs of Aboriginal and non-Aboriginal POIs. This change in rate ratio provides evidence that differences in rates of ARCs between Aboriginal and non-Aboriginal POIs significantly varied among communities for both males (P ≤ 0.05) and females (P ≤ 0.05). Individual and community characteristics associated with ARCs In terms of individual characteristics, the same pattern of results was observed for Aboriginal and non-Aboriginal POIs: rates of ARCs were statistically significantly higher in age categories younger than the reference group (38–42-year olds) and statistically significantly lower in age categories older than the reference group (Table 3). Table 3. Adjusted rate ratios of ARCs among Aboriginal and non-Aboriginal POIs in NSW, resulting from the multi-level Poisson regression model Aboriginal (N = 43,590) Non-Aboriginal (N = 286,473) Males Females Males Females RRa 99% CI RRa 99% CI RRa 99% CI RRa 99% CI Individual characteristics (age POI)  13–17 1.63** (1.53–1.74) 1.63** (1.47–1.80) 2.20** (2.15–2.26) 2.74** (2.59–2.89)  18–22 2.97** (2.79–3.17) 2.47** (2.24–2.73) 5.12** (5.01–5.24) 3.79** (3.60–4.00)  23–27 2.46** (2.39–2.63) 1.81** (1.62–2.02) 2.78** (2.71–2.85) 1.91** (1.80–2.02)  28–32 2.04** (1.90–2.19) 1.56** (1.39–1.75) 1.68** (1.64–1.73) 1.30** (1.22–1.38)  33–37 1.73** (1.60–1.86) 1.48** (1.39–1.75) 1.37** (1.34–1.41) 1.30** (1.23–1.39)  38–42 (ref) 1.00 1.00 1.00 1.00  43–47 0.69** (0.62–0.75) 0.68** (0.59–0.78) 0.72** (0.69–0.74) 0.73** (0.68–0.78)  48–52 0.44** (0.39–0.50) 0.40** (0.33–0.48) 0.40** (0.39–0.42) 0.39** (0.36–0.43)  53–57 0.26** (0.22–0.31) 0.30** (0.22–0.41) 0.24** (0.23–0.25) 0.22** (0.20–0.25)  58–62 0.19** (0.14–0.24) 0.25** (0.14–0.44) 0.14** (0.13–0.15) 0.15** (0.12–0.17)  63–67 0.16** (0.11–0.25) 0.21** (0.10–0.46) 0.08** (0.07–0.09) 0.15** (0.12–0.19)  68+ 0.05** (0.03–0.11) 0.28* (0.08–1.02) 0.03** (0.03–0.04) 0.05** (0.04–0.06) Community characteristics SES quintile  1 (least disadvantaged) (ref) 1.00 1.00 1.00 1.00  2 1.10 (0.79–1.54) 0.83 (0.54–1.26) 1.13 (0.91–1.39) 1.23* (0.98–1.53)  3 1.40* (1.01–1.95) 0.83 (0.55–1.24) 1.53** (1.25–1.88) 1.66** (1.34–2.07)  4 1.33* (0.94–1.88) 1.05 (0.69–1.61) 1.88** (1.51–2.34) 2.09** (1.66–2.64)  5 (most disadvantaged) 1.32* (0.93–1.88) 0.91 (0.59–1.39) 1.81** (1.45–2.26) 2.08** (1.65–2.63) Remoteness  Major City (ref) 1.00 1.00 1.00  Regional 1.38* (1.05–1.82) 1.51* (1.10–2.06) 1.39** (1.16–1.67) 1.40** (1.14–1.68)  Remote 2.41** (1.41–4.11) 2.10* (1.20–3.67) 1.47* (1.00–2.16) 1.39* (0.91–2.11) Liquor licences  Hotels 0.79** (0.67–0.93) 0.97 (0.78–1.21) 1.04 (0.95–1.45) 1.30** (1.15–1.46)  On-venue 1.18** (1.11–1.25) 1.14* (1.06–1.23) 1.17** (1.13–1.21) 1.12** (1.07–1.17)  Packaged 0.91 (0.81–1.03) 0.92 (0.79–1.07) 0.99 (0.93–1.07) 1.01 (0.93–1.10) Population characteristicsb  Young males 1.03 (0.99–1.08) 1.06* (1.01–1.12) 0.98 (0.95–1.01) 0.97* (0.94–1.00)  Aboriginal 1.03** (1.01–1.05) 1.02* (1.00–1.04) 1.01 (1.00–1.02) 1.03** (1.01–1.04) Aboriginal (N = 43,590) Non-Aboriginal (N = 286,473) Males Females Males Females RRa 99% CI RRa 99% CI RRa 99% CI RRa 99% CI Individual characteristics (age POI)  13–17 1.63** (1.53–1.74) 1.63** (1.47–1.80) 2.20** (2.15–2.26) 2.74** (2.59–2.89)  18–22 2.97** (2.79–3.17) 2.47** (2.24–2.73) 5.12** (5.01–5.24) 3.79** (3.60–4.00)  23–27 2.46** (2.39–2.63) 1.81** (1.62–2.02) 2.78** (2.71–2.85) 1.91** (1.80–2.02)  28–32 2.04** (1.90–2.19) 1.56** (1.39–1.75) 1.68** (1.64–1.73) 1.30** (1.22–1.38)  33–37 1.73** (1.60–1.86) 1.48** (1.39–1.75) 1.37** (1.34–1.41) 1.30** (1.23–1.39)  38–42 (ref) 1.00 1.00 1.00 1.00  43–47 0.69** (0.62–0.75) 0.68** (0.59–0.78) 0.72** (0.69–0.74) 0.73** (0.68–0.78)  48–52 0.44** (0.39–0.50) 0.40** (0.33–0.48) 0.40** (0.39–0.42) 0.39** (0.36–0.43)  53–57 0.26** (0.22–0.31) 0.30** (0.22–0.41) 0.24** (0.23–0.25) 0.22** (0.20–0.25)  58–62 0.19** (0.14–0.24) 0.25** (0.14–0.44) 0.14** (0.13–0.15) 0.15** (0.12–0.17)  63–67 0.16** (0.11–0.25) 0.21** (0.10–0.46) 0.08** (0.07–0.09) 0.15** (0.12–0.19)  68+ 0.05** (0.03–0.11) 0.28* (0.08–1.02) 0.03** (0.03–0.04) 0.05** (0.04–0.06) Community characteristics SES quintile  1 (least disadvantaged) (ref) 1.00 1.00 1.00 1.00  2 1.10 (0.79–1.54) 0.83 (0.54–1.26) 1.13 (0.91–1.39) 1.23* (0.98–1.53)  3 1.40* (1.01–1.95) 0.83 (0.55–1.24) 1.53** (1.25–1.88) 1.66** (1.34–2.07)  4 1.33* (0.94–1.88) 1.05 (0.69–1.61) 1.88** (1.51–2.34) 2.09** (1.66–2.64)  5 (most disadvantaged) 1.32* (0.93–1.88) 0.91 (0.59–1.39) 1.81** (1.45–2.26) 2.08** (1.65–2.63) Remoteness  Major City (ref) 1.00 1.00 1.00  Regional 1.38* (1.05–1.82) 1.51* (1.10–2.06) 1.39** (1.16–1.67) 1.40** (1.14–1.68)  Remote 2.41** (1.41–4.11) 2.10* (1.20–3.67) 1.47* (1.00–2.16) 1.39* (0.91–2.11) Liquor licences  Hotels 0.79** (0.67–0.93) 0.97 (0.78–1.21) 1.04 (0.95–1.45) 1.30** (1.15–1.46)  On-venue 1.18** (1.11–1.25) 1.14* (1.06–1.23) 1.17** (1.13–1.21) 1.12** (1.07–1.17)  Packaged 0.91 (0.81–1.03) 0.92 (0.79–1.07) 0.99 (0.93–1.07) 1.01 (0.93–1.10) Population characteristicsb  Young males 1.03 (0.99–1.08) 1.06* (1.01–1.12) 0.98 (0.95–1.01) 0.97* (0.94–1.00)  Aboriginal 1.03** (1.01–1.05) 1.02* (1.00–1.04) 1.01 (1.00–1.02) 1.03** (1.01–1.04) CI = confidence interval; POI = person of interest; SES = socio-economic status. aAdjusted incidence rate ratios. bPercentage of young males (20–29) and Aboriginal Australians per POA. *P ≤ 0.05; **P ≤ 0.01. Table 3. Adjusted rate ratios of ARCs among Aboriginal and non-Aboriginal POIs in NSW, resulting from the multi-level Poisson regression model Aboriginal (N = 43,590) Non-Aboriginal (N = 286,473) Males Females Males Females RRa 99% CI RRa 99% CI RRa 99% CI RRa 99% CI Individual characteristics (age POI)  13–17 1.63** (1.53–1.74) 1.63** (1.47–1.80) 2.20** (2.15–2.26) 2.74** (2.59–2.89)  18–22 2.97** (2.79–3.17) 2.47** (2.24–2.73) 5.12** (5.01–5.24) 3.79** (3.60–4.00)  23–27 2.46** (2.39–2.63) 1.81** (1.62–2.02) 2.78** (2.71–2.85) 1.91** (1.80–2.02)  28–32 2.04** (1.90–2.19) 1.56** (1.39–1.75) 1.68** (1.64–1.73) 1.30** (1.22–1.38)  33–37 1.73** (1.60–1.86) 1.48** (1.39–1.75) 1.37** (1.34–1.41) 1.30** (1.23–1.39)  38–42 (ref) 1.00 1.00 1.00 1.00  43–47 0.69** (0.62–0.75) 0.68** (0.59–0.78) 0.72** (0.69–0.74) 0.73** (0.68–0.78)  48–52 0.44** (0.39–0.50) 0.40** (0.33–0.48) 0.40** (0.39–0.42) 0.39** (0.36–0.43)  53–57 0.26** (0.22–0.31) 0.30** (0.22–0.41) 0.24** (0.23–0.25) 0.22** (0.20–0.25)  58–62 0.19** (0.14–0.24) 0.25** (0.14–0.44) 0.14** (0.13–0.15) 0.15** (0.12–0.17)  63–67 0.16** (0.11–0.25) 0.21** (0.10–0.46) 0.08** (0.07–0.09) 0.15** (0.12–0.19)  68+ 0.05** (0.03–0.11) 0.28* (0.08–1.02) 0.03** (0.03–0.04) 0.05** (0.04–0.06) Community characteristics SES quintile  1 (least disadvantaged) (ref) 1.00 1.00 1.00 1.00  2 1.10 (0.79–1.54) 0.83 (0.54–1.26) 1.13 (0.91–1.39) 1.23* (0.98–1.53)  3 1.40* (1.01–1.95) 0.83 (0.55–1.24) 1.53** (1.25–1.88) 1.66** (1.34–2.07)  4 1.33* (0.94–1.88) 1.05 (0.69–1.61) 1.88** (1.51–2.34) 2.09** (1.66–2.64)  5 (most disadvantaged) 1.32* (0.93–1.88) 0.91 (0.59–1.39) 1.81** (1.45–2.26) 2.08** (1.65–2.63) Remoteness  Major City (ref) 1.00 1.00 1.00  Regional 1.38* (1.05–1.82) 1.51* (1.10–2.06) 1.39** (1.16–1.67) 1.40** (1.14–1.68)  Remote 2.41** (1.41–4.11) 2.10* (1.20–3.67) 1.47* (1.00–2.16) 1.39* (0.91–2.11) Liquor licences  Hotels 0.79** (0.67–0.93) 0.97 (0.78–1.21) 1.04 (0.95–1.45) 1.30** (1.15–1.46)  On-venue 1.18** (1.11–1.25) 1.14* (1.06–1.23) 1.17** (1.13–1.21) 1.12** (1.07–1.17)  Packaged 0.91 (0.81–1.03) 0.92 (0.79–1.07) 0.99 (0.93–1.07) 1.01 (0.93–1.10) Population characteristicsb  Young males 1.03 (0.99–1.08) 1.06* (1.01–1.12) 0.98 (0.95–1.01) 0.97* (0.94–1.00)  Aboriginal 1.03** (1.01–1.05) 1.02* (1.00–1.04) 1.01 (1.00–1.02) 1.03** (1.01–1.04) Aboriginal (N = 43,590) Non-Aboriginal (N = 286,473) Males Females Males Females RRa 99% CI RRa 99% CI RRa 99% CI RRa 99% CI Individual characteristics (age POI)  13–17 1.63** (1.53–1.74) 1.63** (1.47–1.80) 2.20** (2.15–2.26) 2.74** (2.59–2.89)  18–22 2.97** (2.79–3.17) 2.47** (2.24–2.73) 5.12** (5.01–5.24) 3.79** (3.60–4.00)  23–27 2.46** (2.39–2.63) 1.81** (1.62–2.02) 2.78** (2.71–2.85) 1.91** (1.80–2.02)  28–32 2.04** (1.90–2.19) 1.56** (1.39–1.75) 1.68** (1.64–1.73) 1.30** (1.22–1.38)  33–37 1.73** (1.60–1.86) 1.48** (1.39–1.75) 1.37** (1.34–1.41) 1.30** (1.23–1.39)  38–42 (ref) 1.00 1.00 1.00 1.00  43–47 0.69** (0.62–0.75) 0.68** (0.59–0.78) 0.72** (0.69–0.74) 0.73** (0.68–0.78)  48–52 0.44** (0.39–0.50) 0.40** (0.33–0.48) 0.40** (0.39–0.42) 0.39** (0.36–0.43)  53–57 0.26** (0.22–0.31) 0.30** (0.22–0.41) 0.24** (0.23–0.25) 0.22** (0.20–0.25)  58–62 0.19** (0.14–0.24) 0.25** (0.14–0.44) 0.14** (0.13–0.15) 0.15** (0.12–0.17)  63–67 0.16** (0.11–0.25) 0.21** (0.10–0.46) 0.08** (0.07–0.09) 0.15** (0.12–0.19)  68+ 0.05** (0.03–0.11) 0.28* (0.08–1.02) 0.03** (0.03–0.04) 0.05** (0.04–0.06) Community characteristics SES quintile  1 (least disadvantaged) (ref) 1.00 1.00 1.00 1.00  2 1.10 (0.79–1.54) 0.83 (0.54–1.26) 1.13 (0.91–1.39) 1.23* (0.98–1.53)  3 1.40* (1.01–1.95) 0.83 (0.55–1.24) 1.53** (1.25–1.88) 1.66** (1.34–2.07)  4 1.33* (0.94–1.88) 1.05 (0.69–1.61) 1.88** (1.51–2.34) 2.09** (1.66–2.64)  5 (most disadvantaged) 1.32* (0.93–1.88) 0.91 (0.59–1.39) 1.81** (1.45–2.26) 2.08** (1.65–2.63) Remoteness  Major City (ref) 1.00 1.00 1.00  Regional 1.38* (1.05–1.82) 1.51* (1.10–2.06) 1.39** (1.16–1.67) 1.40** (1.14–1.68)  Remote 2.41** (1.41–4.11) 2.10* (1.20–3.67) 1.47* (1.00–2.16) 1.39* (0.91–2.11) Liquor licences  Hotels 0.79** (0.67–0.93) 0.97 (0.78–1.21) 1.04 (0.95–1.45) 1.30** (1.15–1.46)  On-venue 1.18** (1.11–1.25) 1.14* (1.06–1.23) 1.17** (1.13–1.21) 1.12** (1.07–1.17)  Packaged 0.91 (0.81–1.03) 0.92 (0.79–1.07) 0.99 (0.93–1.07) 1.01 (0.93–1.10) Population characteristicsb  Young males 1.03 (0.99–1.08) 1.06* (1.01–1.12) 0.98 (0.95–1.01) 0.97* (0.94–1.00)  Aboriginal 1.03** (1.01–1.05) 1.02* (1.00–1.04) 1.01 (1.00–1.02) 1.03** (1.01–1.04) CI = confidence interval; POI = person of interest; SES = socio-economic status. aAdjusted incidence rate ratios. bPercentage of young males (20–29) and Aboriginal Australians per POA. *P ≤ 0.05; **P ≤ 0.01. In terms of community characteristics, more socio-economic disadvantage in the community was associated with higher rates of ARCs for non-Aboriginal POIs and Aboriginal males, but not for Aboriginal females. For both Aboriginal and non-Aboriginal POIs, rates of ARCs were higher in remote and regional communities than in major cities. A higher per capita rate of on-venue licences within a community was significantly associated with higher rates of ARCs, regardless of Aboriginal status. For non-Aboriginal females and Aboriginal males, a higher per capita rate of hotel licences in the community was also associated with higher rates of ARCs. For non-Aboriginal females, a higher proportion of young males in the community was significantly associated with lower rates of ARCs. For Aboriginal females a higher proportion of young males was associated with higher rates of ARCs. A small, but statistically significant, association was found between a higher proportion of residents in a community identifying as Aboriginal and a higher rate of ARCs, regardless of the Aboriginal status of the POI, except among non-Aboriginal males, where this was not statistically significant. Nevertheless, the marginal nature of this association (e.g. a 1% increase in Aboriginal Australians in a community lead to a 3% increase in ARCs for Aboriginal males), suggests its statistical significance is a consequence of the large sample size, rather than a practically meaningful association. DISCUSSION This study found that age standardized rates of ARCs were statistically significantly higher for Aboriginal POIs, young people (aged 13–37 years) and on weekends. Differences in the rates of ARCs for Aboriginal and non-Aboriginal POIs varied significantly across communities. ARC rates were significantly higher in remote or regional communities, in communities with a higher per capita rate of on-venue licences, and for more socio-economically disadvantaged communities. For national- or jurisdictional-level policy frameworks, these findings highlight the need to ensure legislation continues to target young people, such as the enforcement of minimum legal drinking age and potentially further consideration of increasing the legal minimum drinking age, especially considering those aged 18–22 had the highest rates of ARCs (Cobiac et al., 2009). Minimum Unit Pricing (MUP) policies are an additional policy strategy that has shown to reduce alcohol consumption and related assaults in countries such as Canada, United Kingdom and Australia (Boniface et al., 2017). However, more evidence is required for the evidence of MUP for sub-groups, such as Indigenous Peoples and young people (Seaman et al., 2013). Similarly, the importance of continuing to target alcohol consumption on weekends is also highlighted by these findings. Currently, policies restricting night time alcohol sales in central entertainment districts have been successfully implemented in cities in Norway, California and Australia (Kypri, 2015). Evaluation of these liquor licensing restrictions in Australia and Norway has shown significant reductions in weekend night time assaults (Menéndez et al., 2017; Rossow and Norström, 2012). Even though all communities included in this study are subject to the same jurisdictional-level framework, there are unique community characteristics that contributed to higher rates of ARCs. First, rates of ARCs were higher in communities with a higher per capita rate of on-venue licences and hotel licenses for Aboriginal males and non-Aboriginal females. Currently, there is no official threshold for the number of liquor licenses allowed in any one community in NSW. Applications for new licenses have to provide a community impact statement and local council feedback, which covers the number of licenses in their community. While these factors are considered in the decision-making process by the NSW government office of Liquor and Gaming, implementing a jurisdiction-wide threshold to cap the density of alcohol licences in any one community may help reduce the variability in rates of ARCs across different communities. Second, rates of ARC were significantly higher in remote and regional communities and in lower SES communities, both of which have been previously associated with higher rates of alcohol harms (Australian Institute for Health and Welfare, 2015; Collins, 2016). Given the impracticality of legislative approaches for communities based on their geographical location or SES, this finding highlights the need to develop a mechanism that effectively engages with vulnerable communities to assist them to effectively reduce their rates of ARC. Previous research has identified multi-component approaches that can effectively help regional communities reduce ARCs, such as increasing licensees’ awareness and police activity (Shakeshaft et al., 2014). In addition to benefiting these high-risk communities, such a process would have jurisdictional-level benefits given the resources expended in responding to ARC, such as policing and the provision of emergency and health care services following serious assaults, are met at the jurisdictional-level rather than by local councils. Despite the finding that the risk factors for ARC were mostly the same for Aboriginal and non-Aboriginal Australians, this study found rates of ARCs were significantly higher for Aboriginal Australians. Two possible explanations for this are: Aboriginal status is being confounded with other factors, and/or there are characteristics not measured in this study that are specific to Aboriginal Australians that increase their risk of experiencing ARC. The findings of this analysis are consistent with the possibility of confounding by remoteness, because both the proportion of Aboriginal Australians and rates of ARCs increase with greater remoteness (Australian Bureau of Statistics and Australian Institute of Health and Welfare, 2008; Gmel et al., 2016). They are also consistent with the possibility of confounding by socio-economic status: although rates of ARC were significantly higher for more socio-economically disadvantaged communities, this was not true for Aboriginal females and less pronounced for Aboriginal males which most likely reflects the high level of disadvantaged experienced by Aboriginal Australians, relative to non-Aboriginal Australians (Australian Bureau of Statistics and Australian Institute of Health and Welfare, 2008). Given this study did not identify any characteristics uniquely associated with ARCs for Aboriginal Australians, it is unlikely that the variables included in this study explain the differences in rates of ARC between Aboriginal and non-Aboriginal Australians. Furthermore, a study by Wundersitz (2010) identified similar factors associated with (non-alcohol-related) violent offending by Aboriginal Australians as our study identified for alcohol-related crimes for Aboriginal Australians, including males, young people and remoteness of community. Additionally she identified childhood abuse and alcohol abuse as predictors of violent offending, with the latter being the number one predictor. Given there are similar factors associated with violent offending and ARCs and similar characteristics for Aboriginal and non-Aboriginal Australians, it is likely that there are other, fundamental drivers to violence among Aboriginal Australians, whether alcohol-related or not. Previous studies have identified that issues regarding race, ethnicity and history can influence rates of ARCs, such as experience of racism or transgenerational trauma resulting from the history of colonization and dispossession (Wundersitz, 2010; Cunningham and Paradies, 2013; Osborne et al., 2013). This highlights the need to develop specific strategies for Aboriginal communities that are devised and implemented by the community themselves and evaluated in partnership with researchers. Particularly strategies that support Aboriginal communities to empower themselves to address ARCs can be promising to reduce ARCs (Tsey, 2008). More work is needed in this space, as was highlighted in a systematic review of Australian Aboriginal community-based studies (Snijder et al., 2015). Only 31 evaluations were published between 1990 and 2015 (one addressed substance use and related harms), with only a moderate level of participation of Indigenous Peoples, especially in the design and evaluation phases of these studies (Snijder et al., 2015). Methodological considerations There are several methodological limitations due to the use of police data that are not collected for research purposes. First, the Aboriginal status for 58% of POIs in the dataset was recorded as unknown, these POIs were excluded from the analyses because it is unclear whether Aboriginal people are more or less likely to be recorded as unknown in police incident data (Thompson et al., 2012). Second, while the proxy measure for ARCs has not been explicitly validated for use with Aboriginal Australians, the proxy has been validated with the NSW population, which included Aboriginal Australians (Breen et al., 2011). Currently, no published research indicates that patterns of ARCs differ between Aboriginal and non-Aboriginal Australians. The findings of this study suggest that these patterns might be very similar and an additional analysis of the times at which ARCs flagged by the police occur also indicated that times are similar for Aboriginal and non-Aboriginal Australians. Regardless, the validation of the proxy measure with Aboriginal populations would help improve future estimates for Aboriginal Australians (Chikritzhs et al., 2004). Third, there are potential biases that influence the police recorded data, such as institutional racism (Cunneen, 2006; Wundersitz, 2010). Previous studies identified that 40% of Aboriginal Australians reported to have experienced racism within the legal system, possibly leading to an artefactual overrepresentation of Aboriginal Australians in police records (Cunneen, 2006; Cunningham and Paradies, 2013). Fourth, the use of POA is not a complete measure of communities; ten POAs in this study were not included in POA classification, either because they covered two or more areas, or only partially covered one area (Australian Bureau of Statistics, 2012). Despite its limitations, POA was considered the most appropriate classification to use, because of the availability of community characteristics data by POA and the availability of postcodes in the datasets containing ARCs. Finally, the possibility of reoffending makes it not possible to identify whether characteristics of POIs contribute to a higher proportion of these groups involved in ARCs, or whether the group has more repeat offenders. The findings of this study can nevertheless identify at risk groups that can benefit from targeted policies or programs. CONCLUSION In 2016, The Lancet published an Indigenous-led paper that called for governments internationally to improve Indigenous health by developing targeted strategies that are grounded in data collected in national surveillance systems (Anderson et al., 2016). The current study operationalized this by utilizing routinely collected police data to identify that there is a need for programs that are tailored to vulnerable people and communities, specifically young people, socio-economically disadvantaged and Aboriginal communities. Current best-evidence suggests such programs are likely to be most effective if they are led by communities themselves, and directly respond to issues of transgenerational trauma, racism and social exclusion, in addition to ARC (Gray and Wilkes, 2011). The lack of published evaluations of such programs highlights the need to develop a practical framework on which partnerships between communities and researchers can be developed (Snijder et al., 2015). ACKNOWLEDGEMENTS The authors thank the NSW Ministry of Health for funding the body of research of which this study is one part. Mieke Snijder thanks the NDARC Education Trustees for funding her doctoral research, which includes this paper. The authors thank BOCSAR for providing the data. FUNDING This study was funded by the NSW Ministry of Health as part of the ‘Aboriginal Injury Prevention and Safety Promotion demonstration Grants Program’. M.S. was funded by a postgraduate scholarship from the NDARC Educationan Trustees. CONFLICT OF INTEREST STATEMENT None declared. REFERENCES Aboriginal Health and Medical Research Council . ( 2013 ) AH&MRC Guidelines For Research Into Aboriginal Health: Key Principles . Sydney : Aboriginal Health and Medical Research Council of New South Wales . Anderson I , Robson B , Connolly M , et al. . ( 2016 ) Indigenous and tribal peoples’ health (The Lancet-Lowitja Institute Global Collaboration): a population study . Lancet 388 : 131 – 57 . Google Scholar CrossRef Search ADS PubMed Australian Bureau of Statistics . ( 2007 ) Australian Standard Geographical Classification (ASGC) Remoteness Area Correspondences, 2006 . Canberra : Australian Bureau of Statistics . Australian Bureau of Statistics . ( 2011 ) Census of Population and Housing . Canberra : Commonwealth of Australia . Australian Bureau of Statistics . ( 2012 ) Postcodes & Postal Areas [Online]. Canberra. Available: http://www.abs.gov.au/websitedbs/censushome.nsf/home/factsheetspoa?opendocument&navpos=450 (16 May 2018 date last accessed). Australian Bureau of Statistics . ( 2013 a) 2033.0.55.001 – Census of Population and Housing: Socio-Economic Indexes for Areas (SEIFA), Australia, 2011 . Canberra : Australian Bureau of Statistics . Australian Bureau of Statistics . ( 2013 b) Standard Population for Use in Age-Standardisation . Canberra : Australian Bureau of Statistics . Australian Bureau of Statistics and Australian Institute of Health and Welfare . ( 2008 ) The Health and Welfare of Australia’s Indigenous and Torres Strait Islander Peoples . Canberra : Commonwealth of Australia . Australian Institute for Health and Welfare . ( 2015 ) The Health and Welfare of Australia’s Aboriginal and Torres Strait Islander Peoples: 2015 . Canberra : AIHW . Boniface S , Scannell JW , Marlow S . ( 2017 ) Evidence for the effectiveness of minimum pricing of alcohol: a systematic review and assessment using the Bradford Hill criteria for causality . BMJ Open 7 : e013497 . Google Scholar CrossRef Search ADS PubMed Breen C , Shakeshaft A , Slade T , et al. . ( 2011 ) Do community characteristics predict alcohol-related crime? Alcohol Alcohol 46 : 464 – 70 . Google Scholar CrossRef Search ADS PubMed Chikritzhs T , Gray D , Stockwell T , et al. . ( 2004 ) Applying National Indicators of Alcohol-related Harms to Indigenous Australians: A Discussion Paper . Perth : National Drug Research Institute, Curtin University of Technology . Cobiac L , Vos T , Doran C , et al. . ( 2009 ) Cost‐effectiveness of interventions to prevent alcohol‐related disease and injury in Australia . Addiction 104 : 1646 – 55 . Google Scholar CrossRef Search ADS PubMed Collins SE . ( 2016 ) Associations between socioeconomic factors and alcohol outcomes . Alcohol Res 38 : 83 – 94 . Google Scholar PubMed Cunneen C . ( 2006 ) Racism, discrimination and the over-representation of Indigenous people in the criminal justice system: some conceptual and explanatory issues . Curr Issues Criminal Justice 17 : 329 – 46 . Cunningham J , Paradies YC . ( 2013 ) Patterns and correlates of self-reported racial discrimination among Australian Aboriginal and Torres Strait Islander adults, 2008–09: analysis of national survey data . Int J Equity Health 12 : 47 . Google Scholar CrossRef Search ADS PubMed d’abbs P . ( 2015 ) Widening the gap: the gulf between policy rhetoric and implementation reality in addressing alcohol problems among Indigenous Australians . Drug Alcohol Rev 34 : 461 – 6 . Google Scholar CrossRef Search ADS Gmel G , Holmes J , Studer J . ( 2016 ) Are alcohol outlet densities strongly associated with alcohol-related outcomes? A critical review of recent evidence . Drug Alcohol Rev 35 : 40 – 54 . Google Scholar CrossRef Search ADS Gray D , Wilkes ET . ( 2011 ) Alcohol restrictions in Indigenous communities: an effective strategy if Indigenous -led . Med J Aus 194 : 508 . Kypri K . ( 2015 ) Evidence of harm from late night alcohol sales continues to strengthen . Addiction 110 : 965 – 6 . Google Scholar CrossRef Search ADS PubMed Lister G , Mcvey D , French J , et al. . ( 2008 ) Measuring the societal impact of behavior choices . Soc Mar Q 14 : 51 – 62 . Google Scholar CrossRef Search ADS Manning M , Smith C , Mazerolle P . ( 2013 ) The societal costs of alcohol misuse in Australia . Trends Issues crime Criminal Justice 545 : 1 – 6 . Matthews S , Chikritzhs T , Catalano P , et al. . ( 2002 ) Trends in Alcohol Related Violence in Australia, 1991/92–1999/00 . Perth : Curtin University of Technology . Menéndez P , Kypri K , Weatherburn D . ( 2017 ) The effect of liquor licensing restrictions on assault: a quasi-experimental study in Sydney, Australia . Addiction 112 : 261 – 8 . Google Scholar CrossRef Search ADS PubMed NSW Bureau of Crime Statistics and Research . ( 2015 ) Dg1513103 . Sydney : BOCSAR . NSW Bureau of Crime Statistics and Research . 2016 . Crime Statistics: Definitions and Explanations: Incidents [Online]. Available: http://www.bocsar.nsw.gov.au/Pages/bocsar_crime_stats/bocsar_glossary.aspx#incidents [16 May 2018, date last accessed]. NSW Office of Liquor Gaming and Racing . ( 2015 ) Liquor Premises List 01/10/2015 . Sydney : NSW Office of Liquor, Gambling and Racing . Osborne K , Baum F , Brown L . ( 2013 ) What Works? A Review of Actions Addressing the Social and Economic Determinants of Indigenous Health . Canberra/Melbourne : Australian Institute for Health and Welfare/Australian Institute for Family Studies . Palk G , Davey J , Freeman J . ( 2007 ) Prevalence and characteristics of alcohol-related incidents requiring police attendance . J Stud Alcohol Drugs 68 : 575 – 81 . Google Scholar CrossRef Search ADS PubMed Perry SW . ( 2004 ) American Indians and Crime: A BJS Statistical Profile, 1992–2002 . Washington, DC : Department of Justice, Bureau of Justice Statistics . Pink B . ( 2011 ) Socio-Economic Indexes for Areas (SEIFA). Technical Paper. Canberra: Australian Bureau of Statistics. Rehm J , Mathers C , Popova S , et al. . ( 2009 ) Global burden of disease and injury and economic cost attributable to alcohol use and alcohol-use disorders . Lancet 373 : 2223 – 33 . Google Scholar CrossRef Search ADS PubMed Rossow I , Norström T . ( 2012 ) The impact of small changes in bar closing hours on violence. The Norwegian experience from 18 cities . Addiction 107 : 530 – 7 . Google Scholar CrossRef Search ADS PubMed Sacks JJ , Gonzales KR , Bouchery EE , et al. . ( 2010 ) National and state costs of excessive alcohol consumption . Am J Prev Med 49 : e73 – 9 . Google Scholar CrossRef Search ADS Seaman P , Edgar F , Ikegwuonu T . ( 2013 ) The role of alcohol price in young adult drinking cultures in Scotland . Drugs (Abingdon Engl) 20 : 278 – 85 . Google Scholar PubMed Shakeshaft A , Doran C , Petrie D , et al. . ( 2014 ) The effectiveness of community action in reducing risky alcohol consumption and harm: a cluster randomised controlled trial . PLoS Med 11 : e1001617 . Google Scholar CrossRef Search ADS PubMed Snijder M , Shakeshaft A , Wagemakers A , et al. . ( 2015 ) A systematic review of studies evaluating Australian indigenous community development projects: the extent of community participation, their methodological quality and their outcomes . BMC Public Health 15 : 1154 . Google Scholar CrossRef Search ADS PubMed StataCorp . ( 2015 ) Stata Statistical Software: Release 14 . College Station, TX : StataCorp LP . Teece M , Williams P . ( 2000 ) Alcohol-related assault: time and place . Trends Issues Crime Criminal Justice 169 : 1 – 6 . Thompson SC , Woods JA , Katzenellenbogen JM . ( 2012 ) The quality of Indigenous identification in administrative health data in Australia: insights from studies using data linkage . BMC Med Inform Decis Mak 12 : 133 – 43 . Google Scholar CrossRef Search ADS PubMed Tsey K . ( 2008 ) The control factor: a neglected social determinant of health . Lancet 372 : 1629 . Google Scholar CrossRef Search ADS PubMed Wiggers JH , Hacker A , Kingsland M , et al. . ( 2016 ) Facilitating police recording of the alcohol-related characteristics of assault incidents: a stepped wedge implementation trial . Drug Alcohol Rev 35 : 30 – 9 . Google Scholar CrossRef Search ADS Williams P . ( 2001 ) Alcohol, Young People and Violence . Canberra : Australian Institute of Criminology . Wundersitz J . ( 2010 ) Indigenous Perpetrators of Violence: Prevalence and Risk Factors for Offending . Canberra : Australian Institute of Criminology . © The Author(s) 2018. Medical Council on Alcohol and Oxford University Press. All rights reserved. This article is published and distributed under the terms of the Oxford University Press, Standard Journals Publication Model (https://academic.oup.com/journals/pages/about_us/legal/notices)

Journal

Alcohol and AlcoholismOxford University Press

Published: May 26, 2018

There are no references for this article.

You’re reading a free preview. Subscribe to read the entire article.


DeepDyve is your
personal research library

It’s your single place to instantly
discover and read the research
that matters to you.

Enjoy affordable access to
over 18 million articles from more than
15,000 peer-reviewed journals.

All for just $49/month

Explore the DeepDyve Library

Search

Query the DeepDyve database, plus search all of PubMed and Google Scholar seamlessly

Organize

Save any article or search result from DeepDyve, PubMed, and Google Scholar... all in one place.

Access

Get unlimited, online access to over 18 million full-text articles from more than 15,000 scientific journals.

Your journals are on DeepDyve

Read from thousands of the leading scholarly journals from SpringerNature, Elsevier, Wiley-Blackwell, Oxford University Press and more.

All the latest content is available, no embargo periods.

See the journals in your area

DeepDyve

Freelancer

DeepDyve

Pro

Price

FREE

$49/month
$360/year

Save searches from
Google Scholar,
PubMed

Create lists to
organize your research

Export lists, citations

Read DeepDyve articles

Abstract access only

Unlimited access to over
18 million full-text articles

Print

20 pages / month

PDF Discount

20% off