Close proximity to alcohol outlets is associated with increased serious violent crime in New Zealand

48 AUSTRALIAN AND NEW ZEALAND JOURNAL OF PUBLIC HEALTH 2012 vol. 36 no. 1© 2012 The Authors. ANZJPH © 2012 Public Health Association of Australia

There is considerable evidence that

alcohol consumption makes a

significant contribution to the global

burden of disease and to economic and social

costs. It is estimated that 3-8% of all global

deaths and 4-6% of global disability-adjusted

life-years are attributable to alcohol.1 For the

consumer, alcohol consumption poses an

increased risk of serious short-term harms

such as death and injury; in the longer-term

excessive alcohol consumption is associated

with an increased risk of more than 60 types

of diseases and conditions, including diabetes,

gastro-intestinal conditions, heart and liver

disease, a range of cancers, and mental and

fetal disorders.2 For broader society, excessive

alcohol consumption contributes to a variety

of harms such as crime, antisocial behaviour

and victimisation as well as unemployment,

absenteeism and relationship breakdowns.3-5

In New Zealand (NZ), the adverse effects

of excessive alcohol consumption place

a considerable burden on the health care

system and society. There are an estimated

1,000 deaths per year that are alcohol-related,

with more than half due to injuries from road

trauma, accidents and self-inflicted harm.6

A high number of deaths are also evident in

young people and Māori; almost 15% of all

deaths in young people and 8% of all deaths

among Māori are attributed to alcohol.6

Between 2002 and 2006 nationally there were

5,413 alcohol-related hospital admissions of

young people aged 15-24 years, with 35% of

these admissions being for alcohol-related

injuries.7 As in many other countries, alcohol

is recognised in NZ as a contributor to a range

of societal harms such as crime and antisocial

behaviour. It is estimated that almost one-

third of all recorded crime in NZ involved an

offender who had consumed alcohol prior to

offending.8

A large number of studies over the past 20

years have examined the relationship between

area-level crime and access to alcohol outlets.

Researchers have typically applied cross-

sectional study designs and found positive and

statistically significant relationships between

crime rates and alcohol availability.9-16

Other studies using longitudinal data and

employing time-series designs have found

positive associations.17-22 In most of these

studies, alcohol outlet density (the number

of alcohol outlets per square kilometre (km2),

square metre (m2) or per population) is used

as a surrogate for alcohol availability. Several

studies have measured alcohol access as the

number of alcohol outlets within a certain

radius of a study respondent’s home;23,24 others

have used the number of alcohol outlets per

roadway mile.12,25 Distance-based measures

have also been employed by using the closest

distance from a respondent’s home to an

alcohol outlet as their access measure.23 In

one study using a multilevel analysis, both

individual distance to the closest alcohol

outlet and the aggregated mean distance of

respondents to outlets was used as the measure

of alcohol availability.26 Individual distance to

the closest alcohol outlet was found to be

unrelated with alcohol consumption, but the

higher the mean distance to the closest alcohol

outlet, the lower the mean level of alcohol

consumption.26 Studies measuring alcohol

Abstract

Objective: To examine the association

between geographic access to alcohol

outlets and serious violent crime in New

Zealand.

Methods: A national study of alcohol outlet

access and serious violent crime used a

cross-sectional ecological analysis. Serious

violence offences recorded between

2005 and 2007 were aggregated for 286

police station areas. Using Geographical

Information Systems (GIS), 9,320 licensed

premises were geocoded and road travel

distances to the closest alcohol outlet

type/category were calculated for each

area. Negative binomial regression models

measured the association between the

distance to the closest alcohol outlet and

the number of serious violent offences in

each police station area, controlling for

area-level measures of social deprivation,

Māori population, young males 15-29 years

and population density.

Results: There were significant negative

associations between distance (access)

to licensed outlets and the incidence of

serious violent offences with greater levels

of violent offending recorded in areas with

close access to any licensed premises

compared to those areas with least access

(IRR 1.5, 95% CI 1.10-2.03); with on-

licensed premises (IRR 1.6, 95% CI 1.16-

2.08); and off-licensed premises (IRR 1.4,

95% CI 1.05-1.93).

Conclusion: Having greater geographic

access to alcohol outlets was associated

with increased levels of serious violent

offending across study areas.

Implications: Alcohol availability and

access promoted under the current

liberalised licensing regime are important

contextual determinants of alcohol-related

harm within New Zealand communities.

Key words: alcohol availability, crime,

violence, Geographical Information

Systems, New Zealand

Aust NZ J Public Health. 2012; 36:48-54

doi: 10.1111/j.1753-6405.2012.00827.x

Submitted: May 2011 Revision requested: October 2011 Accepted: December 2011Correspondence to: Peter Day, GeoHealth Laboratory, Department of Geography, University of Canterbury, Private Bag 4800, Christchurch 8140, New Zealand; e-mail: [email protected]

Close proximity to alcohol outlets is associated with

increased serious violent crime in New Zealand

Peter Day, Gregory Breetzke, Simon Kingham, Malcolm CampbellGeoHealth Laboratory, Department of Geography, University of Canterbury, New Zealand

Alcohol Article

2012 vol. 36 no. 1 AUSTRALIAN AND NEW ZEALAND JOURNAL OF PUBLIC HEALTH 49© 2012 The Authors. ANZJPH © 2012 Public Health Association of Australia

outlet density have also stratified alcohol outlets by licence type

(i.e. on-site and off-site).12,15,17,24,27 For example, Gruenewald and

colleagues12 examined the influence of alcohol consumption at bars,

restaurants and at the homes of drinkers and their friends on drinking

and driving rates; while Treno and colleagues examined how both

on-premise alcohol establishments such as bars and restaurants

and off-premise establishments such as liquor and grocery stores

affected individuals’ self-reported injuries.24 Other work has

examined whether higher alcohol outlet densities influence health

outcomes such as suicide rates,28 sexually transmitted diseases,25

motor vehicle accidents20 and pedestrian injury collisions.29 Almost

without exception, the increased availability of alcohol using various

measures of alcohol outlet density is associated with an increase

in alcohol-related harm such as crime or adverse health outcomes.

A number of studies have emerged from NZ investigating the

relationship between the density of alcohol outlets, and alcohol

consumption and related harms. One study examined the relationship

between physical, socio-economic and social environments and

alcohol consumption patterns of drinkers aged 12-17 years.30 Using

multilevel modelling, they found alcohol outlet density in Auckland

to be associated with greater amounts of alcohol consumption

among teenage drinkers, as was neighbourhood deprivation. A

cross-sectional study found positive associations between alcohol

outlet density and individual drinking and related problems across

six university campuses throughout NZ, and these associations

remained after they controlled for various demographic variables

and pre-university drinking.31 One recent study found statistically

significant associations between the density of off-licences, bars and

clubs and the prevalence of binge drinking.32 After controlling for

demographic and socio-economic variables, the association between

binge drinking and off-licences remained. The authors reported “a

4% increase in binge drinking associated with each extra off-licence

within 1 km of home”.32 Other studies relating access to alcohol

outlets and neighbourhood deprivation have found greater access

to alcohol outlets in more socially deprived neighbourhoods.33,34

Studies examining the relationship between alcohol outlet density

and injury and criminal offending in NZ are less forthcoming,

with only two previous ecological studies known. One investigated

the association between single-vehicle night-time motor vehicle

accidents and alcohol outlet density in Auckland.35 It found positive

and significant associations between restaurant and club density

and vehicular accidents in certain District Health Boards in the

city. The other study found that off-licence alcohol outlet density

was significantly associated with higher levels of anti-social

behaviour, drug and alcohol offences, family violence, and motor

vehicle accidents in Manukau City, while on-licence densities of

certain types of outlets such as clubs and bars were also associated

with a number of criminal offences and motor vehicle accidents.36

Other anecdotal evidence suggests an association between alcohol

outlet density and alcohol-related offending in Queenstown and

Tauranga.8,37 Such has been the concern in NZ about the impact

of excessive alcohol consumption on individuals and society as a

whole that, in 2009, the National Government mandated the New

Zealand Law Commission (an independent crown entity) to review

the country’s liquor laws. Among the Commission’s focus areas

was an examination of the adequacy of the current criteria used

for granting a liquor licence as well as the impact of the density of

alcohol outlets on alcohol-related harms. Relying on predominantly

international evidence, the Commission found that geographical

access to alcohol outlets could have an impact on levels of alcohol-

related harm in NZ, although it recommended that further research

be undertaken to tease out the exact nature of this relationship.37

In this national level study, we begin to address this gap in NZ

literature by examining the association between geographic access

to alcohol outlets and serious violent offending. We use a novel

measure for access to alcohol outlets – median road travel distances

to closest alcohol outlet − as a measure of alcohol availability as we

believe this measure accurately represents both access to alcohol

outlets and alcohol outlet density. Recently, researchers have

proposed that the effect of alcohol outlet densities can be separated

conceptually into a proximity effect (how easily one can access

alcohol) and an amenity effect (how outlets influence the quality

and characteristics of surrounds within the local community).38 In

terms of the former, there is a concept that increasing the alcohol

outlet density makes alcohol more accessible (each new store makes

someone closer to an alcohol outlet).38 While this may be true, the

real effect on the overall access to alcohol in that area is likely to be

negligible, especially if the new alcohol outlet is near other outlets;

which is often the case (i.e. spatial clustering of outlets). The use of

a geographic measure to measure access to alcohol outlets is more

suitable for this purpose since the road travel distance from each unit

area to the new alcohol outlet will be calculated and incorporated

into the new measure.

In the study we also examine the association between access to

alcohol outlets by licence type and category and levels of violent

offending. In particular, we examine whether areas with better

geographic access to alcohol outlets exhibit higher rates of violent

offending compared to areas with poorer access, after controlling

for important area-level influences on violent crime such as social

deprivation.

While national policing authorities perennially identify the

availability of alcohol as a contributor towards higher crime rates

and alcohol-related harm in our communities, this relationship

has rarely been empirically assessed in NZ, and certainly not at a

national level.

Methods We examined the association between distance to nearest alcohol

outlet, grouped by licence category/type, and serious violent

offences recorded between 2005 and 2007 for all 286 police

station areas across NZ. These areas ranged in size from 1 km2 to

12,199 km2 with a median of 566 km2. The relationship between

geographic access to alcohol outlets and area-level violent offending

was tested using negative binomial regression models. A number

of area-level socio-demographic predictors of violent crime were

Alcohol Alcohol availability and violent crime


also controlled for in the models. These predictors have been shown

to be potential area-level predictors of crime in previous studies

in New Zealand and include social deprivation,39 the proportion

of Māori population,40 the proportion of young males 15-29 years

and population density.37

DataThe trading names and addresses of premises licensed (as at July

2006) to sell liquor across the 74 District Licensing Agencies of New

Zealand were provided by the Liquor Licensing Authority which

has statutory authority to collect licensing information. Licensed

premises with on, off, and club licence types and supermarket,

grocery store, bottle store, hotel, tavern, chartered club, and

restaurant licence categories were included in the analysis. Overall,

9,320 of these licensed premises (97% total: 96% on-licensed,

and 99% off-licensed) were successfully geocoded. Data on the

number of violent offences recorded in the calendar years 2005-

07 at police station area level (n=286) were provided by the New

Zealand Police (the data were not available at a smaller geographic

scale). Serious violent offences included homicide (completed and

attempted), robbery, grievous assault and serious assault. There

was an average of 23,119 serious violent offences (5.7 per 1,000

population) recorded over the study period.

The 286 police station areas comprised 41,393 census meshblocks

(the smallest geographic census unit with an average population

of 100 people). Census meshblocks were used for calculating

geographic access to alcohol outlets at the smallest scale available

to determine each police station area’s overall median geographic

distance (access) to a range of alcohol outlet types, and for

aggregating socio-demographic data to police station areas. Census

demographic data for 2006 were sourced from Statistics New

Zealand on the number of Māori aged 15 years and over, the number

of young males 15-29 years, total usual resident population, and

the New Zealand Index of Deprivation score (NZDep2006) which

comprises of nine weighted demographic measures of material

deprivation.41

AnalysisGeographic access using road travel distance to the closest alcohol

outlet for on, off, and for either licence type, and hotel, tavern and

club, supermarket, bottle store, convenience store, and restaurant

licence categories were calculated for each census meshblock

using a Geographic Information System (GIS). A meshblock

was represented by its population-weighted centroid, the centre

of population based on address points (rather than its geometric

centroid) and the travel distance (in metres) to the nearest alcohol

outlet along the road network was calculated using GIS road network

functionality. The median distance meshblock to the closest alcohol

outlet for each licensed premise type and category was calculated

from the meshblocks in each police station area and these median

distances were categorised into quartiles, with 1 being the closest

median distance through to 4 being the furthest distances. Similar

GIS methods have been used in previous New Zealand studies

measuring geographic access to health-related community resources

and the contextual effects on health of access to gambling venues.42,43

To control for important population-level influences (confounders)

on serious violent crime, the proportion of the total population who

were Māori (15+ years), young males (15-29 years), population

density (usual resident population per km2), and the NZ Deprivation

index (NZDep2006) scores were derived for each police station

area by aggregating meshblock level data. The Māori and young

male population measures for each area were then categorised

into quartiles, population density into tertiles, and NZDep2006

categorised into quintiles.

To account for over-dispersion in the crime data (i.e. greater

variance than expected) negative binomial regression models were

used to model the relationship between serious violent offending

and access to alcohol outlets. The analysis was grouped by outlet

type and category. We adjusted for the area-level confounders to

more clearly determine the relationship between serious violent

offending and alcohol outlet access. Because police station areas

can be contiguous, the crime data can be expected to exhibit

spatial autocorrelation, producing spatially correlated errors in

the normal negative binomial regression. However, calculations of

Moran’s I statistics indicated a lack of statistically significant spatial

autocorrelation, indicating that the inclusion of a spatially lagged

variable in the models was not warranted. The first stage baseline model measured the relationship between

serious violent crime and alcohol outlet access. The model at stage 2

controlled for area deprivation (NZDep2006), while stages 3, 4 and

5 controlled for the proportion of young males, the proportion of

Māori population (15+ years), and population density respectively.

The incidence rate ratios (IRRs) and 95% confidence intervals (CIs)

for serious violent offending were calculated for quartile measures

of alcohol outlet access for each type and category of licensed outlet

with the furthest distance quartile being the comparator with an

IRR of 1. The IRRs were obtained by exponentiating the regression

coefficients, i.e. exponentiating the relative change in the incidence

rate brought by a one-unit change in an independent variable holding

other independent variables constant. For the GIS analysis we used

the software package ArcGIS and for the statistical analysis we used

SPSS and STATA 10.

ResultsMedian travel distances to the closest alcohol outlet varied by

police station area with median travel distances being 3-4 times

greater in the quintile of areas with the lowest rates of serious violent

offences compared to the quintile of areas with the highest rates of

serious violent offences. The median travel distance to the closest

off, on or any licensed outlet steadily decreased from 4.5 km, 4.3

km and 4.1 km respectively in areas with the lowest rates of serious

violent offences, to 1.1 km, 1.3 km and 0.9 km in those areas with

the highest rates of serious violent offences (Figure 1).

The population demographics and distribution of alcohol outlets

across police station areas are presented in Table 1, with police

Day et al. Article


station areas grouped into quintiles based on rates of serious

violent offences. Those areas in the highest quintile had the highest

proportions of Māori population (15+ years), percentage of young

males, and greatest population density, and were predominantly

urban areas (75.4%) compared to other police station areas. Police

station area rates of serious violent offences were also strongly

patterned with quintiles of area level social deprivation. The quintile

of areas having the greatest rates of violent offences also had the

highest level of deprivation with the opposite being true for those

areas with the lowest violent offence rates. The number of alcohol

outlets located in police station areas consistently increased with

increasing quintiles of serious violent offence rates. This was evident

for all types and categories of outlets, with 2.2-5.5 times more

outlets in the quintile of areas with the highest violent offence rates

compared to the quintile of areas with the lowest rates (Table 1).

Table 2 presents the results of our investigation into the

relationship between access to alcohol outlets and serious violent

offending. The IRRs and 95% CIs for serious violent offending

predicted from quartiles of distances to the closest alcohol outlet

and adjusted for confounders are shown for each outlet type/

category. The results from the full models including the IRRs and

95% CIs of the confounder variables are not shown due to space

constraints. We found strong evidence of a negative relationship

between distance to closest outlets and serious violent offending

after controlling for area-level confounders. The IRRs in the closest

distance quartile were consistently significantly greater than 1 and

gradually decreased to non-significance with quartiles of increasing

distance. The exception here was off-licence premises which

consistently had significant IRRs over 1 for all distance quartiles.

These findings suggest that geographic access to alcohol outlets is

a significant predictor of serious violent offending and that there is

a negative gradient in the relationship between distance (access) to

alcohol outlets and levels of serious violence.

DiscussionAlcohol is a major contributor to crime, anti-social behaviour and

victimisation in NZ and is responsible for a range of social problems

directly affecting the health and well-being of both offenders and

Table 1: Population demographics and alcohol outlets for Police Station areas (n=286) stratified by quintiles of serious violence offence rates per 1000 population 2005-07.

Serious violence offence rates per 1000 population quintile Category 1

(0- 2.3)

2

(2.4- 3.6)

3

(3.7- 4.8)

4

(4.9- 6.9)

5

(7.0- 31.6)

Total

Population 2006Usual resident (n) 526,125 612,612 760,842 930,489 1,197,459 4,027,527

% Maori 7.7 8.6 11.3 15.5 21.6 14.4

% young males 9.2 9.2 9.5 9.1 10.8 9.7

Population density per km2 7 13 16 17 30 15

% Urban police station areasa 36.2 49.1 66.7 73.7 75.4

NZDEP quintile score for police station areasb 1 2 3 4 5

Alcohol outlets (n) Total 1,053 1,238 1,528 1,897 3,604 9,320

On-licencesc 637 755 948 1,228 2,615 6,183

Restaurants 300 415 540 618 1,333 3,206

Hotel/Tavern/Club 337 340 408 610 1,282 2,977

Off-licences 416 483 580 669 989 3,137

Supermarkets 138 151 188 221 341 1,039

Bottle Stores 93 158 187 199 364 1,001

Hotel/Tavern/Club 185 174 205 249 284 1,097

a Based on Statistics New Zealand urban and rural area classification where 50% or more of the population were living in urban classified areas.b Composite NZDep2006 estimate using meshblock level data (1 low; 5 high deprivation)c Includes Club licence category

Figure 1: Median distance to closest alcohol outlet by license type for Police station areas by quintiles of serious violence offence rates per 1,000 population.


O�-licence

On-licence

All outlets

Police station area serious violence o�ence rate quintile(Low) (High)

1 2 3 4 5

5000

4500

4000

3500

3000

2500

2000

1500

1000

500

0

Dis

tanc

e to

clo

sest

alc

ohol

out

let (

m)

O�-licence

On-licence

All outlets

Police station area serious violence o�ence rate quintile(Low) (High)

1 2 3 4 5

5000

4500

4000

3500

3000

2500

2000

1500

1000

500

0

Dis

tanc

e to

clo

sest

alc

ohol

out

let (

m)


victims. Current levels of violent offending in part reflect the effects

of greater access to alcohol (consumption) which has become

increasingly affordable and available since deregulation through

the Sale of Liquor Act 1989. The availability of alcohol, promoted

through the physical location of outlets, trading hours, and price

of alcohol, is recognised as an important contextual influence on

levels of neighbourhood violence and other alcohol-related harms.44

Our study found a strong negative association at the police

station area level between distance to alcohol outlets and serious

violent crime after controlling for area-level confounders such as

social deprivation and vulnerable populations. Those areas with the

greatest geographic access to alcohol outlets, regardless of licence

type and category, recorded the highest incidences of serious violent

crime, with the incidence of violent crime lessening in areas with

increased travel distances to alcohol outlets. The exception was off-

licence premises, which were a significant predictor of area-level

violent crime incidence, regardless of geographic distance. These

findings demonstrate that geographic access to alcohol outlets is an

independent predictor of area-level violent crime after controlling

for confounders and that outlet licence type and category tends not

to alter the nature of this association (Table 2). Another notable

finding of this national study is that there was a clear social gradient

in the degree of area-level violent crime in NZ as evidenced from

the socio-demographic characteristics of areas when categorised

by violent crime rates. Those areas with the highest rates of violent

offences also had the highest level of social deprivation (Table

1). The findings also show the importance of locational access to

alcohol outlets as the number of outlets was clearly correlated with

area-level violent crime rates (Table 1).

The results of this research are generally congruent with previous

international research investigating this relationship. For example,

in the US researchers also found an increase in crime with an

increase in access to off-premise alcohol outlets although in their

study the researchers used the number of outlets per population as

their measure of alcohol availability.45 Researchers have attempted

to explain these findings by examining the types of neighbourhoods

that typically have greater access to alcohol outlets and by examining

the types of social interactions that occur in and around off-premise

alcohol outlets.46 Research in Australia has also found positive

associations between access to off-premise alcohol outlets and

crime. In Melbourne, the density of liquor licences was found to

be positively associated with rates of domestic violence over time.

The effects were particularly large for packaged alcohol outlets,

suggesting a need for licensing policies that pay more attention

to off-premise alcohol availability.47 Similarly, New South Wales

based research found a significant relationship between violence

and density of off-premise outlets in urban areas.48

The fact that an increase in alcohol outlet density leads to an

Day et al. Article

Table 2: Incidence rate ratios (95% confidence intervals) for serious violent offences predicted from quartiles of distance to closest alcohol outlets by type/category for baseline and full models after adjustment for confounders. Any licence On licence Off licenceDistance to closest outlet quartile

Baseline Baseline Baseline

Closest (<0.9 km) 2.4 (1.81-3.16) (< 1.2 km) 2.0 (1.50-2.67) (<1.1 km) 2.4 (1.78-3.14)

Closer (0.9-1.6 km) 1.2 (0.92-1.60) (1.2-2.1 km) 1.3 (1.01-1.76) (1.1-1.9 km) 1.3 (1.00-1.75)

Further (>1.6-4.5 km) 1.2 (0.93-1.63) (> 2.1-5.6 km) 1.1 (0.83-1.48) (>1.9-4.8 km) 1.3 (0.97-1.70)

Furthest (>4.5 km) 1.0 (> 5.6 km) 1.0 (>4.8 km) 1.0

Full model1 Full model1 Full model1

Closest (<0.9 km) 1.5 (1.10-2.03) (< 1.2 km) 1.6 (1.16-2.08) (<1.1 km) 1.4 (1.05-1.93)

Closer (0.9-1.6 km) 1.3 (0.98-1.65) (1.2-2.1 km) 1.2 (0.96-1.60) (1.1-1.9 km) 1.3 (1.00-1.65)

Further (>1.6-4.5 km) 1.3 (1.01-1.63) (> 2.1-5.6 km) 1.2 (0.96-1.55) (>1.9-4.8 km) 1.4 (1.08-1.73)

Furthest (>4.5 km) 1.0 (> 5.6 km) 1.0 (>4.8 km) 1.0

On and off licence categories On license Baseline Full model1 Off licence Baseline Full model1

Hotel/tav/club Hotel/tav/clubClosest (<1.4 km) 2.2 (1.65-2.91) 1.4 (1.05-1.82) (<1.7 km) 2.1 (1.57-2.78) 1.3 (0.99-1.66)

Closer (1.4-2.7 km) 1.3 (0.96-1.67) 1.2 (0.91-1.51) (1.7-3.1 km) 1.2 (0.89-1.55) 1.0 (0.78-1.28)

Further (>2.7-6.2 km) 1.2 (0.88-1.56) 1.2 (0.95-1.53) (>3.1-6.3 km) 1.0 (0.77-1.36) 1.0 (0.80-1.28)

Furthest (>6.2 km) 1.0 1.0 (>6.3 km) 1.0 1.0

Restaurants Bottle storesClosest (<1.6 km) 2.0 (1.49-2.63) 1.5 (1.15-2.07) (<1.4 km) 1.8 (1.39-2.43) 1.6 (1.21-2.07)

Closer (1.6-3.4 km) 1.2 (0.88-1.54) 1.1 (0.86-1.44) (1.4-3.3 km) 1.0 (0.77-1.33) 1.3 (1.00-1.59)

Further (>3.4-9.9 km) 1.1 (0.85-1.50) 1.2 (0.93-1.33) (> 3.3-9.2 km) 0.9 (0.65-1.14) 1.2 (0.95-1.54)

Furthest (>9.9 km) 1.0 1.0 (> 9.2 km) 1.0 1.0

Supermarket grocery storeClosest (<1.4 km) 2.5 (1.86-3.25) 1.3 (1.00-1.77)

Closer (1.4-3.2 km) 1.3 (0.99-1.73) 1.2 (0.94-1.54)

Further (>3.2-8.4 km) 1.2 (0.91-1.57) 1.1 (0.86-1.39)

Furthest (>8.4 km) 1.0 1.01 Full models include adjustment for area-level social deprivation, % young males 15-29 years, Māori population 15+ years and population density.


increase in area-level crime and alcohol-related harms is certainly

not new.9,13,35,36,44 The results here extend this earlier work by showing

how violent crime, access to alcohol outlets and alcohol outlet density

are significantly associated regardless of licence type (Table 2).

The results also show that as serious violent crime rates increase,

alcohol outlet density and social deprivation increase (Table 1); we

also provide this evidence at a national level. Previous studies of this

nature have been context specific and the results have been limited

to small geographic areas rather than at a national level. Moreover,

alcohol outlet density has typically been operationalised as the

number of alcohol outlets per administrative area (i.e. census unit)

which does not take into account travel patterns and distances. To our

knowledge, this is the first study that examines alcohol outlet density

and crime at a national level using this measure of alcohol availability.

Another notable contribution of this study relates to the

implications for local policy makers as it provides support for

measures that limit the number of alcohol outlet licences within

geographic areas especially those with socially deprived and

vulnerable populations and high levels of alcohol-related violence

and harm. The current Sale of Liquor Act does not permit a licence

to be declined on the basis of the number of existing outlets in an

area, thus there are numerous high outlet density areas across NZ.37

Under the current Act, which governs who may sell alcohol, to whom

and the hours at which sales can be made, liquor licence applications

must ensure proposed licensed premises meet the requirements

of the Resource Management Act 1991 (RMA) and the Building

Code. However, the RMA is an environmental statute and does not

function as a licensing tool. Other than influencing liquor licence

trading hours and their effect on surrounding amenities, the RMA

is not well suited to imposing restrictions on the location, number

or size of licensed premises to control the availability of alcohol

as this is not necessarily for an environmental purpose.37 Alcohol

availability encompasses not only the geographic proximity of

outlets but also trading hours and the price at which alcohol is

sold. The significance of each of these may vary across different

environments but evidence suggests that these factors and their

influence on each other impacts on alcohol consumption and levels

of alcohol-related harm at the community level.44 This study has

shown how the confounding effects of social deprivation and young

and vulnerable populations combined with outlet density is likely

to lead to greater levels of serious violent crime.

The full range of options available to NZ policy makers centre

around the Law Commission’s issues paper on liquor licensing law

reform which calls for a complete overhaul of the current Sale of

Liquor Act 1989 and makes 153 recommendations on strategies for

dealing with alcohol-related harms.37 The three broad areas where

these fall into are supply controls, demand reduction and problem

limitation. These recommendations are all-encompassing and

attempt to address the complexities of the causes of and solutions

to alcohol-related harm rather than a piecemeal approach of, for

example, simply recommending limitations in the number of alcohol

outlet licences within geographic areas. The Law Commission issues

paper recommends that all Territorial Local Authorities (TLAs)

adopt a local alcohol policy which provides a mechanism for local

communities to have a say in licensing decisions.37 To reduce the

wide variation in alcohol availability across TLAs, the Commission

stated that policies should include strategies for reducing alcohol-

related harms such as local restrictions on licence trading hours and

the identification of areas which have reached ‘saturation levels’

in terms of outlet density and the cumulative impact of licensing

but without specifying what these levels are. While such proposed

strategies are admirable, their implementation is likely to be met

with some resistance, particularly from existing and would-be

liquor licence holders. By providing clear empirical evidence that

an increase in alcohol outlet density within a region increases

alcohol-related harms such as violent offending, this may provide

TLAs with important information to help develop local alcohol

plans for liquor licence submissions.

This study has limitations. First, given the ecological study design

the association between alcohol outlet access and violent crime is

strongly related to geographical scale. A different geographical scale

may result in changes in the degree of association between alcohol

outlet access and violent offending. The extent of causal inference

one can conclude from the results is limited given the ecological

study design. Second, alcohol outlet size and licence operating

hours are important factors in the availability of alcohol and are

strongly linked to alcohol-related harms however the relative size of

premises and licence hours were not included in the available data

for our analysis. Third, a number of confounders were not adjusted

for in the analysis as data were not available including the level of

policing which may influence reported crime, the level of alcohol

sales and alcohol consumption. Fourth, we have used recorded

violent offences data regardless of whether or not these were alcohol

related as data linking alcohol involvement with individual offences

were not available. However, given the high degree of alcohol

involvement associated with violent offending, the data are likely

to be highly indicative of alcohol-related violence. Finally, we have

not included health outcomes such as Emergency Department and

hospital admissions that result as a consequence of violent crime.

The numbers of area-level violent offences are suggestive only of

the geographic variation in the burden of alcohol-related harm on

the health care system and health inequalities.

This study has established that in NZ there is a strong association

between geographic access to alcohol outlets and area-level

serious violent crime, with high incidence of crime in areas with

greatest locational access to alcohol outlets and reduced crime

incidence in areas where access is less. The high and socially

disparate density of alcohol outlets promote environments which

foster social conditions that may impact on alcohol-related harms

and contribute to inequalities in health. Policies to reduce the

availability of alcohol should involve greater efforts to improve local

alcohol retail environments including interventions restricting the

establishment and trading hours of alcohol outlets, particularly in

socially disadvantaged neighbourhoods and neighbourhoods with

vulnerable populations.



AcknowledgementsWe are grateful to Bruce Holmes with the Liquor Licensing

Authority for the alcohol licence data and Gavin Knight and Obert

Cinco from the Statistics Section, New Zealand Police for the

provision of crime data. Peter Day is funded through the GeoHealth

Laboratory, a collaboration between the University of Canterbury

and the New Zealand Ministry of Health.

References1. Rehm J, Mathers C, Popova S, Thavorncharoensap M, Teerawattananon Y,

Patra J. Global burden of disease and injury and economic cost attributable to alcohol use and alcohol-use disorders. Lancet. 2009;373(9682):2223-33.

2. World Health Organization. Expert Committee on Problems Related to Alcohol Consumption. Second Report. WHO Technical Report Series No.: 944 [Internet]. Geneva (CHE): WHO; 2007 [cited 2010 Dec]. Available from: http://www.who.int/substance_abuse/expert_committee_alcohol_trs944.pdf

3. Dale C, Livingstone M. The burden of alcohol drinking on co-workers in the Australian workplace. Med J Aust. 2010;193(3):138-40.

4. Karvonen S, Rimpela A. Urban small area variation in adolescent’s health behavior. Soc Sci Med. 1997;45(7):1089-98.

5. Pridemore W. Vodka and violence: alcohol consumption and homicide rates in Russia. Public Health Matters. 2002;92(12):1921-30.

6. Connor J, Broad J, Jackson R, Vander Hoorn S, Rehm J. The Burden of Death, Disease and Disability due to Alcohol in New Zealand. Wellington (NZ): Alcohol Liquor Advisory Council; 2005.

7. Craig E, Jackson C, Han D, NZCYES Committee. Monitoring the Health of New Zealand Children and Young People: Literature Review and Framework Development. Auckland (NZ): Paediatric Society of New Zealand, New Zealand Child and Youth Epidemiology Service; 2007.

8. Rebecca Stevenson. Police National Alcohol Assessment [Internet]. Wellington (NZ): New Zealand Police; 2009 April [cited 2010 Dec]. Available from: http://www.police.govt.nz/sites/default/files/Police-National-Alcohol-Assessment.pdf

9. Britt H, Carlin B, Toomey T. Neighborhood level spatial analysis of the relationship between alcohol outlet density and criminal violence. Environ Ecol Stat. 2005;12:411-26.

10. Freisthler B, Midanik L, Gruenewald P. Alcohol outlets and child physical abuse and neglect: applying routine activities theory to the study of child maltreatment. J Stud Alcohol Drugs. 2004;65(5):586-92.

11. Freisthler B, Needell B, Gruenewald P. Is the physical availability of alcohol and illicit drugs related to neighbourhood rates of child maltreatment? Child Abuse Negl. 2005;29(9):1049-60.

12. Gruenewald P, Johnson F, Treno A. Outlets, drinking and driving: a multilevel analysis of availability. J Stud Alcohol Drugs. 2002;63(4):460-8.

13. Gyimah-Brempong K. Alcohol availability and crime: evidence from census tract data. South Econ J. 2001;68(1):2-21.

14. Gyimah-Brempong K, Racine J. Alcohol availability and crime: a robust approach. Appl Econ. 2006;38(11):1293-307.

15. Lipton R, Gruenewald P. Spatial dynamics of violence and alcohol outlets. J Stud Alcohol Drugs. 2002;63(2):187-95.

16. Livingston M. Alcohol outlet density and assault: a spatial analysis. Addiction. 2008;103(4):619-28.

17. Gruenewald P, Remer L. Changes in outlet densities affect violence rates. Alcohol Clin Exp Res. 2006;30(7):1184-93.

18. Norström T. Outlet density and criminal violence in Norway1960-1995. J Stud Alcohol Drugs. 2000;61(6):907-11.

19. Scribner R, Theall K, Ghosh-Dastidar B, Mason K, Cohen D, Simonsen N. Determinants of social capital indicators at the neighborhood level: a longitudinal analysis of loss of off-sale alcohol outlets and voting. J Stud Alcohol Drugs. 2007;68(6):934-43.

20. Treno A, Johnson F, Remer L, Gruenewald P. The impact of outlet densities on alcohol-related crashes: a spatial panel approach. Accid Anal Prev. 2007;39(5):894-901.

21. Trolldal B. An investigation of the effect of privatization of retail sales of alcohol on consumption and traffic accidents in Alberta Canada. Addiction. 2005;100(5):662-71.

22. Yu Q, Scribner R, Carlin B, Theall K, Simonsen N, Ghosh-Dastidar B, et al. Multilevel spatio-temporal dual changepoint models for relating alcohol outlet destruction and changes in neighbourhood rates of assaultive violence. Geospat Health. 2008;2(2):161-72.

23. Pollack C, Cubbin C, Ahn D, Winkleby M. Neighbourhood deprivation and alcohol consumption: does the availability of alcohol play a role? Int J Epidemiol. 2005;34(4):772-80.

24. Treno A, Gruenewald P, Johnson F. Alcohol availability and injury: the role of local outlet densities. Alcohol Clin Exp Res. 2001;25(10):1467-71.

25. Cohen DA, Ghosh-Dastidar B, Scribner R, Miu A, Scott M, Robinson P, et al. Alcohol outlets, gonorrhoea, and the Los Angeles civil unrest: a longitudinal analysis. Soc Sci Med. 2006;62(12):3062-71.

26. Scribner R, Cohen D, Fisher W. Evidence of a structural effect for alcohol outlet density: a multilevel analysis. Alcohol Clin Exp Res. 2000;24(2):188-95.

27. Treno A, Grube J, Martin S. Alcohol availability as a predictor of youth drinking and driving: a hierarchical analysis of survey and archival data. Alcohol Clin Exp Res. 2003;27(5):835-40.

28. Escobedo L, Oritz M. The relationship between liquor outlet density and injury and violence in New Mexico. Accid Anal Prev. 2002;34(5):689-94.

29. LaScala E, Johnson F, Gruenewald P. Neighborhood characteristics of alcohol-related pedestrian injury collisions: a geostatistical analysis. Prev Sci. 2001;2(2):123-34.

30. Huckle T, Huakau J, Sweetsur P, Huisman O, Casswell S. Density of alcohol outlets and teenage drinking: living in an alcogenic environment is associated with higher consumption in a metropolitan setting. Addiction. 2008;103(10):1614-21.

31. Kypri K, Bell M, Hay G, Baxter J. Alcohol outlet density and university student drinking: a national study. Addiction. 2008;103(7):1131-8.

32. Connor J, Kypri K, Bell M. Alcohol outlet density and alcohol-related harm. Proceedings of the Australasian Epidemiological Association; August 31-September 1, 2009. The Australasian Epidemiologist. 2009;16(2):25.

33. Hay G, Whigham P, Kypri K, Langley J. Neighbourhood deprivation and access to alcohol outlets: a national study. Health Place. 2009;15(4):1086-93.

34. Pearce J, Day P, Witten K. Neighbourhood provision of food and alcohol retailing and social deprivation in Urban New Zealand. Urban Policy Res. 2008;26(2):213-27.

35. Mathieson A. Alcohol in Auckland − Reducing Associated Harm. Auckland (NZ): Auckland Regional Public Health Service; 2005.

36. Cochrane W, Cameron M, McNeill K, Melbourne P, Morrison S, Robertson N. A Spatial Econometric Analysis of Selected Impacts of Liquor Outlet Density in Manukau City. Hamilton (NZ): Population Studies Centre, University of Waikato; 2010.

37. New Zealand Law Commission. Alcohol in Our Lives: An Issues Paper on the Reform of New Zealand’s Liquor Laws (NZLC IP15, 2009) [Internet]. Wellington (NZ); Government of New Zealand; 2009 [cited 2010 Dec]. Available from: http://www.lawcom.govt.nz/

38. Livingston M, Chikritzhs T, Room R. Changing the density of alcohol outlets to reduce alcohol-related problems. Drug Alcohol Rev. 2007;26:557-66.

39. Maynard K, Coebergh B, Anstiss B, Bakker L, Huriwai T. Toward a new assessment: The identification of cultural factors which may pre-dispose Mäori to crime. Soc Policy J N Z. 1999;13:43-58.

40. Newbold G. Crime in New Zealand. Wellington (NZ): Dunmore Press; 2000.41. Salmond C, Crampton P, Atkinson J. N Z Dep2006 Index of Deprivation.

Wellington (NZ): Department of Public Health, University of Otago; 2007.42. Pearce J, Mason K, Hiscock R, Day P. A national study of neighbourhood access

to gambling opportunities and individual gambling behaviour. J Epidemiol Community Health. 2008;62(10):862-8.

43. Pearce J, Witten K, Bartie P. Neighbourhoods and health: a GIS approach to measuring community resource accessibility. J Epidemiol Community Health. 2006;60(5):389-95.

44. Popova S, Giesbrecht N, Bekmuradov D, Patra, J. Hours and days of sale and density of alcohol outlets: impacts on alcohol consumption and damage: a systematic review. Alcohol Alcohol. 2009;44:500-16.

45. Graham K. Isn’t it time we found out more about what the heck happens around American liquor stores? Addiction. 2006;101:619-20.

46. Freisthler B, Midanik L, Gruenewald PJ. (2004). Alcohol outlets and child physical abuse and neglect: Applying routine activities theory to the study of child maltreatment. J Stud Alcohol Drugs. 2004;65:586-92.

47. Livingston M. A longitudinal analysis of alcohol outlet density and domestic violence. Addiction. 2011;106:919-26.

48. Stevenson RJ, Lind B, Weatherburn D. The relationship between alcohol sales and assault in New South Wales, Australia. Addiction. 1999;94:397–410.

Day et al. Article

Documents

Close proximity to alcohol outlets is associated with increased serious violent crime in New Zealand