Increasing the use of alcohol screening and brief intervention in New Zealand

Abstract

New Zealand has high rates of acute alcohol-related harm relative to chronic harm, and when compared with other countries. Acute harm is typically caused by binge drinking. New Zealand alcohol use surveys consistently find that about 20–25% of adult drinkers binge drink. There is substantial overseas evidence that alcohol screening and brief intervention (SBI) in primary health care is effective and cost-effective for reducing alcohol harm among binge drinkers. However, while there is also growing evidence that alcohol SBI can successfully be implemented in New Zealand, the approach has been significantly under-utilised. This article provides a brief overview of New Zealand research that supports a case for increasing the use of alcohol SBI in primary health care. It also highlights that when looking to successfully implement alcohol SBI in other settings, it is important to ensure that the intervention is simple and purpose-built for both the setting and recipient.

Keywords:

• alcohol
• intervention
• screening and brief intervention
• research
• hazardous consumption
• binge drinking
• primary health care

Introduction

Hazardous consumption of alcohol, or ‘binge drinking’,1 in New Zealand places a significant burden on the nation's health, social and justice systems and, more importantly, an unacceptable burden of harm on individuals, family/whānau and communities.

There is a substantial body of international evidence on alcohol screening and brief intervention (SBI) and, in particular, on the effectiveness of alcohol SBI for reducing hazardous drinking (see, for example, Moyer et al. 2002; Ballesteros et al. 2004; Babor & Kadden 2005; Riper et al. 2009). However, there is less New Zealand-based research on alcohol SBI, specifically on how it can work in real-time primary health care and other settings.

This has created a conundrum. In the New Zealand health system very little attention has been given to how we could better identify the 20–25% of drinkers who consume alcohol hazardously, and intervene early and effectively to reduce their consumption. As it is internationally, the vast majority of funding for alcohol treatment in New Zealand has consistently been allocated to the specialist addiction sector, which sees less than 1% of drinkers yearly (National Committee for Addiction Treatment 2008). A paradigm shift is needed, so that the focus of alcohol intervention is extended to capture the much larger group of hazardous drinkers who are incurring the lion's share of harm from New Zealand's binge drinking culture.

Primary health care is an important setting for reaching hazardous drinkers, given that most New Zealanders see their doctors at least once a year (Ministry of Health 2008). However, these drinkers need to be prompted about their alcohol consumption, because it is unlikely that they will visit their doctor for the sole purpose of discussing their concerns about their drinking.

This article summarises recent New Zealand research, which has shown that alcohol SBI that are fit for purpose for both practitioners and recipients can be successfully and routinely implemented, and can reduce binge drinking levels. It also highlights evidence of the cost benefits of alcohol SBI in primary health care and emergency departments, and the potential for applying and testing this intervention in other settings.

Alcohol use and related harms in New Zealand

Globally, alcohol causes on average 3.2% of all deaths and 4% of all disability-adjusted life years (DALYs; World Health Organization 2010). In New Zealand, it causes 3.9% of all deaths and 7.4% of all DALYs (Connor et al. 2004).2 Despite the limitations often associated with inter-country comparative analyses, these studies and others indicate that New Zealand has a very high rate of ‘acute’ alcohol-related harm3 on an international scale. Acute harm is typically caused by drinking large amounts of alcohol on a single occasion or hazardous/binge drinking.

National surveys consistently show that approximately one in four New Zealand drinkers aged 16–64 years binge drink on a typical drinking occasion (Ministry of Health 2009; Fryer et al. 2011). This approximation is similar to estimates of potentially hazardous drinking4 in New Zealand, with just over one in five (21.1%) drinkers aged 15 years and over reporting that they drink in a potentially hazardous way (Ministry of Health 2008). In addition, New Zealand's 2007/2008 national alcohol and drug use survey found that 61.6% of New Zealand drinkers aged 16–64 years reported binge drinking at least once in a 12-month period, and 12.6% reported binge drinking weekly (Ministry of Health 2009).

Hazardous/binge drinking is most common among young people aged 18–24 years, and males (Ministry of Health 2008). A high proportion of Māori and Pacific drinkers also consume alcohol in a potentially hazardous way (Ministry of Health 2008, 2009), in part reflecting the youthfulness of the Māori and Pacific population and the disproportionate amount of alcohol-related harm among Māori and Pacific drinkers.

Injuries are responsible for about half of New Zealand's alcohol-attributable deaths (Connor et al. 2004). In a 2007 international study of alcohol and injury in emergency departments, the World Health Organization found that, of the 12 countries providing data on consumption of alcohol before injury, New Zealand had the second highest proportion of alcohol-related injuries (36% compared with the average 20.4%; World Health Organization 2007).

A large number of alcohol-related injuries involve violence of some kind. For example, of the 2,581 patients presenting with facial fractures between 1996 and 2006 at Christchurch Hospital's specialist Oral and Maxillofacial Surgery Service, almost half (49%) had alcohol-related injuries. Inter-personal violence was a factor in 78% of all alcohol-related facial fractures in this period. In comparison, inter-personal violence was a factor in only 12% of non-alcohol-related facial fractures (Lee 2009). In a study conducted in Hawke's Bay Hospital's emergency department during May 2008, alcohol was found to be a contributing factor in 82% of all injuries due to intentional harm by third parties (Jones 2009).

An international study comparing the costs of harmful alcohol use in several countries estimated that between 25% and 51.6% of New Zealand's total spend on public order and safety was alcohol-related. This figure was significantly higher than the costs incurred by the other nine countries reviewed (Baumberg 2006).

Alcohol screening and brief intervention

There is strong international research, and a growing New Zealand evidence base, to show that alcohol SBI in primary health care is effective in reducing hazardous alcohol consumption and harm for individuals (Moyer et al. 2002; Ballesteros at al. 2004; Babor & Kadden 2005; Riper et al. 2009), and for the broader population if practised widely enough (Chisholm et al. 2004). The consistent message from international research is that alcohol SBI has a statistically significant effect on the hazardous drinking behaviour of male and female drinkers of all ages who are not alcohol dependent (Moyer et al. 2002; Ballesteros 2004; Babor & Kadden 2005).5

Some meta-analyses have found a 10–13% reduction in alcohol consumption within a six-month period for recipients of brief interventions compared to control groups, and a 16–34% reduction within a 12-month period (Anderson et al. 2004; Whitlock 2004; Bertholet et al. 2005; Babor et al. 2007; Kaner et al. 2007). Other meta-analyses have shown decreased emergency department visits and hospitalisations (Irvin 2000; D'Onofior & Degutis 2002) and a halving of the odds of sustaining an alcohol-related injury in the 6 to 12 months following presentation at an emergency department and receiving a brief intervention (Havard et al. 2008).

A review of trials with pregnant women found that brief intervention is effective in reducing alcohol consumption for this group of women. One study found that the brief intervention group was five times more likely than the control group to be abstinent by the end of the third trimester of pregnancy. The brief intervention also had a positive effect on birth outcomes of their children, as mothers who received brief intervention had babies with higher birth weights, higher birth lengths and lower fetal mortality rates compared with the control group individuals (Nilsen 2009).

Research further suggests that about 1 in 5 to 1 in 8 high-risk drinkers who receive alcohol SBI will reduce their drinking to low-risk levels (Babor & Grant 1992; Babor at el. 1994). This compares favourably with smoking cessation, where a change of behaviour occurs for approximately 1 in 20 who receive treatment (and 1 in 10 if nicotine replacement therapy is included; Department of Health 2005).

There is also evidence that the briefest of alcohol interventions can be just as effective in reducing hazardous drinking and alcohol-related harm as its more intensive counterparts (Richmond et al. 1995; Wutzke et al. 2002; Kaner et al. 2007; Anderson et al. 2009; Secretariat World Health Assembly 2010). In addition, international studies and a recent New Zealand ‘potential cost savings’ analysis suggest that alcohol SBI in primary health care and emergency departments is a cost-effective strategy for improving health outcomes and reducing alcohol-related harm (World Health Organization 2009; Love et al. 2011).

A typical alcohol SBI approach

A core feature of alcohol SBI is that it captures hazardous drinkers who may not seek help for alcohol problems. This means that alcohol SBI is primarily an opportunistic treatment intervention that leverages off people's presentations to services for reasons generally unconnected with alcohol problems. Alcohol SBI is, therefore, more often than not delivered by generalists rather than experts in alcohol or alcohol treatment.

A typical alcohol SBI starts with some form of screening to identify whether the individual has a drinking problem requiring intervention. Alcohol screening tools are largely short and structured questionnaires that take between one and five minutes to administer. Of these tools, the Alcohol Use Disorders Identification Test (AUDIT) is the most well known and used.6

Once assessed as requiring an intervention, the hazardous drinker either receives some brief advice (usually a short, structured session to encourage change in their drinking behaviour) or is referred, as required, to a more specialised and targeted alcohol intervention.

The intervention itself aims to moderate individuals’ drinking patterns to less hazardous levels and motivate those requiring further treatment to engage with appropriate programs (Kaner et al. 2007; Andrews 2010). An intervention will normally incorporate some or all of the following elements:

•	feedback on the person's alcohol use and any alcohol-related harms
•	clarification on what constitutes low-risk alcohol consumption
•	information on the harms associated with hazardous drinking
•	details of the benefits of reducing alcohol intake
•	an analysis of high-risk situations for drinking, and coping strategies
•	development of a personal plan to reduce consumption (Kaner et al. 2007).

Alcohol SBI is usually a face-to-face interaction, but telephone, mail and, more recently, computers have been used.

New Zealand research on alcohol SBI in primary health care

Primary health care services, and particularly general practices, are an important setting for delivering opportunistic alcohol SBI. This is because a large proportion (an estimated 80%) of New Zealanders go to general practitioners (GPs) at least yearly (Ministry of Health 2008), meaning that many hazardous drinkers are likely to have contact with primary health care services at some time.

While the 2006/07 New Zealand Health Survey data revealed no significant differences in access rates by ethnicity, it did highlight gender differences, with women significantly more likely than men to have seen GPs in the previous 12 months (83.4% of women compared with 76.6% of men; Ministry of Health 2008). There is also some research to suggest that, in general, patients are open to receiving preventative lifestyle advice from their doctors (Slama et al. 1989).

Two recent studies supported by the Alcohol Advisory Council of New Zealand (ALAC) found that alcohol SBI can work effectively in New Zealand primary health care settings. These studies (outlined below) also confirm international research on the efficacy of opportunistic alcohol SBI for reducing hazardous drinking (Moyer et al. 2002; Ballesteros et al. 2004; Babor & Kadden 2005).

A web-based alcohol SBI in university primary health care

The first study trialled a web-based alcohol SBI (e-SBI) with young people aged 17–29 years attending the University of Otago's student primary health care service. The researchers conducted a randomised controlled trial in which students who scored in the hazardous or harmful range of the AUDIT tool and consented to the trial (576 of the 975 students who completed the screening) were randomised to receive:

•	an information pamphlet (the ‘control group’)
•	a web-based motivational intervention (the ‘single-dose e-SBI group’) or
•	a web-based motivational intervention with further interventions one and six months later (the ‘multi-dose e-SBI group’) (Kypri et al. 2008).

The study found that the e-SBI was effective in both the uptake of alcohol screening and reducing alcohol use and harm. Eighty-seven per cent of the students who participated in the study completed the screening, and those receiving a brief intervention reported a 20–30% reduction in episodic risky drinking and alcohol-related problems compared with the control group when followed up and reassessed six months (Kypri et al. 2005) and 12 months later (Kypri et al. 2008). The study also found that the single-dose e-SBI group reduced their hazardous drinking to levels similar to those in the multi-dose e-SBI group. In other words, the e-SBI in itself was sufficiently effective for participants to reduce hazardous drinking and maintain the effects for 12 months (Kypri et al. 2008).

An ABC alcohol SBI approach in general practice settings

The second study trialled whether an approach used in New Zealand general practices for routinely screening for tobacco use and providing brief advice or referral could be successfully adapted for alcohol use.

Known as the ABC alcohol approach, it involved:

•	A: asking all patients about their alcohol consumption.
•	B: offering brief advice to those identified as risky drinkers.
•	C: referring patients to a nurse or for specialist alcohol counselling if appropriate.

The approach was trialled by the Whanganui Regional Primary Health Organisation (WRPHO) in its 15 general practices from May 2010 to January 2011. The Whanganui region has a population of about 64,000, of whom just over a third live in rural communities. Twenty-four per cent identify as Māori (significantly higher than the 14% national average) and the region has a high level of social deprivation (Newton & McMenamin 2011).

Just over 90% of the region's population were enrolled in the WRPHO during the trial period, of whom 67% identified as Pākehā/European, 18% Māori and 5% ‘other ethnicity’. Thirty-nine per cent of the enrolled population were classified as having high socio-economic needs (Newton & McMenamin 2011).

During the trial, 43% of the WRPHO's enrolled population (15 years and over) were asked about their alcohol consumption using the AUDIT-C screening tool, with one practice reaching 74% of its patients.

Of the 43% who were asked about their alcohol use, 24% were recorded as consuming above the recommended drinking guidelines and of these 36% were given brief advice or referred to specialist services (Newton & McMenamin 2011). While fewer Māori than Pākehā were asked about their drinking (36% Māori compared with 46% Pākehā), they were more likely to receive a brief intervention (40% of those Māori who were above the recommended drinking guidelines received a brief intervention compared with 34% Pākehā; Newton & McMenamin 2011).

This WRPHO trial showed that primary health care practitioners were able to successfully screen and, in a number of cases, offer brief advice or refer to specialists if needed.

The high rates of screening can be partly attributed to an innovative patient reminder system and user-friendly IT tools that prompted frontline clinicians (GPs and practice nurses) to ask about their patients’ drinking and to then lead them through a brief, easy and evidence-based intervention. The IT tools were developed to integrate with the practices’ patient management systems and were fit for purpose in primary health care. In addition, a process evaluation of the trial concluded that the alcohol ABC model could be replicated in other regions with relative ease if practices adopted the necessary integrated IT systems, protected nurse time and commitment to addressing lifestyle issues (Cvitanovic & Gifford 2011).

The evaluation also raised the need for skills-based training to support practitioners to provide brief advice, as they were less comfortable in doing this (Gifford et al. 2012). A training package has since been developed to address this. The implementation of alcohol SBI in primary health care settings, and in particular the provision of brief advice, will therefore be enhanced by the improved training package. This training will also contribute to addressing inter-practice variance, at least to some degree.

Given that almost half of the WRPHO's population aged 15+ years were screened within the trial's first year, the ABC alcohol approach could reach 80% of the 15+ population of any primary health care practice within two to three years.7 This means it has the potential to reduce alcohol-related harm among a significant number of individuals (who would not normally seek help or be identified as risky drinkers) and their families/whānau, as well as improve health outcomes for a wide range of medical conditions.

Potential cost savings

As well as having the potential to impact positively on the lives of thousands of New Zealanders, there are cost benefits involved in implementing alcohol SBI.

A recent New Zealand report on the potential cost savings of alcohol SBI in primary health care services and emergency departments concluded a likely $1.74 return for every dollar spent on alcohol SBI in general practice over a three-year period, and a $2.13 return over three years for every dollar spent in an emergency department setting. The report also highlighted the potential for cost savings in other sectors, particularly justice (Love et al. 2011).

While the report noted the proportionately greater return on investment in an emergency department setting, it suggested that the absolute gains are potentially greater from implementation in primary health care services because of the much more frequent contact these services have with the wider population (Love et al. 2011). Overall the study suggested that, based on a modelling of the available literature, alcohol SBI is likely to both improve health outcomes and save the health sector money.

The findings of this report are consistent with international studies of the cost-effectiveness of alcohol SBI and the relatively low cost of implementing it in these settings (Fleming et al. 2002; Kraemer 2007; Andrews 2010). For example, a recent report published by the UK Department of Health found that the savings after seven years exceeded costs by a factor of nearly 12 to one. It concluded that, in terms of public expenditure, the interventions offered good value for money, as the combined savings for the health and criminal justice systems exceeded the cost of the interventions by a factor of more than three to one (Knapp et al. 2011).

Increasing the use of alcohol SBI in New Zealand

While there is evidence that alcohol impacts on more than 60 medical conditions (World Health Organization 2011), patients presenting to primary health care and other health settings with one or more of these conditions are typically not being routinely asked about their drinking, or being provided with appropriate health information on their alcohol consumption. This is in spite of the strong evidence of the effectiveness and cost-effectiveness of alcohol SBI in primary health care.

This lack of consistent and routine alcohol SBI in primary health care is likely to be related to several factors. For instance, Moriarty et al. (2011) argue that alcohol SBI tools and guidelines developed for primary health care do not make allowance for barriers such as the competing demands (and practitioners’ mindful prioritisation of these), time constraints, the sensitive nature of the alcohol topic and that patients are likely to be presenting for concerns unrelated to their drinking. Others also cite similar issues. For instance, Dew (2012) argues that practitioners in primary health care commonly resist implementing public health initiatives, such as opportunistic screening in clinical consultations, because of the perceived constraints this places on clinical autonomy and the risk to the doctor–patient relationship. Practitioner fear of damaging the doctor–patient relationship is also highlighted by other researchers as a major barrier to having the alcohol conversation (McCormick 2006; Sheridan & Butler 2008).

However, given the significant impact of hazardous drinking on the health and well-being of New Zealanders, and the effectiveness of alcohol SBI in primary health care for reducing this impact, it is important to find ways around these barriers where possible.

Further, while alcohol SBI is increasingly being implemented in primary health care and emergency departments internationally, there has been no similar level of implementation in New Zealand.

Since 2009 ALAC has been working with others to improve the New Zealand evidence on the efficacy of alcohol SBI and to identify a model that would work well in a New Zealand context. The WRPHO and university e-SBI trials demonstrate that fit for purpose alcohol SBI can work well in New Zealand and can overcome the usual barriers to implementation. The e-SBI trial utilised a web-based approach for techno-savvy university students, and the WRPHO trial demonstrated that alcohol SBI can also work well in primary health care, particularly if it is fully integrated into patient management systems and practitioners have a simple, evidence-based process that they know can make a difference to patient outcomes and are provided with skills-based training. This work has been pivotal in getting alcohol SBI on the New Zealand government's agenda as a key component of any strategy for reducing alcohol harm. In addition there also appears to be an increasing recognition by the government of the need to screen for problematic alcohol use within a variety of public services (justice, health and other social services) and to intervene early.

In accord with the findings of the WRPHO and e-SBI trials, McCormick and others (2010) argue that for alcohol SBI to work successfully in different settings it must be fit for purpose for those settings.

Further, attempts to introduce alcohol SBI in a way that expects practitioners to develop skills that are very different to, and outside of, what they see as their legitimate role are highly likely to fail (e.g. doctors and nurses are not employed to be alcohol and other drug counsellors). As such, it is important to ensure that:

•	practitioners can be true to their own professional orientations
•	the alcohol SBI is simple and integrated into current systems and practices
•	there is a good reminder system and tools to support changed practice.

Consequently, because emergency departments differ significantly from primary health care services, an alcohol SBI is likely to look very different in emergency department settings and utilise quite different support mechanisms. For example, an intoxicated injury patient would be identified as ‘at risk’ straight away and could be referred to an alcohol helpline. The helpline could then phone the patient within a day or two to provide a brief assessment, advice and information, a more intensive brief intervention or referral to another service, depending on what is required.

Similarly, specific types of intervention are likely to work better with certain groups. For example, a web-based alcohol SBI is likely to be effective in reaching techno-savvy university students, and there is growing evidence that this approach is successful in reducing alcohol use among some students (Kypri et al. 2004).

Other settings such as sexual health clinics, justice and social services also need interventions that fit well with their roles and relationships with individuals, as well as easy access to a more intensive intervention (such as a helpline) to refer the individual to if required.

Research is needed to test the efficacy and likely success of implementing alcohol SBI in these other settings in New Zealand. This is because, while the evidence of the effectiveness of alcohol SBI in primary health care is strong, it is less convincing in other service settings.

Conclusions

There is substantial overseas evidence that using alcohol SBI in primary health care is effective and cost-effective for reducing alcohol-related harm among binge drinkers. However, while alcohol SBI is increasingly being implemented in primary care internationally there has, until recently, been no provision for similar systematic implementation in New Zealand. This is in spite of the significant burden of hazardous drinking on New Zealand society.

Recent efforts to increase the New Zealand evidence base have clearly shown that alcohol SBI can be successfully applied in real-time primary health care settings. The University of Otago's e-SBI study found its use within a university primary health care setting was effective in both the uptake of alcohol screening and reducing alcohol use and harm, while the WRPHO demonstrated that it had the capacity, capability and willingness to ask patients routinely about their alcohol use. There is also evidence that alcohol SBI is likely to improve health and other social and justice sector outcomes, as well as saving taxpayer money.

This research has provided a good case for both government and primary health care practitioners to explore how they can best introduce opportunistic alcohol SBI in New Zealand primary health care settings. The research also prompts consideration of the extent to which, and how, alcohol SBI could be applied to other health, social and justice settings.

The Whanganui research, in particular, highlighted the importance of ensuring that alcohol SBI is simple and purpose-built within the settings in which it is used and that practitioners receive appropriate skills-based training. Attempts to introduce alcohol SBI in such a way that practitioners feel they are expected to work outside their roles and skills is likely to fail. Innovative and simplified models of practice that fit specific settings and are tailored to recipients should be trialled and evaluated for effectiveness, particularly in non-health settings where there is a dearth of efficacy research.

While it is important to increase knowledge of the effectiveness of alcohol SBI in non-health settings, it is even more critical that efforts and resources are first put into strengthening New Zealand's approach to alcohol SBI in primary health care settings, where the evidence clearly shows good uptake and effectiveness. On this front, progress is already being made, supported by a growing interest among primary health care practitioners in responding more effectively to the need for alcohol management within their areas.

Notes

1. For the purposes of this article, the consumption of large amounts of alcohol, or binge/hazardous drinking, is defined as drinking more than six (for men) or four (for women) standard drinks on a single drinking occasion.

2. One DALY can be thought of as one ‘lost’ year of healthy life (see http://www.who.int/healthinfo/global_burden_disease/metrics_daly/en/). DALYs are lost through premature death or living with less than perfect health (World Health Organization 2010).

3. Acute harms occur in a single event, such as injuries, deaths, road trauma, crime and disorder, drowning, burns, alcohol poisoning and unwanted pregnancies.

4. ‘Potentially hazardous drinking’ is defined as an established pattern of alcohol consumption that carries a high risk of future damage to physical or mental health, but may not yet have resulted in significant adverse effects. This is in line with the international definition of hazardous drinking as an Alcohol Use Disorders Identification Test (AUDIT) score of 8 or more (Ministry of Health 2008).

5. However, there is some uncertainty about the actual effect size and how it can be interpreted in terms of individuals’ changed risk behaviour.

6. Developed by the World Health Organization, the AUDIT consists of 10 questions focused on hazardous drinking and alcohol use disorders. It has been well validated across different cultural groups in a variety of countries and takes about five minutes to complete. There is also a shorter version (AUDIT-C), which takes about one to two minutes to administer and comprises three questions focusing mainly on drinking quantity, frequency and patterns. There is evidence to suggest that AUDIT-C has the same specificity as the full AUDIT. Two other screening tools, which take about one minute to administer and are specifically developed for use in emergency department settings, are Paddington Alcohol Test (PAT) and Fast Alcohol Screening Test (FAST).

7. It should be noted that the screening rates in Whanganui have continued to increase on a steady basis post-trial (C. Newton, Project coordinator for the WRPHO trial, 2012, pers. comm.).