Public Health Wales / High Impact interventions – Brief interventions in alcohol misuse
Prevention and Promotion TechnicalTeam
High impact interventions – screening and brief interventions for alcohol misuse
Author:Fiona Kinghorn, Locum Consultant in Public Health
Date:16 September 2010 / Version:1
Publication/ Distribution:
  • Prevention and Promotion Programme Board
  • Public Health Wales (Intranet)

Review Date:
Purpose and Summary of Document:
This paper focuses on the issue of alcohol misuse, a major preventable cause of morbidity and mortality in Wales. Specifically, the paper will focus on provision of brief advice, or brief interventions, with adults for alcohol misuse by health professionals, where the evidence of positive impact on health service use and health outcomes is strongest. Provision of brief interventions for alcohol misuse in emergency care settings will be mentioned.
The paper makes the case for the need to further develop the adult brief interventions in primary care programme across Wales as an important contribution to delivering cost effective interventions within a clear timescale which will show demonstrable effect on population health outcomes as well as impacting on health service use.
Work Plan reference:

CONTENTS

1Public Health Impact......

2NHS Service Impact......

3Intervention......

4Intervention Impact......

5Cost effectiveness information......

6How would the intervention be delivered......

7Conclusions......

8Recommendations......

9References......

Appendix 1 – Variability in content......

Appendix 2 – Nice costing model applied to Wales......

ACKNOWLEDGEMENTS
Thanks to the following Public Health Wales staff for their contribution to the preparation of this paper:
  • Craig Jones and Su Mably
  • Andrea Gartner and Nathan Lester

© 2010 Public Health Wales NHS Trust.

Material contained in this document may be reproduced without prior permission provided it is done so accurately and is not used in a misleading context.

Acknowledgement to Public Health Wales NHS Trust to be stated.

1Public Health Impact

Whilst alcohol consumption forms a normal part of UK culture, alcohol misuse is a major public health issue. Excessive drinking is a significant cause of morbidity, mortality and social problems in developed and developing countries.1 It is estimated as the third highest out of twenty-six risk factors for morbidity in the European Union, after tobacco and high blood pressure and ahead of overweight/obesity.2

Alcohol is causally related to more than 60 medical conditions, including malignant cancers, diabetes mellitus, cardiovascular diseases, gastro intestinal diseases, neuropsychiatric disorders, unintentional injury (such as motor vehicle accidents and falls) and intentional injury (such as self-inflicted injuries).3 Alcohol also causes societal problems - it is estimated that there were 18,000 incidents of violent crime attributable to alcohol in Wales in 2007 – 08.4

There is an upward trend in both alcohol-related and alcohol-attributable hospital admission rates in Wales – alcohol-related rates for males are around twice as high as that for females.4The average annual number of alcohol-related hospital admissions for residents in Wales accounts for around 1.5% of all admissions.4

There are around 1,000 deaths attributable to alcohol (deaths due to conditions in part attributable to alcohol) each year in Wales.4 In men alcohol-attributable deaths accounted for 4.3% of all male deaths (average between 2002 – 2006) compared to 260 alcohol-related deaths (causes of death most directly related to alcohol). In women the figures were 344 (2% of all female deaths) alcohol–attributable deaths and 146 alcohol-related deaths for the same time period.4

In most cases there is a dose-response relationship between alcohol consumption and risk of disease.5Alcohol consumptionin the UK has almost doubled since 1960.6 In 2005-06 a higher percentage of 13-year-olds reported drinking alcohol at least once a week in Wales than in England and Scotland and out of forty European countries, Wales had the third highest percentage amongst 13-year old girls and fourth highest percentage amongst 13-year old boys.7

More men than women in Wales reported either drinking above recommended guidelines or binge drinking in 2007. For men, this behaviour was highest in 25 – 34 year olds whilst in women the highest percentage was in 16 – 24 year olds. For both men and women, such behaviour decreases with age.8

Most alcohol-related harm is caused by excessive drinkers whose consumption exceeds recommended drinking levels, not the drinkers with severe alcohol dependency problems.1

2NHS Service Impact

It is estimated that the cost of treating alcohol-related health problems in NHS Wales is between £70 – 85 million per year.9

A crude illustration of one element of cost to the NHS can be found by considering costs of alcohol-related hospital admissions. Applying an average general medical admission cost per episode (£1,467) from one Health Board in Wales (Cardiff and Vale) to 2006 data on male and female alcohol-related admissions in Wales residents (n = 15,034), the estimated cost of alcohol-related admissions in Wales in 2006 is over £22 million. However it should be noted that:

  • The general medical admission cost per episode may differ between Health Boards in Wales
  • The general medical admission cost per episode includes total Health Board costs and does not equate to releasable bed days or avoidable cost – this would require a different calculation not possible within the timescale of preparing this paper
  • Costs used here relate to the specialty of medicine. It is acknowledged however that alcohol-related admissions will span a number of specialties
  • The alcohol-related admissions data used contain both in-patient data (emergency or elective involving at least one overnight stay) and day case data (elective admissions admitted for day care, where no overnight stay is required) but the general medical admission cost used relates to an in patient stay, which will in part be an overestimate in costs given that the cost of a day case is less
  • However, alcohol-related admissions data (as opposed to the wider alcohol-attributable admissions data) will not reflect the true burden of alcohol in terms of health services usage and therefore this figure is likely to be a significant underestimate of the cost of alcohol admissions to the NHS

3Intervention

Brief advice regarding alcohol can be summarised as ‘a short, evidence-based structured conversation about alcohol consumption with a client to motivate and support the individual to think about and/or plan a change in their drinking behaviour in order to reduce their consumption’).10 Brief interventions are aimed at people who are not specifically seeking advice on alcohol consumption.

Brief advice is targeted at individuals who could benefit from such an intervention. Targeting requires ‘identification’ or ‘screening’, where a practitioner needs to ask a set of questions about an individual’s drinking in order to make an objective assessment of their level of risk of harm linked with their drinking behaviour.11 Details of what a brief intervention can include can be found in appendix 1.

Providing brief advice, also called ‘brief interventions’, for alcohol misuse in primary care has been well researched and promoted as a one way forward to reduce alcohol consumption in a community.1 Brief interventions for alcohol misuse lower alcohol consumption – a Cochrane review included studies whose participants drank on average 306 grams of alcohol (over 30 standard drinks per week). Results showed that after one year or more, people who received the brief intervention drank less alcohol than people in the control group (with an average difference of 38 grams/week, range 23 to 54 grams). Longer counselling showed little additional benefit. Further work is required for certain sub groups at risk, namely young people, older people and ethnic minorities.1

The effectiveness review underpinning recent NICE guidance on preventing the development of hazardous and harmful drinking 12 included 27 systematic reviews which provided a strong set of evidence for the effectiveness of brief interventions for alcohol misuse mainly in primary care on a range of health and health-related outcomes, including alcohol consumption, morbidity (such as hypertension), mortality, alcohol-related injuries, the social consequences of alcohol (as measured by the Drinker Inventory of Consequences) and healthcare resource use (for example new injuries requiring readmission to trauma services).

Limited evidence was found in 3 of the systematic reviews for the effectiveness of brief interventions for alcohol misuse in emergency care settings.12

Primary studies in the overarching review were mainly drawn from the USA, although some were drawn from the UK – authors of the review conclude that this evidence base therefore has some applicability to a UK-based setting. Most reviews focused on the primary care setting but limited evidence was also found for effective brief interventions in emergency care settings. Reviews spanned the adult population of mixed age from 12 – 70 years old. Evidence of effect was found in both men and women. Socio-economic status did not influence the effectiveness of brief interventions. Most studies focused on Caucasian populations; studies with ethnic minority populations were poorly reported in the included literature.

4Intervention Impact

The Health England pilot project13 outlined in the cost effectiveness section of this report, assumed that brief interventions for alcohol misuse delivered in primary care would reduce the average level of alcohol consumption by 40% after 12 months in all who received the intervention. The project highlights that a 40% decrease in the level of drinking is associated with an additional 0.0233 quality adjusted life years (QALYs) per person.

This work assumed that individuals receiving the intervention were 45 years old. Drinking was assumed to be associated with 18 diseases.

Modelling work completed by the University of Sheffield14to support development of recent NICE guidance has also been completed to assess the cost effectiveness of combined screening and brief interventions for alcohol misuse. This was completed for intervention at next GP registration (where a Practice Nurse delivers the intervention) and intervention at next GP consultation. Data are also available for interventions within Accident and Emergency departments but are not examined in detail in this Project Board paper.

Modelled scenarios for providing an intervention at next GP registration assume that screening would be applied to 39% of the population in England over a 10-year period, so that a third of England’s hazardous and harmful drinkers would be screened, detected and given a brief intervention. Results show that the estimated costs of delivering the intervention are outweighed by financial savings due to subsequent reduced burden of illness. The QALYs gained for this intervention over a 10-year screening programme, depending on the screening tool used, range from 30,800 – 32,700. It should be noted that both the QALY and cost figures in the section below are based on a 30-year time horizon which was seen as sufficient to measure the outcomes of a 10-year programme. 14

Modelled scenarios for providing an intervention at next GP consultation, by a General Practitioner, assume that 96% of the population would be screened over the next 10 years, with 70 – 79% of hazardous and harmful drinkers receiving a brief intervention within the time period. The result is between 108,000 to 117,000 QALYS gained over 10 years.

5Cost effectiveness information

The Matrix Knowledge Group developed a pilot project for Health England to ‘develop and demonstrate a prioritisation method to inform investment in preventative health interventions’. 15‘Multi-Criteria Decision Analysis’ was used prioritise interventions, which included:

  • Identifying interventions to evaluate (these needed to be of interest to decision-makers and have been included in a review of effectiveness and/or cost effectiveness)
  • Identifying criteria against which to evaluate the interventions – this included cost effectiveness (measured by cost per QALY gained, including healthcare costs avoided); the proportion of the population eligible for the intervention; the distribution of benefits (ratio of the proportion of the most disadvantaged 20% of the population eligible for the intervention to the proportion of the population as a whole eligible for the intervention); affordability (the budget required to fund the intervention if all eligible people received the intervention); certainty (confidence in the evaluation of the intervention, based on an assessment of the quality of the method and data used in the evaluation
  • Measuring the interventions against the criteria
  • Combining the scores for all criteria

Seventeen interventions were included, covering alcohol misuse, smoking, mental health, obesity and sexually transmitted infections. Use of statins was included as a benchmark. Figure 1 highlights results for several of the interventions in the priority list. It shows that brief interventions for alcohol misuse delivered in GP surgeries came 6th out of 17 in the priority ranking. Within these calculations, 15.9% of the population affected would be reached, including a greater proportion of the disadvantaged population compared to those not living in disadvantage. The net cost per QALY gained is -£750. Whilst this intervention is less affordable than for example increasing tax by 5% on cigarettes, the benefits of the intervention nevertheless outweigh the costs in terms of quality and length of life.

Figure 1

Intervention / Problem targeted / Priority ranking / Priority score
(% decision makers rank as top priority) / Reach
(%
population affected) / Inequality score
(% disadvantaged affected / % all affected) / Cost-effectiveness
(Cost per QALY gained) / Affordability
(*** <£100; **£100-
£1 bn;
*>£1 bn) / Certainty
(***high quality evidence;
**good quality evidence;
*low quality evidence)
Increase tax by 5% / Smoking / 2 / 9.8 / 22.3% / 1.55 / -£3,320 / *** / ***
Brief interventions delivered in GP surgeries / Alcohol / 6 / 8.6 / 15.9% / 1.78 / -£750 / ** / **
Statins for primary prevention / Statins / 12 / 4.2 / 6.2% / 1.08 / £9,858 / * / **

The NICE costing report underpinning NICE guidance on preventing alcohol-use disorders16 identifies screening and brief interventions with adults in primary care as having significant resource impact (in terms of a material cost or saving). However the report highlights regional variation in results so that calculations may differ for Wales in comparison to results of different English regions. The University of Sheffield economic modelling report 14 provides several scenarios in screening and brief intervention for alcohol misuse which demonstrate not only cost effectiveness, but also cost saving (in that they provide additional health benefits as well as an overall reduced health service cost).

6How would the intervention be delivered

Clinical and cost effective work on brief interventions, and details in appendix 1, highlight variation in length and content of screening and brief interventions for alcohol misuse as well as the type of health professional providing the intervention. This paper focuses on screening and brief interventions provided in the primary care setting either by a GP or Practice Nurse. There is no evidence for increased effectiveness depending on the health professional providing the intervention and the evidence is unclear regarding any additional benefit from increased exposure to a brief intervention (for example follow up sessions) or including motivational interviewing principles. 11

With regard to current delivery, the Welsh Assembly Government has commissioned Public Health Wales, at a cost of £118,000, to provide a brief interventions in primary care work stream from April 2010 – March 2011 consisting of the following actions:

  • Training for 60 primary care professionals in Wales on Identification and Brief Advice for alcohol misuse, and development of resources/training pack. The training course looks at alcohol becoming a regular topic of conversation for health professionals with patients (an identification tool, called ‘AUDIT’(alcohol-use identification test), will be used) and if necessary the provision of 1 opportunistic 5 – 10 minute intervention, with further shorter interventions by the same staff member as required during ongoing general consultations
  • Development of a Welsh Medical Resource Centre (WeMeReC) package of support materials, in partnership with WeMeReC colleagues
  • Development of guidance on brief interventions for Health Board planners
  • Establishment of a professional network for alcohol and information sharing

The Royal College of General Practitioners (RCGP) also delivers brief interventions for alcohol misuse to primary care professionals in Wales. To date in the 2010-11 financial year the RCGP has delivered this training to 90 primary care professionals.

The WAG also funds the University of WalesInstitute, Cardiff (UWIC) to provide training on brief interventions for alcohol misuse to Accident and Emergency staff in Health Boards in Wales. This programme is based on the research carried out by Professor Jonathan Shepherd and colleagues in CardiffUniversity’s Violence and Society Research Group (VRG) which demonstrated that it is possible to detect alcohol misuse and treat it using brief interventions when patients return to trauma and maxillofacial clinics for standard injury care. Eight NHS emergency departments are currently familiar and engaged in ‘primary prevention’ of violence through ‘data sharing’ – a feature of the VRG’s Cardiff Model; the UWIC training enables the ‘secondary prevention’ component of the Cardiff Model to be fulfilled. The UWIC contract aims to train approximately 150 band 5-7 hospital nurses, from trauma and maxillofacial departments in eight NHS hospitals on how to identify alcohol misuse and deliver brief interventions to reduce risky drinking.

Considering this intervention in the primary care setting, there are currently 1,940 WTE GPs in Wales. If we estimate that a total of 240 primary care professionals, mainly GPs will have been trained by March 2011, this amounts to a reach of 12% of the GP profession. Cost effectiveness information for England included in this report is based on 2 different overarching scenarios and presumably assumes that all GPs and Practice Nurses are trained in screening and brief intervention to enable the indicated population reach if applied systematically. The 2 scenarios consist of:

  • Intervention at next GP registration, which would screen 39% of the English population over the 10-year period and where one-third of England’s hazardous and harmful drinkers would have been screened, detected and given a brief intervention
  • Intervention at next GP consultation, where 96% of the population would be screened and between 70 – 79% of hazardous and harmful drinkers would receive a brief intervention within the 10 years

Application of NICE costings for each of these scenarios, applied to Wales, can be found in appendix 2. It should be noted however that definitions of drinking behaviour and the population percentages considered, differ between England and Wales so that calculations are not wholly comparable between England and Wales.