The measurement and recording of alcohol-related violence and disorder
A report of research commissioned by the Portman Group
Executive summary
This report concerns research commissioned by the Portman Group and conducted by the Social Issues Research Centre between March and October 2001.
The research examined in detail the current procedures used by police, hospital accident and emergency (A&E) departments, town centre managers, representatives of crime reduction partnerships (CRPs) and others to record and collate information on alcohol-related violence and disorder. Attention was paid to the varying definitions of ‘alcohol-related’ that were employed.
Qualitative methods, in the form of interviews, focus groups and discussions with the major stakeholders in this area, including representatives of drinks trade bodies, were used throughout England, Scotland and Wales to provide balanced and representative perspectives. Three questionnaire studies and a telephone survey were also conducted to provide quantitative data. A small comparative study was conducted in The Netherlands.
Principal Findings
Police data
- There is very widespread variation in both the definition and the recording of alcohol-related violence and disorder among police forces and command units. Methods vary from direct computer database input to entries on printed forms and officers’ notes.
- The data that are available contain substantial sources of potential error and inconsistency because of the methods employed and are collated for policy or strategic purposes in only a minority of cases.
- Only 10% of police forces are able to compare their data directly with those collated by other forces.
- Interpretation of police data is often very difficult because both offenders and the victims of alcohol-related assaults are included in the figures and the sources of the information vary from reports of crime to arrest records.
- Data relating to offences in and around licensed premises, while often collected in a more systematic fashion, are prone to a number of errors and include offences which may not be alcohol-related in any meaningful way.
- Due to these and other factors, claims that 70% or 80% of all late-night violence in town centres is attributable to alcohol consumption cannot be empirically substantiated by police data. This was acknowledged by many officers themselves.
Accident and emergency data
- A&E data concerned with injuries sustained in alcohol-related incidents are recorded and collated in ways that are even more variable than those employed by police officers. They vary from routine assessments using clinical or blood alcohol concentration (BAC) tests to occasional diagnosis when perceived as clinically relevant. The manner in which the data are recorded ranges from computer form entry to clinical notes.
- Comparability of data across A&E departments is very restricted by variations in recording practices and data retrieval systems.
- Integration of A&E data with that obtained by police forces is hampered by incompatible methods and systems and by perceived data protection concerns.
- A&E data relate almost entirely to the victims of violence rather than to alcohol- related offending.
- The large majority of A&E consultants felt that the procedures for assessing alcohol consumption and data recording were in need of substantial improvement.
- While A&E consultants typically thought that between 70% and 80% of violent injuries might be related to alcohol consumption, empirical studies conducted within hospitals suggest that the figure is substantially lower.
Crime reduction partnerships
- For the large majority of CRPs the amount of work directed at alcohol-related issues amounted to less than 30% of their total activity.
- Where the reduction of alcohol-related violence and disorder was seen as a priority, the lack of empirical evaluation methods very often made it difficult to assess the true effectiveness of initiatives in this area.
- Less than one quarter of partnerships had formal documents outlining their aims, objectives and measures of achievement.
- The lack of proper evaluation of partnership activities makes it almost impossible to identify examples of best practices which might be recommended for inclusion in future schemes.
Town centre management
- Town and city managers have a particular role to play in the reduction of alcohol-related violence through their function in planning and development issues, their involvement in CRPs and their management of closed circuit television systems.
- In the majority of cases around 10% of town centre managers’ work was devoted to alcohol-related issues.
- Fewer than 5% of town centre managers compiled reports of alcohol-related incidents in order to allow objective evaluation of measures aimed at combatting such problems.
Drinks trade bodies.
- In contrast with the views expressed by police, A&E consultants and others, a number of representatives of trade bodies felt that alcohol-related disorder was probably decreasing.
- Some felt that there was insufficient recognition of the role that their organisations had played in initiatives aimed at reducing such problems.
- Most organisations were of the strong view that current sources of data concerned with the scale and pattern of alcohol-related violence and disorder were inaccurate, unreliable and open to varying interpretations.
- They felt that the concept of ‘alcohol-related’, which often implied direct cause and effect, was misleading.
- Clarification of definitions and improvements to existing recording and data collation methods were seen as being essential to the development of more appropriately targeted initiatives aimed at the reduction of alcohol-related problems.
Conclusions and recommendations
- It is clear from our research that current data recording practices are quite inadequate to enable the nature and scale of alcohol-related violence and disorder to be assessed with any degree of accuracy. The varying definitions of ‘alcohol- related’ add further confusions.
- This lack of reliable and valid data similarly makes objective evaluation of initiatives aimed at the reduction of alcohol-related problems very difficult and, in some cases, quite impossible.
- The collating and publishing of existing data is unsystematic and is sometimes motivated by quite separate concerns - e.g. as a means of opposing liquor licence applications or to make a case for increased resources.
- The extent to which those who consume alcohol are over-represented among the category of violent offenders is unknown. In most town centres on Friday and Saturday nights it is clear the large majority of people have consumed alcohol, but only a small minority commit offences.
- Given that the causal connection between alcohol and violence cannot be defined on the basis of current evidence, the claim that between 70% and 80% of all violence and disorder in town and city centres is ‘related’ to alcohol consumption is quite meaningless.
Recommendations
- In the light of the points noted above, the report makes four specific recommendations.
1. Model recording procedures, which demonstrate objective assessment, systematic recording and ease of data retrieval and analysis, should be developed for both police and A&E departments . Such models should be tested in pilot schemes in various parts of the UK with assistance from external specialists and appropriate resources. Revised models derived from these experimental trials should, after appropriate consultation with relevant stakeholders, be considered by appropriate government departments for implementation nationwide.
2. In parallel with recommendation 1 specific evaluation models, based on existing, well-developed professional guidelines, should be developed for use within CRPs. After experimental trials which demonstrate the utility of such models, their deployment should be seen as a condition for public funding of CRP initiatives.
3. Research that directly measures the degree to which alcohol is a risk factor in violent crime and injury should be conducted. This would compare the alcohol consumption levels of offending and non-offending groups in town centres on Friday and Saturday nights. It would enable the issue of cause and effect to be examined more clearly and would identify the specific types of individual who may present a risk following drinking. This, in turn, would enable initiatives aimed at tackling the problems to be more clearly and appropriately targeted.
4. In order to avoid the ‘selective’ use of data on alcohol-related violence and disorder, whether derived from existing or improved procedures, publication of such data should be undertaken at fixed intervals in consistent formats. These reports should be available to all relevant stakeholders as a matter of course.
Dr Peter Marsh, 29 October 2001.