Counting the Cost - download the full report in pdf format Click on the image above to download and read the full document using Adobe's Acrobat Reader.

Quick links


Motherhood in Western Europe

Insights from Western European Mothers

The changing face of motherhood — Western Europe

The accompanying reports combine a review of existing literature with an analysis of original quantitative data derived from a poll of 9,582 mothers from 12 countries in Western Europe, making it one of the largest studies of this kind ever conducted

Child Obesity and Health

An analysis of the latest available data from the Health Survey for England (HSE)

Child Obesity and Health — download the full report in pdf format

In this ‘National Childhood Obesity Week’, the SIRC report, Children, obesity and heath: Recent trends, holds up a true mirror, accurately reflecting the trend towards slimmer, healthier children. more

The Future of Freemasonry

An examination of the role of Freemasonry in the 21st century


This report is, as far as we know, an account of the first ever study that has been commissioned by Freemasons from a non-Masonic body. None of the SIRC members involved in the project are Freemasons, a fact that evoked surprise and welcome in equal measure from the Lodge members we met. more

The Changing Face of Motherhood

Insights from three generations of mothers


The report seeks to answer some specific questions about the changing face of motherhood and determine the extent to which modern ‘solutions’ to motherhood are more or less beneficial than the solutions of the past. more

The measurement and recording of alcohol-related violence and disorder
A report of research commissioned by the Portman Group

Qualitative research

Interviews and discussions of both a formal and informal nature were conducted with a wide range of respondents representing the key stakeholders. The principal aims of these were to identify current practices regarding the definition, recording and collating of information on levels and patterns of alcohol-related problems. In addition, the interviews focused on perceived problems inherent in such practices with regard to validity, reliability, accuracy, etc. The final parts of the interviews and discussions were concerned with identifying potential improvements to such systems and their potential for integration with measures of alcohol-related violence deriving from other sources.

We have to say from the outset that, although we expected to find a considerable degree of inconsistency in recording practices, both between the various agencies and in different parts of the country, we were unprepared for the pattern which emerged in the very early stages of the research. Put simply, we have been unable to discover any extant procedures that can provide anything more than rough indications of the level and pattern of alcohol-related violence and disorder in even the most localised contexts. All existing procedures, in our view, have such serious conceptual and methodological weaknesses that they are unable to provide truly objective and reliable data in this context.

We noted earlier that given the imprecise conceptual status of the term ‘alcohol-related’ it is unlikely that any measurement and recoding procedures will yield data which are unequivocal. Nonetheless, it is clearly recognised by most of the stakeholders themselves that improvements to existing systems are urgently required if the debate on the patterns of alcohol-related problems is to become meaningful and if appropriate measures for ameliorating these problems are to be developed and properly targeted.

Police data

Lack of data

A number of our interviews and discussions with police officers in different parts of the country highlighted the fact that a substantial number of forces and command units keep no records at all of the extent to which certain types of crime are deemed to be ‘alcohol-related’. This was confirmed by the questionnaire survey data showing an absence of record keeping in this context in nearly 30% of cases (see Section 3.1.1). The lack of data was most evident in Scotland. At a meeting of senior officers from all of the Scottish forces none was able to say that they could provide accurate statistics on alcohol-related crime in their areas.

In the UK generally there was a tendency among a number of police officers to refer to data which related to ‘street crime’ — certain types of assault and public disorder offences. Such crimes were, in their opinion, largely alcohol-related (up to 80%) and changing levels of alcohol-related crime could, therefore, be gauged from such figures. This argument that ‘street crime’ statistics can somehow stand in place of those measuring alcohol- related offences is, however, clearly spurious since the role played by alcohol in these crimes remains unclear. A number of officers, when asked to justify assertions that up to 80% of these offences were attributable in part to the consumption of alcohol, either admitted that it was just a guess or referred vaguely to ‘experience’.

This widespread lack of data arose from two main factors. In some cases the consumption of alcohol was not recorded on either report notes or charge sheets. Some custody officers went so far as to claim that it was impossible for them to obtain such information from those in their charge since it would constitute an ‘interview’ — something which they are not permitted to do. In other cases a note of whether an offender had been drinking was made on charge sheets or custody records. The form of the notes, however, was inconsistent and did not involve formal coding, making collective data impossible to retrieve.

This pattern is very similar to that observed in the original Drinking and Public Disorder project conducted over 10 years ago. In the majority of cases we found it impossible to obtain meaningful indications of how many offenders had consumed significant amounts of alcohol prior to their offending without hand sorting paper records and notes. Even then the variations in notes provided by different officers made objective evaluation difficult.

A number of senior officers and licensing officers accepted that the current situation was unsatisfactory for a number of reasons. The lack of data, for example, made policy decisions regarding proactive policing of certain areas very difficult. In other cases they referred to the difficulty of evaluating their own initiatives aimed at curbing alcohol-related crimes. In the Isle of Man, for example, the inspector who had worked to support the recent relaxation of licensing hours on the island was concerned that it would be very difficult to demonstrate the true benefits of such ‘liberal’ approaches in the absence of objective data.

Inconsistent data

While it is true that a majority of forces and command units did keep records of one sort or another concerning the involvement of alcohol in offending, it was difficult to detect any consistent pattern in such recording. This, again, was acknowledged by police officers themselves:

“There is a definite lack of standardisation. There is no common reporting system for all police forces, so data isn’t collated in a reliable way.”

This sentiment has also clearly been expressed by HM Inspector of Constabularies in the On the Record report. This notes that, in an earlier report, Review of Crime Recording Procedures (1996), many forces had no identifiable individual with responsibility for crime recording. Four years later, little improvement in this area was observed. These serious inconsistencies in crime recording generally are greatly magnified in the area of alcohol-related crimes where there is not even a commonly agreed means of defining and subsequently noting such types of crime.

The only exceptions to these inconsistencies arose when alcohol consumption was part of the formal definition of the crime itself — e.g. drink-driving, drunk and disorderly, etc. In these cases there is a formal code to identify the crime that can be entered quite simply on existing reporting forms and computer database input screens. Alcohol-related violence and disorder, however, has no such formal code and many reporting systems are not designed to enable recording of this information

Measuring and recording tools

Many of the problems noted above derive not only from the lack of clear responsibility for maintaining accurate records of alcohol-related crime, but from the lack of usable recording systems. Many command units employ distinctly dated computer systems such as PAYFEC and Blue 8, described succinctly by a number of people we spoke to as ‘crap’. These often include GIS mapping routines for the identification of ‘hot spots’ of offending. They are, however, relatively poor data analysis tools because of their inflexible formats. (One police force had been able to purchase only one half of a system, the part which produced GIS maps, but not the one that enabled any other data analysis at all.) This means that, even if there is systematic recording of alcohol-related crime data at a local level, retrieving such information for subsequent analysis and reporting is far from straightforward.

The need for more ‘user-friendly’ computer systems, shared by all police forces, which would enable more systematic recording of relevant data, was identified by the large majority of officers with whom we spoke. We return to this subject in Section

Lack of resources

Police officers frequently referred to the lack of resources as the main reason why reliable data on alcohol-related offences were often unavailable.

“The ability to record data will depend on resources, and it’s unlikely that these will increase.”

Lack of manpower and high staff turnover were seen as directly responsible not only for the lack of maintenance of adequate reporting systems but also for the failure to obtain relevant information in the first place. They referred to ‘time lags’ in the transfer of data within the organisation and inefficient inter-departmental communication systems. Many patrol officers, for example, with first-hand information about the contexts and nature of alcohol-related incidents, were often prevented by time constraints in passing this on to those responsible for more formal crime recording. Shift patterns within some command units also made ‘hand-overs’ of relevant information problematic.

Problems of definition

It was clear from the research that police officers differed widely in what they felt was meant by the term ‘alcohol-related’. Some thought that the term should only be applied if alcohol consumption was ‘relevant’ to the offence. This, however, begged the question of what was meant by ‘relevant’. Others saw the term as applying to all crimes where the perpetrator or the victim or both had consumed alcohol. Yet another group thought that ‘alcohol-related’ applied when an offence was committed by someone who was drunk.

This lack of consistency regarding definition leads to further inconsistencies in recording of crime. In some cases arrest records include a ‘Sober’ / ‘Had Drink’ / ‘Drunk’ category box. Others simply had a tick-box marked ‘Alcohol’ while some required a free text entry if prior drinking was to be indicated. Elsewhere, other sources of ambiguity were evident:

“The current form which is submitted upon the arrest of an individual has a tick box which states alcohol, yes or no. This is ambiguous as officers tend to tick yes if it’s theft of a bottle of beer and to damage of an off-licence window for example. New forms expected 2001.”

None of the definition and recording procedures that we encountered made any reference to the causal role of alcohol. As one officer put it:

“Unless the offence is specific to drinking — i.e. drink-driving — it is very difficult to quantify the effect of alcohol as a causation factor.”

Most officers felt that simply recording whether an offender was inebriated was about the best that could be done. Anything else would require unacceptable time and effort and rely ultimately on the ‘judgement calls’ of individual arresting officers. None thought that it would be possible to establish any degree of consistency in such judgements. As one senior officer put it:

“A large amount of data rely on judgment calls of individuals and so are not always accurate or standardised. The best system in the world would be defeated by this. Bobbies are not good at judgment.”

The most senior officers we interviewed tended towards the most pessimistic prognoses for developing more meaningful definitions and subsequent recording of alcohol-related crime. There was a consensus that simply obtaining a reasonably objective assessment of whether or not an offender was ‘under the influence’ was probably the limit of any procedure.

Data loss and distortion

Many police officers emphasised that much data relating to alcohol-related offences was ‘lost’ at one point or another. Firstly, many alcohol-related incidents are never reported to the police in the first place and so no possibility of recording exists. This ‘hidden’ figure is impossible to quantify, although a number of officers thought that the recorded data might represent only 10% of all alcohol-related violence and disorder. Drunken violence within the home was thought to be particularly prone to under-reporting.

There was also a number of ways in which data on alcohol-related offences can be lost even when such offences have been reported to the police. In the past there has been a tendency within some police forces to ‘overlook’ a number of reported crimes when, for example, the offender is not apprehended, or when a complaint is not pursued. This practice was criticised by the HM Inspector of Constabulary, Keith Povey, in his July 2000 report On the Record. He was quoted4 as suggesting that many police forces massaged their crime figures and detection rates to ‘put them in the best possible light’ and called for an end to the practice.

While this source of data loss has been largely remedied, some rather spurious ‘increases’ in alcohol-related violence and disorder have appeared in reports generated from police statistics. A senior Lancashire Constabulary officer, for example, was of the view that, prior to the Povey report, up to 20% of all reported alcohol-related incidents had gone unrecorded because they never reached the stage of being defined as crimes. Unsurprisingly, therefore, figures generated using the new system, which includes such reports, appear to indicate an apparent rise of around 20% in alcohol-related offences in many parts of the country. The fact that these ‘illusory’ rises have been used by some police forces to justify policy and strategy changes is clearly regrettable.

Other sources of data distortion arose from changes to operational procedures in some police forces. In a few cases, for example, proactive policing policies had been introduced to intervene in drunken and rowdy behaviour in town and city centres at an earlier stage. This often involves making arrests for relatively minor offences (which previously might have been overlooked) with a view to preventing more serious incidents occurring later in the night. The net result of such procedures, however, has been an apparent increase in alcohol-related minor assaults and public order offences.

As we noted earlier, crime statistics obtained from police forces always reflect policing policy as well as patterns of criminal behaviour. It is difficult to see how this factor could ever be removed from data sources or even controlled for in any systematic way. Local policing strategies must always be guided to some extent by local conditions and issues. Indeed, one of the most useful functions of data on alcohol-related offending is to guide such policies and ensure targeted police responses. We return to this issue in Section 4.

Agenda-driven data

There was an implicit recognition by a number of police officers that the extent to which alcohol-related crime data were obtained and recorded depended on a number of ‘policy’ factors. The point noted above regarding the interpretation of apparent increases in alcohol-related violence is one clear example of such agenda-driven use of data.

In other cases it appeared that in a few police forces and command units the encouragement to record and collate data in the first place derived from, for example, a determination to ‘crack down’ on certain licensed premises or to demonstrate the need for increased resources. It also appears that in at least three cases the figures relating to alcohol-related offences have only been collated and published in order to support oppositions to new liquor licence applications.

While the development and use of police statistics for such purposes may be justified on a number of grounds, the practice strongly mitigates against the potential for standardisation of recording and reporting practices. The integration of data at regional and national levels can only be achieved if they are obtained in a routine and systematic manner and are demonstrably independent of the uses to which they might be put. We discuss this issue further in Section 4.6.

Licensed premises data

Recording and collation of data relating to offences committed in or near to licensed premises was conducted by the majority of police forces and command units that we visited. This is also borne out by the questionnaire data (see Section 3.1.7). A number of the recording systems also used GIS mapping programmes, allowing easy identification of ‘hot spots’ and particularly troublesome venues. In most cases the licensing officer was responsible for obtaining and collating this type of information, ensuring a degree of consistency over time and standard practices within a command unit.

These data, however, while considerably more detailed and systematic than those relating to alcohol-related offences in general, are still not without problems. Firstly, there is often no direct way of telling whether the offences to which the data relate did, in fact, involve alcohol consumption. The fact that they occurred in or around licensed premises may suggest that they were more likely to be alcohol-related to some degree, but some offences clearly might not be so related.

A second problem, noted by a number of senior police officers, arises from the source of the data. This is often in the form a crime report which has the location of the incident noted. If a licensed premises is mentioned in the report, it is collated by the licensing officer. In a number of cases, however, it is unclear whether the offence was in any way related to the nearby premises or, indeed, was alcohol-related in any way. In town centres on Friday and Saturday nights the easiest way to describe to the police where a disturbance is occurring is to refer to easily available landmarks — e.g. the pubs and clubs which occupy much of our urban centres. (We note in passing that in giving directions to the SIRC office we describe it as being next door to the Angel and Greyhound pub.)

A further problem was identified by a number of police officers which arose even when a reported offence occurred within licensed premises or by people entering or leaving the premises. This was to do with the readiness of management to report such offences in the first place. It was generally agreed that the best-run premises were not necessarily the ones which appeared to have fewest incidents of violence and disorder, but those where the management was prepared to liaise effectively with the police to deal responsibly with such incidents. Less responsible operators might simply ‘deal’ with the offenders themselves rather than reporting them to the police. The distortions created by these factors may be quite substantial. Where data collated by licensing officers are used to create a ‘points system’ for pubs and clubs, providing material which the police might use at licensing sessions, there are clear disincentives for some operators to report incidents of violent or disorderly behaviour to the police.

Other sources of data relating to incidents in and around licensed premises are sometimes collated from arrest reports when those charged with alcohol- related ‘street’ offences were asked where they had been drinking. The reliability of these, however, was also questioned by a number of police officers and even some licensing officers themselves. The temptation for offenders to ‘mislead’ the police with regard to their local drinking places is clearly present.


Many police officers, particularly when speaking ‘off the record’, were highly critical of the procedures employed in gathering data on alcohol-related offences. Several acknowledged that their own figures could not accurately reflect the ‘real’ problems that they faced in routine policing activities and saw an urgent need for more reliable and integrated recording systems. A number of senior officers, however, felt that this would not be achieved without firm directives from ‘on high’:

“A Home Office initiative requiring all police forces to supply information in a standard form is required if we are to get meaningful figures.”

Others emphasised the desirability of consistency between police forces:

“Any system for recording and collating [alcohol-related crimes] should be standardised nationally to enable direct comparisons to be made. It can then act as a barometer of change to gauge the effect of initiatives.”

“We need more standardised computer records with a specific, mandatory field for entry of details about drinking.”

In addition to such standardisation, however, some officers pointed out the need for better assessments of the role of alcohol and clearer definition. One, for example, suggested that the concept of ‘alcohol-aggravated’ offences might be more meaningful than the current notion of ‘related’. Others felt that more qualitative information regarding either a victim’s or an offender’s drinking prior to incidents would be more helpful than a simple tick box, providing courts and other agencies with better insights into the role that alcohol might have played.

In addition to specific ideas for improving the value of alcohol-related offence data a number of officers pointed to the need for integrating more fully better recording methods into operational procedures. Not only should there be more reliable figures, but they should be used in more intelligent ways. Proper analysis of the data on, say, a weekly basis, should be used to guide the deployment of resources, activities of licensing officers and proactive approaches to identified trouble spots.

Empirical studies

There are few empirical studies that have examined levels of alcohol consumption among persons detained by the police. Perhaps the most significant and relevant is the Home Office-funded study by Graham Robertson and his colleagues at the Institute of Psychiatry in London5. This focused on offences of drunkenness, but also assessed the states of intoxication of those arrested for a number of other offences in a Metropolitan Police district. The assessments were based on visual examination and, therefore, subject to some degree of error. The results, however, are most interesting. Of a total of 186 offenders charged with assault, only 15% were judged to be suffering from any alcohol impairment and only in 5% of cases was there any ‘definite’ or ‘major’ impairment. The proportions were higher for grevious bodily harm, with 31% impaired to some degree by alcohol consumption and 14% suffered from ‘definite’ or ‘major’ impairment.

These findings raise a number of important issues. While the methods used in the study may be seen as rather ‘unscientific’, they are little different from those employed by the majority of police officers. It is also important to note that Robertson and his colleagues were not simply measuring whether offenders had consumed any alcohol prior to their arrest, but rather whether such consumption had resulted in any degree of obvious impairment of cognitive or motor functioning. In Section 4.5 we discuss more fully the implications of the two types of measure in establishing the role of alcohol in patterns of offending.

Accident and emergency data

While there is currently no formal directive that police forces record and collate data on alcohol-related issues, there is even less requirement for accident and emergency (A&E) departments to do so. Nonetheless, the telephone survey of A&E consultants showed that the large majority do, in fact, record information about patients’ alcohol consumption either on a routine or case-by-case basis (See Section 3.2.1).

Perceptions of alcohol-related problems

A&E consultants were deeply divided in their views concerning the extent to which alcohol consumption was a factor in the aetiology of injuries. Some were of the view that the majority of injuries treated on Friday and Saturday nights resulted, either directly or indirectly, from states of intoxication. Some put the proportion as high as 70% or more:

“We have recently calculated that over 75% of trolley cases are alcohol-related.”

Others, however, took a very different view:

“Everyone has the impression that A&E is swimming with drunks on a Saturday night, it’s not really that bad.”

“ A figure of 70% is rubbish. Maybe 60% have drunk alcohol but they have not been involved in violent incidents — maybe it’s a pedestrian knocked over — all are lumped together as alcohol-related, but this is not a meaningful figure.”

These variations in perception could not be explained in terms of the regional settings of the hospital departments or the local populations that they served. Rather, the views seemed to reflect more the types of recording practices employed and the forms of data presentation. It was our clear impression that A&E consultants who made the highest assessments of the number of cases that were alcohol-related tended to have what were, in our view, the weakest recording procedures.

Variations in data collection

There was widespread variation in terms of the measurement of alcohol consumption, the point at which assessments were made (and by whom) and in the manner in which data were recorded. In some hospitals assessments were made by nurses at the triage stage. Elsewhere assessment of prior drinking was made by doctors or consultants, often some time after admission.

At a recent meeting of A&E professionals at senior executive level in Lewisham it was identified that there are currently 18 different opportunities to record data relating to the prior alcohol consumption of patients and 26 different ways to record it.

Variation in definitions

The definition of an alcohol-related injury was seen as problematic by a number of clinicians. The simple fact that a patient might have been drinking alcohol prior to their attendance at an A&E department was seen by many as being insufficient to establish the notion of ‘related’:

“Do you put in ‘alcohol-related’ if someone has had a glass of wine with a meal or when behaviour has been altered by alcohol. Doctors need to be more accurate when defining an injury as being ‘alcohol-related’.

Other consultants, however, insisted that all one could do was use an objective measure of whether alcohol had been consumed — anything else would involve subjective judgements. In some cases the location of the incident which led to the injury was seen as the defining factor. If the injury had been sustained in or near licensed premises then it was reasonable to class it as alcohol-related. Other A&E consultants disagreed strongly with this approach, with one describing it as ‘quite mad’.

There was also considerable disagreement concerning whether the injury to a sober A&E attendee inflicted by a drunk assailant should be classified as alcohol-related. Many consultants viewed their obligations as being solely directed toward the patient and his or her well-being. They were not police officers and, thus, such antecedents of the injury or the state of the assailant were not their concern. Others, however, felt that in order to compile meaningful records details such as these should be ascertained.

Variations in measurement

The forms of assessment varied from simple inspection — e.g. ‘smells of alcohol’ or ‘slurred speech’ — to the use of diagnostic tests (CAGE, PAT6, etc.) and alcometers. Recording procedures similarly varied from written notes to designated field entries on computer records.

The use of BAC estimates such as breath tests was seen as very controversial by the large majority of A&E consultants. In two departments such testing was conducted routinely as part of normal admissions and diagnostic processes. One consultant said that taking a breath test was just like measuring blood pressure. Elsewhere, however, consultants thought that patients would resent BAC testing and some even thought it might raise legal and human rights issues. From research evidence7, however, it is clear that few patients actually refuse alcometer tests and such tests, some consultants argued, are essential for distinguishing between intoxication and head trauma.

Variations in recording procedures

The use of tick boxes to record the presence of alcohol, either on written forms or, more frequently, computer input procedures, was the more common approach in A&E departments. Here, however, there were further sources of inconsistency. In some cases the box could be ignored if was felt to be not relevant, while in others a positive or negative response was required in all cases. Some consultants felt that systems in which the alcohol assessment was not required by a compulsory entry often led to underestimations of the number of patients who had been drinking — either because the nurse or doctor simply forgot to enter a code or, pressed for time, ignored it.

In a number of hospitals no formal recording systems were used at all. Instead, the prior consumption of alcohol was recorded only if it was seen as relevant and entered on a patient’s notes.

Variations in data retrieval and analysis

There were one or two systems in operation in A&E departments where it was seemingly possible to enter data relating to alcohol consumption, but not get it out again. A number of other systems had very limited analytical procedures and were primarily oriented around the retrieval of individual case notes.

“Our current computer system does not allow us to access data as a whole, only individual records. So we are not able to view alcohol- related attendances as a group.”

“At the moment our computer system contains masses of data, but no meaningful analysis. A more advanced system would allow more effective access to our data.”

With computerised note systems it was sometimes possible to retrieve data by free text searching for terms such as ‘alcohol’, ‘drinking’ etc. In most cases, however, this was difficult and rarely were analyses of this type conducted.

Resource limitations

Many A&E consultants and doctors referred to limited resources as a reason for limited data recording and/or analytical procedures.

“We do have a computerised records system but we don’t have the resources with which to accurately record and input alcohol-related attendances.”

“Resources are already stretched — if you want to collect more data you need to pay staff more — needs leadership to ensure accurate and reliable data”

“At present, it would be impossible to get the nurses to do more and so I would question the reliability of data collected by triage. A possible way of initiating data collection systems would be to contract individuals for that specific task — run these as pilot schemes. The problem arises that once staff have been upgraded then they can’t then be downgraded.”

Other consultants pointed to the relatively high levels of staff turnover in A&E departments as being an obstacle to developing more reliable recording systems:

“A major problem with data collection is staff turnover. The department has 20 to 30 full-time doctors who tend to change over every 6 months. No sooner has one brigade been trained to diligently collect data, than the team moves on and data become inconsistent.”

It was quite evident from these and other comments from consultants and from our own observation of activity in A&E departments that the development of more accurate, reliable and meaningful data on alcohol-related injuries will have some resource implications. While the task of recording data may be accommodated within existing triage and diagnostic procedures, collating and analysing the data will require not only improved computer software but also staff with time to undertake such tasks.

Violence against accident and emergency staff

There has been considerable publicity in recent years regarding violence and aggressive behaviour by intoxicated A&E attendees directed towards staff in those departments. In our interviews with consultants and doctors, however, this issue was rarely seen as being of particular significance. Some consultants mentioned that staff sometimes received verbal abuse from drunks and, on occasions, threats of violence. Physical assaults, however, were very rare. Attendees who had consumed drugs rather than alcohol were generally seen as presenting a greater risk.

Information sharing

A&E consultants again varied widely both in their motivation to share data on alcohol-related injuries with the police and concerning the ways in which this might be done. A number of consultants felt that only in cases where serious crimes appeared to have been committed should doctors release information to the police. Others, however, felt that a more cooperative role in data sharing could help to tackle some of the root causes of the injuries that they saw on a regular basis.

This sharing of data to tackle specific problems most often took the form of identifying specific pubs and clubs which were associated with a disproportionate level of injury or intoxication:

“I have personally telephoned police about a succession of underage girls who were drinking too much. All came from same pub, which has now been closed. I saw 16 in a period of a week – vomiting, bumped heads, minor injuries etc.”

In other cases data-sharing with the police was conducted on a more formal basis and there were a few hospitals in the UK where this was done routinely. In most cases the aim was to add to existing police data an indication of the level of alcohol-related violence which had been unreported and undetected in order to provide more comprehensive information. Some of these initiatives had received special funding from government and other agencies. (See also the A&E telephone interview data presented in Section 3.2)

The issues of confidentiality and data protection regulations were raised by almost all consultants and many were concerned about the implications of the Caldicott Committee report8. While some felt that these imposed serious constraints on sharing data with other agencies, others thought that simply removing patients’ names from the records could suffice.

Further discussion on inter-agency data sharing and integration is contained in Section 4.


Many consultants and doctors saw the need for improved methods of recording data on alcohol-related injuries. More efficient and flexible computer systems that could generate data reports in formats compatible with other systems were often mentioned. Virtually all, however, as noted earlier, saw such developments as being difficult to achieve within current staffing and resource levels. A number implied that there were, perhaps, more important things for A&E departments to be doing.

The idea of pilot schemes and/or external audits of alcohol-related injuries, conducted within a sample of A&E departments, was mentioned by a number of consultants — particularly those who had been involved in academic studies of alcohol-related issues. Through these, they felt, one could obtain much more systematic and meaningful data than in even the best of routine, in-house procedures where several different nurses and doctors are responsible for data recording.

We consider these recommendations more fully in Section 4.2.

Empirical studies

In contrast to the very variable and often clearly inaccurate recording of alcohol-related injury data in many A&E departments, a few empirical studies have been conducted which indicate that it is possible to establish more robust procedures.

A study by Little et al. (1980)9 involved breath alcohol analysis of patients attending the A&E department of a large teaching hospital during the evening. This revealed that 40% of patients had consumed alcohol prior to attending and 32% had BACs exceeding the limit for driving.

These figures are of some interest since they are considerably lower than the often quoted ‘70% drunk’ level. The highest levels of intoxication were found not only among the age group that most frequently is to be found in town and city centres in the evening but also among older patients in the 40—49 age category. In fact, the average BAC level varied little across the 20—59 years range.

There were, as we might expect, higher numbers of intoxicated patients on Friday and Saturday nights, but here the maximum was 44% — again contrasting with anecdotal reports of those attending at this time being ‘universally’ drunk.

Studies by Professor Yates at the University of Manchester10 provide similar figures. While 60% of all assaulted patients had consumed some alcohol, a significantly smaller proportion (35%) had BACs in excess of 80 mg/100 ml, the upper limit for legal driving.
A recent report by the Scottish Trauma Audit Group (STAG) analysed data relating to alcohol consumption by a large sample of A&E patients attending between 1994 and 1996. The assessment of alcohol consumption was more subjective than in the Little et al. study, often relying on clinical examination, and no BAC levels were reported. It was found that 26% of all injuries had been preceded by alcohol consumption. The age group most likely to have consumed alcohol prior to injury was 40—49 years and males were more highly represented that females. These proportions, however, increased significantly on weekend nights, reaching a maximum of 52%.

The STAG study, like that of Little et al, again indicates that even basic measurement of alcohol consumption levels and unsophisticated analysis of the data can generate useful information which is much more consistent than the anecdotal reports of consultants and nurses. Alcohol is clearly highlighted as a factor in the aetiology of injury, but at a lower level than many might imagine.

The sharp increase in alcohol consumption among attendees at weekend nights, shown in both of these studies, has to be considered in the context of alcohol consumption levels among the populations as a whole at these times. We should note here the finding of Jonathan Shepherd11 that ‘young male victims of assault may not be distinguishable from other young males on the basis of habitual or binge alcohol consumption.’ This is an issue which we consider in Section 4.5.

Crime reduction partnerships and projects

Crime reduction partnerships (CRPs) have been developed in most areas of the country and aim to draw together relevant local agencies to combat crime more effectively. They typically consist of forums of police officers, local authority officials and voluntary agencies. In some areas trade associations, residents associations and A&E departments are also involved.

A substantial number of CRPs, particularly those in rural areas, had relatively little on alcohol-related crime and disorder. Rather, their efforts are directed towards the particular types of crime which are more typical in the neighbourhoods, which might be anything from shoplifting to sheep rustling. Our qualitative work was directed, therefore, towards CRPs and specific projects where a concern for reducing alcohol-related problems was a dominant focus, leaving a more general survey of CRPs to the quantitative questionnaire study (see Section 3.4).

Types of crime reduction partnership strategies

From our interviews and discussions it was clear that there were two distinct types of CRP initiative that aimed to reduce alcohol-related violence and disorder. First, there were those initiatives that included broad approaches such as the coordination of Pubwatch and Doorwatch schemes, Door Supervisor registration, etc. Secondly, there were smaller-scale initiatives which relied heavily on local knowledge and increased cooperation between local agencies. Examples of these included responses to hot spots identified by police GIS mapping. In one case moving a late-night kebab van to a different location had resulted in a dramatic reduction in the number of offences committed in its immediate vicinity. In another city, having refuse collectors remove bottles from public rubbish bins during Friday and Saturday evenings had resulted in a similarly marked effect.


At a more general level, of course, the CRPs provide a valuable opportunity to exchange ideas and perspectives on alcohol-related problems among the main stakeholders and coordinate activities in sensible ways. Where A&E departments were involved in the partnerships, additional information concerning alcohol-related assaults — many of which might not come to the direct attention of the police — became available.

In very few cases, however, were we able to identify any significant sharing of quantitative data on alcohol-related problems within the partnerships. In one city plans were being made by the CRP for combining available A&E and police data in a single report as a way of monitoring the effectiveness of specific initiatives. Elsewhere there were some attempts to use police data in more meaningful ways — e.g. directing planning and environmental considerations. In most cases, however, there were either no objective data available within the CRPs or what data there were did not relate in any direct way to the activities of the partnerships.

Members of CRPs, however, argued that, while numerical data might not always be available, the sharing of local knowledge, perceptions and views by participants in the partnerships enabled more ‘intelligence-based’ initiatives to be undertaken. They also tended to view this as adding to the potential to evaluate the effectiveness of both large- and small-scale initiatives (but see Section 2.3.4 below).

Best practices

In the past the CRPs have tended to set their own agendas and pursue their own targeted initiatives. This, indeed, is still the case in many areas. There are, however, now more clearly defined guidelines issued by the Home Office. The Crime Reduction Tookit, for example, offers extensive advice to CRPs and provides an indication of best practices and a focus on violent crime is seen as being one area of ‘best value’ and is strongly encouraged. The increased uptake of such guidance should lead to greater uniformity in partnership composition and activity, while still enabling a focus on specific, localised issues.

Evaluation of crime reduction partnership activity

It was striking that very few of the members of Crime Reduction Partnerships with whom we met described evaluation procedures which involved truly objective methods. Some referred to police data concerning alcohol-related crimes, but conceded that these were rarely used in a formal evaluation context. Most relied on internal and often quite subjective assessments of the effectiveness of initiatives provided by the members of the CRPs themselves. This picture was confirmed by the questionnaire survey (see Section 3.4.5) with fewer than 20% of partnerships setting measurable performance targets.

Where objective evaluation of CRP initiatives was undertaken using, for example, police crime statistics, further problems were evident. Principally, the design of the CRP programmes and the evaluation studies were such that it was difficult to demonstrate that any changes in levels of offending were directly attributable to the CRP initiatives.

This point concerning the proper design of evaluation studies is cogently made in the excellent paper by Michael Hough and Nick Tilley at the Home Office Police Research Group. While they focus on in-house police evaluations their points apply equally to evaluation of CRP effectiveness, where the procedures are, in fact, mostly carried out by the police as members of those partnerships. They noted that:

“Even when real changes in crime occur, accurately attributing cause and effect is hard. Minimally, one needs some form of comparison groups against which to compare trends in the crime problem, and some form of circumstantial evidence about the mechanisms which brought on the effect. Where a fall in crime can be attributed to a preventative strategy, some checks are needed to see whether the crime has simply been displaced across place, time or offence type.” (Hough and Tilley, 1998:39).

The points made by Hough and Tilley are particularly relevant in the context of evaluations of initiatives aimed at reducing alcohol-related crimes. In some cases it appeared that the focus was not so much on the crimes themselves, but on the extent to which they were alcohol-related. The point here is that just tackling the alcohol component does not always lead to any significant change in the level of the crime itself, only the extent to which the crime is alcohol-related.

A good example of this effect was observed in a study by MCM Research in 1990, commissioned by the Home Office and the West Midlands police. A bye-law prohibiting drinking in public was introduced in June 1989 in Coventry — the first of its kind in the country. Police data indicated a significant decrease in alcohol-related crimes of violence and disorder following the introduction of the bye-law. The level of the offences themselves, however, irrespective of whether they were alcohol-related or not, had continued to rise at exactly the same rate that would have been predicted from previous trends. It is difficult to see, therefore, what had really been achieved by way of crime reduction.

The large majority of CRPs, of course, are not concerned with alcohol-related crime per se, but with a wide range of patterns of offending. Evaluations of their effectiveness, therefore, are unlikely to be quite so distorted as that of the Coventry bye-law. Nonetheless, it was clear from both our qualitative and quantitative research that more professional and scientific evaluation methods need to be employed in order to determine their effectiveness in tackling, among other things, alcohol-related violence and disorder, and to identify the specific ‘best practices’ in this area.

Specific projects

In addition to activity of the several hundred CRPs around the country a number of projects have been developed which focus specifically on alcohol-related violence and disorder. These, like the broader CRPs, involve partnerships between a variety of agencies. Notes on examples of projects that we visited in the course of our research are provide below.

Project Amethyst

This project aims to ‘reduce both the incidents and fear of violent crime and disorder associated with the misuse and abuse of alcohol within the county of Cornwall and the Isles of Scilly.’ It received substantial funding (£900,000) from the government’s Crime Reduction Programme.

The principal partners in this project are the police, probation departments, youth service, youth offending team, health workers, together with other bodies and agencies.

The police maintain a searchable database of alcohol-related offences and prior to the start of the project claimed that 63% of all persons arrested between the ages of 18 and 34 were under the influence of alcohol. Details concerning the measurement of such ‘influence’ were not provided.

Project Amethyst has set as its target a reduction by 30% of violent crime committed by identified repeat offenders in licensed premises. It also seeks to reduce the number of violent crimes committed in public places between the hours of 10 pm and 4 am by 10%. Further objectives include a reduction in criminal damage and the number of repeat referrals to A&E departments by chronic drinkers.

Included in the list of ‘proposed targets’ is the rather confusingly worded: ‘Develop a set of appropriate multi-agency targets to monitor the impact of the project to alcohol related issues.’ In the most recent project report, however, we find the statement:

“… we now await a decision concerning the funding of an independent research programme into the effects of Project Amethyst, as the project will no longer be centrally evaluated.”

We have to say that we are surprised that a project of this scale did not have objective evaluation procedures built in to its design from the outset. It would seem that if additional funding for such purpose is not forthcoming it will be extremely difficult to learn any lessons or identify potential best practices from what is otherwise a valuable project.


‘Targeting Alcohol-related Street Crime’ is a project focused on the city centre and bay areas of Cardiff. It is operated by a partnership of South Wales Police, Cardiff County Council, the A&E department, licensing magistrates, licensees forum and others. The project employs the services of a support nurse who conducts follow-up interviews with attendees at A&E whose injuries were thought to be alcohol-related. It also has a dedicated data analyst who is responsible for collating data and presenting reports.

The main aims of the project are those of “implementing and evaluating various measures designed to a) reduce levels of violence, b) increase the chances of violent offenders being brought to justice and c) increase access for victims to Victim Support and mental health services.”

A further element of the project is to identify the extent to which police data under-represent the true scale of alcohol-related violence, For this purpose there is close data-sharing between police and the A&E department. This appears to show, according to one source within the project, that police are aware, either through crime report or arrest data, of 86% of all attendance at A&E for alcohol-related injuries. Some other sources, however, suggested that the figure was rather lower. There is, however, some doubt cast by TASC’s data on the claim by Jonathan Shepherd (Shepherd et al., 1989) that as much as 75% of all alcohol-related violence is undetected by the police.

TASC serves as a very useful model of focused inter-agency cooperation and intelligent use of data. There is, however, a major obstacle to evaluating its effectiveness that was not identified at the outset. The number of licensed premises in Cardiff has increased quite substantially in recent years, resulting in larger numbers of people frequenting the city centre on Friday and Saturday nights. No accurate measures of the changing population sizes are available. This means that it is very difficult to interpret the statistics of alcohol-related violence and disorder collated by the data analyst. Increased numbers of people will inevitably result in increased numbers of crime since there are more people available to be either the perpetrators or victims of crime. (For this reason it always necessary to express crime figures as a proportion of population.)

In a recent TASC report this inability to undertake empirical evaluation is quite evident in the ‘General Synopsis and Discussion’ section:

“In the light of the evidence discussed it is reasonable to tentatively conclude, despite raw figures suggesting otherwise, that the interventions implemented by T.A.S.C have had a relatively positive effect on alcohol-related incidents of violence and public order.”

This is clearly, to a large degree, akin to guesswork. In the absence of comparable data from prior to the beginning of the project to the present time, no objective evaluation can be undertaken. And for data to be comparable the sizes of the population at whom the TASC initiatives are directed need to be known.

Another obstacle to evaluation which is evident in the case of TASC is presented by initiatives which seek to encourage people to report alcohol- related incidents to the police, including those attending A&E departments. If the initiative is successful then police figures based on such reports will appear to show a rise in alcohol-related crimes. It is very difficult, therefore, to evaluate the success of parallel initiatives aimed at reducing alcohol-related crimes since one cannot identify which of the reported incidents would have gone unreported prior to the implementation of the initiative.

Project Jupiter

This project, Joined Up Partnerships in the East Midlands Region, is a Government Office for the East Midlands (GO:EM) initiative to create a data exchange and crime mapping network across 40 East Midlands partnerships. It is focused primarily on ways in which data can be recorded, collated and shared between agencies involved in crime reduction initiatives. In this regard it relies heavily on the work of Stephen Radburn, Data Exchange and Crime Mapping.

The project has examined the priorities of CRPs in the East Midlands region and at a national level with a view to identifying what data sources are required in order to target initiatives most effectively. We can see from Figure 2.1, taken from Feasibility Study for Information Exchange — Final Report, that alcohol-related violence and disorder ranked as a priority for over half of CRPs. However, in Figure 2.2, derived from a survey of CRPs conducted by Project Jupiter, only 15% of partnerships receive data relating to this type of crime.

Figure 2.1 National and Regional Stategic Priorities set in 1999 (source:PRCU)

              Figure 2.1

Figure 2.2 Data Received (by Crime Type) and Percentage Geocoded

Figure 2.2

Commenting on the current state of data availability and reliability the Project Jupiter report concludes:

“The lack of data availability questions the degree to which strategic decisions reflect the real issues facing partnerships. Much of the data collected by partnerships are incompatible owing to differences in time periods, geography (coterminosity) or data format (both unit of collection e.g. beat, ward and media e.g. electronic, paper). Partnerships need to collect benchmarking data in addition to local data with significant potential for unnecessary replication of effort across a region. By empowering local ownership of the data collection and analysis process you also encourage diversification — while this might be applauded in conducting audits, it has significant implications for developing common data exchange and G.I.S.”

By way of remedy for the current fragmented approach to data collection and sharing the report proposes:

“An alternative solution would be for all the Partnerships/County councils involved to access a single server running a single copy of the required applications by modem dial-in with all users sharing a central data repository and software. This will then eliminate the need for data transfers as all updates will be applied directly to the central database and allow the whole application to scale according to future demand. The central database can be configured to allow each partnership to view/update their own data or even allow them to view each of the other partnerships data.”

Project Jupiter represents a very significant advance in thinking about the ways in which data can be ‘cleaned’, distributed and used more effectively in the evaluation of crime reduction initiatives. Ultimately, however, the value of systems designed to achieve these ends will depend very much on the quality of data which is entered in the first place. Given that the definition of alcohol-related violence and disorder is not without significant problems, inaccurate or irrational measurement of such crimes will not be remedied by even the most sophisticated of shared data servers.

Project LION

This project (London Information On the Net) is currently at the ‘proof of concept’ stage and has focused almost entirely on the Borough of Lewisham. A further five London boroughs have now been included.

The project again seeks to facilitate the exchange of crime data between the relevant agencies. The aims are:

The partners involved in the scheme are police, probation officers, healthcare workers, social services and others concerned with crime reduction.

On our visit to Project LION at New Scotland Yard the main benefits of the project seemed to be the integration of several data sets relating to crime or injuries in a particular area, aided by mapping programmes. One could examine, for example, ‘hot spots’ indicated by police data, ambulance reports and other sources of information.

While the project exists primarily as a demonstration of what might be achieved through data sharing in this way it has already shown that agencies which are sometimes reluctant to pass their data on to others can be persuaded to do so by the merits of a system that has several built-in safeguards to preserve privacy and anonymity.

The major problem encountered by Project LION is the need for extensive data cleaning and integration of information contained in very different types of database. Software is being developed for this purpose.

At the moment the only source of alcohol-related incident data available to the LION database is that provided by the ambulance service. This, unfortunately, is not the most systematic or reliable of data sets. There is also the problem noted with regard to Project Jupiter of data which might derive from inconsistent or inappropriate measures of, say, the extent to which acts of violence or injuries were alcohol-related.

Despite these problems Project LION indicates the potential to develop much more detailed information regarding alcohol-related crime and disorder than currently exists. The project is to be evaluated by academics at the London School of Economics but no details of the methods to be used have yet been released.

City and town centre managers

There has been a growing emphasis on the safety of city and town centres over the past 15 years. The development of the night-time economy has contributed to an increased fear of crime and disorderly behaviour. Many centre managers, therefore, are involved in CRPs and other initiatives aimed at reducing street crime, assaults and anti-social conduct.

Perceptions of alcohol-related problems

Managers were generally consistent in their perceptions of the level of alcohol-related problems in their towns and cities. While most recognised that there were problems on Friday and Saturday nights, their greatest concerns were with issues such as retail theft and commercial development. Some saw the rise of the night-time economy as making a positive contribution to their town and city centres, at least in commercial and revenue terms. This consensus is consistent with questionnaire data indicating that most managers viewed alcohol-related violence as being above the ‘medium’ level but rarely in the ‘serious’ category. See Section 3.3.1.

Glasgow is an example of a city centre that has become increasingly fashionable, due in a substantial degree to the expansion of the ‘night-time economy’. The style of drinking establishments and the clientele that they attract has changed markedly and the number of licensed premises has increased dramatically in the last 10 years. Such development was again seen as being beneficial, although the problems that it presented were also recognised. A number of managers elsewhere felt that failure to tackle predictable problems of late-night disorder was often due to a lack of resources and police under-manning. In some areas, it seemed, the developments had risen at a pace which had outstripped any increases in such resources.

In Brighton, however, the town centre manager had a rather less sanguine view of alcohol-related problems. He was of the view that half of all violence occurred on the streets of Brighton and felt that most of this was associated with drinking. Despite this, he was in favour of the proposals for 24 hour licensing and felt that more focused initiatives to combat alcohol misuse were the way forward.

Sources of data on alcohol-related problems

Relatively few city centre managers collected statistics on alcohol-related violence and disorder, although some had access to such data through their involvement in partnership schemes. This is consistent with the questionnaire data reported in Section 3.3.3. In a few cases the CCTV camera operators supplied information directly to managers which included reference to drunkenness and similar incidents. Rarely, however, were such data systematically recorded.

Work with partnerships

All of the managers interviewed in the qualitative research were involved in some degree of work with other agencies and bodies that included a focus on alcohol-related problems. Many were active in Pubwatch and Clubwatch schemes, anti-violence steering groups, restaurateurs’ organisations and similar bodies. Elsewhere there was involvement in local transport initiatives, including those to ensure that late-night drinkers had the means to get home. Some managers saw the ability to empty the town and city centres at night as a critical factor in reducing alcohol-related crime.

Measures to combat the problem

By far the most common strategy supported by town centre managers was the introduction of CCTV systems and their expansion to meet the demands of the night-time economy. Most such schemes were run jointly with the local police force, although quite often staffed by civilians. There was a consensus that police monitoring of the CCTV system was preferable since they tended to be more ‘crime aware’ as a result of their experiences on the beat. Lack of police resources, however, meant that it was often impossible to release officers for such duties.

While the introduction of CCTV was generally thought to have a positive effect, few town centre managers could provide data to support such a claim. In many cases no evaluation had been undertaken at all.

In one case (Oxford) evaluation had been undertaken by an independent researcher at Oxford Brookes University. He noted the commonly experienced effect of an apparent rise in street crime immediately after the installation of the cameras. This, however, is most easily attributed to the increased detection of crime that the system provides rather than any change in the level of criminal behaviour itself.

A particular focus of the evaluation study was on the extent to which CCTV systems reduce the levels of fear of crime in city centres. The logic is that people will feel safer if they see cameras that they believe might act as deterrents to crime. Examination of people’s fear of crime both 6 months before the installation of cameras and 6 months after, however, showed no significant change. Nor was it possible to conclude that the cameras had significantly increased the detection of crimes and the apprehending of offenders. The issue of geographical displacement of crime, often used as a criticism of CCTV systems, was also not included in the Oxford evaluation.

The Netherlands

Research in The Netherlands was included in the original study conducted 10 years ago to establish some comparisons with the UK in terms of levels and patterns of alcohol-related problems and the method of tackling them. It is generally thought that the Dutch and the British have similar cultural traditions and styles of drinking and this was evident from our, admittedly limited, research there.

It was partly on the basis of our examination of the liberalisation of licensing hours in The Netherlands that we recommended similar changes in the United Kingdom — recommendations which have now been acknowledged and adopted in the government’s White Paper Time for Reform. The purpose of the more recent visit was to obtain first-hand accounts of current alcohol-related problems and the methods employed to tackle them. We were also particularly interested in identifying data collection procedures and evaluation methodologies.


Interviews were conducted with senior representatives of the three main agencies in The Netherlands that are directly concerned with alcohol misuse and violence12. These were as follows:

Perceptions of alcohol-related problems

Wouter Buwalda pointed out that there has been a political shift to the right in The Netherlands in recent years. At the same time a large degree of ‘liberalism’ exists, reflected in the 24 hour opening of shops and bars. There is, he suggests, an increased trend towards viewing these liberal aspects of society as being problematic. Alcohol-related disorder, for example, is now less tolerated, even though there is no evidence to suggest that it has increased either in the last decade or as a result of the more liberal approaches to licensing. He thought that alcohol had now overtaken drugs as a perceived cause of social harm, including violence.

Wim van Dalen generally agreed with Bouwalda’s assessment but stressed that at present it is not possible to tell if there is an actual increase in alcohol- related problems or just an increase in public concern. Nonetheless, he was personally convinced that ‘most individuals who are the source of problems have been drinking a lot’.

Data (un)availability

While van Dalen was clear that little or no quantitative evidence concerning alcohol-related crimes was available to his department (NIGZ), Buwalda (of LPTGS) insisted that the work by NIGZ clearly showed that alcohol-related crime had significantly increased. He also referred to data derived from police records. There has, however, only been one piece of research conducted and that looked at the amount of police time spent on alcohol-related matters rather than changes in offending rates. We were unable to discover any research which relied on systematic recording of alcohol-related offences.

Buwalda commented that lack of measurement is a common and crucial problem in all crime research but added:

“To me it is not important if the figures are 30%, 60% or 90% — there is an issue anyway … all scientists look for best practices and figures, but in my opinion the effectiveness of measures must always be seen within a social and cultural context of a particular country or population”.

This view was very familiar to us from previous research on alcohol-related issues in The Netherlands. Substantial weight is attached to public perceptions of crime while less attention is given to generating empirical data. Essentially, the notion is that if people think there is a problem, then there is a problem and ways of tackling it must be developed.

Dr Burt Bieleman produced a report of qualitative research conducted by INTRAVAL on youth, drugs, alcohol and aggression. The research involved interviews with both customers and owners of bars, cafes, discos, etc. It showed that 15—23 year olds tend to be the highest consumers of alcohol. This is reflected in trends among this age group to have more disposable income and to visit bars more often.

The report also indicated mixed drinks were becoming increasingly popular, as was drinking in the home before going out. There was also reference to ‘fighting intentions’ when going out drinking among males in this group.

Bieleman emphasised, however, that quantitative data was largely unavailable and that which existed was problematic. Police reports, he suggested, were more representative of what the police were doing rather than what was actually happening in bars and on the streets. In addition, very few hospitals record alcohol-related details in any systematic manner.

He pointed to further difficulties in interpreting police reports, including the inability in many cases to distinguish between offenders and victims in recorded alcohol-related incidents. There was also the problem of subjectivity in recording procedures: ‘Who defines what as aggressive?’ Bieleman also referred to familiar problems in researching alcohol-related violence — it is actually quite hard to find violent incidents at night, even when specifically looking for them.

Bieleman noted inconsistencies between what the police measure and record and what they tell the newspapers. Their files, he claimed, contain very little about alcohol- and drug-related violence, yet they readily blame disorder on young people drinking. He gave an example of trouble after a recent football match, instigated by visiting fans, where local people out drinking in bars in the city were blamed.

Finally, he noted that there are very few mechanisms for data collection on a national basis in The Netherlands. The country is much more regionalised than the UK with greater powers being delegated to local government. He is currently trying to encourage these local governments to collect data in formats which will, ultimately, enable a national picture to be established. He noted in this context, however, that some regions would be reluctant to record alcohol- related offences in this way because they were more motivated to have their local policies appear effective.

Wim van Dalen thought that it was not difficult to identify sources of useful data on alcohol-related issues. The new legislation, for example, which requires much more stringent checking of age in bars means that more people are in a position to identify and quantify problems. It was the lack of a clear policy requiring such data to be collected that was the real problem. He suggested that it would be quite easy, given the political will, to require hospitals to collate the information many already have on the alcohol level of patients on admission. ‘We will be more able to confront and solve problems when we have better figures’.

Tackling the problems

As noted above the minister for health affairs has increased the enforcement of existing legal measures such as those relating to the serving of intoxicated people and under-age drinkers. The latter is a special priority given the increased level of alcohol consumption — both in terms of quantity and strength of drinks — among youth in The Netherlands.

Wouter Buwalda referred to current discussions regarding proposals to increase the rate of VAT on alcoholic drinks. A recent NIGZ conference on alcohol had concluded that the best way of reducing the negative consequences of alcohol was to limit its availability or increase its price.

He was also in favour of what many might see as a rather strange change in licensing laws. ‘In my opinion the age limit should be developed according to the aggression potential — i.e. the aggression related to a specific drink.’ He argued that it was wrong that the legal drinking age for beer (16) was lower than that for whisky (18) when aggressive males mainly drank beer rather than spirits. It was, he opined, the CO2 in beer that leads to greater intoxication and should, therefore, be restricted to older drinkers.

A somewhat less controversial approach was advocated by Dr Burt Bieleman. He felt that it was necessary to obtain more objective measures of the problem, perhaps on a small, local scale. Initiatives based on such more evidence-based assessments could serve as pilot schemes which could then be evaluated and best practices identified.

Wim van Dalen was generally of a similar view and identified specific issues such as those relating to door staff training and transport facilities. He conceded, however, that with the current density of pubs and bars in some areas it might be difficult to achieve change immediately. For that reason a focus on the ‘reform’ of offenders might be the way forward. He referred to schemes in place in some parts of the country where those arrested for alcohol-related offences are given the opportunity to seek advice or go on a course instead of being charged.

Dr Burt Bieleman recognised that it was much easier to control customer behaviour in discos and clubs than on the streets. What happened in such venues had a direct impact on activity in the streets after they left the premises. This was why current strategies had been directed towards improving the operation of the premises themselves.

Clubs in Amsterdam now have the option of closing later (7 am), but the last 2 hours are now defined as ‘cooling down’ time. In these 2 hours they are not permitted to serve alcohol and lights have to be turned up, music levels down etc. As an unforeseen result of this, however, clubs tend to try and get people out as soon after 5 am as possible, since they are not selling any drinks and many customers, in fact, want to leave anyway if there is no alcohol. This has created a new problem of people hanging around on the streets for longer, waiting for taxis, etc.

Nuisance law

In most regions of The Netherlands there exist what are rather quaintly known as ‘nuisance laws’. Such laws enable the mayor of a city to close down a pub or club or restrict its hours of operation if it can be established that it is causing a nuisance. This occasionally happens, but usually only following a serious incident such as a stabbing.

It is possible that legislation along these lines might be seen as appropriate in some parts of the UK, with local authorities exercising powers similar to those of Dutch mayors. New British legislation already allows for police officers of the rank of inspector or above to close premises for 24 hours in certain circumstances. Although this is very much a controversial issue we may be able to learn some lessons from the Dutch experience, in the way that we followed their lead in working towards 24 hour licensing.


There was little in what we encountered in The Netherlands which surprised us. While data recording and collating procedures with regard to alcohol- related problems might be deficient in the UK, they are almost non-existent in The Netherlands. In that country, however, there is greater recognition that approaches to alcohol are primarily political. They are based on people’s perceptions of what the problems are rather than on what objective measures might reveal those problems to be.

From what we could gather there has been little change in either the patterns or the levels of alcohol-related violence and disorder in The Netherlands during the past 10 years. What has changed, however, is popular tolerance of such behaviour. It is now seen as being of greater concern and that is sufficient to warrant new strategies.

It is our view that what we see quite clearly in The Netherlands is a phenomenon which is also very characteristic of the UK, but less often made explicit. Our concerns with alcohol-related violence and disorder are less driven by objective information but by our increasing, and at times irrational, fears of crime. We return to this point in the concluding Section 4.

The industry view

Towards the end of the 6 month project interviews were arranged with representatives of the major drinks industry bodies and associations13. The intention here was to obtain comment on the issues which had arisen in the course of our research and to elicit perceptions of the problems and the best ways of tackling them.

Perceptions of the problem

There was a general perception, expressed particularly by the BLRA, that alcohol-related disorder was probably decreasing. It was accepted, though, that the police would probably state the opposite. The ALMR felt that there was no real evidence one way or the other. It was emphasised, however, that was also little evidence of a direct causal connection between alcohol and violence. The ALMR also felt that many of the current problems arose from off-licences etc. as much as from pubs and bars. Comment was also made regarding some published research on alcohol-related violence which was seen as ‘unsophisticated’ but ‘headline grabbing’.

There was a general consensus that more meaningful definitions of ‘alcohol-related’ needed to be established — ones which made clearer the causal connection or lack of it between alcohol and violence. There was also concern that often the source of the alcohol that was supposed to be related to violent and disorderly behaviour was often unknown. It was felt, particularly by the BII, that it was unreasonable to blame the pub trade for the problems if, in fact, the drinks that had been consumed had come from, for example, cheap ‘smuggled’ imports.

The BII recognised the need for more systematic and consistent methods of recording alcohol-related violence and disorder which, in turn, would enable more focused approaches to tackling the problems. Reference was made to the institute’s ‘social responsibility initiative’. The BII clearly recognises that alcohol-related crime is an issue, but they argue that local authorities and the police do not have any solid, reliable figures with which to work. Instead, these groups rely on the ‘saturated market argument’, claiming that there are too many pubs and clubs, without recognising the positive role that can be played by responsible operators.

The issue of alcohol-related disorder was not seen as a core issue for members of the BHA. Fast food outlets are the sector most likely to encounter problems of this nature but the BHA is hoping that the extension of licensing hours may go some way to resolving this issue.

The SELP argued that it was difficult to assess whether disorder in and around licensed premises was due to alcohol or other factors. They stressed that the general perception of the society was that incidents of alcohol-related disorder were decreasing due to measures such as bus and taxi provision, staggered closing and door staff registration schemes.

The representative of BEDA claimed that since the launch of the White Paper Time for Reform there had been a steady flow of ‘bad news’ about alcohol-related problems, including the British Crime Survey. BEDA’s members, particularly those in cities such as Manchester and Liverpool, recognised that there were substantial problems of alcohol-related violence, but these were not helped by policing policies and resources. There was, they claimed, a lack of police manpower to deal adequately with the increased numbers of people in city centres on Friday and Saturday nights.

There was also an increasing trend of problems arising from people being refused entry to clubs because they were already drunk when they arrived. The problems, as BEDA might be predicted to say, often arose more from excessive drinking in pubs than in the nightclubs of their members.

Industry reputation

The BII stressed that they had a duty to their members to defend the reputation of the industry and to counteract the distorted image that often prevailed. They suggested that if we were to believe everything that we read about the drinks industry then we would conclude that it existed simply to sell shots of liquor for 50p to 16 year olds.

Representatives of the other bodies also felt that there was a need for a more balanced view of the industry. They accepted that there were problems from time to time and were determined to tackle them. But they felt that not enough recognition had been given to the initiatives that they had already undertaken to reduce alcohol-related violence and disorder.

On a more positive note the BII welcomed the move of licensing matters from the Home Office to the Department for Culture, Media and Sport (DCMS). This, they felt, indicates that the government now recognises that alcohol is not just something which is associated with crime but with entertainment, tourism and other forms of enjoyment.

Specific causes of problems

The issue of the discounting of drinks in pubs was noted by BEDA who were seeking to have the practice regulated by government. They claimed that most licensees are reluctant to sell alcohol at discount prices, partly because it attracts the ‘wrong’ sort of people. The market, however, is so competitive that they often feel left without choice.

The BLRA have debated the idea of fixing a minimum price in a given geographical area in order to curb irresponsible promotion practices. They thought, however, that this might lead to accusations of cartel behaviour and price fixing from the Office of Fair Trading. Instead they have published a guide: Promotions and Happy Hours: A Good Practice Guide for Pub Owners and Licensees.

There was also recognition by trade bodies that other ‘unhelpful’ trading patterns, particularly in venues catering almost exclusively for young people under the age of 25, were sometimes evident — the result of highly competitive conditions. They felt that the role of their organisations was to cooperate with other agencies in raising standards of professionalism and to eradicate such practices.

Working with partnerships

The BLRA hosted a feedback seminar on its partnership initiative in June of 2001 which was chaired by Assistant Chief Constable Robert Taylor of ACPO. The views were quite mixed. There was often, it was felt, confusion about the role of the industry bodies within the schemes. The BLRA chief executive concluded that there needed to be a clearer focus for the industry and the role that it played — e.g. liaising with local authorities and the police, as opposed to involvement with health and alcohol treatment.

Local authority partnerships were often primarily concerned with town centre pubs and the disorder associated with specific outlets. The industry bodies accepted that they had a significant role to play in ensuring better management practices, but felt that local authorities, transport companies and the Home Office, etc., needed to do their part as well. The general consensus was that partnership schemes were a positive development but there was a suggestion that some had a tendency to get ‘carried away’.

Among specific suggestions that had been made within crime reduction partnerships was that the drinks industry should sponsor police officers on a local level. There have been very mixed responses to this idea.

One issue which the BII felt was overlooked by most crime and disorder partnerships was that of ‘bootleg booze’. According to the BII the illegal importation of drinks across the channel results in each local authority in the UK having the equivalent of 1 million pints of beer per year in addition to that sold through legitimate outlets. A greater emphasis on removing this factor could lead to reduced levels of alcohol-related street disorder.
The BII are particularly keen to be involved more in local partnership schemes that seek to ‘reward’ pubs that have a good record of preventing alcohol- related violence. An example of such a scheme is the ‘Wandsworth Safer Pubs Committee.’ They also see the development of qualifications such as their door supervisors’ ‘passport’ as being of significant benefit. (They are currently applying to the QCA to make this a fully recognised scheme.)

A number of representatives of trade organisations were concerned about the difficulties of evaluating CRP initiatives in which they played a role. It was noted that when a partnership scheme is put in place there is often encouragement given to licensees to call the police when violent or disorderly incidents arise. There was concern that this should not be taken to indicate that the number of incidents itself had increased. Licensees should also not find themselves receiving ‘black marks’ from the police for reporting such offences.

SELP sees its role as working with police licensing officers, solicitors, barristers, other trade organisations and local authority licensing departments. It is, therefore, less involved in CRPs than some other bodies.

While most trade bodies welcomed the opportunity to be involved with other agencies in seeking to reduce alcohol-related problems, some felt that many of the partnerships lacked proper structure and did not necessarily focus on the most important issues. More extensive guidance from bodies and departments such as the Home Office was required to ensure greater consistency and properly targeted initiatives.