| Annual Report of Her Majesty's Chief Inspector of Constabulary 1998/99 | ||||
|
Chapter 4 COMPARISION OF POLICE PERFORMANCE 4.1 The Police and Magistrates Courts Act 1994 introduced Key Objectives measured by Key Performance Indicators through which the Home Secretary could prioritise particular services provided by the police each year. This year will be the fifth year in which police performance is reported against these objectives in our annual report. 4.2 Since their introduction the national key objectives have largely remained the same. However, this year has seen a significant refocusing of the objectives to reflect the Home Secretary's current priorities; in particular the targeting of young offenders and the improvement of the quality of life by reducing the level of local crime and disorder (FIGURE 7). Many developments linked to these two objectives have been introduced in the past two years including the Crime and Disorder Act, which brought in to being the statutory community safety partnerships and introduced the Youth Justice Board with supporting youth offending teams. 4.3 The setting of key objectives has achieved considerable success in focusing effort on particular policing problems. This will continue in 1999/2000 in the context of Ministerial priorities which will form part of the overall framework for police performance measurement. However, at the same time, the framework for measuring police performance is in the process of undergoing a fundamental reform which will come into effect from 1 April 2000 in conjunction with Best Value introduced by the Local Government Act 1999. 4.4 Following the Comprehensive Spending Review (CSR), which reported in April 1998, the Home Office set out the aims and objectives for the criminal justice system as a whole, and in support of these a set of overarching aims and objectives were developed for the Police Service. In order to provide a high level view of how the Police Service is performing against the overarching aims and objectives, a suite of indicators has been developed in consultation with ACPO, APA, HMIC, HM Treasury and the Audit Commission. For policing this has meant a recognition of the multiplicity of police activities that impact on each of the objectives, the inter-relationships between the activities and the difficulty in measuring success. The development of the performance indicators was underpinned by certain criteria. Ideally, the indicators should be:
4.5 Although in theory the performance measures should be outcome based, in reality it is often difficult to measure outcomes directly. In many cases therefore, output indicators or proxy measures have to be used as an interim measure while better measures are developed and the ability to collect such data becomes feasible. KEY OBJECTIVE 1 - DEALING SPEEDILY WITH YOUNG OFFENDERS 4.6 The principal aim of the new youth justice system is to prevent offending by young people. One of the six key objectives supporting that aim is the swift administration of justice so that every young person accused of breaking the law has the matter resolved without delay. In support of this aim, the Government has pledged itself to halving the average time taken from arrest to sentence for persistent young offenders from an average period of 142 days in 1996 to 71. In September 1998, the Home Office and the Lord Chancellor's Department issued a joint circular entitled 'Measuring performance to reduce delays in the youth justice system'. This circular set an initial target of completing at least 50% of persistent young offender cases within 71 days. A joint inspection into how long youth cases take, involving HMIC, CPS Inspectorate and HM Magistrates' Courts Service Inspectorate found that in their sample of persistent young offender cases 62.5% were finalised within the target time. Although performance across sample sites was variable, the study suggests that the target is achievable. 4.7 There are a number of ways in which the police can contribute to this process including speeding up the process of file submission for all young offender cases and with fast tracking schemes for persistent young offenders in conjunction with other agencies. The Police Service and the Crown Prosecution Service (CPS) have for the last two years been involved in a comprehensive project to monitor and improve the timeliness and quality of files submitted for prosecution under the Joint Performance Management (JPM) initiative. To put momentum into the process, files should be passed to the CPS within specific time guidelines. Further, when passed to the CPS, it is important that the quality of the file is satisfactory or at least of sufficient quality and completeness that the prosecution can proceed without further work being required on the file. 4.8 In light of the concern about young offenders, from April 1998 youth files have been monitored separately. As this data request is new, not all forces were able to provide the data in the required format initially. However, by the last quarter of 1998/99 only one force (North Yorkshire) was still unable to provide all the data requested due to the introduction of a new information technology system that had yet to become fully operational. 4.9 In 1998/99, the number of files that related to young offenders totalled approximately 70,000 constituting 13% of total files (Table 1 Appendix C). This low proportion is a reflection of the fact that that many young offenders are dealt with by way of caution or other diversionary tactic. Another factor is that the data collection exercise is in its early stages and in the first year not all forces have been able to count youth files separately for the whole of the year. Some forces have only been able to provide sample data or data for the latest quarter. The accuracy and completeness of the data will gradually improve once new data collection processes and information systems are fully implemented. 4.10 Nationally, in 1998/99, the proportion of files that were processed within the time guidelines averaged 76% with the average of all forces being 74%, ranging from 42% (Nottinghamshire) to 97% (North Wales) (FIGURE 8). The majority of forces (25) achieved a level of over 70% of files processed within target times. Following the 'Review of Delay in the Criminal Justice System' (the 'Narey Report') a number of initiatives intended to expedite case progress from charge to disposal were piloted in six sites viz: Tyneside, Croydon/Bromley/Sutton, North Staffordshire, North Wales, Blackburn and Burnley, and Northamptonshire. Between October 1998 and March 1999, these sites piloted the following initiatives:
FIGURE 8 Percentage of Youth Files within PTI Guidelines 1998/99 4.11 North Wales Police show the highest proportion of youth files processed within time guidelines which may reflect the success of the Narey pilot projects sited in the area. The evaluation of the pilot sites by Ernst and Young, found that the six initiatives together reduced the average times between charge and disposal from 89.5 to 37.9 days for young offenders. Other areas with pilot sites have not achieved such high levels, but quarterly data shows marked improvements in performance in the last two quarters of 1998/99 in all pilot forces. For example Northamptonshire's performance rose from 58.6% within time guidelines for April to June 1998 to 84% in January to March 1999. 4.12 To sum up, forces have responded well to the requirement to monitor this data and by the end of the year only one force was unable to provide data on youth files separately. The forces in the top quartile of performance are achieving a level of at least 20 percentage points higher than those in the bottom quartile. This demonstrates the potential for improvement through the adoption of good practice as described in the Youth Justice Board Report 'Speeding up Youth Justice' issued in May 1999. 4.13 As with KPI 1i this is the first year that this data has been collected specifically for youth files and consequently there is not data from previous years with which to compare performance. All forces were able to provide this data although some have provided sample data. This indicator relates to all those files relating to young offenders that were deemed satisfactory or, at least sufficiently satisfactory for the case to proceed when passed to the CPS for prosecution. Throughout England and Wales, the proportion of files that met the quality standards was 92% with the average of all forces being 91%. The proportion ranged from 99% (North Yorkshire) to 72% (City of London) with the majority of forces (31) achieving at least 90% (FIGURE 9). This is a very high level of performance which is to be congratulated. However, the key performance indicator examines those files that are either fully satisfactory or just sufficient to proceed which is a slightly lower standard of quality. When the proportion of those files which are classed as fully satisfactory alone are examined, performance is more varied ranging from 33% (Hampshire) to 80% in Dyfed-Powys and is an area where improvements can be made. FIGURE 9 Percentage of Youth Files either Fully Satisfactory or Sufficient to Proceed 1998/99 4.14 The Crime and Disorder Act 1998, introduced statutory community safety partnerships to work together in reducing local levels of crime and disorder. The Act placed a responsibility for the partnership to produce a crime and disorder strategy by 1 April 1999. Consequently, in the first year of their establishment, forces were asked to have mechanisms in place for measuring success in tackling the specific crime and disorder problems identified by the partnerships. 4.15 At present the District Audit Service is embarking on a programme of visits to audit the strategy and progress of the partnerships. The two lay Assistant Inspectors of Constabulary are taking part in a number of these visits in order to gain information for future HMIC inspections. In addition to this work, during the summer of 1999 HMIC, in partnership with the Crime Reduction Unit of the Home Office, has pursued this work through a comprehensive series of visits to all forces. HMIs, accompanied by senior Home Office officials, have discussed local crime reduction targets and partnership performance with chief constables and their senior management teams, together with many of forces' local partners. The approach throughout has been to listen and learn so that best practice can be shared, resources focused, and policy developed to best help these new partnerships achieve their objectives. 4.16 One of the most important outcomes of police activity is crime reduction to which all operational policing contributes. There is no easy method of measuring success in the prevention of crime and the only measure of crime reduction at force level currently available, is the absolute level of recorded crime per 1000 population. This measure has however, been affected by the changes in the counting rules introduced on 1 April 1998, which broadened the classifications of crimes classed as notifiable, so increasing the numbers of crimes included in the national statistics for 1998/99. Work to determine what that effect has actually been, in terms of numbers, is still on-going but indications are that the overall trend is down by 1% which means the sixth consecutive fall in recorded crime figures. The Number of Crimes Recorded per 1000 Population 4.17 Without examining the trend, it can be seen that the number of crimes recorded per 1000 population varies dramatically across forces from below 60 in a few non-urban forces (Dyfed-Powys, Surrey and Hertfordshire) to almost 150 in some more urban areas (Humberside and GMP) excluding the City of London which has a very small resident population (FIGURE 10). In general, forces covering a predominantly rural area have a lower level of crime per head of population than those covering a predominantly urban area. In recognition of the different levels of crime experienced by forces, the funding formula takes account of socio-demographic indicators that have been found to be linked to, or to predict, high levels of crime, (e.g. young male unemployment). FIGURE 10 Number of Recorded Crimes per 1000 Population 1998/99* 4.18 Nevertheless, within groups of similar forces, there are still wide disparities in levels of crime per head of population that are difficult to explain. The potential to decrease the level of recorded crime per 1000 population is illustrated by the fact that force performance often differs greatly from their most similar force averages. 4.19 There are a number of forces with a level of recorded crime per 1,000 population that appears to be out of the normal range for forces of similar nature. In particular, Humberside has a significantly higher level than other forces. Her Majesty's Inspector has investigated the reason for this high level, which has followed the same pattern for a number of years, but the results are inconclusive. One reason claimed by the force is an exceptionally high level of integrity in recording of crimes and there is some evidence from the last inspection to support this view. Nevertheless the force is well aware of the need to reduce crime and efforts to that end were seen during the recent inspection not least in the area of partnerships. Her Majesty's Inspector recommended in the latest inspection report that the force improve as a matter of urgency its crime management system. This would allow more confidence to be placed in its crime figures and also give increased scope for analysing trends and patterns and so through directed policing give real opportunity for crime reduction (Table 2, Appendix C). KEY OBJECTIVE 3 - TO TARGET DRUG-RELATED CRIME IN PARTNERSHIP WITH OTHER LOCAL AGENCIES 4.20 Developing an outcome based performance indicator to measure success in tackling drug-related crime that is both meaningful and for which the data is easily available has proved difficult. One of the problems with measuring success against this objective is the measurement of drug-related crime itself. Research consistently points to the link between drug misuse and crime, for example in a study for the Home Office on English arrestees and drug misuse, nearly half the arrestees who reported taking drugs within the last year said their drug use was connected with their offending. But it is still only possible to make estimates of the effect on crime that drug misuse has. 4.21 In April 1998, the Government's ten-year strategy for tackling drugs misuse 'Tackling Drugs To Build A Better Britain' was published. The strategy highlights the cross-cutting nature of the drugs problem and recognises that most Government departments and local agencies have a part to play in tackling the drugs problem. This strategy aimed to set clear, consistent and rigorous targets to achieve the aims of the strategy in the first year and these have now been published. The objective that the police will contribute most to is: 'to protect our communities from drug-related anti-social and criminal behaviour'. 4.22 The key performance target for this objective is 'to reduce levels of repeat offending amongst drug misusing offenders by 50% by 2008 and 25% by 2005'. To this end the Police Service is encouraged to make maximum use of arrest referral schemes and the target set for 1999/2000 is to double the number of face-to-face arrest referral schemes and the number of arrestees referred to and entering treatment programmes. The key performance indicator for 1998/99 is an interim indicator while more appropriate indicators to measure the police contribution to the ten-year drug strategy are developed which can link into the overarching aims and objectives. FIGURE 11 Number of Arrests for Supply or Possession with Intent to Supply per 10,000 Population 4.23 This is a new indicator for which forces have been required to institute a new data collection exercise. However, all forces have been able to provide this information (Table 3 - Appendix C). The differences between the performance of forces against this indicator are striking. Nationally, forces average 4 ranging between 1 (City) and 9 (Cumbria) arrests per 10,000 population (FIGURE 11). The forces with a higher level of arrests vary greatly in terms of size, geography and population such as Cumbria to West Yorkshire. As the disparity cannot be explained by such factors the difference probably reflects the priority given to drug-related crime and the methods used to combat it within forces. Further research is required in order to understand the differences better. KEY OBJECTIVE 4 - TO MAINTAIN AND, IF POSSIBLE, INCREASE THE NUMBER OF DETECTIONS FOR VIOLENT CRIME 4.24 Violent crime is a consistent concern of the public and consequently has remained a priority for the Police Service. In recent years recorded violent crime has been rising nationally with significant increases recorded in many forces. This has been attributed in some way to changes in recording practice - particularly in relation to encouraging the recording of domestic violence. In addition to this, from 1 April 1998, the counting rules for recorded crime changed in order to improve the recorded crime statistics and simplify the rules. The main principle was that there should be one crime for each victim. The results of the changes were to expand the notifiable offences to include almost all indictable and Triable Either Way (TEW) offences. One of the effects of these changes has been to increase the number of recorded crimes that are classed as violent crime in particular 'common assault' which previously was non-notifiable and now is included in the notifiable offences. 4.25 The changes to the counting rules appear to have had a differential effect in forces with six forces experiencing a rise of over 100% and others seeing rises of less than 25%. Currently, it is not possible to identify the trend in violent crime because of the varying effects in forces. However, work is in progress within the Home Office to provide some estimate of overall effect. KPI 4 - The percentage of violent crimes detected 4.26 Nationally, the detection rate has fallen slightly from 69% in 1997/98 to 66% which is the level achieved in 1996/97 and so, superficially, does not appear to have been unduly affected by the change in the counting rules (Table 4 Appendix C). In 1998/99 performance in detecting violent crime varied greatly across the country averaging 78% but ranging from 39% (Metropolitan Police) to 97% (Dyfed-Powys). This wide spread of performance hides the fact that the majority (31) of forces achieved a performance of between 70 and 90 percent (FIGURE 12). The variation in performance has increased this year with a difference of 58 percentage points between the highest and lowest performers compared to 43 in 1997/98. FIGURE 12 The Percentage of Violent Crime Detected 1998/99 4.27 One factor that may have influenced force detection rates for violent crime is the introduction of the change in the recorded counting rules. Although the national detection rate has not changed greatly in 1998/99, some forces have seen considerable changes this year which may well relate to this factor. At the top end of the scale, three forces showed increases of 10 or more percentage points in their detection rate (City of London, Greater Manchester and Lancashire) whilst at the other end the Metropolitan Police and Sussex experienced decreases of 13 and 10 percentage points respectively (FIGURE 13). Until the analysis of the changes in recorded crime has been completed it is not possible to determine what the exact effect has been on detection rates and why it has been affected.
FIGURE 13 The Change in Percentage of Violent Crime Detected Between 1997/98 and 1998/99 4.28 Although the change in the counting rules may explain in part the change in detection rates seen in 1998/99, it does not explain how it is that a small number of forces have detection rates some 20 percentage points below the performance of the top quartile of forces. Within the forces in the bottom quartile are most of the large metropolitan forces, which demonstrates a possible link between performance in detecting violent crime and the nature of area policed. One of the distinguishing features of the large urban forces in terms of violent crime is that there tends to be a higher proportion of street robbery recorded. Such offences are often more difficult to detect as the offender is usually a stranger to the victim as proved by the national detection rate of 22.6%. Consequently, the level of robbery can affect the proportion of violent crime detected in those areas. Robbery constitutes an average 11% of violent crime nationally, but ranges from 21% (Cleveland) to 1% (Dyfed-Powys and Gwent). All metropolitan forces have proportions of robbery in the upper end of the scale (9% or above) and there is a correlation between the proportion of robbery recorded as violent crime in the force area and the violent crime detection rate. 4.29 Despite this, it can be seen that forces with similar socio-demographic profiles achieve quite different levels of performance and have produced significantly different levels of improvement over the past three years. The greatest increase in the detection rate for violent crime was demonstrated by West Midlands which improved by 20 percentage points over a 3 year period, which is an achievement to be congratulated. Conversely, Nottinghamshire and Sussex showed decreases of 17 and 11 percentage points over the same time period (FIGURE 13A). FIGURE 13A Percentage Point Change Between the Detection Rate for Violent Crime in 1996/97 and 1998/99 4.30 Whilst there may be reasons for low or decreasing detection rates for violent crime linked to recording practices and operational pressures, further work must be done to drive up performance, particularly from those forces with detection rates in the bottom quartile. KEY OBJECTIVE 5 - TO INCREASE THE NUMBER OF DETECTIONS FOR BURGLARIES OF PEOPLE'S HOMES 4.31 Since 1994/95, there has been a continued focus on domestic burglary which remains one of the greatest concerns of the public. The targeting of this problem has had significant results illustrating that success can be achieved through consistent effort and intelligence-led policing. This can be seen by the fact that for the fifth year running the level of recorded domestic burglary has decreased with a fall in 1998/99 of 5.7% (down by 28,393 crimes). Over the past two years since 1996/97 forces have averaged a decrease of 19% with three forces achieving a decrease of over 30% (Bedfordshire, Cumbria and Leicestershire). No force has shown an increase in recorded burglary dwellings over this period but for some the decrease has been less marked that than for others. Sussex showed least success in this area of performance by maintaining the same level and six other forces have shown a decrease of less than 10%. KPI 5 - The percentage of burglaries of dwellings detected by primary means and overall 4.32 In addition to monitoring the level of crime reduction in this area, the Key Performance Indicator measuring the performance of forces in addressing domestic burglary is the percentage of those crimes detected both in total and by primary means. Primary detections being those where an offender has been charged, summons or cautioned or where the offence has been previously recorded as a crime and the offender has had the offence taken into consideration when sentenced. FIGURE 14 The Percentage of Burglary Dwelling Detected in Total and that Made up of Primary Detections 1998/99 4.33 Nationally, the proportion of crimes that were cleared up by primary means is low at 15% and has not improved on the previous year (Table 6 Appendix C). The disparity in the levels of primary detections that are achieved ranging from 43% (Dyfed-Powys) to 10% (Greater Manchester) is evidence of the potential for improvement (FIGURE 14). Only seven forces achieved a primary detection rate of 20% or greater but the fact that these forces varied from urban forces like South Wales and more rural forces like Cumbria illustrates that all forces should be able to achieve equivalent performances. 4.34 The overall detection rate varies enormously from nearly 60 per cent to just over 10 per cent but the use of secondary detections as a clear up method is determined largely by force policy. Consequently, these detections can be used to contribute from zero per cent (City of London and Cleveland) to over half of the total detection rate in eight forces. Nationally these detections make up 38% of all detections for domestic burglary, the largest proportion being seen in Norfolk where 64% were obtained by secondary means (or other means) (Table 7, Appendix C). Many secondary detections result from post sentence interviews which should be conducted with the primary object of gathering intelligence in line with guidance issued by HMIC. From 1 April 1999, the counting rules for detecting crime have been amended to remove the distinction between primary and secondary clear ups, whilst strengthening the rules to ensure that the same strict criteria are required, in terms of evidence, for all detections. Detections will, where offenders are already in prison, be allowed when there is corroborative evidence to support any admission made and the CPS and a senior police officer agrees that it should be classed as a genuine detection. The new rules should remove any inconsistencies or ambiguities that existed before and generally improve the integrity of the detection figures as a whole. FIGURE 15 Percentage Point Change in Primary Detection Rate for Burglary Dwelling between 1996/97 and 1998/99 4.35 Since 1996/97 the national primary detection rate has risen from 13% to 15% which is still very low. In that time three forces have shown notable increase in the primary detection rate - Dyfed-Powys, Gloucestershire and West Midlands with increases of 10, 8 and 8 percentage points respectively (FIGURE 15). Both Gloucestershire and West Midlands have shown steady increases from a low base over the past four years to over 17% which is above average. Dyfed-Powys has maintained and improved on its consistently high performance. All three forces have also shown decreases in recorded burglary dwelling at the same time as improvements in primary detection rates. 4.36 Conversely two forces have shown serious decreases in the primary detection rates namely Cleveland and Norfolk with reductions of 8 and 7 percentage points respectively. Cleveland has shown considerable success in reducing domestic burglary with a decrease of 17.6% over the three years while the decrease seen in Norfolk at 4.6% is well below the average decrease. 4.37 The performance of a few forces in achieving significant reductions in recorded burglary dwelling, high or improving primary detection rates or both reflects the potential performance that could be achieved by other forces. Nevertheless, the focus on burglary dwelling over the past four years has been a factor in a greatly reduced level of domestic burglary nationally and a slightly improved primary detection rate both of which are a welcome success. KEY OBJECTIVE 6 - TO RESPOND PROMPTLY TO EMERGENCY CALLS FROM THE PUBLIC 4.38 As ever, the response to 999 calls forms an important first impression to the customer and the public expects a high level of service delivery. Over the past 4 years the number of 999 calls has risen by 16% nationally and in some forces by as much as 77%. This is a result of many influences including improvements in technology which allow more calls to be handled by a force and increased demand which itself is affected by the increased mobile phone ownership. 4.39 However, as seen from research (The Use of Call Grading, PRG Paper 13, 1994) the proportion of 999 calls that turn out to require an immediate response is often quite low ranging from 18% to 61% in the period of the study. Consequently, many forces are attempting to improve demand management through the use of public service desks and helpdesks and through local advertising on the use of the 999 call system. Nevertheless, as each individual 999 call could be an emergency, the need to answer the call as quickly as possible is still of fundamental importance. 4.40 Once the call has been answered, the next step is to respond to that incident. All forces operate a graded response system whereby calls are classified according to their urgency and the most appropriate response deployed. This can mean an immediate response or the transfer of the call to another department for an appointment to be made. This is another key area that catches the public imagination. The public expects that once they have dialled 999 with an emergency, the police will arrive instantly. KPI 6 i - The percentage of 999 calls answered within the local target time 4.41 Over the past 5 years performance in answering 999 calls within target has remained consistently high averaging 89% in 1998/99. The service itself has improved through the lowering of targets (no force now has a target time for answering of over 15 seconds). In 1998/99 local targets ranged from 5 seconds (North Yorkshire) to 15 seconds (14 forces). This year only one force (Derbyshire) changed its target by decreasing it from 15 seconds to a more challenging 10 seconds. 4.42 The level of performance achieved against these targets is almost without exception very high. Only four forces (Bedfordshire, Hertfordshire, Dorset and Essex) achieved lower than 80% of calls answered in target and 21 forces achieved over 90% the highest being Cumbria at 99% (FIGURE 16). Of the four forces with the lowest performance against this indicator all are involved in projects to review or restructure their call handling systems. The decrease in performance of Hertfordshire in answering 999 calls has been the cause of concern to HM Inspector. The decrease has been attributed to a new telephone and subsequently a new monitoring system for call answering which has resulted in setbacks in improving performance. In 1997, Essex introduced a force information room that handles all emergency phone calls and all radio traffic and response deployment in the county. This was an ambitious project which has presented some difficulty in implementation and suffered delay. Indications are that performance is now beginning to improve. This demonstrates that with a few exceptions usually related to changing the process of the technology or organisational structure of call handling, the public in general is now receiving a consistently high level of service in this area. However, it is only the first step in the process and the action taken following the answering of the call will also have just as great an impact on public perceptions as the initial answering. FIGURE 16 The Percentage of 999 Calls Answered within Local Target Time 1998/99 4.43 Since 1996/97 a number of forces have shown noticeable improvements in performance most notably Merseyside, Leicestershire, Norfolk and Humberside where performance rose by 19, 15, 14 and 14 percentage points respectively (FIGURE 17). In Merseyside, Norfolk and Leicestershire projects to introduce centralised call handling functions experienced many difficulties which caused significant dips in performance. In all cases, considerable effort has been put into improving performance, in the case of Norfolk this has involved a costly recovery project. In Merseyside the force has developed incident management units which combine crime management units, intelligence and call handling facilities through which all non-immediate calls are routed. In Humberside, HM Inspector of Constabulary has previously found elements of call handling performance to be of significant concern and so it is to its merit that such progress have been achieved. FIGURE 17 Change in Percentage of 999 Calls Answered in Target between 1996/97 and 1998/99 4.44 Two forces, Dyfed-Powys and Hampshire were not able to supply data in 1996/97 due to the introduction of new technology. While Hampshire has maintained a level of performance of 83% of calls answered on target over the past two years, Dyfed-Powys has demonstrated a decrease of around 12 percentage points in 1998/99 compared to 1997/98 (Table 8, Appendix C). This has been attributed to a combination of factors: the introduction of the new technology and a reduction in the target time. However, early reports on the current year's figures indicate an upturn in performance and a recent review of staffing has resulted in improved efficiency in achieving targets. 4.45 In response to the incoming Best Value regime and as a result of the need to make efficiency gains many forces have been examining innovative methods of providing the call handling function. A number of forces are examining the sharing of communications rooms with other forces and with other emergency services. HMIC welcomes this response to the efficiency agenda and hopes that further progress in the sharing of facilities will be made. Overall, the performance in answering 999 calls has been maintained at a high level with one or two exceptions that should be resolved in the coming year. There is no room for complacency however, and forces are urged to continue to achieve this increase and to consider how their response to all calls not just 999 calls could be improved. 4.46 In responding to these incidents classed as requiring an immediate response, the speed of the response possible is determined by the size and geography of the force, and targets are set to take account of these factors. In general, response times for rural areas are set at 20 minutes and urban areas are set at 10 minutes. Few forces vary from these targets but a number set one target for the whole force area. All such comparisons are complex but for ease of illustration all forces have been shown with their overall performance against all targets set, to give a summary view (FIGURE 18). FIGURE 18 Percentage of Incidents Requiring an Immediate Response Answered within Local Target Time 1998/99 4.47 Performance against the urban and rural targets varies but as a whole, there is a very high rate of response within the target times. Nationally, the proportion answered within target averages 88% ranging from 76% (Norfolk) to 98% (North Yorkshire) (FIGURE 18). As mentioned before, Norfolk has experienced problems in centralising call handling and control room functions. In addition, there has been some concern about the over-grading of incidents as immediate. In 1997/98, HM Inspector of Constabulary recommended that Norfolk should identify where accountability lies for incident response performance and make arrangements to hold to account those responsible. The force has done this and has also chosen this area as a pilot for the force programme of Best Value process reviews as it is acutely aware of the need to continue to improve performance. Only three forces achieve less than 80% (Cambridgeshire, Norfolk and Warwickshire). In Warwickshire, although the migration to the new central communication centre in 1996 was free of major technical difficulties some working practice difficulties remain. Possibly related to both are the low staff morale and the high sickness levels found in this area. It is obvious that there is room for improvement in aspects of call handling and HM Inspector has recommended that action be taken. FIGURE 19 Percentage Point Change in Immediate Response Incidents Answered within Local Target Time 1996/97 to 1998/99 4.48 Over the past two years, a number of forces has shown notable improvements including Merseyside with an increase of 19 percentage points and Leicestershire, Hampshire, Suffolk and Humberside also showing notable improvements (FIGURE 19). In Hampshire new guidelines defining which type of incident requires which graded response were issued in November 1998 significantly reducing the proportion of incidents that require an immediate response from 18 to 4.5 per cent. The increase over two years in Suffolk reflects the restructuring of the call handling facilities over the past four years. In Leicestershire, as with 999 call handling, considerable effort has gone into improving performance in this area after a decrease in achieving target times following the introduction of centralised call handling. The only force showing a sizeable reduction in performance is Dyfed-Powys which is related as explained previously to the introduction of new call handling technology and indications are that improvements are now occurring. 4.49 In our thematic inspection report Lost Time on the management of sickness absence and medical retirements published in November 1997, the cost of sickness absence and medical retirements was highlighted. We suggested that the high levels of, and wide disparity between forces in levels of medical retirements for both police and civilian support staff confirmed the need for targets for improving performance in this area. Consequently, we proposed that by 1999/00, a sustained management effort should lead to all forces achieving at least the performance achieved by the top quartile in 1996/97 i.e. no more than 33% of all police retirements being a result of medical reasons. Those forces that had already achieved this level or better were urged to maintain and improve on their current performance. For civilian support staff, it was suggested that the Audit Commission target of 25% proposed for Local Government should be adopted. FIGURE 20 Percentage of Police Medical Retirements 1998/99 4.50 As illustrated last year in my Annual Report, the initial results indicated that forces were making substantial progress, with a reduction in the actual number of medical retirements achieved of 185 resulting in 38% of all retirements being due to medical reasons nationally. Once again, this year (FIGURE 20 AND 21)) improvements have been made in many forces with 29 forces reducing or maintaining the number of medical retirements recorded leading to a reduction of 185 of medical retirements nationally (Table 10, Appendix C). However, in terms of percentages this has actually created a slight increase in the national figure to 39% due to a drop in ordinary retirements. Over the two year period, the number of medical retirements has fallen by 370 nationally, equating to a fall of 6 percentage points with forces averaging a fall of - 5 percentage points. In comparison to the targets set, only 17 forces have now achieved a level of 33% or lower as compared to 12 in 1996/97 which is disappointing (FIGURE 20). In fact, in nine forces medical retirements still constitute half or more of all police retirements. 4.51 However, examining percentages alone can be deceptive, as they can be affected by the level of ordinary retirements. Since 1996/97, 27 forces have achieved a decrease in the numbers of medical retirements, equating to a fall of 10 or more percentage points in sixteen forces. Some of the decreases achieved have been substantial most notably in Merseyside, West Midlands and Avon and Somerset where reductions of 125, 59 and 42 medical retirements respectively have been seen. Of the fourteen forces that have shown an increase in the percentage of medical retirements, the most notable increase has been seen in Suffolk (up 29 percentage points) (FIGURE 21). However, this illustrates the fact that in forces with low numbers of retirements in total, small changes in medical retirements can cause substantial changes to the percentage - the rise in Suffolk being due to an increase of just seven medical retirements. In terms of numbers of medical retirements, the Metropolitan Police has shown the greatest increase (up 53 medical retirements) however, due to a rise in the number of ordinary retirements of 4 percentage points. South Wales has also shown an increase of 21 medical retirements - an increase of 10 percentage points. So overall, although fewer forces than expected have already managed to meet the targets set in terms of percentages, many have shown significant commitment to reducing the numbers of medical retirements and may meet the target next year. 4.52 It is difficult to estimate the cost of the savings these reductions have made, without knowing the rank, length of service of the officer and any injury award made. It will however equate to a substantial sum. As an area in which significant savings can be made, it is essential that further effort is made to ensure that the good practice highlighted in our thematic report is implemented - particularly by the 9 forces where over half of all police retirements are still due to medical causes. 4.53 In comparison civilian medical retirements have fallen by 221 over the past two years equating to a decrease of 11 percentage points (Table 11, Appendix C). Only nine forces managed to achieve the target of 25% or less so far but 31 forces reduced or maintained the numbers of civilian staff medical retirements some like West Midlands significantly so, with a fall of 46 (decrease of 42 percentage points). 4.54 A reduction of sickness absence is another area that has the potential to provide substantial efficiency savings. From the national figures (FIGURES 22 AND 23), it can be seen that average sickness absence per police officer and per civilian employee have both slightly reduced in 1998/99. Comparisons between forces are problematic due to the methods of counting sickness absence. Many forces can only count the absence in days, but most forces run shift systems which have days of longer than 8 hours. If an officer is off sick for one day but works a 10 hour shift system, the amount of working time lost is equivalent to a day and a quarter. However, comparisons within forces over time are valid. Over the two year period from 1996/97 less than half of forces (20) have managed to reduce average police officer sickness. 4.56 Since 1997/98 the level of working days lost to sickness per police officer has fallen by approximately a day nationally. This equates to a reduction of 128,860 days in total which in itself is the equivalent of over 560 extra officers being in work. This illustrates what substantial differences actually underlie the figures in Figure 22 and the extra officers on the streets it would create if all forces could reduce their sickness levels the way some of the better performers have (FIGURE 22). FIGURE 22 Average Number of Working Days Lost to Sickness per Police Officer
FIGURE 23 Average Number of Working Days Lost to Sickness per Civilian (Including Traffic Wardens) 4.57 In 1996/97 some forces have made significant improvements in sickness management. In particular, Merseyside has had phenomenal success in reducing sickness from a very high average of 18.6 days per officer to 11.3, a decrease of over 7 days on average. Other forces that have had great success in this area are the Metropolitan Police (down 3.5 days), Humberside (down 3.6 days) and Durham and Hertfordshire (both down by 2.9 days). Conversely, Essex and North Wales have seen increases of 3.6 and 3 days per officer respectively. HM Inspector has recommended that a review of sickness management be undertaken in Essex. 4.58 In comparison with police sickness, civilian working days lost to sickness are slightly higher although there has been a decrease since 1996/97 with 23 forces demonstrating lower levels this year (FIGURE 23). Once again Merseyside has shown outstanding improvements with a decrease of 7.5 days per civilian over the past 2 years. Hertfordshire, Greater Manchester, Cambridgeshire and Humberside have also shown sizeable improvements. Bedfordshire has shown a steady increase over the past two years with a total of six additional days per civilian on average in 1998/99 compared with 1996/97. However, the force commissioned an independent report to look into this area in recognition of the problem. The report was published in March 1999 and has recommended a force-wide sickness management and monitoring group and the rewriting of all policies and procedures. It is hoped that this project will bring improvements. Overall, some progress has been seen in sickness management but there is still great potential for reduction. The fact that many forces have included reducing sickness in their efficiency plans for 1999/2000 may have an impact on the current year's performance. 4.59 For a number of years the Police Service has been urged to improve its performance in dealing with racial incidents. In April 1998, ACPO published basic standards for the reporting of racial incidents and other organisations have published detailed guidance. In order to ensure that appropriate action is taken in cases where the crime or incident has a racial motivation, it is vital that the community has the confidence to report the incident to the police in the first place. During our follow-up inspection on police and community relations Winning the Race Revisited, each force visited acknowledged that there was significant under-reporting of racial incidents. This picture was confirmed by the Macpherson Inquiry, and the report recommended that all possible steps should be taken by police forces to encourage reporting of racist incidents and crimes. So, although the ultimate aim is to reduce the number of racist incidents that take place, until the Service is confident that all such incidents are reported, a measure of improved police response and increased community confidence can be illustrated by an increase the number of such incidents recorded. 4.60 The new Macpherson definition of a racist incident has now been adopted by the police service - it is not different in substance but simpler, clearer and gives primacy to the perception of the victim: "A racist incident is any incident which is perceived to be racist by the victim or any other person" FIGURE 24 Racist Incidents 4.61 In 1997/98 the numbers of racist incidents rose by just 6% with just over half of forces (25) increasing the numbers recorded. The figures for 1998/99 are due to be published by the Home Office in December but early indications are that there has been a substantial improvement. Provisional data shows that the national figure has risen by 75% with all but one force showing an increase in the numbers of racist incidents recorded. Some forces including the Metropolitan Police appear to have doubled their figures. However, wide variations in the level of incidents recorded are still seen in forces with similar minority ethnic populations. Although the differences in these populations, geography and socio-demographic factors could explain some differences, they provide an indicator of differences in the systems for reporting and recording racist incident. When the numbers of racist incidents are standardised by the minority ethnic population of the force area, the numbers reported per 1000 minority ethnic population vary less, but still range between 22 (Cumbria) and 1 or less (Bedfordshire, City and Lincolnshire). From initial indicators, the emphasis on the reporting and recording of racist incidents following the Macpherson Inquiry appears to have had a significant effect on the reporting and recording of racist incident.
Anyone who wishes to comment on an inspection, a report HM Chief Inspector of Constabulary
Details on the role of the Inspectorate, copies of
individual reports
HM Inspectorate of Constabulary
Tel: 0171 273 2279 This and other HMIC reports are also on our website:
|
||||
