- Email: [email protected]
Developing mental health indicators in England
Public Health (2008) 122, 897e905
www.elsevierhealth.com/journals/pubh
Original Research
Developing mental health indicators in England John Wilkinson*, Jenny Bywaters, Stuart Simms, David Chappel, Gyles Glover North East Public Health Observatory and the National Mental Health Observatory, University of Durham Queen’s Campus, Wolfson Research Institute, Stockton on Tees TS17 6BH, UK Received 28 July 2007; received in revised form 24 September 2007; accepted 26 October 2007 Available online 3 March 2008
KEYWORDS Mental health; Indicators; Regions
Summary Objectives: This paper reports on the construction of a set of indicators for mental health and the publication of a report for England’s Chief Medical Officer. The report was the seventh in a series of reports aimed at initiating public health action to improve health at a regional level in England. Study design: Observational study using routine data. Methods: A set of over 80 indicators was constructed by an expert group. These indicators were then populated from routine datasets. Commentary was provided on each indicator in the report. Results: A small team compiled this large set of indicators from routine data working in a public health observatory in one region of England. Conclusions: It is possible to produce a large array of indicators about mental health. The conclusion from examining these indicators is that there are many unexplained differences in mental health across the regions of England. Many of these indicators are closely linked to deprivation. Some indicators show a lack of difference across the country, and in many instances, service provision is inconsistent and does not always relate to need. In some cases, there was a worrying absence of data, e.g. data on ethnicity and the mental health of offenders. ª 2007 The Royal Institute of Public Health. Published by Elsevier Ltd. All rights reserved.
Introduction In 2004, the Chief Medical Officer for England commissioned the UK Association of Public Health Observatories to produce a series of reports highlighting issues of public health importance,
with the aim of highlighting regional differences and stimulating and empowering public health practitioners (working at these levels) to take action across a range of sectors to improve health. This paper will highlight the issues in relation to the seventh report in the ‘Indications’ series, on
* Corresponding author. Tel.: þ44 191 334 0400; fax: þ44 191 334 0391. E-mail address: [email protected] (J. Wilkinson). 0033-3506/$ - see front matter ª 2007 The Royal Institute of Public Health. Published by Elsevier Ltd. All rights reserved. doi:10.1016/j.puhe.2007.10.012
898 mental health. The paper will describe the selection of indicators, list them, and also discuss presentational issues. This report focused on adults (aged 18e64 years). Indicators on mental health in older people will be included in the next planned report in the series on older people. A planned report on adolescent mental health is subject to the priorities agreed with the commissioners.
Background Mental health is identified as one of the six national priorities for action in the White Paper, ‘Choosing Health’.1 In the past, indicators of mental health have been difficult to find. Targets for mental health improvement have largely concentrated on suicide rates, which, although important, give a limited picture of the mental health of a community. This paper reports on the production of the UK Association of Public Health Observatories’ report on mental health.2
Policy context One in six people in England suffer from a mental health disorder, equating to around eight million people being affected in England today. The National Service Framework (NSF) for Mental Health,3 together with the National Suicide Prevention Strategy,4 provide the key policy context within which the report considers the mental health of adults of working age. Many of the data for the report are drawn from information sources such as the annual service mapping and financial mapping which were commissioned by the Department of Health so that progress with NSF implementation could be monitored in detail. The report aimed to inform the implementation of national and regional priorities. The main audiences include national policy makers, regional directors of public health, strategic health authority and primary care trust (PCT) chief executives and boards, and leaders in other health and social care organizations.
Methods A small steering group was established to select the indicators. The membership of this group included representatives of service providers, the voluntary sector, the Department of Health and the public health observatory. The choice of indicators was, in part, guided by key issues, recommendations and targets identified in a range
J. Wilkinson et al. of policies including the NSF for Mental Health3 and ‘Choosing Health’.1 This provided an opportunity to bring together a number of indicators that have been collated by different projects over the past few years (such as ‘Mental Health Outcomes’,5 ‘Better Metrics’6 and the ‘Local Basket of Inequalities Indicators’7), and to develop indicators utilizing new data sources that have become available recently [such as the Mental Health Minimum Data Set (MHMDS)8 and new data from primary care9] or have been underutilized in this area (such as from the Health Survey for England10 and employment information11). A full list of the indicators selected is shown in Table 1. The report presents a wide range of data on the factors that can give rise to poor mental health, the mental health status of populations, provision of interventions of care for mental illness, service user experience, and traditional outcomes such as suicide. The authors had the advantage of access to the Durham Mental Health Mapping Service, which provided a systematic overview of mental health services in England until 2006,12 and were also able to make use of the MHMDS8 which was developed by one of the authors of the report (GG). The report is structured in a similar way to previous reports and looks at indicators of:
risk and protective factors and determinants; population health status; interventions; effectiveness of partnerships; service user experience; and workforce capacity.
Ethnicity Ethnicity is an important issue in mental health. A recent service user census13 found considerable variation in the number of people in hospital from different ethnic groups. There may be variations in underlying morbidity, diagnosis and management that need further investigation. There are also good data on ethnic mental health from the EMPIRIC study.14 However, for this report, which compares English regions, the data were not broken down further by ethnicity. While there is some interest in whether a particular ethnic group has different experiences in different parts of England, the fourth ‘Indications’ report15 showed that the data are not robust enough to make these comparisons. Ethnic variations will continue to be best analysed at national level at the moment. A comparison of ethnic coding quality was included within the MHMDS, as this was not possible when the fourth ‘Indications’ report15
Developing mental health indicators in England Table 1
899
List of indicators contained in the report.
Risk and Protective Factors and Determinants Deprivation % of super output areas in each region, by quintile of deprivation Mental health needs Mental health ‘needs’ indices by region employment e % of people of working age in employment Employment of people with mental health problems % of adults of working age with a mental health problem in employment Incapacity benefits Mental and behavioural disorders incapacity benefit claimant rate per 100,000 population aged 16e59 years Limiting long-term illness % of adults with a limiting long-term illness Alcohol Standardized % of alcohol consumption above ‘sensible’ daily limits, 2005 Drugs People in contact with structured drug treatment, crude rate per 100,000 Physical activity % of adults (aged 16 years and over) participating in 20 or more days of moderate intensity sport and active recreation (averaging five or more times per week) in the last 4 weeks Healthy eating % of adults who consume five or more portions of fruit and vegetables per day Social capital e participation in civic society % of adults participating in any civic activity in the previous 12 months Religion e % of population reporting having a religion, 2001 Social support e % of adults who could get a lift somewhere if needed Social networks e % of adults speaking to relatives by phone Neighbourliness e % of adults speaking to neighbours Education % of pupils aged 15 years in schools maintained by the local authority achieving five or more grade A)eC GCSEs or equivalent Learning and development % of working age adults in job-related training in the past 13 weeks (working and unemployed) Violence and safety % of the population with a ‘high level of worry about violent crime’ Total incidence of violence per 10,000 adults reported by people in the British Crime Survey All violent reported crime recorded by the police forces per 10,000 population Gambling Household expenditure on gambling payments, 2002e2005 Ethnic coding % of records in Mental Health Minimum Dataset (MHMDS) which include a valid and usable ethnic group code Population Health Status Psychiatric disorder % of adults with a GHQ12 score of 4 or more Psychiatric morbidity % of population scoring 18 or more on the Clinical Interview Schedule Usage of specialist mental health services People aged 18e64 years documented as receiving care in the MHMDS, rate per 100,000 population Residential and nursing home care Council supported residents aged 18e64 years with mental health problems, rate per 100,000 population Severe mental illness on general practitioners’ register % of service users registered with a general practitioner who have severe long-term mental health problems Suicide Directly age-standardized mortality rates for suicide and injury undetermined, persons aged 15 years and older Directly age-standardized mortality rates for suicide and injury undetermined, males 15 years and older Directly age-standardized mortality rates for suicide and injury undetermined, females 15 years and older (continued on next page)
900
J. Wilkinson et al.
Table 1 (Continued) Self-harm admissions Directly age-standardized Directly age-standardized Directly age-standardized Directly age-standardized
self harm hospital admission rate per 100,000 population self harm hospital admission rate per 100,000 population hospital admission rate per 100,000 population for poisoning hospital admission rate per 100,000 population for poisoning
Interventions Mental health promotion % of local implementation teams (LITs) self-assessed as ‘GREEN’ for having a mental health promotion lead officer % of LITs self-assessed as ‘GREEN’ for having a mental health promotion strategy and action plan Assertive outreach Rate per 100,000 population aged 18e64 years reported by primary care organizations to be currently receiving mental health assertive outreach care Crisis resolution Rate per 100,000 population aged 18e64 years reported by primary care organizations to be currently receiving crisis resolution service care Early intervention for psychosis Rate per 100,000 population aged 15e34 years reported by primary care organizations to be receiving psychosis early intervention care Day care % of LITs with National Health Service (NHS) day hospitals Whole-time-equivalent care staff in NHS day care facilities, rate per 100,000 population aged 18e64 years % of LITs with at least one day centre or resource centre, 2005e2006 Rate of drop-in facilities per 100,000 population aged 18e64 years Admissions for depression and anxiety Directly age-standardized hospital admission rates for depression per 100,000 population aged 15e74 years Directly age-standardized hospital admission rates for anxiety disorders per 100,000 population aged 15e74 years, 2001e2002 Admissions for schizophrenia Directly age-standardized hospital admission rates for schizophrenia per 100,000 population aged 15e74 years Prescribing in primary care Average daily quantity (ADQ) of antidepressants prescribed by region ADQ of antipsychotics prescribed by region ADQ of hypnotics and anxiolytics prescribed by region Suicide audits % of primary care trusts with completed suicide audit Administration of electroconvulsive therapy (ECT) Individuals administered ECT, rate per 100,000 population aged 16e64 years Mental Health Act Commission Service users detained under the Mental Health Act 1983 resident in NHS facilities and independent hospitals, rate per 100,000 population >18 years Effectiveness of Partnerships Mental health expenditure e non-statutory % in total planned investment in adult mental health services by main provider type Multidisciplinary community mental health teams % of community mental health teams reported as achieving full local integration between NHS and social services partners Service User Experience Complaints % of complaints regarding mental health trusts/service that are referred back to trust Annual survey of patient experience % of Healthcare Commission survey respondents that rated the care they had received as either excellent, very good or good % of Healthcare Commission survey respondents that had received at least one care review % of Healthcare Commission survey respondents that had an out-of-hours contact telephone number
Developing mental health indicators in England
901
Table 1 (Continued ) Workforce Capacity Clinical psychology staff Total whole-time-equivalent clinical psychology staff, rate per 100,000 population aged 18e64 years Clinical psychology staff by grade Medical staff Total whole-time-equivalent medical staff, rate per 100,000 population aged 18e64 years Medical staff by grade, 2005 rate per 100,000 population aged 18e64 years Psychiatric nurses Total whole-time-equivalent psychiatric nursing staff, rate per 100,000 population aged 18e64 years Psychiatric nursing staff by grade Occupational, art, music and drama therapists Total whole-time-equivalent occupational and creative therapy staff, rate per 100,000 population aged 18e64 years Occupational and creative therapy staff by therapy staff group Counsellors and psychotherapists Total whole-time-equivalent counselling and psychotherapy staff, rate per 100,000 population aged 18e64 years Counselling and psychotherapy staff by staff group Graduate workers % of LITs with graduate workers Gateway workers % of LITs with gateway workers Community development workers for Black and minority ethnic communities Whole-time-equivalent Black and minority ethnic community development workers, rate per million population aged 18e64 years Whole-time-equivalent Black and minority ethnic community development workers, rate per million Black and minority ethnic population aged 18e64 years Support time and recovery staff Whole-time-equivalent support time and recovery staff, rate per 100,000 population aged 18e64 years
was published, and an indicator on Black and minority ethnic community development workers was also included.
Drugs and alcohol Only one indicator was published in relation to drugs and alcohol as this is the subject of a future report.
Displaying the data Bar charts were used to display regional indicator values, and most include 95% confidence intervals to identify whether regional values differ significantly from the national average. Traffic light colours were used when making a judgement about the performance of a region against the national average, based on whether and how the regional value differed significantly from the national average. Alternative colours were used when it was felt that a judgement about desirability could not be made. These indicated whether the regional value was significantly low, consistent or significantly high compared with the national average.
National target or recommended levels, where these have been clearly stated, were highlighted using a vertical red line. Stacked bar charts and graphs in different colours were also used to illustrate comparisons. Each subsection of the report includes: the regional graph(s); a brief description of each indicator; the rationale for including the indicator(s) and background information, including reference to national targets, monitoring requirements and recommendations; and a summary of the main findings and a regional commentary.
Results Many of the risk factors for mental illness are linked to deprivation, so a general pattern was seen with the three northern regions (North East, North West and Yorkshire and Humber) showing worse measures than the three southern regions (South East, South West and Eastern England) and
902
J. Wilkinson et al.
the two midlands regions (West Midlands, East Midlands). London has a very inconsistent pattern, appearing at different places on different indicators. In contrast, the service-based indicators do not often show this pattern, with the northern regions often doing better. The findings fall into three main categories: where there appear to be important differences between the regions, e.g. alcohol consumption; where regional differences appear to be unimportant or insignificant, e.g. expenditure on gambling; and
issues where there is an absence of data, e.g. accident and emergency attendances for selfharm in recent years, or the data are too poor to interpret, e.g. ethnic coding. An example of one indicator is shown in Figure 1, and a section of the summary table is shown in Figure 2.
Discussion There are areas where there is scope for tackling the determinants of poor mental health (e.g. drugs, alcohol, physical activity). The report also identified
North East North West Yorkshire & Humber East Midlands West Midlands East of England London South East South West
England 0
20
40
60
80
100
120
Directly age-standardised finished consultant episode rate
Figure 1 Illustration of one indicator. Directly age-standardized hospital admission rates for schizophrenia per 100,000 population aged 15e74 years, 2001e2002. Indicator description: Hospital admissions for schizophrenia (primary diagnosis ICD10 codes F20, F21, F23.2, F25) per 100,000 population aged 15e74 years. Data are taken from the Clinical and Health Outcomes Knowledge Base (http:// www.nchod.nhs.uk/). Rationale and background: Hospital episode statistics are based on consultant episodes (a period of care under one consultant within one provider) and not admissions, hence some overcounting may occur. There may be variation in completeness of hospital records, accuracy of diagnoses, and quality of coding. The data may also reflect variations in symptomatology and diagnostic criteria. Data from the independent sector is not included, which may be significant for this condition. Another artefact to be aware of is National Health Service trust reconfiguration, which can artificially create a large number of admissions from one trust to its successor. Schizophrenia is one of a number of psychotic illnesses and there may be some diagnostic variation across the country. There has probably been less discretion during a severe episode of schizophrenia as to whether a person would be admitted than with conditions such as neuroses. However, the development of a range of new services such as assertive outreach, crisis resolution, home treatment and early intervention teams, as well as the quality and quantity of community-based services, will have an impact on the admission rate. Commentary: A NortheSouth divide is less apparent compared with neuroses. Possibly there are higher rates in more urbanized regions. There still seems to be about a two-fold variation across the country. Comparison with the indicator for neuroses admissions and total number of available beds suggests that there is not a simple explanation of more beds available reducing admission threshold or a substitution of one type of admission for another.
903
East of England
London
South East
South West
Local authority data available
54.7
53.5
53.8
51.2
52.5
54.2
56.6
55.2
57.5
56.0
Y
Percentage of working age adults in job-related training in the past 13 weeks (working and unemployed), 2004/05
25.6
26.6
24.5
25.4
26.0
24.3
24.8
25.7
27.4
26.1
Y
Percentage of working age adults currently working towards or studying for any qualifications, 2004/05
14.6
14.7
13.7
15.2
14.5
14.6
13.1
16.0
14.3
15.1
Y
Percentage of the population with a 'high level of worry about violent crime', 2004/05
16.6
14.6
18.3
16.0
16.0
17.0
14.3
25.5
13.2
10.4
N
Total incidence of violence per 10,000 adults reported by people in the British Crime Survey, 2004/05
563.0
473.8
613.2
507.1
530.6
654.8
4 70.9
556.5
611.5
559.2
N
All violent reported crime recorded by the police forces per 10,000 population, 2004/05
225.5
183.1
239.6
225.0
212.4
226.4
176.0
342.3
183.1
190.2
N
Percentage of records in Mental Health Minimum Dataset which include valid and usable ethnic group code, 2004/05
59.3
77.7
78.6
60.2
62.9
66.7
70.8
66.9
36.3
47.2
N
Percentage of adults with a GHQ12 score of 4 or more, 2004
13.2
17.5
14.2
12.1
12.9
12.8
11.2
14.8
13.0
11.5
N
Mortality rates (age-stand) for suicide and injury undetermined (persons aged 15 years and older), 2003-05
10.8
13.1
11.9
11.3
10.7
10.4
10.0
10.4
10.4
11.0
Y
Mortality rates (age-stand) for suicide and injury undetermined (males aged 15 years and older), 2003-05
16.4
20.4
18.5
17.3
16.4
15.8
15.0
15.4
15.7
16.9
N
Mortality rates (age-stand) for suicide and injury undetermined (females aged 15 years and older), 2003-05
5.4
6.4
5.6
5.6
5.2
5.2
5.1
5.7
5.3
5.4
N
Self harm hospital admission rates (age-stand) per 100,000 population, 2003
142.4
251.2
175.6
206.7
141.5
153.4
112.8
71.1
123.9
151.1
Y
Self harm hospital admission rates (age-stand) per 100,000 population, 2004
158.8
273.6
199.7
212.9
172.5
152.5
124.5
79.5
140.8
189.7
Y
Hospital admission rates (age-stand) for poisoning, rates per 100,000 population, 2003
165.3
283.3
208.9
221.4
160.7
197.9
123.7
95.5
145.3
157.8
Y
North East
England
East Midlands
Percentage of pupils aged 15 years in local authority schools achieving 5 or more grade A*-C GCSEs or equivalent, 2004/05
Indicator
North West
West Midlands
Yorkshire & Humber
Developing mental health indicators in England
RISK AND PROTECTIVE FACTORS AND DETERMINANTS (continued)
POPULATION HEALTH STATUS
Figure 2
‘Traffic light’ indicators.
important variations in the provision of mental health services between regions in relation to need. There is a need to undertake more work at a subnational level on mental health data. Work needs to be undertaken to develop suitable mental health indicators that can be used as part of the Local Area Agreement process. Initially, subregional analysis needs to be undertaken in order to gain better understanding of the local position.
There is an urgent need to develop new indicators of mental health to enable comparison of European countries, also paying attention to the possibility of comparisons at a subnational level. At the present time, mortality data provide one of the few ways of making any informed comparisons; suicide is commonly used, but very different approaches are used in European countries, thus making comparisons difficult.16
904 Some important differences were identified in the report that are highlighted between regions. However, some of the usual patterns that are often seen when making health comparisons are very different in mental health. Readers of the report will be struck by the range and nature of the indicators used; this reflects the complexity of the subject and also the wide range of determinants of good mental health. However, a major gap in this report is the absence of data on mental health from the criminal justice sector. There is a need to address this as a matter of urgency. The North East Public Health Observatory (NEPHO) has suggested that one way to fill this gap is through the creation of a National Observatory for Offender Health. Recent initiatives, such as the one in England entitled ‘New Ways of Working’,17 propose a number of new models for the delivery of mental health services in the future. New indicators will need to be considered to monitor these objectives. However, the MDMDS in England already collects data on patient activity from specialist mental health services and not just on inpatient treatment activity. The report was aimed at regional directors of public health, strategic health authorities, directors of public health in PCTs and staff in care services improvement partnerships. It is hoped that directors will find the report a valuable resource in making decisions, and in holding to account those responsible for the delivery of mental healthcare services and improving the mental health of the population. Early feedback following the publication of the report suggests that some of these hopes may be being realized.
Acknowledgements The authors would like to thank the following for their contributions to this report: Susan Panrucker, Business Manager, North East Public Health Observatory; Megan Newark, Health Information Analyst, North East Public Health Observatory; Susan Walrond, former Senior Information Manager, North East Public Health Observatory; Paula Whitty, Clinical Effectiveness Lead (Better Metrics), Healthcare Commission; Andrew McCulloch, Chief Executive, Mental Health Foundation; Jane Parkinson, NHS Health Scotland; and all staff who contributed from within the Healthcare Commission and the Prescription Pricing Authority.
Ethical approval None sought
J. Wilkinson et al.
Funding The North East Public Health Observatory is funded by the Department of Health. The Mental Health Observatory is funded by the National Institute of Mental Health (NIMHE) and NHS R&D.
Competing interests None declared.
References 1. Department of Health. Choosing health e making healthier choices easier. Cm 6374. London: Department of Health. Available from: URL: http://www.dh.gov.uk/en/Publicationsandsta-tistics/Publications/PublicationsPolicyAndGuidance/DH_4094550; 2004 [accessed 28.07.07]. 2. Wilkinson JR, Glover G, Chappel D, Bywaters J. Indications of public health in the English regions: mental health. York: Association of Public Health Observatories Available from: http://www.apho.org.uk/apho/publications/mental health.pdf; 2007 [accessed 28.07.07]. 3. Department of Health. National service framework for mental health. Modern standards and service models. London: Department of Health. Available from: http://www.dh.gov. uk/en/Publicationsandstatistics/Publications/Publications PolicyAndGuidance/DH_4009598; 1999 [accessed 28.07.07]. 4. Department of Health. National suicide prevention strategy for England. London: Department of Health. Available from: http://www.dh.gov.uk/en/Publicationsandstatistics/Publications/PublicationsPolicyAndGuidance/DH_4009474; 2002 [accessed 14.12.07]. 5. Healthcare Commission. Better metrics. London: Healthcare Commission. Available from: www.healthcarecommission.org. uk/_db/_documents/Urgent_care_metrics_master11Dec06. pdf; 2006 [accessed 16.07.07]. 6. Healthcare Commission. The better metrics project. London: Healthcare Commission. Available from: http://www. healthcarecommission.org.uk/serviceproviderinformation/ bettermetrics.cfm; 2007 [accessed 01.01.07]. 7. London Health Observatory. Local basket of indicators. London: London Health Observatory. Available from: http:// www.lho.org.uk/HEALTH_INEQUALITIES/Basket_Of_Indica tors/BasketOfIndicators.aspx; 2004 [accessed 7.07.07]. 8. Glover G. Data quality in the mental health minimum dataset. Stockton upon Tees: North East Public Health and Observatory Available from: http://www.nepho.org. uk/index.php?c ¼ 1372; 2006 [accessed 28.07.07]. 9. NHS Information Centre. Quality and outcomes framework (QOF). Leeds: NHS Information Centre. Available from: http:// www.ic.nhs.uk/services/qof; 2007 [accessed 7.07.07]. 10. Health and Social Care Information Centre. Health survey for England 2004. Leeds: Health and Social Care Information Centre. Available from: http://www.dh.gov.uk/en/ Publicationsandstatistics/PublishedSurvey/HealthSurvey ForEngland/index.htm; 2005 [accessed 28.07.07]. 11. Office for National Statistics. NOMIS e official labour market statistics. Durham: Durham University. Available from: http://www.nomisweb.co.uk/; 2007 [accessed 7.07.07]. 12. Glover G, Dean R. Durham mental health mapping service. Durham: Durham University. Available from: http://www.dur. ac.uk/service.mapping/amh/; 2007 [accessed 28.07.07].
Developing mental health indicators in England 13. Mental Health Act Commission, Care Services Improvement Partnership, National Institute of Mental Health in England. Count me in. Results of the 2006 national census of inpatient mental health and learning disability services in England and Wales. London: Healthcare Commission. Available from: http://www.healthcarecommission.org.uk/_db/_documents/ Count_Me_In_2006.pdf; 2006 [accessed 16.07.07]. 14. National Centre for Social Research. Ethnic minority psychiatric illness rates in the community (EMPIRIC) e quantitative report. London: The Stationery Office; 2002. 15. Fitzpatrick J, Jacobson B, Aspinal P. Indications of public health in the English regions: ethnicity and health. Stockton on Tees: Association of Public Health Observatories.
905 Available from: http://www.apho.org.uk/apho/publications/Ethnicity.pdf; 2005 [accessed 28.07.07]. 16. Corcoran P. National programme for suicide prevention: can they only rely on suicide statistics? Psychiatr Danub 2006; 18:127. The 14th European Symposium on Suicide and Suicidal Behaviour: from Greenwich to the Caucasus, from the Urals to Iberia. 17. Royal College of Psychaitrists, National Institute of Mental Health in England. New ways of working for psychiatrists: enhancing effective, person-centred services through new ways of working in multidisciplinary and multi-agency contexts. London: Department of Health; 2005.
Available online at www.sciencedirect.com
www.elsevierhealth.com/journals/pubh
Original Research
Developing mental health indicators in England John Wilkinson*, Jenny Bywaters, Stuart Simms, David Chappel, Gyles Glover North East Public Health Observatory and the National Mental Health Observatory, University of Durham Queen’s Campus, Wolfson Research Institute, Stockton on Tees TS17 6BH, UK Received 28 July 2007; received in revised form 24 September 2007; accepted 26 October 2007 Available online 3 March 2008
KEYWORDS Mental health; Indicators; Regions
Summary Objectives: This paper reports on the construction of a set of indicators for mental health and the publication of a report for England’s Chief Medical Officer. The report was the seventh in a series of reports aimed at initiating public health action to improve health at a regional level in England. Study design: Observational study using routine data. Methods: A set of over 80 indicators was constructed by an expert group. These indicators were then populated from routine datasets. Commentary was provided on each indicator in the report. Results: A small team compiled this large set of indicators from routine data working in a public health observatory in one region of England. Conclusions: It is possible to produce a large array of indicators about mental health. The conclusion from examining these indicators is that there are many unexplained differences in mental health across the regions of England. Many of these indicators are closely linked to deprivation. Some indicators show a lack of difference across the country, and in many instances, service provision is inconsistent and does not always relate to need. In some cases, there was a worrying absence of data, e.g. data on ethnicity and the mental health of offenders. ª 2007 The Royal Institute of Public Health. Published by Elsevier Ltd. All rights reserved.
Introduction In 2004, the Chief Medical Officer for England commissioned the UK Association of Public Health Observatories to produce a series of reports highlighting issues of public health importance,
with the aim of highlighting regional differences and stimulating and empowering public health practitioners (working at these levels) to take action across a range of sectors to improve health. This paper will highlight the issues in relation to the seventh report in the ‘Indications’ series, on
* Corresponding author. Tel.: þ44 191 334 0400; fax: þ44 191 334 0391. E-mail address: [email protected] (J. Wilkinson). 0033-3506/$ - see front matter ª 2007 The Royal Institute of Public Health. Published by Elsevier Ltd. All rights reserved. doi:10.1016/j.puhe.2007.10.012
898 mental health. The paper will describe the selection of indicators, list them, and also discuss presentational issues. This report focused on adults (aged 18e64 years). Indicators on mental health in older people will be included in the next planned report in the series on older people. A planned report on adolescent mental health is subject to the priorities agreed with the commissioners.
Background Mental health is identified as one of the six national priorities for action in the White Paper, ‘Choosing Health’.1 In the past, indicators of mental health have been difficult to find. Targets for mental health improvement have largely concentrated on suicide rates, which, although important, give a limited picture of the mental health of a community. This paper reports on the production of the UK Association of Public Health Observatories’ report on mental health.2
Policy context One in six people in England suffer from a mental health disorder, equating to around eight million people being affected in England today. The National Service Framework (NSF) for Mental Health,3 together with the National Suicide Prevention Strategy,4 provide the key policy context within which the report considers the mental health of adults of working age. Many of the data for the report are drawn from information sources such as the annual service mapping and financial mapping which were commissioned by the Department of Health so that progress with NSF implementation could be monitored in detail. The report aimed to inform the implementation of national and regional priorities. The main audiences include national policy makers, regional directors of public health, strategic health authority and primary care trust (PCT) chief executives and boards, and leaders in other health and social care organizations.
Methods A small steering group was established to select the indicators. The membership of this group included representatives of service providers, the voluntary sector, the Department of Health and the public health observatory. The choice of indicators was, in part, guided by key issues, recommendations and targets identified in a range
J. Wilkinson et al. of policies including the NSF for Mental Health3 and ‘Choosing Health’.1 This provided an opportunity to bring together a number of indicators that have been collated by different projects over the past few years (such as ‘Mental Health Outcomes’,5 ‘Better Metrics’6 and the ‘Local Basket of Inequalities Indicators’7), and to develop indicators utilizing new data sources that have become available recently [such as the Mental Health Minimum Data Set (MHMDS)8 and new data from primary care9] or have been underutilized in this area (such as from the Health Survey for England10 and employment information11). A full list of the indicators selected is shown in Table 1. The report presents a wide range of data on the factors that can give rise to poor mental health, the mental health status of populations, provision of interventions of care for mental illness, service user experience, and traditional outcomes such as suicide. The authors had the advantage of access to the Durham Mental Health Mapping Service, which provided a systematic overview of mental health services in England until 2006,12 and were also able to make use of the MHMDS8 which was developed by one of the authors of the report (GG). The report is structured in a similar way to previous reports and looks at indicators of:
risk and protective factors and determinants; population health status; interventions; effectiveness of partnerships; service user experience; and workforce capacity.
Ethnicity Ethnicity is an important issue in mental health. A recent service user census13 found considerable variation in the number of people in hospital from different ethnic groups. There may be variations in underlying morbidity, diagnosis and management that need further investigation. There are also good data on ethnic mental health from the EMPIRIC study.14 However, for this report, which compares English regions, the data were not broken down further by ethnicity. While there is some interest in whether a particular ethnic group has different experiences in different parts of England, the fourth ‘Indications’ report15 showed that the data are not robust enough to make these comparisons. Ethnic variations will continue to be best analysed at national level at the moment. A comparison of ethnic coding quality was included within the MHMDS, as this was not possible when the fourth ‘Indications’ report15
Developing mental health indicators in England Table 1
899
List of indicators contained in the report.
Risk and Protective Factors and Determinants Deprivation % of super output areas in each region, by quintile of deprivation Mental health needs Mental health ‘needs’ indices by region employment e % of people of working age in employment Employment of people with mental health problems % of adults of working age with a mental health problem in employment Incapacity benefits Mental and behavioural disorders incapacity benefit claimant rate per 100,000 population aged 16e59 years Limiting long-term illness % of adults with a limiting long-term illness Alcohol Standardized % of alcohol consumption above ‘sensible’ daily limits, 2005 Drugs People in contact with structured drug treatment, crude rate per 100,000 Physical activity % of adults (aged 16 years and over) participating in 20 or more days of moderate intensity sport and active recreation (averaging five or more times per week) in the last 4 weeks Healthy eating % of adults who consume five or more portions of fruit and vegetables per day Social capital e participation in civic society % of adults participating in any civic activity in the previous 12 months Religion e % of population reporting having a religion, 2001 Social support e % of adults who could get a lift somewhere if needed Social networks e % of adults speaking to relatives by phone Neighbourliness e % of adults speaking to neighbours Education % of pupils aged 15 years in schools maintained by the local authority achieving five or more grade A)eC GCSEs or equivalent Learning and development % of working age adults in job-related training in the past 13 weeks (working and unemployed) Violence and safety % of the population with a ‘high level of worry about violent crime’ Total incidence of violence per 10,000 adults reported by people in the British Crime Survey All violent reported crime recorded by the police forces per 10,000 population Gambling Household expenditure on gambling payments, 2002e2005 Ethnic coding % of records in Mental Health Minimum Dataset (MHMDS) which include a valid and usable ethnic group code Population Health Status Psychiatric disorder % of adults with a GHQ12 score of 4 or more Psychiatric morbidity % of population scoring 18 or more on the Clinical Interview Schedule Usage of specialist mental health services People aged 18e64 years documented as receiving care in the MHMDS, rate per 100,000 population Residential and nursing home care Council supported residents aged 18e64 years with mental health problems, rate per 100,000 population Severe mental illness on general practitioners’ register % of service users registered with a general practitioner who have severe long-term mental health problems Suicide Directly age-standardized mortality rates for suicide and injury undetermined, persons aged 15 years and older Directly age-standardized mortality rates for suicide and injury undetermined, males 15 years and older Directly age-standardized mortality rates for suicide and injury undetermined, females 15 years and older (continued on next page)
900
J. Wilkinson et al.
Table 1 (Continued) Self-harm admissions Directly age-standardized Directly age-standardized Directly age-standardized Directly age-standardized
self harm hospital admission rate per 100,000 population self harm hospital admission rate per 100,000 population hospital admission rate per 100,000 population for poisoning hospital admission rate per 100,000 population for poisoning
Interventions Mental health promotion % of local implementation teams (LITs) self-assessed as ‘GREEN’ for having a mental health promotion lead officer % of LITs self-assessed as ‘GREEN’ for having a mental health promotion strategy and action plan Assertive outreach Rate per 100,000 population aged 18e64 years reported by primary care organizations to be currently receiving mental health assertive outreach care Crisis resolution Rate per 100,000 population aged 18e64 years reported by primary care organizations to be currently receiving crisis resolution service care Early intervention for psychosis Rate per 100,000 population aged 15e34 years reported by primary care organizations to be receiving psychosis early intervention care Day care % of LITs with National Health Service (NHS) day hospitals Whole-time-equivalent care staff in NHS day care facilities, rate per 100,000 population aged 18e64 years % of LITs with at least one day centre or resource centre, 2005e2006 Rate of drop-in facilities per 100,000 population aged 18e64 years Admissions for depression and anxiety Directly age-standardized hospital admission rates for depression per 100,000 population aged 15e74 years Directly age-standardized hospital admission rates for anxiety disorders per 100,000 population aged 15e74 years, 2001e2002 Admissions for schizophrenia Directly age-standardized hospital admission rates for schizophrenia per 100,000 population aged 15e74 years Prescribing in primary care Average daily quantity (ADQ) of antidepressants prescribed by region ADQ of antipsychotics prescribed by region ADQ of hypnotics and anxiolytics prescribed by region Suicide audits % of primary care trusts with completed suicide audit Administration of electroconvulsive therapy (ECT) Individuals administered ECT, rate per 100,000 population aged 16e64 years Mental Health Act Commission Service users detained under the Mental Health Act 1983 resident in NHS facilities and independent hospitals, rate per 100,000 population >18 years Effectiveness of Partnerships Mental health expenditure e non-statutory % in total planned investment in adult mental health services by main provider type Multidisciplinary community mental health teams % of community mental health teams reported as achieving full local integration between NHS and social services partners Service User Experience Complaints % of complaints regarding mental health trusts/service that are referred back to trust Annual survey of patient experience % of Healthcare Commission survey respondents that rated the care they had received as either excellent, very good or good % of Healthcare Commission survey respondents that had received at least one care review % of Healthcare Commission survey respondents that had an out-of-hours contact telephone number
Developing mental health indicators in England
901
Table 1 (Continued ) Workforce Capacity Clinical psychology staff Total whole-time-equivalent clinical psychology staff, rate per 100,000 population aged 18e64 years Clinical psychology staff by grade Medical staff Total whole-time-equivalent medical staff, rate per 100,000 population aged 18e64 years Medical staff by grade, 2005 rate per 100,000 population aged 18e64 years Psychiatric nurses Total whole-time-equivalent psychiatric nursing staff, rate per 100,000 population aged 18e64 years Psychiatric nursing staff by grade Occupational, art, music and drama therapists Total whole-time-equivalent occupational and creative therapy staff, rate per 100,000 population aged 18e64 years Occupational and creative therapy staff by therapy staff group Counsellors and psychotherapists Total whole-time-equivalent counselling and psychotherapy staff, rate per 100,000 population aged 18e64 years Counselling and psychotherapy staff by staff group Graduate workers % of LITs with graduate workers Gateway workers % of LITs with gateway workers Community development workers for Black and minority ethnic communities Whole-time-equivalent Black and minority ethnic community development workers, rate per million population aged 18e64 years Whole-time-equivalent Black and minority ethnic community development workers, rate per million Black and minority ethnic population aged 18e64 years Support time and recovery staff Whole-time-equivalent support time and recovery staff, rate per 100,000 population aged 18e64 years
was published, and an indicator on Black and minority ethnic community development workers was also included.
Drugs and alcohol Only one indicator was published in relation to drugs and alcohol as this is the subject of a future report.
Displaying the data Bar charts were used to display regional indicator values, and most include 95% confidence intervals to identify whether regional values differ significantly from the national average. Traffic light colours were used when making a judgement about the performance of a region against the national average, based on whether and how the regional value differed significantly from the national average. Alternative colours were used when it was felt that a judgement about desirability could not be made. These indicated whether the regional value was significantly low, consistent or significantly high compared with the national average.
National target or recommended levels, where these have been clearly stated, were highlighted using a vertical red line. Stacked bar charts and graphs in different colours were also used to illustrate comparisons. Each subsection of the report includes: the regional graph(s); a brief description of each indicator; the rationale for including the indicator(s) and background information, including reference to national targets, monitoring requirements and recommendations; and a summary of the main findings and a regional commentary.
Results Many of the risk factors for mental illness are linked to deprivation, so a general pattern was seen with the three northern regions (North East, North West and Yorkshire and Humber) showing worse measures than the three southern regions (South East, South West and Eastern England) and
902
J. Wilkinson et al.
the two midlands regions (West Midlands, East Midlands). London has a very inconsistent pattern, appearing at different places on different indicators. In contrast, the service-based indicators do not often show this pattern, with the northern regions often doing better. The findings fall into three main categories: where there appear to be important differences between the regions, e.g. alcohol consumption; where regional differences appear to be unimportant or insignificant, e.g. expenditure on gambling; and
issues where there is an absence of data, e.g. accident and emergency attendances for selfharm in recent years, or the data are too poor to interpret, e.g. ethnic coding. An example of one indicator is shown in Figure 1, and a section of the summary table is shown in Figure 2.
Discussion There are areas where there is scope for tackling the determinants of poor mental health (e.g. drugs, alcohol, physical activity). The report also identified
North East North West Yorkshire & Humber East Midlands West Midlands East of England London South East South West
England 0
20
40
60
80
100
120
Directly age-standardised finished consultant episode rate
Figure 1 Illustration of one indicator. Directly age-standardized hospital admission rates for schizophrenia per 100,000 population aged 15e74 years, 2001e2002. Indicator description: Hospital admissions for schizophrenia (primary diagnosis ICD10 codes F20, F21, F23.2, F25) per 100,000 population aged 15e74 years. Data are taken from the Clinical and Health Outcomes Knowledge Base (http:// www.nchod.nhs.uk/). Rationale and background: Hospital episode statistics are based on consultant episodes (a period of care under one consultant within one provider) and not admissions, hence some overcounting may occur. There may be variation in completeness of hospital records, accuracy of diagnoses, and quality of coding. The data may also reflect variations in symptomatology and diagnostic criteria. Data from the independent sector is not included, which may be significant for this condition. Another artefact to be aware of is National Health Service trust reconfiguration, which can artificially create a large number of admissions from one trust to its successor. Schizophrenia is one of a number of psychotic illnesses and there may be some diagnostic variation across the country. There has probably been less discretion during a severe episode of schizophrenia as to whether a person would be admitted than with conditions such as neuroses. However, the development of a range of new services such as assertive outreach, crisis resolution, home treatment and early intervention teams, as well as the quality and quantity of community-based services, will have an impact on the admission rate. Commentary: A NortheSouth divide is less apparent compared with neuroses. Possibly there are higher rates in more urbanized regions. There still seems to be about a two-fold variation across the country. Comparison with the indicator for neuroses admissions and total number of available beds suggests that there is not a simple explanation of more beds available reducing admission threshold or a substitution of one type of admission for another.
903
East of England
London
South East
South West
Local authority data available
54.7
53.5
53.8
51.2
52.5
54.2
56.6
55.2
57.5
56.0
Y
Percentage of working age adults in job-related training in the past 13 weeks (working and unemployed), 2004/05
25.6
26.6
24.5
25.4
26.0
24.3
24.8
25.7
27.4
26.1
Y
Percentage of working age adults currently working towards or studying for any qualifications, 2004/05
14.6
14.7
13.7
15.2
14.5
14.6
13.1
16.0
14.3
15.1
Y
Percentage of the population with a 'high level of worry about violent crime', 2004/05
16.6
14.6
18.3
16.0
16.0
17.0
14.3
25.5
13.2
10.4
N
Total incidence of violence per 10,000 adults reported by people in the British Crime Survey, 2004/05
563.0
473.8
613.2
507.1
530.6
654.8
4 70.9
556.5
611.5
559.2
N
All violent reported crime recorded by the police forces per 10,000 population, 2004/05
225.5
183.1
239.6
225.0
212.4
226.4
176.0
342.3
183.1
190.2
N
Percentage of records in Mental Health Minimum Dataset which include valid and usable ethnic group code, 2004/05
59.3
77.7
78.6
60.2
62.9
66.7
70.8
66.9
36.3
47.2
N
Percentage of adults with a GHQ12 score of 4 or more, 2004
13.2
17.5
14.2
12.1
12.9
12.8
11.2
14.8
13.0
11.5
N
Mortality rates (age-stand) for suicide and injury undetermined (persons aged 15 years and older), 2003-05
10.8
13.1
11.9
11.3
10.7
10.4
10.0
10.4
10.4
11.0
Y
Mortality rates (age-stand) for suicide and injury undetermined (males aged 15 years and older), 2003-05
16.4
20.4
18.5
17.3
16.4
15.8
15.0
15.4
15.7
16.9
N
Mortality rates (age-stand) for suicide and injury undetermined (females aged 15 years and older), 2003-05
5.4
6.4
5.6
5.6
5.2
5.2
5.1
5.7
5.3
5.4
N
Self harm hospital admission rates (age-stand) per 100,000 population, 2003
142.4
251.2
175.6
206.7
141.5
153.4
112.8
71.1
123.9
151.1
Y
Self harm hospital admission rates (age-stand) per 100,000 population, 2004
158.8
273.6
199.7
212.9
172.5
152.5
124.5
79.5
140.8
189.7
Y
Hospital admission rates (age-stand) for poisoning, rates per 100,000 population, 2003
165.3
283.3
208.9
221.4
160.7
197.9
123.7
95.5
145.3
157.8
Y
North East
England
East Midlands
Percentage of pupils aged 15 years in local authority schools achieving 5 or more grade A*-C GCSEs or equivalent, 2004/05
Indicator
North West
West Midlands
Yorkshire & Humber
Developing mental health indicators in England
RISK AND PROTECTIVE FACTORS AND DETERMINANTS (continued)
POPULATION HEALTH STATUS
Figure 2
‘Traffic light’ indicators.
important variations in the provision of mental health services between regions in relation to need. There is a need to undertake more work at a subnational level on mental health data. Work needs to be undertaken to develop suitable mental health indicators that can be used as part of the Local Area Agreement process. Initially, subregional analysis needs to be undertaken in order to gain better understanding of the local position.
There is an urgent need to develop new indicators of mental health to enable comparison of European countries, also paying attention to the possibility of comparisons at a subnational level. At the present time, mortality data provide one of the few ways of making any informed comparisons; suicide is commonly used, but very different approaches are used in European countries, thus making comparisons difficult.16
904 Some important differences were identified in the report that are highlighted between regions. However, some of the usual patterns that are often seen when making health comparisons are very different in mental health. Readers of the report will be struck by the range and nature of the indicators used; this reflects the complexity of the subject and also the wide range of determinants of good mental health. However, a major gap in this report is the absence of data on mental health from the criminal justice sector. There is a need to address this as a matter of urgency. The North East Public Health Observatory (NEPHO) has suggested that one way to fill this gap is through the creation of a National Observatory for Offender Health. Recent initiatives, such as the one in England entitled ‘New Ways of Working’,17 propose a number of new models for the delivery of mental health services in the future. New indicators will need to be considered to monitor these objectives. However, the MDMDS in England already collects data on patient activity from specialist mental health services and not just on inpatient treatment activity. The report was aimed at regional directors of public health, strategic health authorities, directors of public health in PCTs and staff in care services improvement partnerships. It is hoped that directors will find the report a valuable resource in making decisions, and in holding to account those responsible for the delivery of mental healthcare services and improving the mental health of the population. Early feedback following the publication of the report suggests that some of these hopes may be being realized.
Acknowledgements The authors would like to thank the following for their contributions to this report: Susan Panrucker, Business Manager, North East Public Health Observatory; Megan Newark, Health Information Analyst, North East Public Health Observatory; Susan Walrond, former Senior Information Manager, North East Public Health Observatory; Paula Whitty, Clinical Effectiveness Lead (Better Metrics), Healthcare Commission; Andrew McCulloch, Chief Executive, Mental Health Foundation; Jane Parkinson, NHS Health Scotland; and all staff who contributed from within the Healthcare Commission and the Prescription Pricing Authority.
Ethical approval None sought
J. Wilkinson et al.
Funding The North East Public Health Observatory is funded by the Department of Health. The Mental Health Observatory is funded by the National Institute of Mental Health (NIMHE) and NHS R&D.
Competing interests None declared.
References 1. Department of Health. Choosing health e making healthier choices easier. Cm 6374. London: Department of Health. Available from: URL: http://www.dh.gov.uk/en/Publicationsandsta-tistics/Publications/PublicationsPolicyAndGuidance/DH_4094550; 2004 [accessed 28.07.07]. 2. Wilkinson JR, Glover G, Chappel D, Bywaters J. Indications of public health in the English regions: mental health. York: Association of Public Health Observatories Available from: http://www.apho.org.uk/apho/publications/mental health.pdf; 2007 [accessed 28.07.07]. 3. Department of Health. National service framework for mental health. Modern standards and service models. London: Department of Health. Available from: http://www.dh.gov. uk/en/Publicationsandstatistics/Publications/Publications PolicyAndGuidance/DH_4009598; 1999 [accessed 28.07.07]. 4. Department of Health. National suicide prevention strategy for England. London: Department of Health. Available from: http://www.dh.gov.uk/en/Publicationsandstatistics/Publications/PublicationsPolicyAndGuidance/DH_4009474; 2002 [accessed 14.12.07]. 5. Healthcare Commission. Better metrics. London: Healthcare Commission. Available from: www.healthcarecommission.org. uk/_db/_documents/Urgent_care_metrics_master11Dec06. pdf; 2006 [accessed 16.07.07]. 6. Healthcare Commission. The better metrics project. London: Healthcare Commission. Available from: http://www. healthcarecommission.org.uk/serviceproviderinformation/ bettermetrics.cfm; 2007 [accessed 01.01.07]. 7. London Health Observatory. Local basket of indicators. London: London Health Observatory. Available from: http:// www.lho.org.uk/HEALTH_INEQUALITIES/Basket_Of_Indica tors/BasketOfIndicators.aspx; 2004 [accessed 7.07.07]. 8. Glover G. Data quality in the mental health minimum dataset. Stockton upon Tees: North East Public Health and Observatory Available from: http://www.nepho.org. uk/index.php?c ¼ 1372; 2006 [accessed 28.07.07]. 9. NHS Information Centre. Quality and outcomes framework (QOF). Leeds: NHS Information Centre. Available from: http:// www.ic.nhs.uk/services/qof; 2007 [accessed 7.07.07]. 10. Health and Social Care Information Centre. Health survey for England 2004. Leeds: Health and Social Care Information Centre. Available from: http://www.dh.gov.uk/en/ Publicationsandstatistics/PublishedSurvey/HealthSurvey ForEngland/index.htm; 2005 [accessed 28.07.07]. 11. Office for National Statistics. NOMIS e official labour market statistics. Durham: Durham University. Available from: http://www.nomisweb.co.uk/; 2007 [accessed 7.07.07]. 12. Glover G, Dean R. Durham mental health mapping service. Durham: Durham University. Available from: http://www.dur. ac.uk/service.mapping/amh/; 2007 [accessed 28.07.07].
Developing mental health indicators in England 13. Mental Health Act Commission, Care Services Improvement Partnership, National Institute of Mental Health in England. Count me in. Results of the 2006 national census of inpatient mental health and learning disability services in England and Wales. London: Healthcare Commission. Available from: http://www.healthcarecommission.org.uk/_db/_documents/ Count_Me_In_2006.pdf; 2006 [accessed 16.07.07]. 14. National Centre for Social Research. Ethnic minority psychiatric illness rates in the community (EMPIRIC) e quantitative report. London: The Stationery Office; 2002. 15. Fitzpatrick J, Jacobson B, Aspinal P. Indications of public health in the English regions: ethnicity and health. Stockton on Tees: Association of Public Health Observatories.
905 Available from: http://www.apho.org.uk/apho/publications/Ethnicity.pdf; 2005 [accessed 28.07.07]. 16. Corcoran P. National programme for suicide prevention: can they only rely on suicide statistics? Psychiatr Danub 2006; 18:127. The 14th European Symposium on Suicide and Suicidal Behaviour: from Greenwich to the Caucasus, from the Urals to Iberia. 17. Royal College of Psychaitrists, National Institute of Mental Health in England. New ways of working for psychiatrists: enhancing effective, person-centred services through new ways of working in multidisciplinary and multi-agency contexts. London: Department of Health; 2005.
Available online at www.sciencedirect.com