Hospital admissions for alcohol-related harm: Understanding the dataset

Transcription

1 Hospital admissions for alcohol-related harm: Understanding the dataset

2 Hospital admissions for alcohol-related harm: Understanding the dataset

3 DH INFORMATION READER BOX Policy HR/Workforce Management Planning Clinical Document purpose Gateway reference Title Author Commissioning IM & T Finance Partnership Working For information Publication date 14 July 2008 Target audience Circulation list Hospital admissions for alcohol-related harm: Understanding the dataset Department of Health and North West Public Health Observatory PCT CEs, SHA CEs, Directors of PH, NHS Directors of Commissioning Description Cross ref Superseded docs Action required Timing Contact details For recipient s use An explanatory note on a change in the way that alcohol-related hospital admissions are calculated Hospital admissions for alcohol-related harm: Technical Information and Definition for National Indicator Set NI39, Vital Signs Indicator VSC26 and Alcohol-attributable fractions for England alcohol attributable mortality and hospital admissions (NWPHO July 08) N/A N/A N/A Alcohol Policy Team Room 628 Wellington House London SE1 8UG Crown copyright 2008 First published 22 July 2008 Produced by COI for the Department of Health The text of this document may be reproduced without formal permission or charge for personal or in-house use.

4 Understanding the dataset Introduction The Department of Health is improving the way it estimates alcohol-related health harms and the way it tracks progress to address them in the NHS. We are doing this because primary care trusts and other commissioners need a better understanding of: the impact of harmful drinking on the health of their local population; the scale of alcohol-related demands on the NHS; the adequacy of their current level of provision; how to reduce the burden of alcohol-related harms on local NHS providers; and how to minimise alcohol-related harms in their local communities and their effects on the wider local economy. The key focus of our improvements has been to provide a much more comprehensive method for recognising, recording and tracking alcohol-related hospital admissions. This takes the form of a new national indicator for hospital admissions. The new indicator measures changes and tracks trends in hospital admissions, locally, regionally and nationally. In the past, hospital admissions data reported via the Alcohol Statistics Bulletin from the Information Centre for Health and Social Care. These typically covered just three principal blocks of data on alcohol-related admissions: for alcoholic liver disease, acute toxic effect of alcohol and alcohol-related mental health disorders. However, it became increasingly clear that these data significantly underestimate the extent to which patients with alcohol-related disease and injury are presenting for treatment within the NHS and generating hospital admissions. Following international best practice, data for the new indicator are now also generated using Alcohol Attributable Fractions (AAFs). These AAFs consider a range of diseases and injuries in which alcohol can play a part and estimate the proportion of cases that may be attributed to the consumption of alcohol. These improvements will encourage the earlier identification of people who drink too much. Taken together with advice and support from GPs or hospitals, they will provide the best way of reducing harmful drinking. Thus, they have the potential to provide the greatest health and economic benefits. 1

5 Hospital admissions for alcohol-related harm This document explains the difference between the new methodology and the old, and summarises what the data tell us about alcohol-related harm and trends in increasing harm in England. Understanding the new national indicator What is the new national indicator? The new national indicator provides local measures of the rate of hospital admissions for alcohol-related harm for every 100,000 members of the population (they have been standardised using the European age profile). They are derived from the Hospital Episode Statistics (HES) and cover the period 2002/03 to 2006/07. How will the data be different from what has been published in the past? The North West Public Health Observatory (NWPHO) has previously published information and statistics on alcohol-attributable hospital admissions in Local Alcohol Profiles for England (LAPE). 1 These profiles contain measures of alcoholrelated hospital admission presented in terms of persons (males and females separately) admitted i.e. ignoring repeat admissions within the same HES year. In addition, the National Community Health Profiles 2 contain measures of persons admitted to hospital for alcohol-specific conditions. Thus, neither of these sources of information compares directly with the NHS Vital Signs Indicator and Local Government Indicator VSC26/NI39 data that count admission episodes attributable to alcohol. However, they are, inevitably, very strongly correlated (Figure 1). 1 Available online at 2 Available online at 2

6 Understanding the dataset Figure 1: Comparison between statistics for VSC26/NI39 admission episodes and LAPE persons admitted related to alcohol for all 354 local authorities in England 1,800 1,600 1,400 Males admitted (alcohol attributable) LAPE 2008 Females admitted (alcohol attributable) LAPE 2008 Persons admitted (alcohol specific) HP 2008 R 2 = Persons admitted/100,000 1,200 1, R 2 = R 2 = ,100 1,300 1,500 1,700 1,900 2,100 2,300 2,500 NI39 admissions/100,000 What does the new indicator tell us about national trends in alcohol-related hospital admission now and over the last five years? There is almost double the number of men admitted compared to women, and all admissions have increased by an average of 15% per year to more than 800,000 in 2006/07 (Figure 2). Figure 2: Trend in the number of hospital admissions related to alcohol in England Number of admissions related to alcohol 600, , , , , ,000 0 Males Females 2002/ / / / /07 3

7 Hospital admissions for alcohol-related harm How do admissions vary regionally? Figure 3: Number of admissions by region 160, / / / / /07 140,000 Number of admissions 120, ,000 80,000 60,000 40,000 20,000 0 North East North West Yorkshire and The Humber East Midlands West Midlands East of England London South East South West As Figure 3 shows, the number of admissions is highest in the North West. All regions have shown increases over the last five years. Figure 4: Rate of admissions by region 2, / / / / /07 1,800 1,600 Admis s ions/100,000 1,400 1,200 1, North East North West Yorkshire and The Humber East Midlands West Midlands East of England London South East South West Figure 4 shows that the North West and North East have the highest overall rate of admission per 100,000 population: 1.7 times higher than the South East region. However, the rate of increase appears to be slowing in all regions, with the possible exception of London, which saw increases in rates in all years except 2006/07. 4

8 Understanding the dataset What are the main diseases that contribute to these figures? Analysis of the conditions associated with alcohol-related admissions can be done in two main ways: by looking at the primary alcohol-related diagnosis and by looking at the overall primary diagnosis (which may or may not be alcoholrelated). Both are examined below. Primary alcohol-related diagnosis As Table 1 shows, diseases of the circulatory system (I00 I99) account for more than half (54%) of primary alcohol-related diagnoses. Their incidence has increased by slightly more than the average for all types of alcohol-related admissions since 2002/03 (89% compared to 71%). Within chapter IX of the International Classification of Diseases (ICD) covering diseases of the circulatory system, the main alcohol-related conditions were hypertensive diseases (I10 I15) and cardiac arrhythmias (I47 I48) (38% and 15% of all admissions respectively). The former has increased by 107% since 2002/03 accounting for 47% of the increase in admissions over the period. Mental and behavioural disorders due to use of alcohol (F10) account for about one in six (18%) alcohol-related admissions. These figures have also increased by slightly more than the average since 2002/03 (80%). The other main ICD chapters to which alcohol contributes significantly are diseases of the nervous system (8%), external causes of morbidity and mortality (8%) and diseases of the digestive system (5%). 5

9 Hospital admissions for alcohol-related harm Table 1: Analysis of the conditions associated with alcohol-related admissions by primary alcohol-related diagnosis ICD Title 2006/07 % of 2002/03 % of Change % incr chapter 2006/ /03 (no.) from total total 2002/03 Total A00 B99 Certain infectious and parasitic diseases n/a C00 D48 Neoplasms D50 D89 Diseases of the blood and blood-forming n/a organs and certain disorders involving the immune mechanism E00 E90 Endocrine, nutritional and metabolic diseases F00 F99 Mental and behavioural disorders G00 G99 Diseases of the nervous system H00 H59 Diseases of the eye and adnexa n/a H60 H95 Diseases of the ear and mastoid process n/a I00 I99 Diseases of the circulatory system J00 J99 Diseases of the respiratory system n/a K00 K93 Diseases of the digestive system L00 L99 Diseases of the skin and subcutaneous tissue M00 M99 Diseases of the musculoskeletal system n/a and connective tissue N00 N99 Diseases of the genitourinary system n/a O00 O99 Pregnancy, childbirth and the puerperium P00 P96 Certain conditions originating in the n/a perinatal period Q00 Q99 Congenital malformations, deformations n/a and chromosomal abnormalities R00 R99 Symptoms, signs and abnormal clinical and n/a laboratory findings, not elsewhere classified S00 T98 Injury, poisoning and certain other consequences of external causes V01 Y98 External causes of morbidity and mortality Z00 Z99 Factors influencing health status and n/a contact with health services U00 U99 Codes for special purposes n/a Primary diagnosis As Table 2 shows, in terms of the primary diagnosis associated with alcohol-related admissions, the most prevalent chapter is injury, poisoning and certain other consequences of external causes (S00 T98) (18%), followed by diseases of the circulatory system (I00 I99) (17%), symptoms, signs and abnormal clinical and laboratory findings, not elsewhere classified (R00 R99) (12%) and diseases of the digestive system (K00 K93) (11%). Injuries and poisonings and diseases of the circulatory system have increased by less than the average for all alcohol-related admissions since 2002/03 (64% and 6

10 Understanding the dataset 47% respectively). Larger increases have been recorded for symptoms, signs and abnormal clinical and laboratory findings, not elsewhere classified (112%), diseases of the digestive system (79%), diseases of the musculoskeletal system and connective tissue (127%) and diseases of the genitourinary system (114%). Table 2: Analysis of the conditions associated with alcohol-related admissions by overall primary diagnosis ICD Title 2006/07 % of 2002/03 % of Change % incr chapter 2006/ /03 (no.) from total total 2002/03 Total A00 B99 Certain infectious and parasitic diseases C00 D48 Neoplasms D50 D89 Diseases of the blood and blood-forming organs and certain disorders involving the immune mechanism E00 E90 Endocrine, nutritional and metabolic diseases F00 F99 Mental and behavioural disorders G00 G99 Diseases of the nervous system H00 H59 Diseases of the eye and adnexa H60 H95 Diseases of the ear and mastoid process I00 I99 Diseases of the circulatory system J00 J99 Diseases of the respiratory system K00 K93 Diseases of the digestive system L00 L99 Diseases of the skin and subcutaneous tissue M00 M99 Diseases of the musculoskeletal system and connective tissue N00 N99 Diseases of the genitourinary system O00 O99 Pregnancy, childbirth and the puerperium P00 P96 Certain conditions originating in the perinatal period Q00 Q99 Congenital malformations, deformations and chromosomal abnormalities R00 R99 Symptoms, signs and abnormal clinical and laboratory findings, not elsewhere classified S00 T98 Injury, poisoning and certain other consequences of external causes V01 Y98 External causes of morbidity and mortality n/a Z00 Z99 Factors influencing health status and contact with health services U00 U99 Codes for special purposes n/a 7

11 Hospital admissions for alcohol-related harm Table 3 shows the number of hospital admissions for alcohol-related harm for strategic health authorities in England. Between 2002/03 and 2006/07, the number of admissions increased by 71%, from 473,529 to 811,443. Table 3: Number of hospital admissions for alcohol-related harm: strategic health authorities in England SHA code 2002/ / / / / /07 (%) Q30 North East Q31 North West Q32 Yorkshire and The Humber Q33 East Midlands Q34 West Midlands Q35 East of England Q36 London Q37 South East Coast SHA Q38 South Central SHA Q39 South West England The Information Centre for Health and Social Care also publishes alcohol statistics, including some based on HES, such as Statistics on Alcohol: England, This report considers admissions for three diseases specifically related to alcohol (i.e. where alcohol is a contributory factor in all cases): mental and behavioural disorders due to use of alcohol (ICD10 code F10); alcoholic liver disease (K70); and toxic effect of alcohol (T51). Together, these three conditions accounted for around 208,000 admissions in 2006/07 (Table 4). This figure is considerably less than the current estimate of around 811,000 admissions for all alcohol-related admissions. The Information Centre for Health and Social Care statistics are not directly comparable with the NI39 indicator, as they rely on a small subset of alcohol-related conditions. Nevertheless, the regional pattern of admissions for diseases specifically related to alcohol is very similar to that seen for admissions for all alcohol-related conditions, and overall numbers have increased by a similar amount (62%) between 2002/03 and 2006/07. 3 Available online at alcohol/statistics-on-alcohol:-england-2008-[ns] 8

12 Understanding the dataset Table 4: Number of hospital admissions for three alcohol-specific conditions: strategic health authorities in England, 2006/07 SHA code 2006/ /07 (%) Q30 North East Q31 North West Q32 Yorkshire and The Humber Q33 East Midlands Q34 West Midlands Q35 East of England Q36 London Q37 South East Coast SHA Q38 South Central SHA Q39 South West England What are the differences between the age, gender and ethnicity of those admitted to hospital? During the period 2002/03 to 2006/07, there were approximately twice the number of admissions among males (2.1 million) than females (1.1 million). The age distribution of admission is shown in Figure 5. Figure 5: Age distribution of alcohol-related admissions 100% 90% 80% 70% 60% 50% 40% 30% 85+y 65 84y 45 64y 25 44y 15 24y <15y 20% 10% 0% Males Females As Figure 5 shows, there were relatively fewer admissions in the and age groups for women than there were for men. Over this period, male admissions rose slightly faster than female admissions (average annual increase of 15.1% for males and 13.7% for females), and rose slightly faster in the over 45s than in younger age groups. The age distribution was broadly similar for all regions. 9

13 Hospital admissions for alcohol-related harm While ethnicity is recorded in the HES dataset, the figures are known to be incomplete, especially for earlier years. This is likely to make interpretation of trends difficult. Information on ethnicity was not extracted for this analysis. What are the 20 areas with the highest and lowest hospital admission rates, and where have rates risen or fallen in recent years? Table 5: Areas with the highest and lowest rate of alcohol-related admissions per 100,000 population (European Age Standardised Rate EASR) Areas with highest admission rates Primary care trust 2006/07 Local authority 2006/07 Liverpool 2683 Liverpool 2682 Heart of Birmingham Teaching 2486 Newcastle upon Tyne 2473 Newcastle 2474 Stoke-on-Trent 2461 Middlesbrough 2423 Middlesbrough 2420 Stoke-on-Trent 2411 Salford 2342 Salford 2349 Gateshead 2327 Gateshead 2338 Halton 2236 Leicester City 2233 Leicester 2232 Knowsley 2226 Preston 2226 Manchester 2223 Knowsley 2225 Wirral 2221 Manchester 2223 Sandwell 2129 Wirral 2220 North Tyneside 2104 Wansbeck 2196 Heywood, Middleton and Rochdale 2081 Sandwell 2128 Ashton, Leigh and Wigan 2057 Copeland 2120 North Staffordshire 2018 Chorley 2106 Blackburn with Darwen 2017 North Tyneside 2103 Halton and St Helens 2002 Burnley 2100 Newham 1973 Rochdale 2081 South Tyneside 1968 Wigan

14 Understanding the dataset Areas with lowest admission rates Primary care trust 2006/07 Local authority 2006/07 Surrey 1054 Eastleigh 847 Dorset 1050 Sevenoaks 846 South Staffordshire 1049 Aylesbury Vale 835 Kensington and Chelsea 1017 West Berkshire 834 West Kent 1010 Dartford 833 Coventry Teaching 1002 St. Albans 818 North East Essex 995 Broadland 810 Hampshire 971 Bexley 804 Southampton City 941 Stevenage 789 Sutton and Merton 932 Ryedale 784 West Hertfordshire 932 East Hertfordshire 778 Mid Essex 931 Welwyn Hatfield 769 Richmond and Twickenham 919 East Dorset 765 Buckinghamshire 910 West Oxfordshire 755 Berkshire West 854 South Oxfordshire 739 Oxfordshire 849 New Forest 718 East and North Hertfordshire 848 Wokingham 714 Bexley 806 Vale of White Horse 713 Isle of Wight National Health Service 687 Isle of Wight 687 Milton Keynes 545 Milton Keynes 537 Table 6: Areas with the highest and lowest average change per year between 2002/03 and 2006/07 Areas with highest rise in rate Primary care trust Change Local authority Change North Staffordshire 37% Staffordshire Moorlands 39% Stoke-on-Trent 37% Stoke-on-Trent 37% Enfield 32% Newcastle-under-Lyme 35% Blackburn with Darwen 29% Enfield 32% Calderdale 29% Worcester 31% Brent Teaching 27% Blackburn with Darwen 29% Tameside and Glossop 27% Shrewsbury and Atcham 29% Leicester City 26% Calderdale 29% Telford and Wrekin 25% Broxbourne 28% Shropshire County 25% Oadby and Wigston 27% Hillingdon 25% Blaby 27% Dudley 25% Hyndburn 27% City and Hackney Teaching 24% Brent 27% Berkshire East 24% Nuneaton and Bedworth 27% Berkshire West 23% Tameside 27% Hartlepool 23% Kettering 27% Worcestershire 23% Leicester 26% Leicestershire County and Rutland 23% Slough 26% Wolverhampton City 23% Oswestry 26% Solihull Care Trust 22% Corby 26% 11

15 Hospital admissions for alcohol-related harm Areas with lowest rise in rate Primary care trust Change Local authority Change Halton and St Helens 7% Derbyshire Dales 5% Portsmouth City Teaching 7% Erewash 5% Derbyshire County 7% North Lincolnshire 4% Somerset 7% Ryedale 4% North Yorkshire and York 7% Berwick-upon-Tweed 4% Sheffield 6% Rushcliffe 4% North East Lincolnshire 6% Nottingham 4% Derby City 6% Wycombe 4% Wakefield District 6% Cotswold 3% Richmond and Twickenham 6% Broxtowe 3% Hounslow 6% Isle of Wight 3% Doncaster 6% Gedling 3% Rotherham 6% Gosport 3% Bournemouth and Poole 6% Mansfield 3% North Lincolnshire 5% Bassetlaw 2% Nottingham City 4% Ashfield 2% Isle of Wight National Health Service 3% Harrogate 2% Nottinghamshire County 3% Newark and Sherwood 1% Bassetlaw 2% Scarborough 2% Milton Keynes 7% Milton Keynes 8% What main types of disease are driving these changes? The 20 local areas with the highest rate of alcohol-related admissions in 2006/07 account for 13% of the total number of alcohol-related admissions in England (compared to 8% of the population). Among these local areas, the rate was, on average, 63% higher than the national figure and nearly twice as high in some cases. The crude rate of admission was higher across these localities for all ICD disease chapters, more so for some conditions than others. For example, the rate for mental and behavioural disorders due to alcohol was 102% higher than the national figure, whereas the rates for hypertensive diseases and cardiac arrhythmias were only 40% and 23% higher respectively. Both mental and behavioural disorders due to alcohol and hypertensive diseases have doubled across these localities since 2002/03, compared to an average increase for all other conditions of 53%. 12

16 Understanding the dataset When will these data be updated? We are currently examining the quality of provisional HES data for 2007/08, with a view to providing updates on a quarterly basis. We expect to be able to release the first batch of quarterly figures in the autumn. Final data for 2007/08 will be released in the early part of Other documents Also published alongside this document is Hospital admissions for alcohol-related harm: Technical Information and Definition for Vital Signs Indicator VSC26, National Indicator Set NI39 and Public Service Agreement Indicator 25.2 (Department of Health/NWPHO, July 2008), which provides a technical description of the methodology for the recently introduced indicator Hospital admissions for alcohol-related harm, included in the three key indicator sets and performance management frameworks. The peer reviewed report on AAFs is published by the NWPHO, Alcohol-attributable fractions for England alcohol-attributable mortality and hospital admissions (NWPHO, July 2008). The NWPHO LAPE website ( contains a number of associated documents and datasets: Profiles for all 354 English local authorities and 152 primary care trusts both showing national indicators alongside 22 other indicators of alcohol-related harm. Local quarterly VSC26/NI39 data and trends from 2002/2003 to 2006/2007. New AAFs for England. 13

17 Crown copyright c 1p July 08 (Web) Produced by COI for the Department of Health