Statistics on Public Health, England 2023

Data on the number of hospital admissions for diseases, injuries and conditions that can be attributed to alcohol consumption are produced by the Office for Health Improvement and Disparities Local Alcohol Profiles for England, which use data from NHS England’s Hospital Episode Statistics.

Estimates of the number of alcohol-related hospital admissions are calculated by applying alcohol-attributable fractions (AAFs) to Hospital Episode Statistics data. An AAF is the proportion of a condition assessed to have been caused by alcohol.

Two measures for alcohol-related hospital admissions have been used:

• Narrow measure – where the main reason for admission to hospital was attributable to alcohol. 
• Broad measure – where the primary reason for hospital admission or a secondary diagnosis was linked to alcohol.

The narrow measure estimates the number of hospital admissions which are primarily due to alcohol consumption and provides the best indication of trends in alcohol-related hospital admissions. These are admissions where an alcohol-related disease, injury or condition was the primary reason for a hospital admission or an alcohol-related external cause was recorded in a secondary diagnosis field.

The broad measure gives an indication of the full impact of alcohol on hospital admissions and the burden placed on the NHS.

This part presents information on the number of hospital admissions that are attributable to smoking. The commentary covers the latest available data (2022/23 for hospital admissions).

Figures presented relate to people aged 35 and over because this age group is most likely to be affected by smoking-related ill health.

• Hospital admissions data is taken from the latest available Hospital Episode Statistics (HES) and is based on the primary diagnosis.

Links to local authority level data data and self-reported health by smoking status are also provided.

Note: Hospital admissions during 2020-21 and 2021-22 were reduced due to COVID-19 disruption in hospital activity as well as potentially undercounting smoking related admissions in cases where people were admitted to hospital with Covid that had another smoking related illness such as lung cancer, but as Covid would have been recorded as the primary diagnosis it will be identified as a non-smoking related admission.

OHID are currently exploring options for producing a more reliable estimate  and outputs will be updated once this estimate is available.

Please note that the charts are based on rounded data. Unrounded data are available in the corresponding data tables.

Hospital admissions estimated to be attributable to smoking, by year

In 2022/23 there were estimated to be 408,700 hospital admissions attributable to smoking.

This is a drop of 8% from a peak in hospital admissions in 2019/20 (446,400), but a 5% increase from the previous year (389,800 admissions in 2021/22).

In 2022/23 this represented 4% of all hospital admissions (which has remained at a similar level during the time series), and 16% of hospital admissions for conditions that can be caused by smoking (down from a peak of 18% in 2018/19).

Hospital admissions estimated to be attributable to smoking, by cause

16% of all admissions for respiratory diseases, 8% of all admissions for cancers, and 7% of all admissions for cardiovascular diseases, were estimated to be attributable to smoking.

25% of admissions for respiratory diseases that can be caused by smoking, and 19% of admissions for cancers that can be caused by smoking, were estimated to be attributable to smoking.

Hospital admissions estimated to be attributable to smoking by region and local authority per 100,000 population are provided in Public Health England Fingertips Tool Local Tobacco Control Profiles.

Adult smoking habits in the UK Statistical bulletins provide data on self-reported health by smoking status.

Hospital admissions: Tables S.1.1 and S.1.2

This part focuses on hospital admissions (inpatient settings only) related to being obese. Data is taken from NHS England's Hospital Episode Statistics (HES) database which contains details of all admissions at NHS hospitals in England.

The association between obesity and increased risk of many serious diseases and mortality is well documented and has led to the National Institute for Health and Clinical Excellence (NICE) developing guidelines on identifying and treating obesity.

Four measures are presented for the number of obesity-related hospital admissions:

1. NHS hospital finished admission episodes (FAEs) with a primary diagnosis of obesity – referred to as admissions directly attributable to obesity. A large proportion of these admissions involve a bariatric surgical procedure (see Tables O.2.1 to O.2.4 in the excel datasets).
2. NHS hospital finished admission episodes with a primary or secondary diagnosis of obesity - referred to as admissions where obesity was a factor. A secondary diagnosis of obesity does not necessarily indicate obesity as a contributing factor for the admission, but may instead indicate that obesity is a factor relevant to a patient’s episode of care. (see Tables O.3.1 to O.3.5 in the excel datasets).
3. NHS hospital finished consultant episodes (FCEs) with a primary diagnosis of obesity, and a primary or secondary bariatric surgical procedure - referred to as obesity-related bariatric surgical procedure.  (see Tables O.4.1 to O.4.5 in the excel datasets).
4. NHS hospital finished consultant episodes (FCEs) with a primary diagnosis of obesity, and a primary or secondary bariatric surgical procedure, but excluding maintenance, revisional, and removal procedures - referred to as obesity-related primary bariatric surgical procedure. In this measure most patients would only be counted once (for their initial procedure).  Though not commented on within the report, it is available in data table O.4.1.

An FAE is the first period of inpatient care under one consultant within one provider.  An FCE is a continuous period of admitted patient care under one consultant within one healthcare provider. Please note that admissions do not represent the number of in-patients, as a person may have more than one admission within the year.

Bariatric surgical procedures encompass a group of procedures that can be performed to facilitate weight loss, although these procedures can also be performed for other conditions. It includes stomach stapling, gastric bypasses and sleeve gastrectomy. In general, such surgery is used in the treatment of obesity for people with a BMI above 40, or those with a BMI between 35 and 40 who have health problems such as type 2 diabetes or heart disease. 

Admissions directly attributable to obesity, by year and sex

Note that many of these admissions will be for bariatric surgical procedures, and changes over time may in part reflect changes in uptake of these procedures, and so are not necessarily prevalence driven.

In 2022/23 there were 8,716 hospital admissions with a primary diagnosis of obesity, a decrease of 22% from a peak in 2018/19 (11,117 admissions).

Around 4 in every 5 patients were female (77%), with this proportion increasing slightly in the last 2 years.

Admissions where obesity was a factor, by year and sex

In 2022/23 there were just over 1.2 million (1,236 thousand) hospital admissions where obesity was recorded as the primary or a secondary diagnosis. This is an increase of 8% on 2021/22, when there were 1,142 thousand admissions.

Some (though not all) of this increase may be due to hospitals being more likely to record obesity as a secondary diagnosis than they were previously. See the Data Quality Statement (coherence and comparability) for more information.

Around 2 in every 3 patients were female (65%) and this is similar over time.

Admissions by age group

For admissions directly attributable to obesity, the number increases to middle age, peaking at 35 to 44, before declining in older age groups.

67% of patients were aged between 35 and 64 and this is similar over time.

Admissions where obesity was a factor by primary diagnosis (top ten diagnoses)

Of those admissions where obesity was a factor, but it was not the primary diagnosis (main reason for the admission), the most common diagnoses related to maternity issues and knee joint issues (arthrosis of the knee). Others in the top ten diagnosis types were the formation of gallstones (Cholelithiasis), heart disease and hip issues (arthrosis of the hip). 

Note that overall there were a large number of different primary diagnoses recorded for admissions where obesity was a factor, and collectively the top ten diagnosis types accounted for less than 20% of all these admissions (240 thousand of 1,236 thousand).

Admissions by deprivation level (rate per 100,000 population)

Admissions were mapped to Index of Multiple Deprivation deciles and rates age standardised using the European standard population.

Rates for both admissions directly attributable to obesity, and for admissions where obesity was a factor increase with the level of deprivation. 

Admissions directly attributable to obesity were four times more likely in the most deprived areas (24 per 100,000 population), compared to the least deprived areas (6 per 100,000 population).

Admissions where obesity was a factor were over twice as likely in the most deprived areas (3,393 per 100,000 population), compared to the least deprived areas (1,430 per 100,000 population).

Admissions where obesity was a factor by Local Authority (rate per 100,000 population)

Admission rates have been age standardised and rounded to the nearest whole number.

Admission rates ranged from 420 to 4,880 per 100,000 population, with the highest admission rate over 11 times greater than the lowest rate. The national rate was 2,225 per 100,000 population.

Luton, Gloucestershire, Southampton, Salford and Newham all recorded admission rates of over 4,500 per 100,000 population.

Bracknell Forest, Windsor and Maidenhead, Wokingham and Slough all recorded admission rates below 650 per 100,000 population.

Admissions rate per 100,000 population by Local Authority where obesity was a factor

Local Authority rates for admissions directly attributable to obesity are available in table O.2.3.

Obesity-related bariatric surgical procedure admissions, by year and sex

In 2022/23 there were 5,099 hospital admissions with a primary diagnosis of obesity and a main or secondary bariatric surgical procedure. This is a increase of 7% on 2021/22 (4,749), but a decrease of 27% from a peak in admissions in 2018/19 (7,011).

77.5% of admissions were for females and this is similar over time.

Obesity-related bariatric surgical procedure admissions, by age group

A third of patients (29%) were aged between 35 and 44 and this is similar over time.

Obesity-related bariatric surgical procedure admissions by deprivation level (rate per 100,000 population)

As with the other measures, obesity-related bariatric surgical procedure admission rates increase with the level of deprivation. 

Admissions for obesity-related bariatric surgical procedures were over 3 times more likely in the most deprived areas (12 per 100,000 population), compared to the least deprived areas (4 per 100,000 population).

Obesity-related bariatric surgical procedure admissions, by Local Authority (rate per 100,000 population)

Admission rates have been age-standardised and rounded to the nearest whole number.

Admission rates ranged from 0 to 32 per 100,000 population. The national rate was 9 per 100,000 population.

South Tyneside (32 per 100,000 population), Stoke-on-Trent (30 per 100,000 population) and Sunderland (65 per 100,000 population) recorded the highest admission rates.

City of London, Isles of Scilly, Lancashire, Liverpool and St. Helens are recorded zero rates.

Note that variation in rates across Local Authorities may reflect differences in provision and uptake of bariatric surgical procedures, and not necessarily differences in obesity prevalence.

Obesity-related bariatric surgical procedure admissions rate per 100,000 population by Local Authority

Other related information is available in the National Obesity Audit publications.

This part presents information on the number of hospital admissions (inpatient settings only) related to drug misuse.

Three measures for the number of drug-related hospital admissions have been calculated using Hospital Episode Statistics (HES) data:

• Measure 1 – hospital admissions with a primary diagnosis of drug-related mental and behavioural disorders – referred to as admissions for drug-related mental and behavioural disorders. 
• Measure 2 – hospital admissions with a primary diagnosis of poisoning by drugs, that are listed as controlled under the Misuse of Drugs Act 1971 (includes both intentional and unintentional poisoning) – referred to as admissions for poisoning by drug misuse.

For measures 1 and 2, a drug-related mental and behavioural disorder, or poisoning by drug misuse were the main reason for the admission (primary diagnosis).

• Measure 3 – hospital admissions with a primary or secondary diagnosis of drug-related mental and behavioural disorders – referred to as admissions where drug-related mental and behavioural disorders were a factor.

Measure 3 is a broader indicator of drug-related admissions. This is because a secondary diagnosis does not necessarily mean that a drug-related mental and behavioural disorder was a contributing factor for the admission, but may instead have just been relevant to a patient’s episode of care. The data quality of secondary procedures has increased over time so increases in admissions, compared to 10 years ago, may partly reflect an improvement in data quality as well as an increase in activity. See the data quality statement for more information.

More detail on the methods used to calculate these measures is provided in Appendix B.1.

Mental Health Trusts and Activity: Mental health providers have historically submitted to both HES and to the Mental Health Services Data Set (MHSDS). To reduce the burden on these providers it was decided that while they are mandated to submit to MHSDS they are not also required to submit the same activity to HES. Some providers who previously submitted to HES have now stopped doing so. As a result, the statistics published in this series are impacted, particularly the trend data relating to Measures 1 and 3 for drug-related mental and behavioural disorders.

As the figures are potentially misleading the time series has been removed from this publication and miitigations will be investigated ahead of any future publication.

COVID-19 and the production of statistics:  Due to the coronavirus illness (COVID-19) disruption, it would seem that there is some effect on the quality and coverage of some of our statistics in relation to the 2020/21 period. It is known that many hospitals are reporting zero or significantly less activity across one or more datasets for March 2020 onwards.

Hospital admissions for drug-related mental and behavioural disorders, and for poisoning by drug misuse

This section covers analysis of measures 1 and 2 (see above), which is those admissions where drug misuse was the primary reason for the admission (whether as a result of drug-related mental and behavioural disorders or poisoning).

Admissions by year

There were 9,690 admissions for poisoning by drug misuse, a 24% decrease on 2021/22 (18,053), and 47% lower than the highest recorded number of admissions of the time series, in 2015/16 (18,128). This measure was subject to a methodological change in 2012/13 and so data is not comparable before then (see the data quality statement for more information). The number of admissions for 2022/23 represents a rate of 17.3 per 100,000 population.

Time series data is available as a rate per 100,000 population in the accompanying data tables, and shows similar trends.

Fig A.13 Admissions for drug-related mental health and behavioural disorders by year has been removed due the impact of the change mentioned above relaing to the reporting of mental health provider activity.

Admissions by sex

More men than women were admitted to hospital for drug-related mental and behavioural disorders (71% male). However, there were similar proportions of men and women admitted to hospital due to poisoning by drug misuse.

Admissions by age

Admissions for drug-related mental and behavioural disorders show a younger age profile than those for poisoning by drug misuse, apart from in the under 16 age group where admissions are higher for poisoning by drug misuse. Levels were highest for younger people (apart from those under 16), peaking between ages 25 and 34. Admissions for drug-related mental and behavioural disorders are lowest for those aged under 16 and over 64.

Although admissions amongst people aged 55 and over are lower (see previous charts), numbers have increased more over time for this age group compared to those aged under 55.

Admissions for drug-related mental and behavioural disorders increased by 39% in those aged 55 and over since 2012/13 (from 272 to 378), compared with a 38% decrease for those aged under 55 (6,277 to 3,921).

Admissions for poisoning by drug misuse increased by 7% in those aged 55 and over since 2012/13 (from 2,091 to 2,238), compared with a 45% decrease for those aged under 55 (13,489 to 7,452).

Admission rates by deprivation level (per 100,000 population) 

Admissions were mapped to Index of Multiple Deprivation deciles and rates age standardised using the European standard population (see Appendix B.1 for more information).

Admission rates for both drug-related mental and behavioural disorders, and for poisoning by drug misuse increase with the level of deprivation. 

Admissions for drug-related mental and behavioural disorders were are almost 8 times more likely in the most deprived areas (19.8%), compared to the least deprived areas (2.5%).

Admissions for poisoning by drug misuse were also 5 times more likely in the most deprived areas (21.2%), compared to the least deprived areas (4.3%).

Admission rates by Local Authority (per 100,000 population)

Admission rates have been age standardised using the European standard population (see Appendix B.1 for more information).

Drug-related mental and behavioural disorders

Havering had the highest admission rate with 31 per 100,000 population, followed by Kingston upon Hull (30), Bristol (26) and Luton and Waltham Forest (both 25).

10 LAs recorded rates of less than 2 per 100,000 population (shown in descending order): Plymouth, Wokingham, Dudley, Cambridgeshire, Devon, Lincolnshire, West Sussex, York, Bracknell Forest, and the Isles of Scilly.

Poisoning by drug misuse

North Tyneside had the highest admission rate with 58 per 100,000 population, followed by the St Helens (55), Middlesbrough (51) and Knowsley (48).

10 LAs recorded rates of less than 7 per 100,000 population (shown in descending order): Barnet, Ealing, Kensington and Chelsea, Newham, Slough, Croydon, Windsor and Maidenhead, Enfield, City of London and the Isles of Scilly.

This section covers analysis of measure 3 (see top section), which is those admissions where a drug-related mental and behavioural disorder was the primary or a secondary diagnosis. A secondary diagnosis does not necessarily mean that a drug-related mental and behavioural disorder was a contributing factor for the admission, but may instead just have been relevant to a patient’s episode of care. 

Admissions by sex

More men than women were admitted. In total, 67% of the patients were male.

Admissions by age

Admissions were highest for people aged between 25 and 54, with those age groups representing 70% of all patients. Admissions are lowest in those aged under 16 and over 75.

Admission rates by deprivation level (per 100,000 population) 

As with the other measures, admission rates increase with the level of deprivation. 

Admissions where drug-related mental and behavioural disorders were a factor were over 8 times more likely in the most deprived areas (416 per 100,000 population), compared to the least deprived areas (48 per 100,000 population).

Admission rates by Local Authority (per 100,000 population)

Hartlepool had the highest admission rate with 613 per 100,000 population, followed by Blackpool (456), and St. Helens (432).

The lowest rates were in Slough (59), Bracknell Forest (40), and Windsor and Maidenhead (37).

Definitions vary between countries. See source publications at the links below for further details.

Scotland

Public Health Scotland publishes data on the number of drug-related hospital stays, the number and characteristics of patients admitted to hospital, the substances involved and the geographical variations within Scotland.

Drug-related hospital statistics

Wales

Public Health Wales publish an annual statistical report on alcohol and drug use from health, social care, education and criminal justice services datasets in Wales. This includes data on hospital admissions involving use of illicit drugs. The latest report is available here:

Data mining Wales: The annual profile for substance misuse 2018 - 19

Northern Ireland

Northern Ireland's Department of Health provides episode based acute hospital inpatient and day case activity data obtained from the hospital inpatient system (HIS), which is a patient level administrative data source:

Episode based activity