England 2015/16 to 2019/20
Published 5 November 2020
Last reviewed and updated 9 December 2020
Key findings
The number of prescription items used primarily in the treatment of diabetes has increased every year between 2015/16 and 2019/20.
The number of patients has also increased every year between 2015/16 and 2019/20 but has not increased as much as items. This has resulted in a greater number of items per patient in 2019/20 than in previous years.
The total cost of diabetes drugs prescribed has increased since 2015/16, driven by large increases in costs of antidiabetic drugs and insulin. These increases are larger than both the number of items prescribed and the number of patients. This has resulted in increased cost per person in 2019/20 than in previous years.
Antidiabetic drugs remain the most prescribed of the diabetes items. Prescribing for all categories increased between 2018/19 and 2019/20, except for diabetic diagnostic and monitoring agents, which decreased by 6%.
Overall demographics of patients have remained steady over the last five years. The median age has remained constant at 64 and the split by sex has remained 55% male and 45% female patients since 2015/16.
Areas of greatest deprivation have the highest number of patients per population receiving diabetes medicines, with 8.6% of people in the most deprived areas receiving diabetes medication, compared to 3.4% of people in the least deprived areas. This pattern is replicated at drug levels, with 7.5% of the population in the most deprived areas receiving antidiabetic drugs, compared to 2.9% of people in the least deprived areas.
1. Introduction
Diabetes is a lifelong condition that causes a person’s blood sugar level to become too high. There are 2 main types of diabetes, type 1 diabetes where the body’s immune system attacks and destroys the cells that produce insulin, and type 2 diabetes where the body does not produce enough insulin, or the body’s cells do not react to insulin. Another form of diabetes is gestational diabetes, where diabetes develops during pregnancy and usually disappears after childbirth.
Chapter Three of the NHS Long Term Plan sets out the NHS’s priorities for care quality and outcomes improvement for the decade ahead. There are also 5 long term highlights for diabetes in the long term plan, aimed at prevention and treatment of this condition.
This publication aims to describe the prescribing of medicines and appliances used for the treatment of diabetes in a primary care setting in England that are dispensed in the community. This does not include data on medicines prescribed and dispensed in secondary care, prisons, or issued by a private prescriber as this is not held by NHSBSA.
These statistics use the BNF therapeutic classifications defined in the British National Formulary (BNF) using the classification system prior to BNF edition 70. Each January the NHSBSA updates the classification of drugs within the BNF hierarchy which may involve some drugs changing classification between years of this publication. Four paragraphs of the BNF are covered within these statistics:
- Insulin
- Antidiabetic drugs
- Treatment of hypoglycaemia
- Diabetic diagnostic and monitoring agents
These medicines are classified by their primary therapeutic indication. However, it is possible that they can be prescribed for other reasons outside of this primary therapeutic indication. For example, metformin which is used to reduce blood sugar for those with type 2 diabetes is also used to treat infertility caused by polycystic ovarian syndrome (PCOS). The clinical indication of a prescription is not captured by NHSBSA during processing, and therefore we cannot determine the reason that a prescription was issued. Due to this, these statistics may not give an accurate estimation of the population who are receiving drugs specifically for diabetes and any inferences made from the data should take this into consideration. Furthermore, while those with type 1 diabetes are typically prescribed insulin this can also be prescribed to those with type 2 diabetes. Therefore, we are unable to distinguish between patients with type 1 or type 2 diabetes and drug types should not be used as a proxy for diabetes type.
The BNF sub paragraph for hypodermic needles is not included as part of this publication as these are often used for other clinical conditions and would unnecessarily skew the data. This is consistent with previous publications released by NHS Digital.
This publication covers the financial years between April 2015 to March 2020. Due to this, the impact of the novel coronavirus (COVID-19) on prescribing has not been explored as the end date of the publication is too close to the start of the pandemic to observe the full impact of COVID-19 and lockdown restrictions being announced by the UK government on 23 March 2020, and to make sound conclusions on the impact of these. The number of items prescribed have increased across BNF paragraphs in March 2020, and a surge was seen in diabetes prescribing in March 2020 as well.
Throughout this publication the term ‘item’, short for ‘prescription item’, means a single instance of a drug that is listed as a separate entry on a prescription form. For example, Insulin 3ml vials x5 would be listed as 1 item, as would Insulin 3ml vials x20.
There are many costs incurred when a dispensing contractor fulfils a prescription. The costs reported in this publication represent the basic price of the item and the quantity prescribed. This is sometimes called the ‘Net Ingredient Cost’ (NIC).
Patient identification
When the NHSBSA processes prescriptions the NHS number of the patient is identified for most, but not all, prescriptions. The proportion of items for which a patient could be identified is shown in table 1. This means that data relating to patient counts represents most, but again not all, patients.
In this release the term ‘patient/s’ is used to mean a unique, valid NHS number that has been captured from a prescription form and subsequently verified by the NHS Personal Demographics Service (PDS).
Where patients are identified, to assign them to a single age band their age is calculated at 30 September of the given financial year, which is halfway through that year. For patients where date of birth has not been captured, they have been included in an unknown category.
Sex information was not available from PDS for a small number of patients in each year, typically fewer than 100. This may be because it was not disclosed by the patient or not recorded by the organisation that collected the data. These patients have been omitted from the respective figures but can be found in the Statistical Summary Tables for this publication.
The English Indices of Deprivation 2019 have been used to provide a measure of patient deprivation. This provides an Index of Multiple Deprivation (IMD) looking at different factors such as income, employment and health deprivation and ranks areas of England in deciles based on their score, where 1 is the most deprived and 10 the least deprived. For our publication, the IMD is derived from the location of the practice where an item has been prescribed.
Where data is provided “per patient”, this refers to only costs or items attributable to identified patients. Therefore, these will not total overall costs or items which refer to all known costs or items.
Table 1: The proportion of items for which an NHS number was recorded for listed BNF sections 2015/16 to 2019/20
Source: Statistical Summary Tables - Patient Identification Rates - Costs and Items
2. Results and commentary
The charts in this report are interactive and allow you to isolate the topics that you are interested in by selecting or deselecting a series from the legend of the chart. For example, if you only want to view data for antidiabetic drugs and insulin, you can click the other drug categories in the legend of the chart to deselect them. If you wish to include them again just click their legend entry again.
Costs and Items
Changes to items do not necessarily result in proportional change in costs. Price changes or temporary price concessions can all lead to different trends in these two measures.
Source: Statistical Summary Tables - Prescribing for Diabetes - Costs and Items (Table 1)
Prescribing for diabetes medicines accounts for 5.1% of all prescribing, up from 4.6% in 2015/16. The number of items prescribed per year that are typically used to treat diabetes has risen from just under 50 million in 2015/16 to almost 58 million in 2019/20. The average increase in the number of items prescribed each year between 2015/16 and 2018/19 was just under 4%. In 2019/20 there was a 4.2% increase in the number of items prescribed compared to 2018/19, with as just over 2 million more items prescribed. This was also impacted by increased prescribing seen in March during the onset of COVID-19. Prescribing was 10% higher in March than the 2019/20 monthly average for diabetes products.
Source: Statistical Summary Tables - Prescribing for Diabetes - Costs and Items (Table 1)
Costs for diabetes medicines have also increased to £1,152 million in 2019/20. This accounts for 12.5% of the total spend on primary care medicine prescribing and is an increase of £76 million (7%) on the previous year. As seen last year, the cost of prescribing for diabetes continues to rise quicker than the number of items.
Source: Statistical Summary Tables - Prescribing for Diabetes - Costs and Items (Table 2)
The number of items has increased steadily overall between 2015/16 and 2019/20. Antidiabetic drugs remain the most prescribed treatment for diabetes, with items increasing an additional 5.1% compared to 2018/19. This brings the total number of antidiabetic drugs to over 42 million this year. Treatment of hypoglycaemia has shown a greater increase than noted in previous years, increasing by 23 thousand items to 343 thousand items after two years of slower rises. Diabetic diagnostic and monitoring agent items have decreased compared to 2018/19 and have reduced by 60 thousand items. This has been primarily due to a reduction in in glucose blood testing reagents, which has been balanced out by a rise in ketone blood testing reagents.
Source: Statistical Summary Tables - Prescribing for Diabetes - Costs and Items (Table 2)
Costs have also followed a similar trend as items for insulin, antidiabetic drugs, and treatment of hypoglycaemia, with steady rises in all 3 categories each year. Diabetic diagnostic and monitoring agents have decreased every year since 2016/17, though this decrease has been more marked in 2019/20, with a 8.3% decrease against 2018/19.
Source: Statistical Summary Tables - Prescribing for Diabetes - Costs and Items (Table 1)
The number of items per patient has steadily increased over the last 5 years, increasing from an average 17.5 diabetic items per patient to 19.0 diabetic items per patient. This does not include any prescribing in other drug categories.
Source: Statistical Summary Tables - Prescribing for Diabetes - Costs and Items (Table 4)
See About these Statistics: Interpretation of a box plot for guidance on interpreting this chart
Prescribing in the community can occur from Clinical Commissioning Groups (CCGs) or from other areas, such as healthcare partnerships or foundation trusts. The figures in this section relate to prescribing from CCGs. The cost per patient varied across CCGs with costs in 2019/20 between £271.09 per patient through to £477.11 per patient. The median cost per patient per CCG was £383.61, a 6% increase from 2018/19 (£361.22) and a 15% increase since 2015/16 (£333.74).
This year, Northumberland CCG’s costs per patient were significantly lower than other CCGs with this being classed as an outlier - that is, more than 1.5 times the interquartile range lower than quartile 1, and statistically significantly lower than other CCGs. There have been no outliers for higher costs across the 5 year period. Cost per patient is spreading further apart and there is more difference between CCGs costs in 2019/20 than in 2015/16.
91 CCGs were within 10% of the national average cost per patient this year. While this is a decrease from 131 last year, this is after a merger of CCGs bringing the number of CCGs to 135, and still accounts for around 2/3rds of all CCGs.
Patient numbers
Source: Statistical Summary Tables - Prescribing for Diabetes - Patient Demographics (Table 1)
Patient numbers have risen every year since 2015/16. The number of identified patients receiving diabetes items has now reached just over 3 million. Over the last 5 years the number of patients receiving diabetes medication through primary care has been consistently around 5% of the national population of England, though this is increasing more quickly than population estimates and is currently 5.3% of the national population.
Patient sex
Source: Statistical Summary Tables - Prescribing for Diabetes - Patient Demographics (Table 3)
While the number of patients has increased year-on-year, the overall split of patients has remained consistent - 45% of patients receiving diabetes medications are female, and 55% are male. Both figures have increased proportionally, which has resulted in a further 170,000 male and 120,000 female patients receiving medicines for diabetes since 2015/16. The number of patients where their sex is Unknown/indeterminate has increased from 34 in 2015/16 to 64 in 2019/20, remaining around 0.001% of prescribing. Patients where their sex was unknown or indeterminate have been grouped together, and can be found in the Statistical Summary Tables.
Source: Statistical Summary Tables - Prescribing for Diabetes - Patient Demographics (Table 3)
At drug level, this split is consistent for most drugs, though females are slightly more likely to receive treatment of hypoglycaemia (47.5% of these patients are female) . The number of female patients receiving treatment of hypoglycaemia has increased more quickly than it has for males over the last 5 years. Patient numbers increased by 12.1% for females between 2015/16 and 2019/20 but only increased 7.4% for males between the same time period. The number of items has increased further still, increasing 25.7% for females from 2015/16 to 2019/20, compared to 21.7% for males in the same period.
Patient age
Source: Statistical Summary Tables - Prescribing for Diabetes - Patient Demographics (Table 4)
The median age of a patient receiving diabetes medication has remained constant over the past 5 years, at 64. Rises in patient numbers have been shown in every age category between 2018/19 and 2019/20,except for ages 0-4, which has decreased 2.7%, and ages 45-49, which has decreased 0.9%. The average rise was 2.5%, with larger rises shown in patients 90+ increasing 6.6%, 30-34 increasing 3.9% and 85-89 increasing 3.4%. Over the past 5 years, patient numbers have dropped 8.5% in 0-4-year olds but risen 22.3% in those aged 90+.
Source: Statistical Summary Tables - Prescribing for Diabetes - Patient Demographics (Table 5)
By drug group, patient trends in the last year have followed the overall 5-year trends. The biggest proportional drop has been seen in antidiabetic drugs in 0-4, decreasing 40% since 2015/16 and the biggest rise has been identified in insulin in over 90s, which has increased 42% since 2015/16. In 15-19-year olds, there has been a drop in all drug types of around 3% except for treatment of hypoglycaemia, which has increased 15% in the last 5 years. This is the opposite trend to the 40-44 and 45-49 age groups, which have seen drops in all categories except antidiabetic drugs which saw a rise of 5% and 7% respectively.
We have included an additional breakdown of prescribing in children. Children are classed as patients aged 17 and under at 30 September for the given financial year.
Table 2: Number of identified child and adult patients - Diabetes patients by Age
Source: Statistical Summary Tables - Prescribing for Diabetes - Patient Demographics (Table 6)
The number of children receiving diabetes medications has increased by 3.4% since 2015/16, increasing every year with the exception of 2017/18 where the number of patients dropped 0.6%. The number of adults receiving diabetes medications has increased each year, and has risen 11.2% since 2015/16.
Patient deprivation
To compare prescribing across deprivation levels, the Index of Multiple Deprivation (IMD) has been used. This takes factors such as income, employment opportunities and health outcomes to rank areas of England from most deprived (1) to least deprived (10). This allows comparison of rates of diabetes as it also provides an estimate of the total number of people in England living in each IMD decile.
Deciles are calculated by ranking the LSOAs from most deprived to least deprived and dividing them into 10 equal groups. These range from the most deprived 10% in decile 1 of small areas nationally to the least deprived 10% in decile 10 of small areas nationally.
Source: Statistical Summary Tables - Prescribing for Diabetes - Patient Demographics (Table 9)
The number of patients receiving diabetes medication decreases as level of deprivation decreases. Furthermore, the percentage of the population receiving diabetes treatment also reduces as deprivation levels decreases. 8,570 per 100,000 of the population of decile 1 (most deprived) are receiving medication for diabetes, by decile 10 this has reduced to 3,420 per 100,000 of the population.
Source: Statistical Summary Tables - Prescribing for Diabetes - Patient Demographics (Table 10)
Across the population of England in 2019/20, antidiabetic drugs are the most common category of diabetes prescribed, though the frequency of prescribing differs across IMD deciles. 7,550 out of every 100,000 people in decile 1 (most deprived) receive prescriptions for antidiabetic drugs, which reduces to 2,850 per 100,000 people by decile 10 (least deprived).
This pattern is similar across the other diabetic drug categories, for example there are around 2,000 people per 1,000 receive prescriptions for insulin in decile 1 compared to 909 per 100,000 in decile 10.
In 2019/20 numbers of items fell for every drug type as each decile decreased in deprivation from 1 to 10. The only exception to this was in decile 9, where there was a small increase in treatment of hypoglycaemia with a 6% increase against decile 8, insulin with a 3% increase against decile 8, and diabetic diagnostic and monitoring agents with a 1% increase against decile 8. In 2019/20 over twice as many items were prescribed for insulin and diabetic diagnostic and monitoring agents in decile 1 compared to decile 10, rising to over 2.5 times as many for antidiabetic drugs.
3. Background
Antidiabetic drugs
Antidiabetic drugs are generally used to treat type 2 diabetes. They are taken by mouth and work in a number of different ways depending on the type of drug for example by increasing the amount of insulin made in the body or by decreasing the production of glucose in the body so lowering blood glucose levels. Patients may be prescribed a single antidiabetic drug or may be prescribed several to work together to achieve the desired control of their diabetes in combination with diet, exercise and lifestyle advice and interventions.
One of the most commonly used antidiabetic drugs is metformin which is usually the first antidiabetic drug a patient with type 2 diabetes will be prescribed. Depending on the response to treatment, additional antidiabetic drugs may be added, and the patient may be prescribed insulin further into treatment if the desired control of the diabetes is not achieved with antidiabetic drugs and diet and lifestyle management alone
You can find out more information on type 2 diabetes medications on the NHS website.
Insulin
Insulin is a hormone that plays a key role in the body’s metabolism, including regulating blood glucose levels. In those with type 1 diabetes, the body produces insufficient insulin to undertake this role effectively. For those with type 2 diabetes, the body does not respond effectively to insulin (known as insulin resistance) or the body does not make enough insulin. People with type 1 diabetes will be prescribed insulin, while only a proportion of those with type 2 diabetes will be prescribed this.
Insulin can be injected or delivered by an insulin pump which regularly infuses insulin into the body. Synthetic insulin or non-synthetic animal insulin are available depending on the needs of the patient. Insulin preparations also vary in how quickly they act and are often used in combinations, depending on the individual requirements of a patient.
You can find out more about insulin on the NHS website.
Diabetes diagnostic and monitoring agents
Diabetes diagnostic and monitoring agents cover a range of monitoring equipment and testing strips that can be used by a person with diabetes to check their diabetic control. Self-monitoring is not routinely suggested for type 2 diabetes but it is an integral part of treatment for people with type 1 diabetes. This can be monitoring blood glucose levels or monitoring ketones (a by-product of the breakdown of fats) in either the blood or urine.
By being able to monitor blood glucose levels, patients can manage their health effectively and prevent hypoglycaemia or hyperglycaemia.
Diabetic patients can be at risk of diabetic ketoacidosis, a serious condition, where the body starts to run out of insulin and ketones build up in the body. Checking ketones levels can be an important early warning of this and monitoring agents allow people to check these levels themselves.
Treatment of hypoglycaemia
Hypoglycaemia is a lower than normal blood glucose concentration. It is the most common side effect of insulin treatment and can cause acute symptoms such as feeling tired and sweating, drowsiness and confusion. This can progress to seizures or unconsciousness if untreated.
Hypoglycaemia can be treated with a sugary drink or snack, though this is not always enough. Fast-acting carbohydrates are prescribed for patients to keep at hand in case of hypoglycaemia, these can be oral liquids and gels, capsules or even injectables.
The NHS Website has additional information on symptoms and treatment of hypoglycaemia.
4. About these statistics
This publication is the first in the series the first in series by us following a public consultation by NHS Digital. Previous versions of this publication are available on the NHS Digital website. Users should always use the figures in the latest publication to ensure they are the most up to date figures available.
Further information on the methodology used in this publication and further background information is available in our Background Information and Methodology supporting document
Patient counts
The patient counts shown in these statistics should only be analysed at the level at which they are presented. Adding together any patient counts is likely to result in an overestimate of the number of patients. A person will be included, or counted, in each category or time period in which they received relevant prescriptions. For example, if a patient received a prescription item of insulin in 2019/20 and also a prescription for treatment of hypoglycaemia in the same year, then adding together those totals would count that patient twice. For this reason, data on patient counts for different BNF paragraphs should not be added together.
Geographies included in this publication
The patient deprivation measures given in these statistics are based upon the LSOA of the postcode of the prescribing organisation as matched to the August 2020 NSPL file. However, higher geographies included in the statistical summary tables of this publication such as CCG use NHSBSA administrative records, not geographical boundaries and more closely reflect the operational organisation of practices than other geographical data sources. Due to the way data is processed, the cost per patient per CCG will be affected by patients who move CCG within the financial year. This is believed to affect around 2% of patients in 2019/20.
Index of deprivation
IMD deciles are calculated by ranking the LSOAs from most deprived to least deprived and dividing them into 10 equal groups. These range from the most deprived 10% (decile 1) of small areas nationally to the least deprived 10% (decile 10) of small areas nationally.
Rates per IMD decile are calculated by using the mid-year population estimates for England to count the estimated number of people in an IMD decile. The number of patients per decile is then divided by this population figure to give a comparable rate across deciles no matter how big or small they are. This allows us to compare the number of people receiving diabetes medications by deprivation.
Planned changes to this publication
This is an experimental official statistic release. Experimental statistics are newly developed or innovative statistics. These are published so that users and stakeholders can be involved in the assessment of their suitability and quality at an early stage. We will regularly be reviewing the methodology used within the statistics.
Further analysis by geographic regions is planned for subsequent releases.
Interpretation of a box plot
Box plots are used to help visualise not just averages, but also how data is spread out. If all data points were arranged from the smallest to the biggest, halfway along this line in the middle would be the median. The middle line in the box is the median. The top section of the box includes the 25% of numbers directly above the median, the bottom includes the 25% directly below the median.
The remaining lowest 25% and highest 25% are usually captured by the whiskers; the whiskers are set to have a maximum length of 1.5 interquartile ranges (the length of the box), the end of each whisker is the most extreme value within this range. Any points that are further than 1.5 interquartile ranges from the top or bottom of the box are classified as statistical outliers and omitted from median and range calculations; these are shown as small circles on the box plot.
The length of the whiskers away from the median show how similar the data is compared to the average; short whiskers indicate that there are no areas extremely different to the average whereas long whiskers show the data is much more spread out and there are bigger differences between the highest and lowest numbers.
In figure 6, the middle line is getting higher which shows the average cost per person increases. The whiskers are also longer in more recent years, showing that the cost per patient is spreading further apart and there is more difference between the CCGs costs than in 2015/16.
5. Statistical disclosure control
Statistical disclosure control has been applied to these statistics. Patient, item, and net ingredient cost have been redacted in the supporting summary tables if they are below five, or a patient count of less than five can be inferred from them. Further information about our statistical disclosure control protocol can be found on our website.
The high-level figures in this statistical summary have been rounded where appropriate for clarity, in most cases to three significant figures. This is to make this narrative as accessible as possible to all readers. The summary tables released with this publication allow users to investigate this data at lower levels of granularity. Figures in the supplementary tables have not been rounded.
6. Accessibility
How to request content in an accessible format
If you need information on this website in a different format like accessible PDF, large print, easy read, audio recording or braille, you can contact us by:
Email: nhsbsa.accessibility@nhs.net
Phone: 0191 203 5318
We’ll consider your request and get back to you in 5 working days.
These contact details are only for accessibility queries. This inbox is not for technical queries or IT problems. If you have a query that is not about accessibility, go to the ‘Contact us’ section of this page.
View our Accessibility statement for Official Statistics Narratives.
7. Revisions to this document
9 December 2020:
- Added link to the Accessibility Statement.
- Minor typographical changes.
8. Feedback and contact us
Feedback is important to us; we welcome any questions and comments relating to these statistics.
You can complete a short survey about this publication to help us improve the Official Statistics that we produce. All responses will remain anonymous and individuals will not be identifiable in any report that we produce.
You can view our privacy policy on our website to see how your data is used and stored.
You can contact us by:
Email: nhsbsa.statistics@nhs.net
You can also write to us at:
NHSBSA - Statistics
NHS Business Services Authority
Stella House
Goldcrest Way
Newburn Riverside
Newcastle upon Tyne
NE15 8NY
Responsible statistician: Bethany Ogle