2015/16 to 2024/25 Annual summary statistics
Date of notice: 15 January 2026
Overview
We have identified an error in the way that we have calculated patient ages for a calendar year. This error has resulted in an overestimate of the number of patients in age-based analyses. Several tables in the supporting calendar year data for each BNF section are impacted, but there are no changes to the key findings, financial year data, or quarterly data.
Description of error
An error has been identified in the calculation of patient age for a calendar year.
Each prescription record was assigned an age based on the patient’s age as of 30 September in the financial year when the prescription was issued. Because of this approach, some patients were assigned 2 different ages in prescription records from the same calendar year.
When prescription records were later grouped by calendar year and age, these patients were counted twice.
Scope
The error directly affected the following tables in the calendar year supporting summary tables for each BNF section:
- Age_Band
- Age_Band_and_Gender
- Prescribing_in_children
To fix the error, we reloaded the data from the NHSBSA Enterprise Data Warehouse. Because gender is a live variable and only a patient’s most recent gender is available, a patient’s recorded gender may have changed between the original publication in July 2025 and the reload in January 2026.
Therefore, the ‘Gender’ table of each BNF section’s calendar year supporting summary tables has also changed.
Solution
Corrected versions of these tables have been produced by calculating ages based on 30 June of the current calendar year, assigning a single age to each patient for the time period.
Impacts
Age_band tables
The number of identified patients has decreased across all known age bands compared to the previous versions:
- BNF 0401 – a 28% decrease
- BNF 0402 – a 40% decrease
- BNF 0403 – a 41% decrease
- BNF 0404 – a 40% decrease
- BNF 0411 – a 43% decrease
The distribution of the number of items and Net Ingredient Costs have changed across age bands, but the total items and total Net Ingredient Costs have remained the same for each BNF Section.
Age_Band_and_Gender tables
The number of identified patients has also decreased across all known age band and gender groups, in line with the changes to the Age_Band tables. These tables only include patients where gender is known, therefore the total items and total Net Ingredient Costs have changed where a patient’s gender has been updated.
Prescribing_in_children tables
These tables contain the same set of patients as the Age_Band tables, but with an alternative grouping of age. Therefore, the total number of identified patients has changed by the exact same amount as for the Age_Band tables. Again, the distribution of the number of items and Net Ingredient Costs have changed across age categories, but the total items and total Net Ingredient Costs have remained the same for each BNF Section.
Gender tables
The distribution of identified patients, number of items, and Net Ingredient Costs has changed due to gender being a live variable in the NHSBSA Enterprise Data Warehouse. However, the total number of identified patients, the total number of items, and the total Net Ingredient Costs have not changed.
There is no change to the key findings, narratives, financial year data or quarterly data included in this release.
We apologise for any inconvenience caused.
2015/16 to 2023/24 Annual summary statistics
Date of notice: 27 June 2025
We have identified an issue with the column names in some tables in the statistical summary narrative. For tables 4, 11, 17, 23, and 30, the column names for the identified patients column and the total prescribed items column were swapped in the data on the table tab. Data in these columns therefore appeared under the incorrect column name.
This affected these 5 tables and data from the data download button for these tables. All charts, other narrative tables, and the supporting summary excel tables were not affected. A narrative containing the corrected column names for these tables has been produced and can be found on the Medicines Used in Mental Health 2023/24 webpage.
We apologise for any inconvenience caused.
Date of notice: 12 July 2024
It's been identified that a category in figure 1 and figure 2 of the statistical summary narrative was incorrectly labelled. The 'Total' category for antidepressants in these 2 charts was labelled as 'Unknown' in the initial release. This has affected data labels for this category in the chart tab and the 'BNF Paragraph Name' column in the table tab and data downloads for figure 1 and figure 2. Chart data for the other antidepressant drug categories was not affected and the numerical data for the 'Total' categories remains the same. Data in the summary table excel files was not affected.
A narrative containing the corrected chart tab, table tab, and data downloads for figure 1 and figure 2 has been produced and can be found on the Medicines Used in Mental Health 2023/24 webpage.
There is no change to the key findings or the narrative text for section 2.1 Antidepressants in this release.
We are sorry for any inconvenience caused.
2015/16 to 2022/23 Annual summary statistics
Date of notice: 15 September 2023
It's been identified that an error was made in this release when training the model of expected items based on pre-pandemic trends. The model was initially trained on the full data instead of only on data from April 2015 to February 2020. This has affected the charts and commentary in section 3 of the statistical summary narrative.
A narrative containing the corrected version of the affected charts and commentary has been produced and can be found in the resources list for this release.
There is no change to the key findings of this release, but the error in model training resulted in incorrect data in the info boxes, charts and commentary in section 3 only.
We apologise for any inconvenience caused.
2015/16 to 2021/22 Annual summary statistics
Date of notice: 2 August 2022
It's been identified that data has been assigned to the incorrect IMD Decile in the statistical summary tables of the latest release. This is due to an ordering error in code used to produce these statistics.
This affects Table 8: Indices of Deprivation (IMD) in each of the statistical summary tables found in the resources list for this release.
Corrected tables have been produced assigning Total Identified Patients, Total Items, and Total Net Ingredient Cost (GBP) to the correct IMD Decile.
There is no change to the key findings of this release and the correct data was used in the statistical narrative.
The mislabelling of this field would contradict the conclusions that would previously have been drawn using data from the statistical summary tables.
We're sorry for any inconvenience this has caused.
2015/16 to 2019/20 Annual summary statistics
Date of notice: 10 December 2020
It has been identified that a column, Identified Patient Flag, was omitted from the statistical summary tables of the latest release.
This affects Table 4 - STP in each of the statistical summary tables found in the resources list for this release.
The tables have been reproduced to include the omitted column.
There is no change to the key findings of this release. The omission of this field would not contradict the conclusions that would previously have been drawn from the data.
We're sorry for any inconvenience this has caused.