Measuring excess mortality: a guide to the main reports
A guide to the major publications related to excess deaths in the UK, who produces them and links to the methods and analysis.
Excess deaths can be defined as the difference between the actual number of deaths in a particular period and the estimated number of deaths expected in that period. Expected deaths can be derived in a variety of ways for different purposes, which means that there may be different estimates of excess deaths published by different organisations and used for those specific purposes. Excess deaths attributable to specific factors - such as hot or cold temperature - can also be estimated by using a statistical model of the relationship between the factor and the number of deaths.
This guidance explains the major publications related to excess deaths in the UK. Details of the methodologies are published by individual organisations.
Producers and publishers of the main reports
The Office for National Statistics (ONS) is responsible for collecting, analysing and disseminating statistics about the UK’s economy, society and population. As such, ONS produces statistics related to the registration of life events which includes publication of Deaths registered weekly in England and Wales.
National Records of Scotland (NRS) and Northern Ireland Statistics and Research Agency (NISRA) are responsible for mortality outputs for Scotland and Northern Ireland, respectively. See:
The Office for Health Improvement and Disparities (OHID) forms part of the Department of Health and Social Care (DHSC) and is responsible for work to improve the nation’s health and reduce health disparities. OHID produced the monthly publication Excess mortality in England until May 2026. This report is currently paused and no further updates have been announced.
The UK Health Security Agency (UKHSA) is responsible for surveillance of infectious diseases and other health related threats to the security of the nation’s health. UKHSA currently produces 4 excess mortality-related products:
- weekly all-cause mortality surveillance in England
- annual surveillance of influenza and other seasonal respiratory viruses in the UK
- annual heat mortality monitoring in England
- annual cold mortality monitoring in England
How deaths are counted
The majority of deaths are not registered on the day they occurred - ONS gives detail on the impact of registration delays on mortality statistics in England and Wales.
Some of the publications listed in this guidance count the number of deaths by the date the death was registered and some count the number of deaths by the actual date when the death occurred.
Publications which use the date of registration avoid uncertainty about deaths that have occurred and not been registered yet, but are less precise about the timing of deaths (for instance, a death which occurred on a Saturday would not be registered until the following Monday at the earliest, and usually later).
Publications which use the date of death occurrence allow more precise attribution of potential causes of changes in mortality rates over time, but usually have only partially complete data due to the delays in registration of deaths.
The main publications and how they can be used
Deaths registered weekly in England and Wales, ONS
If you are looking for the latest figures for deaths registered in England and Wales, then use the ONS deaths registered weekly in England and Wales publication.
Frequency: weekly
How deaths are counted: by date of registration
This publication contains the headline figures on death registrations for England and Wales. The publication also includes numbers of death registrations and excess deaths for the UK as a whole. The statistics are broken down by age, sex, and region.
On 20 February 2024, ONS published a new method for measuring excess mortality.
ONS’s methodology is closely aligned with OHID’s methodology described below. ONS estimates of excess deaths in England are consistent with estimates for other UK countries.
Excess mortality in England, OHID
If you are looking for inequalities in excess mortality statistics within England or a breakdown by cause of death for data between January 2023 and April 2026, use the OHID monthly estimates of excess mortality in England publication.
Frequency: monthly (currently paused)
How deaths are counted: by date of registration
This publication is the most appropriate to use when comparing excess mortality within England for data between January 2023 and April 2026. The publication provides estimated numbers of excess deaths and the ratio of registered to expected deaths. Users can adjust the time periods presented, allowing cumulative estimates to be shown for specific periods. The report includes breakdowns which show inequalities in levels of excess mortality, including by sex, age group and level of deprivation for the English regions and local authorities. OHID’s analysis by cause of death describes which causes were particularly contributing to excess deaths in specific time periods.
OHID published a revised version of its report on 20 February 2024. The report was last updated in May 2026 (April 2026 data). This new analysis and methodology were aligned with ONS’s method where possible. These estimates represent the number of deaths above the number expected, based on trends in mortality over the previous 5 years.
Weekly all-cause mortality surveillance reports, UKHSA
If you are looking for an indication of whether the weekly number of deaths is significantly higher or lower than expected in the 4 UK countries and to make comparisons with other European countries, then use the UKHSA weekly all-cause mortality surveillance publication.
Frequency: weekly (Thursdays)
How deaths are counted: by date of death
All-cause mortality surveillance shows time periods when mortality levels were higher than expected, rather than providing an estimate of the number of actual or expected deaths. When deaths are significantly higher than expected, there is a high mortality signal.
This publication includes details of the weeks in which observed deaths were higher than an expected threshold, which is calculated using a European-wide algorithm called Euromomo. The all-cause mortality surveillance reports include a chart of observed and expected deaths by the date the death occurred across all ages in England. The chart also shows the upper limit for when the number of deaths is considered to be significantly higher than expected.
This publication can be used for:
- making international comparisons
- prompting further investigation to determining the cause of the signal
- assessing the impact of known health events, including high levels of influenza and extreme temperatures
Surveillance of influenza and other seasonal respiratory viruses, UKHSA
If you are looking for estimates of the number of deaths attributable to influenza, COVID-19 and extreme temperature during the winter period, use the UKHSA surveillance of influenza and other seasonal respiratory viruses in the UK publication.
Frequency: annual (in the spring)
How deaths are counted: by date of death
The surveillance of influenza and other seasonal respiratory viruses in the UK statistics are published every year after the winter period.
This publication includes estimates of the number of deaths attributable to influenza, COVID-19 and extreme cold using all-cause mortality data.
This publication can be used to indicate the potential mortality burden from COVID-19, other respiratory diseases and extreme cold.
The report uses an adapted version of the FluMOMO model, which was developed to incorporate COVID-19, details of which are given in UKHSA’s 2023 working paper.
Annual heat mortality monitoring report, UKHSA
If you are looking for statistics on the annual number of deaths associated with the periods of heat in the summer months, use UKHSA’s heat mortality monitoring report.
Frequency: annual
How deaths are counted: by date of death
UKHSA estimates the number of deaths associated with heat in each summer. Heat episodes are defined as periods in which at least one region, or the mean central England temperature is at least 20°C. Deaths during heat episodes are compared with deaths in a baseline period (the 14 days before and 14 days after a heat episode) to estimate heat-associated mortality. Years of life lost are also calculated, as a measure of premature mortality. The data on deaths is stratified by heat episode, region and age group. Since 2023, a breakdown by sex has been included. In 2024, further breakdowns by local resilience forum area, place of death and cause of death were added.
Additionally, in the 2022 report onwards the publication provides a comparison between observed deaths and modelled estimates. The modelled estimates are derived by applying a statistical model of the temperature-mortality relationship in England in the previous 5 years to the temperatures observed in that summer. This allows identification of heat episodes where heat-associated deaths were higher or lower than would be expected for those temperatures.
UKHSA’s heat mortality monitoring provides information on excess deaths observed during episodes of heat each year (known as heat-associated deaths) to inform public health actions and the continuous improvement of the weather-health alerting system in England. The heat mortality monitoring reports provide information on the observed impacts of heat on health, and how these impacts vary depending on temperature and other factors. The publication can be used to understand trends in the number of heat episode days and heat-associated deaths over time. It forms part of monitoring progress against the goals of the adverse weather and health plan for England to reduce mortality and premature mortality related to adverse weather.
Annual cold mortality monitoring report, UKHSA
If you are looking for statistics on the annual number of deaths associated with the periods of cold in the winter months, use UKHSA’s cold mortality monitoring report.
Frequency: annual (in the autumn)
How deaths are counted: by date of death
UKHSA estimates the number of deaths associated with cold in each winter. Cold episodes are defined as periods of 2 or more consecutive days where the average daily mean central England temperature was below 2°C. This is in line with the temperature threshold for issuing a yellow cold-health alert (CHA).
The report estimates the number of cold-associated deaths for each cold episode, based on statistical modelling of how mortality has responded to low temperatures over 5 recent winters. The data on cold-associated deaths is broken down by region, age group, sex, place of death and cause of death.
This report was first published in February 2026 as official statistics in development. UKHSA will gather information about user requirements and further develop the report over the coming years.
UKHSA’s cold mortality monitoring report provides information on deaths observed during episodes of cold each year (known as cold-associated deaths) to inform public health actions and the continuous improvement of the weather-health alerting system in England. The report also informs UKHSA’s winter planning.
The publication can be used to understand trends in the number of cold-associated deaths over time. It can also be used to understand the relationship between temperature and mortality risk. It forms part of monitoring progress against the goals of the adverse weather and health plan for England to reduce mortality and premature mortality related to adverse weather.
Climate-related mortality, England and Wales: 1988 to 2022, ONS
If you are looking for statistics on the risk of death associated with heat and cold, and estimates of the annual number of deaths related to the hottest and coldest days between these dates, use the ONS Climate-related mortality, England and Wales: 1988 to 2025 publication.
Frequency: ad hoc publication
How deaths are counted: by date of death
The publication compares the relative risk of death associated with a range of temperatures, including both cold and heat, with breakdowns for English regions and demographic characteristics.
ONS has worked in collaboration with partners at the London School of Hygiene and Tropical Medicine, DHSC and Wellcome to develop a methodology that adopts techniques used widely in climate research globally (see Gasparrini A, Guo Y, Hashizume M and others. Mortality risk attributable to high and low ambient temperature: a multicountry observational study, The Lancet, 2015: volume 386, issue 9,991, pages 369 to 375). These are official statistics in development. The methods are under development, which means estimates may change.
This publication includes attributable deaths associated with heat and cold. These estimates represent a modelled estimate of excess deaths associated with hotter and colder days, based on temperatures experienced across different locations, while adjusting for other factors influencing mortality. This method uses daily deaths and daily temperatures occurring over 35 years to model the association between outdoor temperature (both heat and cold) and mortality. It adjusts for seasonal and long-term trends, as well as other factors influencing mortality trends, such as air pollution, and was developed based on global scientific research which focuses on the impacts of climate change. This approach provides insight into the impact of a range of temperatures, incorporating delayed effects because studies have shown that temperature-related deaths can occur some time after the exposure to hot or cold weather.
Updates to this page
-
Updated to add UKHSA's cold mortality monitoring report and to reflect the pause in publication of OHID's excess mortality statistics.
-
The information has been updated to better represent the purpose of each publication and to reflect recent changes made to the publications.
-
Change of wording to 'Annual heat mortality monitoring in England' section.
-
First published.