Deaths registered weekly in England and Wales, provisional: week ending 12 January 2024

• In the week ending 12 January 2024 (Week 2), 13,710 deaths were registered in England and Wales (including non-residents), an increase in all deaths compared with the week ending 5 January 2024 (Week 1), when the number of deaths registered was 11,947; week 1 was affected by the New Year's bank holiday.
• In the week ending 12 January 2024 (Week 2), 17.7% of deaths registered involved influenza or pneumonia (2,425 deaths), while 2.8% involved coronavirus (COVID-19) (388 deaths).
• Of deaths registered in the week ending 12 January 2024 (Week 2), 40.5% occurred within the previous seven days; median time from death to registration was eight days.
• The number of deaths was below the five-year average in private homes (3.4% below, 127 fewer deaths), hospitals (15.7% below, 1,151 fewer deaths), and care homes (9.1% below, 299 fewer deaths) but above the five-year average in other settings (1.2% above, 11 more deaths).
• The number of deaths registered in the UK in the week ending 12 January 2024 (Week 2) was 15,651, which was 10.2% lower than the five-year average (1,771 fewer deaths).

In the week ending 12 January (Week 2), 13,710 deaths were registered in England and Wales. Of these, 12,869 were registered in England and 803 were registered in Wales (Table 1).

Try the new way to filter and download these data:

• Deaths registered weekly in England and Wales by age and sex
• Deaths registered weekly in England and Wales by region

We publish timely, provisional counts of death registrations in our Deaths registered weekly in England and Wales, provisional dataset. These are presented:

• by sex
• by age group
• by place of death
• by selected causes of death
• for regions (within England)
• for Wales as a whole

To allow time for registration and processing, figures are published 12 days after the week ends. We also provide provisional updated totals for death occurrences based on the latest available death registrations. With each week's publication, we also update the data for previous weeks within our dataset, for both death registrations and occurrences.

Weekly death registrations are revised over time to provide users with the most accurate data. Changes in numbers of death registrations in each period will usually be minor, but numbers for a given week can either increase or decrease. This can be because of:

• additional deaths having been registered, but not available in the Registration Online (RON) system by the time data is extracted, either because of manual registrations that have not been entered into RON or technical issues

• change of date of registration (because of an error at the registration office)

• deduplication of death records (removal of an accidental double entry

• cancellation of a registration (for instance, because of an error at the registration office)

Revisions to numbers of deaths by cause will likely be more pronounced, because cause of death is not always available at the time of the weekly publication. This is because text from the death registration must be converted to the International Classification of Diseases 10th Revision (ICD-10) codes for cause of death. Most (over 80%) deaths records have cause information available at the time of publication, and the remaining records are updated over time.

As we receive more death registrations from RON over time, the number of deaths that are known to have occurred in a period will increase. We are only informed about a death when it is registered, so numbers of death occurrences are never final; it is impossible to know definitively whether all deaths have been registered. The reported number of death occurrences depends on when data was extracted and increases as time between date of occurrence and data extraction increases.

The proportion of deaths occurring in week that are registered in the same week is affected by the workloads of doctors certifying deaths, bank holidays and other closures of local registration offices, and other circumstances. Because of all these factors, the provisional death occurrence numbers for different weeks are not easily comparable, and numbers of death occurrences for previous weeks will change with each publication.

The five-year average shows us the expected number of deaths per week based on the most recent years and smooths random year-on-year fluctuations. We use the 2015 to 2019 five-year average as a usual non-coronavirus (COVID-19) period to compare with. The further we move away from this period, the less robust the measure is because of changes in population numbers, age and structure.

Deaths registered in 2024 are compared with the 2018, 2019, 2021, 2022 and 2023 five-year average. The number of registration days in a reference period can affect mortality statistics. Bank holidays can affect the number of registrations because registration offices are closed.

We are developing new methodology to calculate the number of excess deaths. This will replace the current method using the five-year-average as a baseline. For more information on different measures of excess death and our current work around excess mortality, please read our blog post: How do we measure expected and excess deaths?

Underlying cause of death versus contributory causes

In this release, we discuss both deaths "involving" a particular cause, and deaths "due to" a particular cause. Those "involving" a cause include all deaths that had the cause mentioned on the death certificate, whether as the underlying or a contributory cause. Deaths "due to" a particular cause refer to the underlying cause of death.

Data coverage

The number of weeks in the year will affect how many days the data cover in the year. Leap years require a 53rd week to be added to the end of the calendar year. The last leap year was in 2020. It is more appropriate to compare 2020 figures with the average for Week 52 than with a single year from five years previously. Read more on the data coverage for our Deaths registered weekly in England and Wales, provisional bulletin in Section 1 of our Coronavirus and mortality in England and Wales methodology.

Registration delays

This bulletin is based mainly on the date that deaths are registered, not the date of death. In this bulletin, we consider deaths to be registered within the previous seven days, when the time between death occurrence and registration is between zero and six days. The number of days between death occurrence and registration depends on many factors and there might be longer delays sometimes, particularly if the death is referred to a coroner. Read more in our Impact of registration delays on mortality statistics in England and Wales article.

For registration delays, we look at deaths registered each week and check how many out of these occurred within the previous 7 days. For instance, for deaths registered on 8 January, this will include deaths occurring between 2 to 8 January. This will be different to numbers of deaths occurring that week.

Classification codes

From the week ending 26 February 2021 (Week 8), new International Classification of Diseases codes for COVID-19 issued by the World Health Organization (WHO) were used for deaths involving COVID-19. Read more in our Coronavirus and mortality in England and Wales methodology.

For further information on data quality, legislation and procedures relating to mortality, and a glossary of terms, view our User guide to mortality statistics methodology.

The weekly figures the Office for National Statistics (ONS) produces are for England and Wales only, and are from the formal death registration process. They are published each week, to provide users with timely data, and to capture seasonal trends. Data for Scotland and Northern Ireland are provided to us by National Records of Scotland and Northern Ireland Statistics and Research Agency, respectively, to produce numbers and rates for the UK overall.

Quality

More quality and methodology information (QMI) on strengths, limitations, appropriate uses, and how the data were created is available in our Mortality statistics in England and Wales QMI.