After several years of higher-than-expected mortality, deaths in England & Wales 2024 were in line with 2019, the last pre-pandemic year. So far in 2025, deaths have been lower than in 2024 – analysis from the Continuous Mortality Investigation (CMI) suggests that mortality rates up to the end of September had improved by around 1.5% compared to 2024, meaning that the first three quarters of 2025 have seen the lowest national mortality rates ever.
Predictions for the rest of 2025
If the end of 2025 continues to see similar levels of mortality to the earlier part of the year, we might expect that 2025 will see a relatively significant improvement in mortality compared to 2024. However, deaths in the winter months tend to be more variable than in other seasons, especially due to deaths from infectious diseases. COVID-19 continues to circulate – based on data from the UK Health Security Agency (UKHSA), there was a wave of infection in September to October which led to a small increase in the number of deaths involving COVID-19, however the wave appears to have peaked in October and the number of new cases is now decreasing. Based on weekly deaths data from the Office for National Statistics (ONS), the numbers of deaths involving flu and pneumonia are significantly higher than those involving COVID-19 so flu and pneumonia are likely to be the main drivers of variability of deaths in the coming months.
Early indications suggest that the 2025/6 flu season may see earlier and / or higher levels of infection than an average year, with flu vaccine coverage looking likely to be similar to 2024
- UKHSA data shows that:
- The percentage of positive tests in England this October is higher than it was in any of the previous 3 years
- Take-up of the vaccine among at-risk groups is so far looking similar to the same point in 2024
- The 2025 Australian flu season, which occurred in the European summer and can often provide an early indication of what to expect in the Northern hemisphere, saw record-breaking numbers of infections
If we do see a relatively severe flu season in 2025/6, we might expect to see elevated mortality at the end of this year compared to 2024. Despite this, across the whole of 2025 we still anticipate a material improvement over the previous year.
Differences between age groups
In 2022 to 2024, the overall mortality experience masked differences by age and sex – 2025 shows a similar pattern, with mortality for men aged below 64 being broadly in line with the average of the last 10 years, while for women of all ages and men over age 65 the rates are among the lowest on record.
There is some evidence from cause of death data suggesting higher excess mortality from circulatory disease among the younger population, which hasn't been seen to the same extent at older ages – but the exact reasons for this are still unclear. Greater understanding of this will be important for future healthcare policy and will help form a view on likely trends in mortality for these age groups.
What might the future look like?
The projected level of longevity improvements in the future will influence pension scheme funding levels as well as insurer pricing when schemes are ready to transact. When considering how longevity expectations may move over time, we typically look at the impact of emerging mortality on industry-standard models, as well as developments in factors which may influence future trends.
Industry-standard models
Each year, the CMI produce a mortality projections model which is calibrated using England and Wales population mortality data. The latest version (CMI_2024) was published at the end of June 2025, and we expect many schemes and insurance market participants will adopt this model for funding and reserving.
Compared to the previous model, the CMI made several technical changes – the main ones are:
- Explicitly allowing for excess mortality in the pandemic period, reducing over time (using the default assumption in the model, there is very little excess assumed in 2025)
- Allowing for different trends in mortality in different age groups
This version of the model gives somewhat higher life expectancies than the previous version, with liabilities typically increasing slightly for most pension schemes adopting the new model. If experience in the last few months of 2025 is roughly as expected, the next version of the model ‘CMI_2025' (which we expect will be published in March 2026) is likely to show a further small increase in life expectancy projections for most pension schemes.
Factors influencing future trends
One factor of particular current focus is the use of GLP-1 drugs for weight loss – these drugs have shown promise in clinical trials and their use has been increasing in the UK, though availability on the NHS is still restricted to a relatively small ‘at-risk' group. Based on the evidence available, it seems likely that weight loss drugs will have some positive impact on life expectancy for those who are able to access and use them effectively, although there are still significant unknowns relating to their impact outside the clinical setting and over any longer-term side effects.
Another developing and uncertain factor is the impact of AI – over the medium to long term its use in healthcare could lead to earlier detection of diseases as well as more personalised treatments, but there have been concerns raised on AI's long-term impact on health due to unemployment and lower levels of social integration.
Looking at population level mortality projections, we think that the impact of weight loss drugs is likely to be positive but relatively small; this view could change if there is further strong evidence on their real-world effectiveness, and if improved pricing / affordability of these treatments results in expanded access to these novel treatments. AI's impact on longevity is likely to be longer term; it may well be positive overall, but the size of the effect is difficult to estimate.
In summary, we expect that 2025 will see the lowest ever mortality rates in England & Wales, although the extent of this will be influenced by flu levels over the remainder of the year. This is likely to lead to a small increase in life expectancies when the next mortality model is published – at present we don't expect pension schemes or insurance market participants to make a further significant adjustment to their best estimate of life expectancy for emerging factors such as GLP-1 and AI, as their impacts are currently very uncertain.
