LeDeR Report 2025 - a first look

 This is a blog post based on my first reaction to the much delayed latest version of the LeDeR report. As you’ll see, I didn’t get as far through it as I planned, for reasons that I hope to explain.

There are a few things about the report that I found a bit odd. The first is that the report is listed as a 2024 report (of LeDeR information in 2023), even though it wasn’t made public until September 2025. When I write anything, the date on it will be when it’s published, not when I handed it in. Listing this as a 2024 report makes invisible the long delay in its publication.

I also didn’t really understand the decision to only include 2021, 2022 and 2023 information in the report (although the authors don’t always stick to this). I get the logic of combining across multiple years to look at some issues where there are smaller numbers of people, but taking as your baseline 2021 (in the second year of the height of COVID-19 pandemic) seems odd when there is more or less national LeDeR information going back to 2018. The reviewing methods have changed over time, but this happened in June 2021 so the 2021 data will include both old and new reviewing methods anyway. Where possible, starting in 2018 would give the reader some handle on what was happening before the COVID-19 pandemic.

The foreword to the main report also says “This year, due to unforeseen pressures on the NHS, there have been somewhat fewer completed reviews than anticipated, affecting reviews during the latter part of 2023 in particular. This limitation may mean there are fewer completed reviews of deaths due to certain conditions, such as seasonal flu”. First, I think these pressures were quite foreseeable, and right from the start of the LeDeR programme it has been a struggle to get reviews completed. Second, I’m not clear if this is a backlog – given the extensive delay to the publication of the report, couldn’t ‘late’ reviews of 2023 deaths have been included in the analyses? Or is it that these missing reviews are never going to happen, which raises real questions about the ongoing feasibility of the LeDeR programme in the absence of focus and investment. As far as I tell from the latest LeDeR report, there were reviews for 77.9% of notifications of deaths in 2021, 83.3% in 2022, and 67.1% in 2023.

Looking through the easy read report against the more detailed reports and infographic, I have some worries about how some important information is shown in the easy read report. I’m particularly thinking about median age of death. The infographic and main report says that this was 62.2 years in 2022 and 62.5 years in 2023, a difference of between 3 and 4 months. The easy read report says “This year, the average age of death is 63. Last year, the average age of age was 62. I know it is one year difference, but it needs to be better”. I don’t think this is right.

When looking at big reports like this, I often start by looking at the detailed data tables right at the end of the report or added as an appendix. This can help me to understand the numbers a bit before reading the words in the main report that interpret what they mean, so I have some ideas and questions that I hope the main report will explain.

I haven’t had the stamina, and I can’t expect you as a reader to have the stamina, to go through all of these, mainly because I got stuck on some straightforward inconsistencies in the first table that I couldn’t get beyond. In the main report this is Table 1.1 (and in the Appendix Table 1.1), which among other things breaks down the number of people with learning disabilities who had died by age group and sex (as recorded at birth). First thing – the number of people in 2023 whose sex was ‘not known’ is different across the two tables – 234 people in the main report table, and 233 people in the Appendix table. Second thing – the Appendix table breaks this down by age group, but the number of people of ‘not known’ sex in each specific age group adds up to 333 people rather than 233 people. I can appreciate that this kind of information is incredibly messy, and also that mistakes like this can happen. I also don’t know that it would make a massive difference in the overall messages to take from the LeDeR dataset. But if a justification for the huge delay in publication is forensic scrutiny of the data from the LeDeR team, NHS England and DHSC…?

I haven’t gone through the whole report checking for these kinds of issues, because there are some other things I need to do before I retire like washing up and sleeping, but it does put me on edge when looking through the rest of the report. Partly to see if there was information going back to 2018 that I could use for longer-term trends, I also looked at the LeDeR report for 2022 (published in 2023). Here I found some other differences that I don’t really understand. The 2025 report said that there were 3,451 notifications of deaths of adults with learning disabilities in 2021, and 3,593 deaths reported in 2022. As far as I can tell, the equivalent data from the 2023 report was 3,096 notifications of deaths in 2021 and 3,044 notifications of deaths in 2022. I don’t know why there are these big differences.

Finally, I want to mention some trends over time about various ways in which the state might deal with, notice, and investigate deaths of people with learning disabilities – these trends aren’t mentioned in the easy read version of the 2025 report. This information is all subject to caveats that not all people with learning disabilities who died have been notified to the LeDeR programme, not all people notified to LeDeR have been reviewed, and not all reviews included this information. They also use information from the two most recent LeDeR reports, which as we have seen can be inconsistent in the information they report.

DNACPRs. The percentage of adults with Do Not Attempt Cardio Pulmonary Resuscitation orders looks like it’s been consistently increasing over time: 69.8% of people in 2018/19; 71.9% of people in 2020/21; 72.5% (or 74.2%) of people in 2021; 72.1% of people in 2022; and 75.5% of people in 2023.

Coroner reporting. There are huge unexplained differences between the figures in the earlier report (22% reported to a coroner in 2018/19; 19% reported to a coroner in 2020/21; 25% reported to a coroner in 2022) and in the 2025 report (40.5% reported to a coroner in 2021; 41.1% reported to a coroner in 2022; 36.4% reported to a coroner in 2023). I don't know why this is.

Police investigation. It looks like information on the percentages of people’s deaths where there was a police investigation has been reported for the first time in the 2025 report: 4.4% of people’s deaths in 2021; 3.8% of people’s deaths in 2022; 2.2% of people’s deaths in 2023.

Safeguarding enquiry. Again, it looks like the percentage of people’s deaths where there was a safeguarding enquiry (I don’t think this is reported anywhere in the main report text, but it is in the Appendix) was reported for the first time in the 2025 report: 16.5% in 2021; 9.2% in 2022; 7.0% in 2023.

Across the piece, it looks like less attention and less scrutiny is being paid of the deaths of people with learning disabilities.

I will stop there for now – this has already turned into a longer blog post than I was planning (no surprise there). I may return to other aspects of the LeDeR report at some point, but I am quite worried about the inconsistencies that have emerged on a cursory look.

All the more recent LeDeR reports can be found here  https://www.kcl.ac.uk/research/leder