Coroners Statistics Annual Bulletin 2018

Ministry of Justice/ONS, 9.5.2019

One death is a tragedy; half a million – well, 541,627 to be precise – is a statistic. 

The Coroners Statistics Annual Bulletin 2018 (which covers the year to end of 2018) has just been released. It is, as always, an invaluable guide to the workload of coroners around the country and possible trends for the future.

The overall bad news is that more of us are dying. Total deaths reported to the ONS have risen year on year, from 484,367 in 2011.  The projected figure of 541,627 deaths in 2018 continues that trend and is set to be the highest total since 1999.

The coroners statistics of course only address those deaths a coroner needs to be told about (those thought to be violent, unnatural or unexplained or of someone in state detention): the good news is that there are less of those.

Of the several things to note in the latest figures the headline is that the removal of the requirement to report deaths of those subject to a Deprivation of Liberty Safeguard (‘DOLS’) authorisation has changed the statistical landscape.

The impact of DOLS

Following the change in the law on 03 April 2017[1] it is no longer an automatic requirement to report the death of a person under a DOLS authorisation to a coroner. Consequently, for the second year in a row since the rule change took effect, the number of deaths reported to coroners has fallen by 4%, from the all-time high of 241,211 in 2016. In 2018 there were 220,648 deaths reported to coroners. This reversal back to 2015 figures suggests that numbers ought to stay reasonably steady at this sort of 2018 figure from now on.

Similarly, for the second year in a row the proportion of deaths reported to coroners, as a percentage of total deaths reported to the Office of National Statistics (ONS) also fell. The Bulletin suggests this too is a reflection of the removal of DOLS deaths from the figures. That may be so, but it isn’t easily verifiable from the figures provided.

What is of interest is that the proportion of deaths reported to coroners as a percentage of total ONS registered deaths remained fairly stable for many years, at around 45% to 46%, from 2007. It then appears to have fallen sharply from 2016 onwards, now being 41% in 2018.

Proportionally far fewer deaths are being reported to coroners than in 2007

Perhaps what is going on is that deaths of vulnerable adults have always featured heavily in deaths reported to coroners. Now the procedural clarity brought about by the introduction of the DOLS reporting requirement, shortly followed by the statutory change removing it, has provided a greater degree of certainty as to how to deal with such cases and so reduced reporting of natural deaths in care homes. Perhaps that is so: it is arguably corroborated by a reduction in conclusions of deaths from natural causes.

Deaths in the custody of the state

In 2018 there were 513 deaths in state detention (excluding DOLS) reported to coroners, 15 less than the previous year. The DOLS deaths that have, of course, complicated these figures in recent years. This welcome decrease in deaths in state detention appears to have been driven entirely by a 13% decrease in deaths of people who were detained under the Mental Health Act (171 deaths in 2018). Concerningly however, the number of deaths in prison custody has risen, by 8%, to 316.

Whilst reporting on deaths in custody has improved over the years there is still much that is unclear. Two key pieces of information are not set out within these coronial figures. First, one cannot ascertain from this Bulletin deaths in custody as a ratio of total prisoner numbers. Second, the eventual conclusions at inquest are not stated[2].

It is possible to track down partial answers to these two questions, though one has to go to other documents. The 2018 UK Prison Population Statistics[3] are of some assistance, but the figures used do not readily track against those contained in the Annual Bulletin.

It appears that whilst the prison population had remained broadly stable for years, there was a notable decrease in the prison population from 2017 to 2018. November 2017 hit a high of 86,327 before falling to 83,430 by May 2018. NSO monthly statistics suggest this reduction continues, with a prison population of 82,578 reported at end of April 2019.

Thus, the increase in deaths in prison custody in 2018 is a real increase, not just a reflection of an ever-increasing prison population. If the prison population were stable a real increase would be concerning enough. But if prison numbers are reducing it must be deeply troubling. It suggests an increasing likelihood of death for each inmate – from 3.4 per 1,000 inmates in 2017 to 3.8 in 2018. That is a significant change.

There is an increasing likelihood of death in prison

However, because the Bulletin cannot categorise deaths by eventual conclusion at inquest it is not immediately evident why so many more prisoners are dying. As all deaths in custody, and not just those that are violent or unnatural must be reported it remains possible that some of the increase is just a reflection of an aging prison population dying of natural causes.

The 2018 UK Prison Population statistics provide some insight, though it is difficult to parse. Incidents of violence and self-harm were all significantly up in the year to 2017[4]. Yet, perhaps unexpectedly, the rate of self-inflicted death fell significantly, to 0.8 per 1,000 in 2017, from a near-high of 1.39 per 1000 only the year before. Until the 2018 deaths’ inquest are completed it will not be possible to say if 2017 was an anomaly.  Although figures from other ONS bulletins suggest it may have been.

The “Safety in Custody Statistics, England and Wales: Deaths in Prison Custody to March 2019” bulletin tells us that of the 317 deaths to March 2019, 87 of them were from self-harm, a worrying increase of 14 self-harm deaths (up 19% from 73 in the previous year)  in the face of a falling prison population.

Why are more people dying in prison and in particular why are self-harm deaths increasing? Answering that question is of course core to the entire purpose of the coronial system.

The Chief Coroner is alive to issues and in his ‘Chief Coroner’s 2017-2018 Annual Report’ details an analysis of a sample of PFD reports arising from death in custody inquests, and notes common themes arising – particularly the inconsistent application of procedures and training needs.

Regional variation

Another curiosity that comes across from the statistics bulletin is the huge variation between regions on a number of data points.

There remain great differences between the number of reported deaths as a proportion of total deaths in some areas. As the Bulletin makes clear, comparisons in this respect contain many anomalies. Various things, such as demographics but also simply where hospitals are (e.g. one coronial region might contain four hospitals, where the next door region contains only two) have a bearing. Plus, deaths are reported to coroners in the area where they occur. ONS death statistics are according to place of residence. Consequently, high commuter areas have much greater incidences of deaths reported to coroners than residence statistics would suggest.

Perhaps more important, is the wide variation between coronial areas as to inquests opened as a percentage of deaths reported. Nationally 13% of deaths reported to coroners went on to an inquest. The regional variation is huge: from lows of 6% (under half the national rate) in Gwent and Nottinghamshire to almost double the national rate (at 24%) in Inner London. In any given year this variation may not mean much. If however patterns are discernible – or if certain regions are persistently more or less inclined to open inquests than others – then that would be of note, although whether the differences are because of local death reporting practices or differing coronial thresholds for continuing on to inquests will not be answered by statistics.

The technological advances in radiological pathology are shown by the huge rise in the number of post-mortems now being conducted using less-invasive techniques (such as enhanced Computerised Tomography (CT) scans) 3,326 post mortems were conducted in this way in 2018, up from 1,671 cases in 2017. More than two thirds of coroner areas in England and Wales carried out at least one less-invasive post-mortem. In Lancashire and Blackburn with Darwen such an approach is now the norm – over a half (55%) of all their post-mortems used less-invasive techniques[5]. This is welcome progress that avoids the distress to the bereaved of their loved one undergoing a surgical autopsy.

Of concern for the bereaved, however, is that the average time to complete an inquest has risen again, from 21 weeks in 2017 to 26 weeks in 2018. The Annual Bulletin puts this down to the removal of DOLS deaths, many of which were dealt with as speedy documentary inquests. The number next year will provide a little more clarity.  Clearly some inquests, such as the 433 jury inquests held in 2018, take some time to prepare. Whatever the reason – an average of six months wait for an inquest is too long the Liverpool and Wirral Jurisdiction manages an average of 8 weeks.


The Coroners Statistics Annual Bulletin is an invaluable guide. Pleasingly it has been getting better and clearer each year. All inquest practitioners should read it carefully. It is a useful high-level summary of what is going on in coroners’ courts around the country that can serve as a basis for identifying possible trends and areas of interest for the future.

Whilst the Bulletin provides good information about deaths in custody there remain many questions. Given the coronial courts investigative duty under Article 2, and the function this over-arching analysis can play in the crucial matter of people dying whilst in the custody of the state, further detail – particularly as to conclusion at inquest – might be thought useful.



[1] Amendment to the Coroners and Justice Act 2009 removed the requirement to report a DOLS death to the coroner as ‘otherwise in state detention’ with effect from 3 April 2017.

[2] Of course many inquests into 2018 deaths would not have been completed when the statistics were compiled.

[3]  Graham Allen and Chris Watson, House of Commons Library, Briefing Paper SN/SG/04334, 23 July 2018

[4] Figures for 2018 are not available at the time of writing.

[5] Black Country, Leicester City and South Leicestershire and South Yorkshire (Western) conducted over a quarter of all their post-mortems using less-invasive techniques (42%, 34% and 29% respectively).