A set of local coroner services with widely varying investigation outcomes fails the bereaved.

Harold Shipman managed to kill around 250 women. The initial investigations of their deaths failed to alert anyone and the whistle was eventually blown by a relative of one of his victims. How many more unnatural deaths are missed by defects in the system of coroner investigation? Before retirement I was head of the homicide team for West Yorkshire Police.

Since retirement I have been researching the work of coroners across England and Wales. I have been shocked by the variation in standards and outcomes.

The investigation and classification of death in England and Wales relies upon the application by medical practitioners of diverse reporting standards set locally by coroners and thereafter upon the effectively unconstrained decision process of those same coroners. There are three main decision points within a coroner’s investigative process – (1) should the death be reported for investigation? (2) should the death advance to inquest? (3) what conclusion (verdict) should be chosen?

Comprehensive analysis of Ministry of Justice and Office for National Statistics data for the period 2000-2010 identified substantial local variation at all three decision points. Rates of reporting deaths to the coroner were found to vary between 12% and 87% of all registered deaths in coroner areas. Those local reporting rates were stable over time, suggestive of a consistent area culture.

Deaths reported to the coroner and advanced to inquest ranged across areas from 6% to 29% and again rates were consistent within areas. When it came to choice of conclusion (verdict), coroners again varied widely in their local ‘profile’ and were consistent in their preferred choices. For example narrative verdicts ranged from almost zero use in one area (Carmarthenshire) to 46% of all the verdicts returned in another (Birmingham and Solihull). Similarly, just 3% of all South Shropshire’s inquest verdicts for the period were recorded as natural causes, while that verdict accounted for a staggering 52% of all inquest verdicts in Sunderland.

Which verdicts seemed to replace each other? Suicide verdicts offered a clue. Suicide verdicts were lowest in Birmingham and Solihull (4% of all verdicts). That same coroner area had the highest proportion of narrative verdicts (46%). Do Birmingham coroners perhaps choose a narrative verdict where others might record a suicide? The psychology behind such a choice may be a perceived preference by relatives of the deceased not to hear that their loved one chose to take their own life. South Shropshire had the highest suicide rate in England and Wales (27% of all verdicts) and a low rate of narrative verdicts (< 7%).

Perhaps citizens there are deemed robust enough to ‘hear it like it is’? The high rate of natural causes verdicts in Sunderland (52%) was matched by a correspondingly low rate of accident/misadventure verdicts. Where most coroners see ‘accident’, does the Sunderland coroner see ‘natural’?

The evidence is surely clear that not all coroner areas can be striking the appropriate balance across the process

The inconsistencies persist. This is not a historical issue. A Statistics Bulletin for 2014 (MoJ, May 2015) shows local rates of reporting deaths vary between 24% and 96% of all deaths and inquests ranging from 5% to 22% of all reported deaths. This Bulletin, for the first time, presents evidence of varying outcomes in choice of conclusion by coroners showing at page 21 that suicide rates for the year varied between 4% (in Peterborough) to 31% (in East Sussex and in Ceredigion).

The evidence is surely clear that, with such wide local variation in investigative outcome, not all coroner areas can be striking the appropriate balance across the process. The proposed introduction of medical examiners in 2018 to review reports of death should bring greater consistency to the first decision point.

Two recent articles in Law Society Gazette have provided further evidence of the difficulties posed by our lack of a national coroner service. Joshua Rozenberg pointed out that not much had changed in 800 years (always the opening line of my past lectures on the coroner to police officers!) and outlined a recent case where the coroner for Inner North London asserted her autonomy by challenging advisory guidance on coroner availability and non-invasive postmortems.

Julian Morris stated that not every jurisdiction has equal support and funding and that the Shipman and Luce enquiries of 2003 had both concluded systems for certification and investigation were out of step for a modern society.

A clear circumscription of the role of the coroner is lacking. The service appears unable to articulate its decision-making principles other than to say ‘each case on its merits’ according to local determination. Research is scarce.

Whilst the chief coroner strives to make progress (and occasionally incurs the public wrath of those coroners who lose their jurisdiction), surely structural change is needed to achieve a consistent approach. Chief coroner Peter Thornton knows of the need to tackle local inconsistency. He promised in his first annual report to address the issue by the second report, but sadly has not achieved that.

Does this matter? Causes of death are the starting point for setting priorities for preventative medicine. Accurate classification leads to the ability to prevent future deaths. The bereaved need information that rises above the distribution of prizes in a postcode lottery. Yes, it matters.

Dr Maxwell McClean is a PHD researcher at the University of Huddersfield