Please note that this website uses cookies necessary for the functioning of our website, cookies that optimize the performance, to provide social media features and to analyse website traffic.

General Medical Council
3 Hardman Street,
M3 3AW  

12 May 2012

My ref:  Dr Richard Shepherd BSc MB BS FRCPath  FFFLM  Consultant forensic pathologist:  The unnatural death of Dr David Kelly 17/18 July 2003

Dear GMC,

Dr Richard Shepherd provided a report dated 16 March 2011 at the request of the Attorney General, (AG) Mr Dominic Grieve QC.

The purpose of this supposedly independent report was to examine the criticisms made by a group, of which I am but one member, to do with the performance of Dr Nicholas Hunt, forensic pathologist, in the examination of the unnatural death of Dr David Kelly MSc CMG.  The AG had received a Memorial in October 2010, and an Addendum in February 2011 from our solicitors, Leigh Day & Co.  We were pleading for the AG to facilitate an inquest, there having been no inquest, by approaching the High Court using S:13 of the 1988 Coroner's Act.   Dr Shepherd's report also dealt with other complaints from outside our group.

It is fair to say that Dr Shepherd's report was central when the AG came to respond to the pleas for an inquest.  

The AG  “I have concluded that the evidence that Dr Kelly took his own life is overwhelmingly strong and that the test set out by Section 13 of the Coroner's Act 1988 is not met.”  Written statement 9 June 2011  

“That has involved help from Dr Richard Shepherd, a leading forensic pathologist, and Professor Robert Flanagan, a distinguished toxicologist.”  Statement by AG to HoC 9 June 2011

In the schedule of responses published by the AG's office – Dr Shepherd's name is mentioned approximately 45 times.

I bring my serious concerns to you about this report that Dr Shepherd made to the AG.  There are in fact many concerns, but I will concentrate on the more serious and end with a most serious charge. I am unable to copy any part from the pdf of Dr Shepherd's report so I will have to refer to paragraphs on individual pages and the relevant headlines in that pdf which I attach (A).  The titles and subtitles are in bold and all that is transcribed from the RS report is in italics.   I will send this letter and all that is relevant both in hard copy to the Triage Team, 3, Hardman Street, Manchester, M3 3AW and by e-mail to This email address is being protected from spambots. You need JavaScript enabled to view it.    I should be grateful for a written, mailed receipt with the details of the attachments because of their importance.

Those responses made by the AG in the above schedule which are relevant to my complaints will be listed by their numbers against each complaint as  AG Schedule -  


Pathological aspects of the death of Dr David Kelly.  Page 3, para 1 – Dr Shepherd refers to Dr Hunt's report of 25 July 2003.  

I have shown that this was

a.  the FINAL report

b.  that the coroner was concerned that a correction and update were to be made to the INITIAL report (“The preliminary cause of death given at the opening of the inquest no longer represents the view of the Pathologist and evidence from him would need to be given to correct and update the evidence already received.”) and

c.  that Lord Hutton referred to this  INITIAL report of 19 July at his inquiry.  (See my letter listing my concerns re the performance of Dr Hunt under - ONE  Failure to reveal at the Hutton inquiry that he changed his opinion as to the cause of death.  Letter to GMC of 17 July 2011 ref GMCFloyd17-07-11 (B))

Surely Dr Shepherd (RS) knew of this initial report given that it was  a.  noted by Lord Hutton  b.  it was noted also in a letter/e-mail I sent to the AG 22 February 2011 and  c.  referred to in paras 29,30 and 31 of the Addendum sent to the AG 28 February 2011 (C).

I have recalled for you under ONE, in the letter to Ms Floyd (C), the case of Dr Kenneth Shorrock.  I quoted this from the GMC summary of the hearing at which Dr Shorrock was found guilty of serious professional misconduct:-

“Professor Vanezis also described the procedure a pathologist should follow when asked to produce a supplementary post-mortem report. Any additional information should be provided to the coroner in the form of a supplementary report which should be read in conjunction with the first report. It should not replace the first report. In addition, reference should be made to the first report’s existence. He further stated that if a pathologist had reason to change his conclusions or opinion, an explanation should be given as to why he has deemed this necessary.

Professor Pounder agreed with this opinion. He too stated that when supplementary reports are issued it is important to make reference to the existence of an earlier report. In Professor Pounder’s opinion Dr Shorrock had a duty to make reference to the existence of the first report. In addition, the second report should have given the reasons for his change of view. …..... RS should have known of its existence as noted by the Oxfordshire coroner and Lord Hutton.  Both had alluded to it.  In bringing his expertise to give an opinion on the performance of Dr Hunt, he was required to examine all the work that was done by Dr Hunt.  

AG Schedule – no comment from the AG although I had included this concern among others in a letter dated 22 February 2011 to him.  (Attached)


Pathological aspects of the death of Dr David Kelly.   
c.  Position of the body   Page 6, para 2.
 “In my opinion, there is no evidence to support the theory that the body was moved after discovery”.  

RS ignores the evidence of the lay searchers, Ms Holmes and Mr Chapman and has no grounds for doing so.  (See transcripts of evidence at Hutton Inquiry)

AG Schedule – Nos  66 95 100 101 102 103 105 112 127 161


Pathological aspects of the death of Dr David Kelly.  
d.  Failure to consider the extent of bleeding at the scene  Page 6, para 6    "In my experience …..I have never known there to have ever been an attempt at quantification of blood loss ….. It is,  in my opinion, an impractical suggestion, devoid of any scientific reality …....."

If there is no attempt to measure, where the need to measure has been acknowledged for many centuries, how is the forensic pathologist able to conclude that a person died from haemorrhage?  The reliance will be on an educated or ill educated guess.  It is acknowledged that a normal adult has to lose at least 2 litres of blood to die from blood loss.  I have established from the literature that blood loss in the field of midwifery can be measured as I had originally suggested.  The circumstances may be very similar and with the same problem of collecting blood from surfaces and from different materials.  This paper is but one example:-

J Midwifery Womens Health. 2010 Jan-Feb;55(1):20-7.

Measurement of blood loss: review of the literature.

Schorn MN.

Vanderbilt University School of Nursing, Nashville, TN 37240, USA. This email address is being protected from spambots. You need JavaScript enabled to view it.

It is important to recognize excessive blood loss during childbirth, which is a significant cause of morbidity and mortality. This article reviews methods to measure blood loss that could be used during childbirth. PubMed, CINAHL, and MEDLINE databases were searched using the phrases "blood loss" and "measurement." The bibliographies of publications were scanned for applicable references. A total of 46 publications are included in this review. The methods used to measure blood loss are categorized into visual estimation, direct measurement, gravimetric, photometry, and miscellaneous. Methods are described and compared. A combination of direct measurement and gravimetric methods are the most practical. Photometry is the most precise, but also the most expensive and complex to use. A variety of miscellaneous methods are presented, but none is a practical or reliable method. Visual estimation of blood loss is so inaccurate that its continued use in practice is questionable and it should not be used in research to evaluate treatment.

The last sentence is based on observations which can be read in several of these papers coming out of research in this field.  I had photometry in mind after the blood was collected, washed/eluted out of the various substrates and the haemoglobin then converted before photometry.  (See attached Opinion (D) as attached to the Memorandum sent to the AG.  RS is responding to this.)  My suggestion, as dismissed by RS, was neither 'impractical' nor devoid of scientific reality.  Without such measurement or measurement of lost blood by volume, there is no scientific basis for concluding that a death was due to exsanguination ie bleeding out.

AG Schedule –  109 110 124


Pathological aspects of the death of Dr David Kelly
c.  Use of Henssge's nomogram to determine the time of death (ToD)  Page 11

i. The weight of David Kelly recorded in the mortuary of 59 kg is incorrect
In my opinion the 59 kg mortuary weight may be low …..

The explanation offered by RS for a difference between 59 kg and 74 kg is risible.  (But see the report that Professor Henssge has kindly provided me with – attached (E)).  Furthermore, there is an acceptance that mortuary scales 'are not especially reliable'.  Why not, and if so why is this accepted by forensic pathologists.  I note the difference in ToD as calculated by the professor:

59 kg  ToD from 19.20 hrs 17 July to 04.20 hrs 18 July

70 kg (4 kg subtracted assuming he was weighed in clothes 8+/- July for medical) -

ToD  16.00 hrs 17 July to 'about' 22.00 hrs 17 July

This latter estimation is compared with the ToD estimation made by Dr Hunt.  Using a post-mortem weight of 59 kg, a rectal temperature of  24 degrees at 19.15 hrs and an ambient temperature of 20.9 degrees he estimated the ToD as lying between -

16.15 hrs on 17 July and 01.15 hrs on 18 July.

(The average air temperature used by the professor was provided by myself and obtained from the University of Oxford Department of Physics.  Other climate recordings were obtained from the Radcliffe Meteorological Centre which is about 8 miles from Harrowdown Hill but they were not used.)  

d.  Other comments concerning estimation of ToD

ii. Blood loss and heat loss.

Prof Henssge dismisses this hypothesis with logic.

iii.  Published versions of Henssge's nomogram

Prof Henssge deals with the apparent drawbacks and describes the proper procedures for deriving an estimate of ToD from the nomogram that bears his name especially in regard to 'different versions'.  RS casts doubt on the method but on uncertain grounds.

iv.  Overall conclusions regarding the ToD of David Kelly

These three paragraphs stand in stark contrast to the opinion of Professor Henssge.  It could be concluded from this opinion of RS that in the investigation of suspicious deaths in the UK no core temperatures are measured, nor ambient or mean temperatures in applying rules that are in essence derived from Newton's law of cooling in order to estimate time of death.  


AG Schedule – 72 81 83 

1. Bleeding to Death from a Transected Ulnar Artery at the Wrist
  I have held from the start that this is not feasible.  (Attached – Morning Star letter of 16 December 2003).   However, Dr Hunt considered this to be the first cause of death in Dr Kelly.  He enlarged upon this in an article in the Sunday Times 22 August 2011 when he spoke for the first time of three big clots up the left sleeve of the Barbour coat.  (It has not been determined if he had the coroner's permission to do that.)

My opinion was backed up by the late Martin Birnstingl FRCS, Consultant in General and Vascular Surgery at St Bartholomew's.  It has been backed up by other surgeons at other times.

A letter to the Times (G) from nine senior doctors August 2010 supported the view of our group in this.  The nine included a past president of the RCS, two forensic pathologists, a vascular surgeon and two with coronial knowledge.

The surgeons based their opinions on observations of arteries that were cut in accidents, in surgery and in war, and what calibre of damaged artery etc would lead to death.  I have stated elsewhere that I cannot recall haemorrhage as being excessive in the young people who came in with wrists they had slashed with razors.  Neither can I recall whether the admission of some was much delayed thus threatening life if Dr Hunt is to be believed.

ANTI  Dr Hunt was supported in his view by other forensic pathologists.  That support continued.

The conclusion by pathologists as to whether bleeding from a transected ulnar artery had lead to death depended on their visual estimate of the blood lost and, I assume, the absence of other causes of death.  That is the volume lost was guessed.  No measurement of the volume has ever been referred to though it is both eminently feasible and desirable for scientific and legal reasons.

All the foregoing, both pro and anti, depends as I say on observation and deduction.  Neither opinion has been proven.  However, Drs Andrew Rouse and Yaser Adi, both public health specialists, researched these two possibilities in Autumn 2003.  They attempted to get their findings published in September of that year, but their letter was refused by three national newspapers in turn.  Their findings and analysis, the facts that is, were published in the 'Rapid Responses' of the BMJ  8 February 2004

Twelve months ago Professor Milroy identified several important issues relating to "expert advice" and the legal system (1). For instance, a learning point from the Sally Clark case appears to be the need for the courts to hear the views of not just the traditional clinicians and forensic pathologists, but "non traditional" experts - in this case the statisticians. We believe that the publication of the Hutton report reinforces this point.

We all now know that the forensic pathology advice presented in Hutton's inquiry was compatible with the view that Dr Kelly died because of a self inflicted wrist injury. However had Lord Hutton asked for expert epidemiological advice he would probably have been told:

Suicide associated with wrist slashing is extremely rare - so rare that the Office of National Statistics does not report wrist slashing as a specific cause of death; it groups such deaths with other uncommon suicide methods such as belly and abdomen stabbings and throat cuttings. (see attached table). This table shows that fewer than five, 55-60 year old men use cutting and piercing instruments to commit suicide annually.

This statistical evidence, combined with the fact that even after searching the medical literature (2) and speaking to medical and surgical colleagues we have not been able to document that wrist slashing can lead to successful suicide, suggests that for all practical purposes wrist slashing suicide does not exist in Britain.        

The Office of National Statistics (OPCS), London 2003.

How can we reconcile such conflicting opinion? The easiest way would be to discredit the epidemiological advice on the basis that it is based on inaccurate or unrepresentative OPCS statistics. This could easily be done.

Would readers send us details of any 55-65 year old males, without a psychiatric history, who have committed suicide by slashing their wrist, during the last 10 years. If we fail to establish that the epidemiological evidence supports the credibility of wrist slashing suicide, we and many others will find it hard to accept that Dr Kelly died by slashing his own wrist.

(1) (2) Ovid Medline online searched 1966 to 2003.

PS: The epidemiological advice reported here was sent to the Secretariat of the Hutton enquiry on September 3rd, 2003.  (No record of this evidence on the Hutton Inquiry web site.  Was receipt acknowledged?)

There followed an inconsequential letter from Professor Milroy, forensic pathologist.

Rouse and Adi again 10 February 2004:-

…..................... Until recently we would have been happy to downgrade our reliance on epidemiological or other data and accept expert assurance on the cause of death. However bearing in mind the Meadows expert testimony debacle, and the poor explanations for every other aspect of Dr Kelly's death we believe that the time is right for interested professionals to publish:
"A case series describing 55-65 year old males, without a psychiatric history, who have committed suicide by slashing their wrist, during the last 10 years"
If we cannot provide hard evidence that slash wrist suicides have occurred with some frequency - as professionals - we will only have ourselves to blame if another conspiracy theory sets hold. We are more than prepared to revise our view that wrist slash death is unlikely and will do so when evidence is published. Specifically, we would like to see case series data published by a reputable author in a peer-reviewed journal such as the BMJ.

The last letter is of interest and is from Professor Forrest

“Who, where, when and how? A second bite at the cherry.

11 February 2004
Alexander R W Forrest,
Professor of Forensic Toxicology, University of Sheffield
Medico-legal Centre, Watery Street, Sheffield S3 7ES
Send response to journal:
Re: Who, where, when and how? A second bite at the cherry.

One of the problems faced by Lord Hutton was that he was not able to examine witnesses on oath because of the way in which his enquiry was set up. It remains open to the Oxford Coroner to resume his inquest touching the death of the late Dr Kelly. A coroner, unlike Lord Hutton, has the power to summon witnesses and to examine them on oath so as to establish who a deceased person was and where, when how he came to his death.

At the very least those who have shared their doubts about how Dr Kelly came to his death with the readers of this journal should consider also sharing their doubts and the reasons for those doubts with the Oxford Coroner.

Competing interests: A R W Forrest is an Assistant Deputy Coroner who retains belief in the efficacy of the Inquest as a means of establishing the means by which a person came to their death.”

Dr Adi found a scientific paper which described 275 attempted suicides.  Geoffrey McKee    Lethal versus Non-lethal Suicide Attempts in Jail   Psychological Reports(1998), vol82.pp.611-14

A quarter used slashing of the arms or wrist.  Only one of the 275 died.


No forensic pathologist has brought forward a case of death caused by bleeding from transection of the ulnar artery at the wrist to Andrew Rouse in the eight years since he and Dr Adi invited doctors to do that.  Neither has a series of such deaths, nor even a single case report, been published.

I have said to my colleagues that if ever the claim was made it would be important to see the post- mortem records in order to show there was no other factor contributing to death ie overdose of depressant drugs.

The only record proving that a death can be caused SOLELY by transection of the ulnar artery at the wrist is cited by Dr Richard Shepherd in his report to the AG - 'in which a young individual died solely as a result of a self inflicted, solitary incision of her left ulnar artery.'  

It is common forensic pathological experience that individuals can and do die as a result of solitary injuries to arteries or other blood vessels whether homicidally, suicidally or accidentally inflicted. l have seen many cases of death from haemorrhagc fullowing incised injuries to the wrist in my career and 1 had a case in the lust year (HSL82) in which u young
individual died solely as a result of a self-inflicted, solitary incision of hcr left ulnar artery.

In my opinion, and in my experience, incised injuries to the ulnar artery may be unusual but they are without doubt potentially lethal injuries.

Previously, I was able to copy the pdf of Dr RS's expert opinion to the AG -

This report became available 9 June 2011 via the web site of the AG.  That was the date on which he stated that he was refusing our plea, and the plea of others, for an inquest.  I asked Mr McGinty, a senior legal officer in the AG's, if he would kindly obtain the post-mortem report of the deceased HSL 82.  He undertook to do so and after three further requests and 78 days later I received it.

(Attached above.  I find that if this is copied, the 'redaction' disappears and the identity etc of the deceased appears)




DR R Shepherd  Consultant Forensic Pathologist  Home Office Pathologist
This report has been subjected to a Critical Conclusions Check in accordance with the Code of Practice for Forensic Pathologist’s held by the Forensic Science Regulator.
Under toxicology RS records -

'I have seen a report dated 2nd August 2010. I note the following significant findings.

Blood alcohol      316 mg / 100 ml

Urine alcohol      384 mg / 100 ml'

Among other important described findings I note that the laceration was of the left wrist whereas there were scars of intravenous injection of drugs on the right forearm.  Given the need for accuracy in getting a needle into a vein, this would strongly suggest that the left arm was dominant.  Is it not surprising that the left wrist was cut?

Dr Richard Shepherd has in

ONE neglected to consider the charge that Dr Hunt omitted to speak of his first post mortem report at Lord Hutton's ad hoc inquiry

in TWO dismisses the evidence of two trained and conscientious lay searchers

in THREE dismisses the possibilities I put forward, and which could be applied from the experience in midwifery, for measuring the volume of lost blood in cases of suspicious death

in FOUR belittles the importance and scientific basis for applying the methods described by Professor Claus Henssge in estimating the time of death.

I am greatly concerned by the opinions he gave in  FIVE.  'I have seen many cases of death from haemorrhage following incised injuries to the wrist in my career …..... '  He provides only assertion.  The post mortem report HSL 82 is advanced as an anecdote to prove that in one case at least a human died SOLELY as a result of an incision of an ulnar artery at the wrist.  The text of this report with the causes of death shows this contention to be completely untrue.

The expert opinion of Dr Richard Shepherd was central or even crucial in the grounds upon which  the Attorney General based his decision to refuse the plea for an inquest.  It was relevant too in the hearing of my plea and the refusal of it in the Administrative Division of the High Court.  I believe that he has not approached his task in an even handed manner and that is unprofessional to say the very least.

In the annexe of his opinion he says -

'I understand that I owe an overriding duty to provide independent assistance, by way of unbiased  opinion in relation to the matters within my expertise and that such advice must be uninfluenced by the exigencies of the case.  I have complied with, and will continue to comply with, that duty.'

David Halpin MB BS FRCS