Response to a Report of a review of the research
framework in North Staffordshire Hospital NHS Trust
(The Griffiths Report), released by the NHS Executive on
May 8th 2000
Dated 23rd September 2000
David P Southall OBE, MD, FRCP, FRCPCH
Professor of Paediatrics
I have written this response as the Principal Investigator of the research
projects discussed in the Griffiths' Report following many visits to review the
original data held by the North Staffordshire Hospital NHS Trust and freely made
available to me. I wish to emphasise the time scale of the studies (for the
neonatal study 1989-1993). The project on premature infants was undertaken
between 7 and 11 years ago when research governance in general was much
less rigorous than now. This response is only possible because most of the
documentation remains available. I acknowledge the difficulties for the Panel in
coping with this time scale.
I must also make it clear that I am writing this as an individual and that the
contents of this report are my responsibility and mine alone. I have attempted to
ensure that all of the facts presented here are true and encompass the whole
picture. They have not been selected or limited to create a misleading
impression of what happened. Any opinions in this report are also my own.
I would also wish to make it clear that I understand the difficulties for my
employer (the North Staffordshire Hospital NHS Trust) with respect to dealing
with the very well organised campaign designed to discredit all of my work and
thereby bring to a halt the child protection activities that I have been undertaking
for the last 15 years.
In writing this document I am addressing a situation in which children have died
or are living lives affected by disability. The parents of children who have
suffered in this way have a terrible burden and I do not wish to make this worse.
The aim of our research was to reduce death and disability and relieve the
suffering of prematurely born infants. When this study started there was a very
high incidence of death and severe disability in such babies, particularly those
born more than 10 weeks before their due time. National statistics for 1990-92
[at the time the introduction of surfactant was just beginning
to make an impact] shows that 55% of the babies born alive but weighing under a
kilogram at birth survived 28 days. Approximately half of them died after birth
without ever leaving hospital. More than a quarter of those alive a year after birth
had at least one permanently disabling problem (blindness, or deafness, cerebral
Prof Southall Page 1
palsy or severe developmental delay). The condition of broncho-pulmonary
dysplasia (chronic lung disease of prematurity) was and continues to be a very
distressing condition for these infants leading to considerable suffering during
periods of prolonged intensive care. Only through medical research will we be
able to make an impact on these major problems and we agree fully with
Professor Griffiths and his Panel that such research must always be of a high
standard and be well structured so that any potentially harmful aspects of the
care given are identified as early as possible.
Contrary to written or implied criticism in the Panel‟s report on research in
children at the N. Staffordshire Hospital:
the two studies being examined (a randomized controlled trial of
Continuous Negative Extrathoracic Pressure (CNEP) in neonatal respiratory
failure and measurements of the effects of airway hypoxia in infants) were
the neonatal CNEP study for ethical reasons employed a scoring system
to facilitate a sequential study design. There was external input, peer review
and stress testing of the score.
there was an effective management system and good documentation for
the nursing input to the neonatal project was experienced and professional
The appointed clinical nurse specialist collected comprehensive data, filed it
carefully, prepared high quality manuals for clinical staff and a booklet for
the study team included a Chartered Statistician. He was closely involved
in developing the protocol and in monitoring the sequential analysis in the
consent was obtained from a parent of all infants prior to their
randomization in the neonatal study. A logbook of the randomization process
was kept but not viewed by the Panel.
parents in both studies were given every opportunity to refuse to enter
their babies in the studies (29 of 248 (12%) of families in the neonatal study
and 27 of 62 (44%) families in the hypoxia study declined). Family wishes
were respected and the parents were supported if they wished to withdraw
their babies during the studies (two did).
examination of the documentation revealed no evidence of treatment
being instituted despite “vigorously refused consent”.
responsibility for consent in the neonatal study was held by the clinicians
caring for the patients in the neonatal unit in Stoke. I was not involved in
seeking consent and therefore parents were not “subjected to the natural
enthusiasm of the researcher” and “…unduly pressured by this”.
Prof Southall Page 2
comprehensive records have been retained and follow up for problems
such as developmental impairment is entirely achievable.
the Scientific Merit Committee did peer review the hypoxia study.
the aims of the hypoxia study were clearly documented and not confused
as stated in the report.
there was no missing information on the hypoxia study; all of the
documentation is available for the study in Stoke, including copies of all the
a randomized controlled study of CNEP in bronchiolitis had been
undertaken in North Staffordshire Hospital at the same time as the neonatal
study; a project the Panel did not acknowledge.
there is a substantial evidence base to support the use of CNEP in infants
and children with a variety of medical problems (see references).
the medical evidence completely refutes the allegations that CNEP
caused brain injury in two patients with bronchiolitis.
The issue raised about allegations of forged consent is impossible for me to
address. It requires a forensic analysis of the consent forms, which are all
available, together with a questioning of the staff involved in the consent process.
I welcome such an investigation.
Contrary to the stated aims of the report, there are 9 direct and 8 implied
criticisms of me as an individual.
There was no recognition of the possibility that the complaints about the research
of which I was a director could be related to a wide-ranging campaign to discredit
the child protection work that I was, and many other paediatricians are,
undertaking. The Panel's exploration of fabricated and induced illness in children
was superficial and outside the terms of reference for their inquiry.
The Griffiths' Panel report  makes extensive criticism of the performance of a
randomised controlled study of negative extrathoracic pressure ventilation in
neonatal respiratory failure , particularly in relation to the administration of the
study, the obtaining of consent and the records kept. The report has been used
as a basis for recommendations on the governance of research. However, the
Griffiths' Panel has made a number of inaccurate statements regarding the
CNEP and hypoxia studies. These need to be corrected to ensure scientific
probity, to maintain the professional reputations of all those who contributed to or
reviewed the research work, and to help prevent the unjustified, or inappropriate,
erosion of public confidence in beneficial therapeutic innovations and in the value
of clinical research in vulnerable subjects such as critically ill infants and children.
The Panel provided little indication of the information they would require to the
members of the research team that they interviewed. The Panel were not given
Prof Southall Page 3
most of the available and relevant documentation, according to a copy of the
information provided to the Panel that was given to me by the Trust in May 2000.
The Report does not state which documents have been examined. The Report
does not provide a complete list of individuals who were interviewed. Drafts of
the report when completed by the Panel were not shown to the members of the
research team who were criticised to ensure that they could respond to issues of
fact before it was made public. I do not understand why it took 8 months after
the conclusion of the interviews with research staff for this Report to be made
It was essential for the Panel to ensure that they had access to complete
information so that their report would be accurate. MHCN Ltd. should carry some
of the responsibility for this. Based on the inaccuracies in the report, I conclude
that documentation was either not supplied or was not examined in sufficient
The panel must have been aware that at least some of the complainants they
interviewed were part of a campaign, to discredit me and other health
professionals involved in the detection of life-threatening child abuse (in
particular illness induced or fabricated in a child). This work involved the
detection by covert video surveillance of intentional suffocation, poisoning, and
other forms of extreme cruelty in 33 families, in a proportion of which previous
children had been killed by their parents. It is extraordinary that the Panel made
no mention of this in their report.
This campaign is led by Mrs Penny Mellor, (whom the Panel interviewed,
producing in their report half a page of points she had made about research
governance) and Mr Brian Morgan, a freelance journalist. I do not know whether
the Panel interviewed him or received evidence from him. I informed the Panel
of the existence of this campaign against me and of the address of the website
heavily contributed to by Mrs Mellor (a self proclaimed child advocate who stated
that she represented a number of families) and Mr Morgan (www.msbp.com).
At approximately the same time that the Griffiths' Panel was established at the
beginning of 1999, allegations about my child protection work, my personal
conduct and my research were formulated in an official letter by a senior
manager in the Trust to Mrs Mellor on 26th January 1999 and shortly afterwards
posted by Mrs Mellor on the Internet. This letter followed allegations made by
Mrs Mellor to a senior manager in the Trust, and entered in a written
contemporaneous file note as follows:
"PM (Mrs Mellor) believes DS (me) was abused by his mother. He does not like mothers.
He seems to go for the boys when they are a certain age. Probably at the age when he
realized he was being abused. Sado masochist. Thinks all boys are at risk from their
mothers. If you look at all the cases he has referred, most will be boys between 9-12.
Did he have long spells in hospital as a child which were mother induced?
Prof Southall Page 4
PM also thinks DS may also be into baby racketeering in Bosnia, with M Samuels. ……
PM thinks we have enough now to suspend DS"
Currently I am suspended by the Trust while they investigate these serious but
unsubstantiated allegations against me. Additional information relevant to this
campaign is given in Appendices 1 and 2.
My response to the Report of the Griffiths' Inquiry is divided into five sections: (1)
the trial of CNEP in neonatal respiratory failure (2) the study of responses of
infants to breathing 15% oxygen (3) the study of CNEP in bronchiolitis (4) the
process by which the Panel reached their judgements, and (5) other issues.
Excerpts of the Griffiths' report are typed in bold, with my comments in plain text.
Prof Southall Page 5
1. The Study of CNEP in neonatal respiratory
1.1 Peer Review of the Study Protocol
“There was no external peer review of the project because the funding
sources did not require it. It is worth contrasting this situation with the
extensive peer review which is a quality characteristic of research funded
by the MRC and the major medical charities.” [7.1.5]
“… if the trial had been subjected to external peer review then there could
have been several aspects of its design and operation that might have been
The Panel did not acknowledge that external peer review for the trial of CNEP in
neonatal respiratory failure had taken place, even though they had access to the
following information. External peer review was obtained as follows:
The Medical Research Council wrote on 22 March 1990 with a confidential
review of project grant application no G8928472 / SA. They described the
application as reflecting “research of high scientific merit, ie of such novelty or
timeliness and promise as is likely to make a significant contribution to
knowledge and / or clinical practice”. This rating fell just short of that needed
to provide funding.
the Clinical Research Committee of the Royal Brompton Hospital awarded a
grant for the CNEP study on 10 July 1989. This was after external peer review
by two consultants at the London Chest Hospital. The funding supported a
research registrar, Dr J Raine, for the first 2 years of the project. Dr Raine
obtained an MD thesis for this work.
an independent consultant neonatologist at the Royal Sussex County Hospital
(ie not from the research centre – the National Heart and Lung Institute – or
the two participating centres, N. Staffordshire and Queen Charlotte‟s
Hospitals) reviewed aspects of the scoring system used in the study.
Staff at the National Perinatal Epidemiology Unit were consulted. They
offered comments and advice on the design of the study and plans for follow-
Information about the study was registered and published in The Oxford
Database of Perinatal Trials
After the study had been completed a requirement of publication in
„Pediatrics‟ was that the Editor and three other experts with relevant
professional experience review the manuscript. The manuscript was
thereafter published in „Pediatrics‟ – one of the highest impact pediatric
journals in world, which has an 80% rejection rate of submitted manuscripts.
No criticisms were subsequently published in the correspondence columns of
Prof Southall Page 6
It is difficult to remember events 11 years ago but the following selection of
information from Dr Samuel‟s diaries in 1989 and 1990 illustrates how widely
we discussed the work we were undertaking with negative extrathoracic
From the diary of Dr Samuels:
Feb 1989 - lecture on CNEP in Leicester Royal Infirmary
March 1989 - lecture on CNEP at Royal College of Nursing meeting
March 1989 - lecture on CNEP at Homerton Hospital
April 1989 - presentation at Annual BPA meeting in York on CNEP in BPD (10 minute talk in Respiratory
June 1989 - Trent Region Neonatal Study Day in Barnsley, discussion regarding CNEP
June 1989 - Chronic lung disease symposium in Hillingdon Hospital
July 1989 - World Congress of Pediatrics in Paris - Dr Samuels attended and gave 2-3 talks; one on CNEP
September 1989 - Dr. David Easa from Hawaii visits National Heart and Lung Institute to discuss CNEP
October 1989 - CNEP presented by Dr Southall at a neonatal meeting in Norwalk, USA
Dec 1989 - Meeting in Washington DC on BPD - Dr Samuels presented use of CNEP in BPD and
randomized controlled trial
January 1990 - presented use of CNEP and INEP at Wolfson Centre, Great Ormond Street Hospital
January 1990 - talk at St Thomas' Hospital on CNEP and INEP
February 1990 - talk on CNEP and INEP to anaesthetists at Great Ormond Street Hospital
March 1990 - talk on use of CNEP in Ann Anderson Lecture Theatre, John Radcliffe Infirmary, Oxford
March 1990 - talk on CNEP and INEP in Rosie Maternity Hospital, Cambridge
APRIL 1990 - neonatal CNEP STUDY STARTS IN NORTH STAFFS. HOSPITAL
1.2 Scoring System used in the study
"A Lancet editorial (entitled The Perils of Paediatric Research) dated 27
February 1999 posed a number of questions. …Clinical progress among
neonates entered into the study was assessed by an unvalidated and
arbitrarily constructed clinical outcomes score". [7.1.1]
The Review Panel asked Professor Southall how the score had been
constructed and were told that the research team had devised it after
discussion. There had been no external input, no peer review by experts,
and there was no sensitivity analysis". [7.1.3]
The principal aim of the trial was to determine whether the use of CNEP
produced less chronic lung disease, a common respiratory complication of being
born very prematurely. The trial confirmed this benefit. Infants treated in the
standard way, without the use of CNEP, needed oxygen therapy for almost twice
Prof Southall Page 7
as long as CNEP treated infants. The outcome score was the primary object of
the study only in a strictly formal statistical sense. The unsigned (anonymous)
Lancet editorial and also the members of the Griffiths panel missed this point
Babies in this study belonged to a very high-risk group for death and disability. It
was not expected that CNEP would produce a dramatic impact on mortality
figures, and in fact the results showed that mortality in both the CNEP group and
the control group were within the national expectations for this high-risk group .
The frequencies of mortality and of cerebral ultrasound scan abnormalities were
not significantly different between the CNEP and the control groups.
In very sick individuals the tracking of relative efficacy and safety of therapies in a
comparative clinical study can be closely interwoven. For this reason it was
decided to use a sequential study design in which either an early indication of
unsafe treatment or early indications of benefit, using a combination of all critical
safety/efficacy parameters, would immediately stop the study. The design of the
study was strongly skewed towards early cessation in the event of evidence of
The clinical outcome score was constructed specifically to facilitate this
sequential study design. It was not intended to be anything other than such a
device. The alternative use of a combined score would have been to halt the
study if any one of the sets of criteria included in the score met an early stopping
criterion. This would have the effect of increasing the probability of an incorrect
early cessation of the trial to an unacceptable level. Moreover, all the measures
used would then be given equal weight. It is important to note that in a
retrospective analysis of the data there was no indication that separate and
simultaneous statistical tracking of mortality and of ultrasonically detected
cerebral abnormalities would have stopped the study. The score combined both
of these safety/efficacy criteria along with other important clinical outcomes.
It is true that the score was not validated as a measure of clinical outcome. This,
as stated above, was not its purpose. However it was far from arbitrary. The
selection of the items to include, and the weightings given to them, were the
subject of collaboration among six paediatricians from three centres, two senior
children‟s nurses and a consultant statistician (Mr John Alexander). The
statistician reviewed the weightings and then stress-tested the scale (a form of
sensitivity analysis) by providing the clinical members of the team with a number
of extreme examples of possible paired outcomes. These experienced clinicians
reviewed these to confirm that the weightings broadly related to the desired
rankings in those cases. Minor adjustments were made to the scale as a result.
Full details of the scoring system were provided to the external reviewers of the
project (see above).
There was a vast amount of clinical and physiological data documented in the
Prof Southall Page 8
study, and the key findings of the analysis of individual components of these data
were reported in the paper in the journal "Pediatrics". No clinical interpretation in
this publication was based on the score values.
Had it been our (the Research Team's) intention to base the principal
conclusions of the study (relating to the value of CNEP) on the score, then a
validation exercise would have been undertaken to ensure that there was a good
correlation between the ranking of groups of clinical cases by doctors, based on
all available clinical data, and the score itself. This was not done because it was
not required to be done, given the limited role to be played by the score.
Nonetheless it was essential that the scoring system be considered by a number
of expert paediatricians to be broadly appropriate for ranking the outcome of
pairs of patients. This evaluation had to determine which items should be
included and how they should be weighted. This was done, as described above.
The risk we took in using this sequential design with the score as a stopping
criterion was that a weakness in the score might have led to an early termination
of the study which on some specific clinical criteria would have been considered
unnecessary, and that as a result the statistical power of the trial would have
been weakened. Ethically, this is preferable to the opposite error, of not using a
sequential design and thus submitting some children to an inappropriate regimen
when there were already sufficient indications in the study to prevent this.
1.3 Organisation And Set Up Of The Study
"CONCLUSIONS ON THE OPERATION OF GUIDANCE COVERING CLINICAL
RESEARCH Local arrangements - Professor David Southall and Dr Spencer
were named as the principal investigators on the LREC application for the
CNEP study … Professor Southall … took relatively few steps to make sure
that the project was in fact run in accordance with the research application.
There was, as far as the Review Panel has been informed, no system of
management and documentation in place that would have given him
assurance that all was running properly. [9.2.6; 9.2.7; 9.2.8]
Obtaining Consent [9.3]
Professor Southall seems to have assumed that the obtaining of consent
was proceeding as intended but does not appear to have ensured that each
member of staff who might be involved in the project was trained or
supervised to ensure that they were doing what the research protocol said
they should. In our interview with Dr Spencer he made clear the steps that
Prof Southall Page 9
he took to ensure that patients he spoke to were properly informed, but he
made no mention of an overall supervisory mechanism" [9.3.1.]
The project protocol was organised as follows by a group of investigators during
the course of many meetings and following considerable correspondence. At the
time of development of the project and during the majority (63%) of the study in
Stoke, Dr Samuels and I were based at the National Heart and Lung Institute
(NHLI) and the Royal Brompton Hospital (RBH). It was agreed that the following
process would govern the conduct of the study.
1. It would start in Queen Charlottes Hospital (QCH) on 12/10/89. It finished
there 15 months later. In this hospital, the research fellow, Dr J Raine
would be responsible for seeking consent. Doctors Modi and Harvey,
Senior Lecturers and Consultant Paediatricians at Queen Charlotte's
Hospital would obtain Research Ethics Committee (REC) approval and
ensure that all clinical issues with the project would be managed in their
tertiary neonatal intensive care unit. Dr Raine also collected the
necessary data, for example on oxygen requirements of the babies,
cerebral ultrasound scans etc.
2. Dr A Spencer and Dr D Brookfield were the senior paediatricians
responsible for the clinical care of neonates in their sub-regional neonatal
intensive care unit in Stoke. Both were consultant paediatricians and Dr
Spencer was also an honorary senior lecturer at the National Heart and
Lung Institute. It was agreed, as would be standard practice in
multicentred studies, that they would have the following responsibilities:
to ensure that the doctors in their unit would seek consent in an
appropriate manner. It was recognized that this was an important
responsibility and that the consent forms would be carefully stored. All
consent forms were available 7-11 years after they were obtained.
to ensure that the process of assessing entry and exclusion criteria for
the study was understood by the professionals seeking consent and
to monitor the clinical progress of the patients in the study and ensure
that there were no apparent harmful side effects of the treatment
to ensure, in their capacity as experts in neonatal medicine, that any
new validated therapies that became available during the period that
the study was in progress were not withheld because of the study (an
example is the use of surfactant).
to ensure that the best possible quality of care was given to infants in
the study regardless of which limb of treatment they were assigned to.
Prof Southall Page 10
to ensure that all parents received the "Parent Information Sheet" a
requirement of the Local Research Ethics Committee (LREC). This
was a page of A4 text, which explained:
the technique of negative pressure ventilation
the need to find out whether or not it was better than usual
that the study involved random allocation of each baby to a
that irrespective of inclusion in the study, babies would receive the
best possible treatment
that their baby could be withdrawn from the study at any time.
3. There was regular contact between Drs Spencer, Dr Brookfield, Dr
Samuels and me about the conduct of the trial, and particularly between
Dr Spencer and me.
During the trial, Dr Spencer produced a clear plan concerning the
adoption of surfactant treatment. This was discussed by members of the
team, including the statistician, John Alexander, and was adopted.
4. There were frequent meetings and telephone contact before and during
the study with our statistician (see below)
1.4 Comments on nursing expertise
"The Review Panel is in no doubt that the nursing sister assigned to
the project worked many more hours than she was contracted for, but she
did not appear to have been provided with a protocol or system of
documentation which made sure that everything was complete for all
patients. Nursing staff and the sister in particular had not been trained or
had adequate research experience for the job that they were being asked to
do. Supervision from the researchers was inadequate, and the staff was
poorly supported by Trust nursing management". [9.3.5]
1. Teresa Wright was the clinical nurse specialist assigned to the project. She
was a skilled and trained co-investigator in the study, and the third of 11 authors
on the final report. Prior to this study, she had been responsible for collating data
in a regional audit into perinatal deaths. She had been a lead researcher into
setting up a randomised controlled study of CNEP in bronchopulmonary
dysplasia in 4 regional neonatal units and had assisted in research comparing
Prof Southall Page 11
the effectiveness of two milks for premature infants. Teresa Wright attended a
number of teaching sessions at both the Royal Brompton and Queen Charlotte's
Hospitals before the study began in Stoke. She was involved in developing the
protocol and the system of documentation for each patient entered into the study.
She collected the following information on each patient and assigned it to a
dedicated folder for each patient in the study:
consent forms (1 sheet)
clinical and physiological record sheets for computer logging (9 sheets)
questionnaires on aspects of the study to be filled in by parents when their
infant had completed the study (3 sheets, including freehand text by parent).
Parents of infants who died were not given this questionnaire.
copies of cerebral ultrasound scans, where performed
in a proportion of cases, data for a temperature regulation study (2 or more
Despite the passage of 7-10 years, the vast majority of this information is
available on all of the patients. Examination of these data will show that Teresa
Wright was professional in her collection and filing of this information on each
child in the study.
2. Teresa Wright was in contact at least once weekly, often more frequently
with either Dr Samuels or me about aspects of the study. For the first 3 months of
the study in Stoke, she was in frequent (every 3-4 days) contact with Dr Raine
and on >4 occasions he visited Stoke to check on the progress of the study.
Teresa Wright usually contacted either Dr Samuels or me whenever a child
became included in the study following the randomisation procedure to ensure
that all was well with respect to the conduct of the study.
3. At the time of the study, it was practice on the neonatal unit that any infant
requiring intensive care would be cared for only by nurses holding the ENB 405
or 409 certificates or by experienced Registered Sick Children's Nurses (RSCN).
Teresa Wright extended this rule to include all infants in CNEP, whether or not
they were intubated.
4. Before the start of the study in Stoke on 2/4/90, CNEP had been used in N
Staffordshire Hospital in neonates with chronic lung disease of prematurity that
were failing to respond to standard therapy.
5. Clinical Nurse Specialist Teresa Wright was funded (initially 1/3 from North
Staffordshire Hospital and 2/3 from the National Heart and Lung Institute, and
from June 1992, fully by the North Staffordshire Hospital; including part funding
from a research grant) in an H graded post to undertake the day to day running
of the study. Her full job description is available and outlines her clinical,
administrative, educational and research responsibilities, including a substantial
commitment to the collection and analysis of data from patients in the CNEP
Prof Southall Page 12
study. She was closely involved in the nursing management of patients in CNEP
and prior to the onset of the randomized controlled study had helped to develop
many of the techniques needed to ensure thermoregulation and neck protection.
She and some of her colleagues thus had considerable expertise in the use of
CNEP in preterm neonates before the study started.
6. Prior to the beginning of the study Teresa Wright made clear her objections
to randomizing individual members of multiple births possibly to different
treatments. She felt that it might be difficult for parents if one twin for example
was seen to be doing less well in one treatment than the other twin receiving
different care. As a consequence of her input, the protocol was changed so that
multiple births were randomized to the same arm of treatment.
7. Teresa Wright was also responsible for preparing materials for nursing and
medical staff on the use of CNEP by staff. She used a detailed and illustrated 34-
page document outlining „The Use of Negative Pressure Ventilation in Infants
and Children‟ for health professionals. This was prepared in conjunction with Dr
Samuels and me, and a Clinical Nurse Specialist from the Royal Brompton
Hospital. This was dated April 1990 and available on the neonatal unit
undertaking the CNEP study in N Staffordshire Hospital from that time. It
covered the following issues:
history of negative pressure ventilation
principles of operation
possible advantages and disadvantages
assessment and monitoring
components of the negative pressure system
making the neck seal and protector
care of the chamber
care of the patient, including
o assessment of nutrition and hydration
o oxygen assessment
o measurement of SaO2
preparation for extubation
preparation for discharge home
equipment and supplies
problems and potential side-effects
This document was not given to the Panel by the Trust or acquired by MHCN.
Prof Southall Page 13
8. Teresa Wright also wrote a number of sections on CNEP for the 'Neonatal
Guidelines Folder' that had been prepared for medical and nursing staff by Dr
Spencer and Dr Brookfield.
9. At the end of 1991, Sister Kate Lucking, a senior nurse on the neonatal unit
supported by Teresa Wright and a research registrar, undertook a small study of
consent utilizing a questionnaire. As a consequence of this study and to further
improve communication with parents, they prepared a 14-page folder entitled
„Negative Pressure Study – Information Booklet for Parents‟ and this was
available in Stoke from February 1992. This folder included 7 colour photographs
and included sections as follows:
negative pressure treatment
handling your baby
weaning from negative pressure
refusal of consent
copy of consent form
parent information leaflet (available to be taken away by parents)
(It should be noted that in relation to comments made elsewhere in the report
that none of the above literature uses the phrase “gentler, kinder treatment”).
This folder was not given to the Panel by the Trust or acquired for it by MHCN. I
drew the attention of the Panel to its existence. Professor Griffiths was reported
in the minutes of my meeting with the Panel on 23/7/99 to have seen this folder.
He was also reported to have stated that the folder did not describe standard
treatment (this is incorrect).
1.5 Statistical issues
"The defect in the research framework is that it should not have been
Professor Southall who was making the decision as to the size and shape
of the study on his own but it appears to have fallen to him by default. In
effect the combination of a slightly complacent LREC, an enthusiastic and
assertive lead researcher and a vacuum in research governance in the
Trust led this trial to run in a less than adequate way". [9.4.4]
1. The research framework was not such that I decided the size and shape of
the study on my own. The study management has been outlined earlier. The
Prof Southall Page 14
study was under the supervision of a chartered statistician who helped determine
the size and shape of the study, and these details were all provided for peer
review of the project before starting the study. Mr John Alexander, is a Chartered
Statistician who had been an adviser to our research at the National Heart and
Lung Institute since 1982. At the time of the study he had no connection with any
companies involved in paediatric ventilatory support. He is currently Deputy
Director, Scientific Support Services, Roche Bioscience, Palo Alto, California.
Mr Alexander undertook regular monitoring of the project. He received
completed data forms as soon as possible after the infants were discharged from
the neonatal unit, reached 56 days of age or died. This was to ensure that we
had not crossed a decision line with regard to completing the study. Regular
meetings were held with Mr Alexander before and during the study, minutes of
which are available. Mr Alexander was not invited to give evidence to the
Griffiths Inquiry, but on hearing of the Inquiry, he wrote from his home in
California on 23 March 2000 to Professor Griffiths to explain statistical issues of
the study. In summary, his letter confirmed:
he took responsibility for the statistical aspects of the study design
the research team would not (and arguably should not) have been aware of
the full statistical rationale and detail underlying the administrative
that randomization was carried out at an administrative centre, remote from
the treatment centre for the first two thirds of the study and that after the move
of my research team to Stoke he was satisfied that bias would not be
introduced if the randomisation procedure was continued and the Academic
Department of Paediatrics maintained clinical independence from the
management of the cases in the trial
that staff seeking consent would not have access to data which allowed them
to guess the treatment to be allocated
that Dr Samuels and I were extremely diligent in contacting him to determine
corrective action in relation to problems that occurred in the randomisation
process. He has cited two specific examples where, in his experience, most
researchers would have taken a local decision without seeking advice, but
where we chose not to do so. The examples referred to the issue of multiple
births when it first arose and the question of how to proceed after accidentally
starting a new pair when an envelope for that pair was already open.
The Griffith's report contained some valuable comments by Professor
Lilford on the management of research see [15.4.7]
Prof Southall Page 15
The introduction of the concept of Data Monitoring Committees (DMC) in large
trials in recent years is to be welcomed, and had this been the practice at the
time we would have welcomed the operation of such a committee in our trial
It should be recognised that, even without the benefit of an independent DMC,
we incorporated stopping levels that differentiated between “harm” and “benefit”
in our trial.
Also, Mr Alexander confirmed in writing to the Panel that randomisation was
absolutely separate from the researchers seeking consent to avoid any possibility
of randomisation bias (see below).
1.6 The process of recruitment to the study and randomising treatment
"An alternative explanation [for parents not recollecting giving consent to
randomisation] might be that in some cases the ward staff sought
randomisation from London and then got consent if CNEP was involved.
Professor Southall was able to describe to the Review Panel in great detail
the way that he and Dr Samuels carried around the randomisation folder
but the subsequent process is much less clear. He appeared to have no
way of checking that consent had been obtained properly". [9.3.4]
The Trust told me I was not to present the randomization logbook for
examination, despite the Panel's request to me that this be made available. I
informed the Panel of the Trust‟s directions to me. An account of this is given in
a transcript of the proceedings returned to me after editing by MHCN on 15 th
November 1999. I will quote from this document : "Professor Southall said that
he was happy to send a report to Professor Griffiths on exactly how
randomization was undertaken. (on a later date Professor Southall stated that
unfortunately permission was subsequently refused by the North Staffs Hospital
Trust and this information was not made available)."
Randomisation was decreed to be undertaken as follows:
1. the appropriate professional on the neonatal unit determined that the
newborn infant with respiratory failure satisfied the eligibility criteria for
inclusion in the study, and did not meet any of the criteria that would
require exclusion from the study.
2. the professional informed the parents about the study using a standard
consent form and patient information leaflet. The professional completed
a consent form with the parent(s). From February 1992 he/she would also
go through the information booklet described above with the parent(s).
Prof Southall Page 16
3. if consent was given to participate in the study, the professional
telephoned either Dr Samuels or me. We were on a rota 24 hours per day,
7 days per week to allocate infants to a treatment group in the study. We
verified by direct questioning that the patient met eligibility criteria and
asked further questions to determine which group of clinical criteria the
patient fulfilled before undertaking randomization to a treatment within that
4. Dr Samuels or I checked the records to determine if there was an opened
envelope for a pair in this group. If so, treatment was assigned to the
second of the open pair. If not, the next envelope was opened and the
treatment for the first pair in that envelope was allocated.
5. Whenever randomization was requested, Dr Samuels or I recorded the
following information on a log sheet:
date and time of call
name of clinician calling
who answered the call (Dr Samuels or I)
patient‟s gestational age, inspired oxygen requirement and whether or
not they were intubated (this determined which group they would fall
into, from a table in the study protocol).
group number (from 1 to 15) that the infant was allocated to
study pair number, and whether they were the 1st or 2nd infant of a pair
information concerning any subsequent withdrawal of the baby from
the study was also noted
The log sheets listing every patient randomised in the study are available; 11 of
13 sheets refer to the patients in Stoke.
The logbook of which doctor or nurse on the neonatal unit sought randomisation,
from whom it was sought, and the date and time of randomisation was not
examined by the Panel.
Thus the hypothesis that staff might have sought randomisation and then
obtained consent was not only incorrect but inappropriate. It was always possible
to track consent and randomisation. If it is known which parents are claiming that
they did not give consent or that consent was fabricated (numbers and details
are not given in the Report), it is possible, from a combination of data from this
log book, from the original medical/nursing records, and from the consent forms
to find the name of the professional that sought consent and the approximate
time and exact date on which consent had been received.
Had the process of seeking consent been undertaken after randomisation, as
suggested by the Panel, then the "refused consent" patients would show as
subsequent deletions in the randomisation log book. No such items appear.
Prof Southall Page 17
Most of the randomised pairs were completed, although there were some
incomplete pairs, as expected, at the end of the study.
The forms for patients in which consent was refused were also sent to our
statistician at the end of the study in a different format. These children received
standard therapy and no further involvement in the study occurred. Mr Alexander
would confirm that these cases were distinct from those included in the active
part of the study.
"There should have been an effective process in place which ensured that
… there were clear opportunities for parents to withdraw, and that they
were given information to take away that made it absolutely clear that their
children were in a trial in which their care had been allocated at random".
One set of parents in Stoke did withdraw their baby from the study. All parents
were given an information leaflet, which explained that treatment was randomly
allocated, and that withdrawal from the study was possible without detriment at
any time. In addition, from February 1992, parents were shown an information
booklet, outlined above, which explained the study, which included one page on
„Refusal of Consent‟. This included the following statement "Should you decide to
consent to the trial and then later change your mind, you may withdraw your
baby from the trial." It also included the statement "All babies treated on this unit
whether or not they are part of the trial will receive the best possible treatment we
1.7 Consent Issues
"What was totally unacceptable to those interviewed was the apparent lack
of adequate explanation, of choice and of consequent properly elicited and
recorded consent, and of involvement in later decision making. [4.2.1]
Some of the consent forms could not be found, some not been signed by
parents or by doctors, others appeared to have signatures that may not
have been those of the parents. Of those who remembered signing,
several remembered agreeing to CNEP but did not recall discussion of
randomisation. It is difficult to escape the conclusion that the process was
Prof Southall Page 18
not managed consistently and no system of management and
documentation was in place to prove that it was". [14.3.8]
The issues regarding explanation and choice have been answered above. Since
the issue of consent was central to the Griffiths inquiry, I report below the Trust's
audit, followed by my own analysis undertaken after publication of the Report.
AUDIT BY THE TRUST
The North Staffordshire Hospital‟s Clinical Audit Department performed an
audit of consent documentation for the CNEP study (dated March 1999). I
was not involved in this analysis. The report is available at the Trust.
MY OWN AUDIT OF CONSENT
On 8th and 12th June 2000, under the direct supervision of and with the support of
a senior manager at the North Staffordshire Hospital, I went through the study
folders, log-book and some of the medical records in an attempt to determine
whether or not I could identify which doctors or nurses had sought consent.
The log-book revealed that 3 patients were withdrawn from the study following
randomization (two had congenital heart disorders and one had septicaemia).
One of these had been randomized to standard treatment and two to receive
The parents withdrew one patient after randomization.
All patients had consent forms but in 6 sets of twins only one form was
completed for both patients.
There was a parent signature in all but one of the consent forms. The remaining
one was a verbal consent taken from the father by the senior registrar and
documented in detail on the consent form. In one consent form, we could not be
certain that there was a parent signature rather than a printed first and second
name for the father.
Of the forms with parent signatures, in 16 it was not possible to determine
whether it was the mother or the father who gave consent.
Of the forms with parent signatures, in 21 a parent did not date their signature. It
was noted that parental signatures were sometimes placed in the „parent‟s name‟
Prof Southall Page 19
In total, 34 professionals were responsible for seeking consent (12 registrars, 1
research fellow, 18 senior house officers, 2 consultants and one clinical nurse
specialist). I did not seek consent. In all but 10 of the consent forms there was a
signature of the professional present. Examination of the log-book, medical
records and nursing notes provided documentation which revealed the name of
the professional who sought consent in these latter cases as summarized below.
Three doctors were named on the consent forms but their signatures were not
present. Examination of the log-book showed that in 2 of these 3 cases, the
name of the professional on the consent form was the same as the name of the
doctor who telephoned for randomization. In the remaining case, a nursing sister
telephoned for randomization. Examination of the medical records in this last
case did not provide any further evidence that it was the doctor named on the
consent form who had sought consent.
In 7 there was no record of who had taken consent on the forms. Examination of
the log-book in these 7 cases showed the following; in one case a named
registrar rang to randomize the patient; in 4 cases a named research fellow; in 1
case a named senior house officer telephoned to randomize the patients.
Examination of the original nursing/medical records on these latter 6 cases
confirmed that the names of the doctors ringing for randomisation were
compatible with the names of the doctors in the medical records who had sought
consent. In the final case a clinical nurse specialist telephoned to randomize the
patient but, on examination of the patient's medical records, it was recorded that
a named registrar had sought consent.
The randomization logbook thus provided in combination with the medical and
nursing notes and the consent forms a means of tracking the professionals
involved in seeking and receiving or not receiving consent, in particular those
who did not correctly complete the consent forms.
The signature of the professional was not dated on 23 of the consent forms.
However, examination of the log-book showed that in all but one of the patients,
the date and time of randomization was completed. In the remaining case, the
date but not the time of randomization was entered.
The Panel did not comment on the fact that there were refusals of consent from
12% of eligible parents and 1 withdrawal from the study in North Staffordshire.
This shows that parents felt able to refuse to enter their babies in the study.
Assessments prior to randomized treatment allocation were due at around 2
hours and 4 hours of age. As agreed in the study protocol, CNEP was begun in
all infants randomised to it after the 4 hours assessment had taken place.
Sometimes consent, randomization and the initiation of CNEP, if this had been
selected, were undertaken some hours after the assessment at 4 hours of age
because of operational difficulties (for example in locating the father or because
Prof Southall Page 20
the medical condition of the mother did not permit the seeking of consent).
CNEP was actually started at between 4 and 10 hours after birth.
"There were allegations that some consent forms, purported to have been
signed by individuals had not been so signed". [4.2.3]
"The consent process
One parent claimed that the “treatment” was instituted in spite of the fact
that she had vigorously refused consent". [14.3.3]
As this information is available, the panel should have tested these allegations by
reference to the appropriate records. Referral to the randomisation logbook and
the research files would have identified the clinicians who sought consent from
these parents (see above), and this clinician's evidence could then have been
tested against the allegations.
Within 4-6 months of the neonatal CNEP study beginning in Stoke, a poster was
placed on each incubator or CNEP chamber indicating which limb of the study
the baby had been placed in. This poster was in the form of a teddy bear picture
that stated either "I am in the CNEP trial receiving standard therapy". (Figure 1)
or “I am in the CNEP trial receiving CNEP therapy”.
During the first few weeks of the study there was a display board on the neonatal
unit describing the project.
Prof Southall Page 21
"It was reported that the information given to parents, both written and
verbal, was incomplete, difficult to understand and inconsistent". [14.3.4]
"The difficulty of understanding and giving a valid consent at a time of
severe physical, psychological or emotional stress was raised by a number
of those interviewed. Apart from the stress of the moment, parents were
also subjected to the natural enthusiasm of the researcher and parents
later felt that they had been unduly pressured by this". [14.3.6]
The strong implication is made that the “natural enthusiasm of the researcher”
unduly pressured parents into consenting to CNEP. However, as the Panel
knew, Dr Samuels and I were not in Stoke for the first two and a half years of the
project, and when we did move there, we were not involved in seeking consent.
This ensured separation of the administrative and treatment aspects of the study.
I acknowledge the stresses on the parents of a newborn infant who is critically ill
and suffering respiratory distress. However, requirement for consent in such
circumstances is unavoidable, whether this is for research reasons or for a
procedure considered necessary to help the child. There is no single best
method for obtaining consent in these circumstances and I welcome debate on
this extremely important issue.
"We met several patients who had a clear recollection of being asked to
allow their children to have CNEP but had no recollection of giving consent
to randomisation in a research project".[ 9.3.2]
"Members of the research team were very ready to tell us that parents were
often poor at remembering what happened and recalling what they were
told. In effect they were asking us to believe that these parents could not
remember consenting to randomization but did remember that CNEP was
"a kinder gentler treatment' ". [9.3.3]
The panel commented that individual parents remembered that CNEP was a
“kinder gentler treatment”. The Panel should identify which member(s) of the
research team, if any, used this phrase.
Prof Southall Page 22
There are research data that show that parents‟ recall for consent undertaken at
the time of critical illness in their child is poor. The Panel does not indicate how
many parents it interviewed for the purpose of its inquiry, nor did it provide
information on how it selected the parents it did interview. The Panel did not
appear to have spoken with parents who had written to the Trust or other bodies
praising the process of the study.
The Panel did not acknowledge the questionnaire given to the parents of 173
infants (to avoid adding to their distress, parents of infants who had died were
excluded) on completion of the CNEP study about parent-infant bonding. This
questionnaire was accompanied by a letter from Dr Raine and me at the National
Heart and Lung Institute. Usually it was given around the time of discharge from
hospital; in some instances it was posted. It contained within it the following
"Dear [NAME] OF PARENT(S),
You will remember that shortly after [NAME] was born you kindly agreed to enroll
him/her into our study comparing negative pressure respiratory support with
standard treatment. As part of this study we have devised a questionnaire which
attempts to compare the effect of these two methods of treatment on the way you
were able to relate to your baby."
The questionnaires from 137 (79%) were returned: the additional freehand
written comments from 22 parents showed that:
13 were positive about the study, nursing care and use of CNEP
5 were neutral, with 2 of these giving practical advice about the CNEP
4 parents were negative about the information given or care received
(including at least one CNEP treated infant and one control infant)
1 family supported CNEP but complained about levels of nursing care on the
neonatal unit for critically ill babies.
The family that raised the initial complaint about consent in the lay press (Mr
and Mrs X) completed a questionnaire and in it wrote a free text statement
supporting the use of CNEP.
Finally, and of further relevance to consent, the Panel did not acknowledge a
contemporaneous questionnaire for a pilot study on CNEP in bronchiolitis
undertaken in Stoke at about the same time (see below for details).
1.8 Follow Up and Side Effects
Prof Southall Page 23
"Some of the apparent side effects, like brain damage, that have caused so
much distress to parents could not be reliably assessed at such a short
period of follow up. [7.1.7]
I agree with the Panel. The initial study used cerebral ultrasound scans as a
marker for brain injuries or abnormalities, and this did not suggest that
neurological injuries were different between the two groups. However, good
quality information on each of the patients in the study was kept available for
later follow up, should that be considered an appropriate exercise. The fact that
7-10 years on the data were available satisfies this need. The information in the
randomization log book and the folders of data allow tracking of hospital records,
even if, as often happens, the names of children are changed by parents.
The Panel also failed to comment on studies into the effects of CNEP on cerebral
function performed before and during the time of the study [5-7]. These studies
have been published in peer-reviewed journals and did not reveal any deleterious
effects. They examined middle cerebral artery blood flow velocity as determined
by Doppler ultrasound in one study, and two others in which cerebral blood
volume and oxygenation and the effects of the neck seal were measured by near
When the Review Panel asked the team about the length of follow up it was
given to understand that this short follow up was determined by the
amount of funding available". [7.1.7]
The primary aim of the study was to determine whether CNEP could reduce the
incidence of chronic lung disease of prematurity, and the follow up was designed
with this in mind. Neurological outcome was addressed in the short term by
ultrasound scanning. To determine the adequacy of neurological outcome in the
longer term would have required a separate study of developmental progress
some years after the study had finished. Funding organizations would not usually
have placed funds "on ice" for a number of years so that such an analysis might
be made and thus a long term follow up study would have needed separate
The neonatal research community has lobbied intensively over the years to try to
ensure that both funding and logistic support for follow up after neonatal trials are
in place nationally. However, the reality, at the time of this study, was that there
was little support for this. Nevertheless, the way in which the data were collected
Prof Southall Page 24
and retained permits a follow up study now that, I would fully agree with the
Panel, is undoubtedly a valuable way forward.
"One of the strengths of a good controlled trial is that it is possible, if the
records have been retained, to go back and attempt further follow
Comprehensive records have been retained and further follow up is perfectly
feasible. This would be scientifically valid and superior to the „audit‟
recommended by the panel.
"At no time did the research team suggest to the complaining parents that
this might be a way of resolving their problems".[7.1.8]
The study ended in October 1993. Mr and Mrs X complained about the study in
April 97. No complaints were received prior to this date. They requested a
meeting with me and I expressed my willingness to agree to this to my line
managers at the Trust. However, this was not agreed by the Trust who handled
the complaint. I informed the Panel of this.
Prof Southall Page 25
2. The study of infant responses to breathing 15%
2.1 Peer review
"Confusion about the research question and hypothesis being tested is
one of the factors that an external peer review might be expected to clarify.
No such review was undertaken because once again none was required in
order to secure the necessary funds". [7.2.3, and see 7.2.2]
Before it started, this study was subjected to peer-review by the medical
committee of a registered charity.
In the light of the sensitive nature of the studies being undertaken, there was
vigorous review of the project by the LRECs of three hospitals (the Royal
Brompton, Crawley and N Staffs Hospitals). I was required to justify the project
to members of the LREC in person at both the Royal Brompton Hospital and at
the N Staffs Hospital (Panel's own statement [6.5.2]). At the Royal Brompton
Hospital, the proposed study was discussed with the chairman of the REC, a
world expert on hypoxia.
On 14th April 1992, I attended the RAF Institute of Aviation Medicine at
Farnborough to discuss the project with a number of experts on hypoxia. They
confirmed that our approach using 15% oxygen at atmospheric pressure was
more appropriate than attempting to study infants either in hypobaric chambers
or on airline flights. The differences between hypobaric and isobaric hypoxia
remain unresolved and this was discussed in our paper and in letters following
the publication of our article in the BMJ. [8-11]
During the time of the study we presented our incomplete data at the Northern
Paediatric Respiratory Forum (1995) for peer review. Again, prior to publication
in the BMJ it was presented at an International Meeting (Apnea of Infancy
Conference, Annenberg Center, California, January 1997 and at the Annual
General Meeting of the Royal College of Paediatrics and Child Health in 1997
. It was subsequently published in a peer-reviewed journal, the British
Medical Journal, along with a number of commentaries [8-9]. The editorial office
of the BMJ informed me that approximately 5000 papers are received each year
of which only 14% are accepted after peer review.
Prof Southall Page 26
"It could be argued that in North Staffordshire Hospital the Scientific
Merit Committee should have done this but they do not appear to have
done so. The Review Panel know that the LREC initially rejected this study
but it was finally approved, with some confusion remaining". [7.2.4]
The Scientific Merit Committee reviewed this project on two occasions. They
requested that a mistake in the calculations relating to sample size be corrected
and this was done.
This was known by the panel who also wrote: "However correspondence
indicates that the project was considered by the Scientific Merit Committee
on 8th June 1992, when a number of questions were raised. [Letter to
Professor Southall from Dr Mucklow Chairman of the Scientific Merit
Committee]. Professor Southall had submitted a detailed protocol to this
Committee. [Enclosure 12 project No 12 8/6/92.]". [6.5.1]
2.2 The design of the trial was confused
"Indeed there is some confusion as to what this trial was supposed to
There is no confusion. As described in the BMJ its "Objective" was "to assess
the response of healthy infants to airway hypoxia (15% oxygen in nitrogen)". The
differences between isobaric and hypobaric hypoxia were raised in the paper 
and in subsequent letters to the editor [10-11]. The Panel could have clarified
this if they had interviewed the responsible co-researchers (Dr Parkins, Mrs
Stebbens, Professor Poets or Dr O'Brien). The four key messages of the paper
are outlined in a box within the paper. 
2.3 Missing information
"The original letter and application form for this project were not available".
Prof Southall Page 27
The Panel could have requested all of the information relevant to this study and it
would have been produced for them. The original letter and research application
form are available. The patient information leaflets and all of the consent forms
for the N Staffordshire patients are available.
Had the Panel reviewed the consenting process for the group of healthy infants,
they would have found that parents (after random selection) were first
approached in writing for an "expression of interest" in enrolling their baby into
the research project. Twenty-seven parents replied that they were not interested
and 35 replied that they were. All parents received written and verbal information
at all stages of consent, and this included information making it clear that they
could withdraw at any stage of the study process. Some parents did decline at
each of the different stages of the project/consenting procedure.
Following our paper in the BMJ, an article was published anonymously in the
Bulletin of Medical Ethics in March 1998 (Editor Dr Nicholson). It was entitled
"Making babies breathe low oxygen" and criticized the ethics of the study. It was
the subject of legal correspondence between the Trust's solicitors and Dr
Nicholson. A later editorial in the same journal  in January 2000 stated "What
excuse is there for the failure of the GMC to have taken any action yet on the
unethical research undertaken in the Paediatric department of the North
Staffordshire Hospital: it has had a large and very thorough dossier of evidence
for nearly two years".
Following the first editorial in the Bulletin of Medical Ethics, the local newspaper
in Stoke published on the 11th May 1998 an article entitled "Expert raps 'risky' cot
death probe". The following day it carried another article entitled "I was misled-
baby test Mum". In a similar fashion to the mothers in the CNEP trial, this parent
alleged that "she would never have allowed her three-month old son [NAME] to
take part in the research if she had known the risks involved. …Mrs [NAME] and
ADDRESS said she only discovered [NAME"S] oxygen had been reduced when
the researchers wrote to her in March to thank her for her cooperation".
When I examined the study records of the consenting process it was
documented that this mother had known from letters she had received and
responded to about the lowering of inspired oxygen before consenting. The
mother was present with her child and the researchers throughout the
administration of 15% oxygen. It was not stated in the Griffiths report whether or
not this mother gave evidence to the Panel. It was recorded that Dr Nicholson
did supply information for which he was praised by the Panel.
Prof Southall Page 28
3. The randomised study and other research into
the effects of CNEP in bronchiolitis
3.1 Lack of research data on CNEP in bronchiolitis
"It is clear that there is no substantial evidence base in support of the use
of CNEP in management of bronchiolitis. A subsequent search of the
literature has shown that there have been no randomised controlled trials
which show that CNEP is more beneficial than conventional therapy for
This statement is incorrect. Between 1990 and 1993 there was a small
randomised controlled study of CNEP in bronchiolitis at North Staffordshire and
Royal Berkshire Hospitals. Twenty-six infants were recruited in Stoke and the
results were published in a peer-reviewed abstract in the European Respiratory
Journal . The data were also presented at an International Meeting (1994
European Respiratory Society). Subsequently, there was a second peer
reviewed abstract about CNEP in bronchiolitis published in Care of the Critically
Ill and presented at a meeting of the Paediatric Intensive Care Society . It
was after our published pilot study, when CNEP appeared to have been well
tolerated by infants and reduced their respiratory distress, that nurses felt it
inappropriate to withhold the treatment from infants with severe bronchiolitis,
particularly those with apnoea. At that point it became a routine method of
respiratory support in our unit.
3.2 Use of CNEP outside a randomised controlled trial
"The Review Panel feels that the lack of trial-based evidence for the
existing conventional treatment, already at that time preferred in almost all
other centres, could not on its own be used as a justification for embarking
on a new variety of treatment outside a framework of proper evaluation,
against the existing conventional treatment, in a trial setting". [11.2.3]
Prof Southall Page 29
The Panel seems to be under the impression that CNEP was used solely as an
alternative to positive pressure ventilation. This is incorrect – it was used as an
adjunct, with positive pressure ventilation being available if and when required.
CNEP is largely a technique applied by nurses to provide non-invasive support
to respiration largely in an attempt to reduce the likelihood of infants needing
intubation and positive pressure ventilation. Our clinical observational results
presented at the Paediatric Intensive Care Society meeting in 1997  and
again submitted for publication in 1998, show how CNEP can help in supporting
respiration (see below).
The evidence base for the use of CNEP for respiratory support [17-32]compares
favorably with the evidence base that exists for more conventional methods of
respiratory support 'for examples' intermittent positive pressure ventilation and
high frequency oscillation, used widely both in North Staffordshire and other
A medical report describing our experience of using CNEP in bronchiolitis was
submitted in 1998 to the medical journal Pediatric Pulmonology who asked us to
make changes to the paper and return it for consideration of publication.
Although the changes recommended were made and the paper was ready to re-
submit to the journal, I was advised against this in April 1999 by the Trust
because of the allegations made by Mrs Mellor about my research and the
ongoing Griffiths Inquiry. In summary this retrospective study analysed data
concerning children admitted to the North Staffordshire Hospital over a 39 month
period. It examined 56 patients admitted to the regional Paediatric Intensive
Care Unit at the hospital with severe bronchiolitis. 15 patients were intubated
prior to arrival in the PICU and 41 received negative pressure ventilation during
transport to or whilst in the PICU. Only 3 of these latter 41 patients were
subsequently intubated. During this same 39 month period, there were 1161
local patients (North Staffordshire) admitted to hospital with bronchiolitis. 17 of
these 1161 were admitted to PICU. However, in total only 5 of 1161 patients
were intubated (2 with congenital heart disease, 1 in theatre after undergoing an
operative procedure, 1 in the Accident and Emergency Department on admission
and 1 in the neonatal unit prior to admission).
3.3 The relevance of the study of CNEP in bronchiolitis to the aims of the
Prof Southall Page 30
My understanding of the remit of the Griffiths' inquiry was that it aimed to
investigate issues of consent with respect to research on children undertaken at
the North Staffordshire Hospital. It is surprising, therefore, that the Panel did not
ask me about the above randomized controlled study of CNEP in bronchiolitis
undertaken between January 1991 and March 1993. This study is highly
relevant to the Panel's comments on the use of this technique in this different
medical condition (see below).
At the end of recruitment to the first phase of this particular study, questionnaires
were sent to all parents seeking their views on the consent process, on CNEP,
on nursing care and on medical support. This documentation related to a
contemporaneous study, is highly relevant to the issue of informed consent, and
is preferable to relying on information from parents about a highly stressful time
ten years ago. Twenty-one questionnaires were returned (81% response rate),
10 of which were from families who had received CNEP for bronchiolitis. Analysis
of the questionnaires indicated that parents had:
adequate or comprehensive information about bronchiolitis 19/20
adequate or comprehensive information about the study 20/20
adequate or comprehensive information about CNEP 9/10
easy or reasonable contact with their child in CNEP 8/10
felt staff helped them relate / interact with their child a lot or moderately 10/10
felt CNEP had helped their baby‟s breathing difficulties 8/10
did not know whether CNEP had helped their baby‟s breathing difficulties2/10
free text from parents in the CNEP group 9/10
free text from parents in the standard group 7/11
In the free text received 7/10 parents in the CNEP group made positive
comments about CNEP; 5 parents strongly supported CNEP. Overall 12/21
parents comment specifically on the high standard of care their child received. Of
these 6/10 were in the CNEP group and 6/11 were in the standard group.
3.4 CNEP as a respiratory support in children
"The use of Continuous Negative Extrathoracic Pressure (CNEP) has been
in routine use in North Staffordshire Hospital NHS Trust as a technique for
respiratory support for cases such as children with bronchiolitis. The
Review has not been able to identify a substantial evidence base to support
this practice". [4.4.1]
Prof Southall Page 31
In the 18 months available to it, the Panel could and should have commissioned
an independent systematic review of the available evidence. Not only did it not
do this, it appears not even to have conducted the most basic search for relevant
evidence, let alone asked the researchers in Stoke (including me) for this. The
information was and is readily available.
The use of negative pressure for respiratory support in infants and children is
long standing, including other groups as well as the team at North Staffordshire
[15-19, 21, 23-24, 27-32].
More recently, there is the paper submitted to the journal Pediatric Pulmonology
which was placed on hold (see above) as well as a clinical case review of the
management of apnoea in acute bronchiolitis from North Staffordshire and
another similar tertiary referral paediatric intensive care unit where CNEP was
not used. This analysis showed that the use of CNEP was associated with a
reduced need for intubation (8 of 31 (26%) in Stoke v 18 of 21 (86%) in the other
unit) and duration of stay on intensive care (median 2 days in Stoke v 7 in the
other unit) These were unpublished data in an abstract that was prepared for
submission and for presentation at the Annual General Meeting of the RCPCH,
April 2000. [We were requested by Managers at the North Staffordshire Hospital
to withdraw this abstract because of the inquiries into my research resulting from
the allegations of Mrs Mellor].
3.4 Evidence that CNEP may cause serious injuries
"Some parents have alleged serious side effects [from CNEP] and in two
cases the Review Panel heard claims that children had suffered serious
brain damage or had died. These claims have been examined by the Trust
who believe that both children had some signs of brain damage before they
underwent CNEP". [4.4.1]
In two of the cases reported to the Review Panel by parents, the Panel
heard claims that babies became brain damaged or had died of brain
injuries following CNEP treatment. [11.1.3]
The Panel should have acknowledged that the Trust's own investigation into
these cases has shown no evidence that CNEP was causally related to the
cerebral problems in these two patients. If they were uncertain about the Trust‟s
findings, the Panel could have instigated an independent medical review on
these two patients.
Prof Southall Page 32
3.5 Proposed audit of the use of CNEP in North Staffordshire Hospital NHS
"Recommendation: that a substantial audit of the use of CNEP in North
Staffordshire Hospital NHS Trust be carried out to see if claims of
significant benefit or damage can be substantiated; …" [4.4.2]
The current evidence base for CNEP as a respiratory support in bronchiolitis is
stronger than exists for many other forms of respiratory support in children that
are currently used routinely in other centres. A randomised controlled study of
CNEP in bronchiolitis was undertaken in North Staffordshire in the early 1990‟s.
This was discontinued because nursing staff had become convinced that CNEP
was superior to standard treatment and did not wish to be involved in a study
where they believed that an effective treatment was being withheld from a
proportion of sick children.
The nature of the „substantial audit‟ of CNEP proposed by the inquiry is unclear,
but the wording of its recommendation suggests that insufficient consideration
has been given to the control of bias in any data collection and analysis
consistent with the term „audit‟. I would support and be happy to collaborate in
follow up studies of children entered into the randomised study described above.
The effects of CNEP have been critically examined; for example, there are
several published physiological studies on the effects of CNEP on cardiac output
, respiratory mechanics and timing , cerebral blood flow velocity ,
cerebral volume and cerebral oxygenation [6-7]. These did not reveal any
evidence that CNEP or the neck seal had deleterious effects on the physiological
Prof Southall Page 33
4. The process underlying the review undertaken
by the panel
4.1 Concerns about procedures
"This Review was conducted as a result of a meeting between Baroness
Hayman, then Parliamentary Under Secretary of State for Health (Lords),
and Mrs Llin Golding MP….. [1.1]
The Review Panel has not sought to determine whether allegations of poor
practice are true, or to apportion blame if practice could have been better;
neither has it attempted to determine whether any actions taken at that time
were wrong. It was not set up to do that because existing mechanisms
such as the NHS Complaints Procedure or the courts exist for that
The terms of reference for the review were “to look into the general
framework for both the approval and monitoring of clinical research
projects in North Staffordshire”. [2.1]
The statements made at paragraph 1.5 of the Report of the Review Panel are
clearly not accurate. The Panel‟s Report contains a significant number of
criticisms directed at the research team and personally at me. I have counted a
total of 9 direct criticisms of me1 and a further 8 implied criticisms of me.2 In
making those criticisms the Review Panel exceeded not only its own stated
objectives (see 1.5 of the Report) but also its terms of reference (see 2.1 of the
Order of events from my perspective
I was informed on 15/1/99 by the Chief Executive of the N Staffordshire
Hospital NHS Trust that Professor Griffiths had been appointed to carry
out a Review into the conduct of research trials in North Staffordshire.
On 16/2/99 I was interviewed by Dr Michael Harrison and Mrs Cynthia
Nash of the Midlands Health Consultancy Network (MHCN) for
approximately one hour.
In a letter from Professor Griffiths dated 22 nd June 1999 I was asked to
attend a meeting with the Review Panel. I will quote from this letter: "the
Expert Panel has decided that it would be helpful for them to meet
See paragraphs 4.5.1; 9.2.7; 9.2.8; 9.3.1; 9.3.4; 9.3.7; 9.4.2; 9.4.3 and 9.4.4 of the Report.
See paragraphs 4.1.1; 4.1.2; 5.6.2; 9.3.5; 10.3.2; 10.5.2; 14.3.5 and 14.3.6 of the Report.
Prof Southall Page 34
informally with a number of key people, including yourself, prior to their
formal consideration of the documentation that has been assembled. ….it
would be helpful if you could let me know urgently if there are any times
that would NOT be convenient for you on the following dates- Friday 23rd
July 1999, Thursday 29th July 1999, Friday 30th July 1999. …I hope it will
be possible to meet with you but I am sorry to say that if none of the dates
are convenient for you, it will not be possible to offer an alternative date
On 12/7/99 Professor Griffiths sent me a letter confirming the place and
time of the meeting.
On 23/7/99 I met with Professor Griffiths and the Review Panel.
On 17/9/99 I was sent the minutes of the meeting with the review panel,
prepared by Mrs Nash of MHCN. I made a large number of corrections to
these minutes and returned them to MHCN. I received the amended
minutes back from MHCN with my corrections within them.
On 17/4/00 I wrote to Professor Griffiths about the questionnaires on the
CNEP in bronchiolitis study described above in section 3.3. I was made
aware of these documents following my suspension for concerns over this
work. I realised that these questionnaires contained information
contemporaneous to consent and considered that the Panel should be
made aware of them.
Professor Griffiths replied on 25/4/00 requesting further information, which
I supplied in my letter of 27/4/00.
(a) Although the Panel says it was not supposed to be investigating me, it is plain
from the published Report that that is what the Panel did. The Review Panel
failed to inform me of what „complaints‟ were being investigated against me. I
was never told the case I had to answer.
(b) Once the Review Panel formulated criticisms, they did not tell me what those
criticisms were and did not (both during the meeting on 23/7/99 and
subsequently) give me the opportunity to respond to them before making their
(c) Neither before nor during my meeting with the Panel on 23/7/99 was I
informed whether or not I had a right to be represented (by a representative of
the BMA, MDU, or by a lawyer) or that it would be appropriate for me to be
represented. I was invited to an "informal" meeting (quotation from the letter to
me from Professor Griffiths on 22/6/99).
(d) My meeting with the Panel on 23/7/99 lasted for some two hours. It was in
my view a "formal" meeting. No one else was present but the Panel members,
Dr Harrison and Cynthia Nash of MHCN and myself. The interrogation of me
Prof Southall Page 35
was intense and at times hostile. The circumstances of that meeting were
oppressive. The interrogation went beyond the issues of research outlined in the
enclosure to the letter to me from Professor Griffiths on 12/7/99 and included
aspects of my child protection and humanitarian aid work.
(e) According to information given to me by the Trust in May 2000, the Panel
were not given the following documents highly relevant to the CNEP study in
The randomization logbook
The parent information booklet
The project study book involving aspects of medical and nursing
The questionnaires and covering letter on mother to child bonding
The Trust's audit of these questionnaires
A sample of the research data collated by Teresa Wright
The CVs of staff working on the project
Though, as the Review Panel knew, the Management at the Trust had refused
permission for one of these documents (the randomisation logbook) to be given
to the Panel, the Panel could/should have insisted on its production. The Report
fails to make reference to the Trust‟s refusal.
I am at a loss to understand why MHCN LTD did not collect most of the
information relevant to the research examined by the Panel. Their contact with
me was minimal.
(f) The Review Panel did not interview the statistician who collaborated in the
research projects and did not acknowledge the sequential analysis which aimed
to minimize the risk of harm should CNEP have been responsible for unforeseen
complications or worsened the respiratory failure.
(g) Although it would have been possible, from the records, for the Panel to have
contacted the parents of every child involved in the research project, it appears
from the Report that the Panel may have only interviewed those parents who
4.2 Issues of natural justice
"The panel very much appreciate the time and trouble that each individual
took in bringing their stories to us, some of which must have been painful
to recall. The Panel hopes the Review has done them justice. The Panel
believe implementation of our recommendations would go some way
towards ensuring that complaints like these would not happen again. In the
event of such complaints, the accountability chain would be clear and
appropriate action could be taken" [16.4].
Prof Southall Page 36
The Panel should not have been seeking to do justice only to those parents
whom it interviewed, but to all parents involved in the studies and to the
professionals conducting the study.
4.3 Origin of the inquiry.
"The Review began because there were complaints about the conduct of
research trials in North Staffordshire". [Part One]
Mr and Mrs X, one set of parents making complaints, have made very public
their views that 1) they were not told that CNEP was being offered to their
children as part of a randomised controlled study, 2) they believe that CNEP
caused brain injury to their child and 3) they had not signed consent, alleging that
it was forged.
It is evident that parents have to cope with extreme stress when their child is
critically ill. Evidence was also presented to the panel outlining that during this
stressful period some parents enrolled into another published study had no
recollection of having signed a consent form. It is therefore possible that Mr and
Mrs X genuinely believed that they had not signed a consent form. However,
some of the following information does not support the claims of this family.
During the recovery of their daughter from neonatal respiratory failure, Mr and
Mrs X received a letter and completed a questionnaire on maternal child bonding.
It is not possible for them to have failed to understand that their daughter had
been entered into a controlled study of CNEP. A letter was sent to them
beginning " "Dear [NAME OF PARENT(S)],
You will remember that shortly after [NAME] was born you kindly agreed to enroll
him/her into our study comparing negative pressure respiratory support with
standard treatment. As part of this study we have devised a questionnaire which
attempts to compare the effect of these two methods of treatment on the way you
were able to relate to your baby."
Mrs X replied to this questionnaire and indicated that in her opinion CNEP was
more effective than standard treatment.
Mr and Mrs X stated in a letter in October 1999 to the British Medical Association
the following: "..this group of parents who have raised concerns about the clinical
Prof Southall Page 37
implications of CNEP, have no connection whatsoever with Mrs Penny Mellor or
Mr Morgan who are apparently campaigning on child abuse issues". The
Guardian newspaper on 13th October 1999 published a correction in response to
a letter from Mr and Mrs X concerning an article in the Guardian of 11 th October
reviewing the campaign against my child protection work. The correction was
written as follows: they "would like to make it clear they have not criticized
Southall's work in child protection and have no connection with those who have
campaigned against Southall's work on Munchausen Syndrome by Proxy".
However, evidence for the existence of a connection is as follows:
Mr and Mrs X's story first appeared in an article by a reporter called Brian
Morgan in the Sunday Independent on 11 May 1997. The same reporter Mr
Brian Morgan is quoted as saying in the Mail on Sunday on April 5 th 1998 the
following: "his and the mothers' aim is to see Southall struck off the medical
register, his work discredited and a public inquiry instigated".
In a letter to my employer on 3rd June 1997 and given to me, Mr and Mrs X
wrote the following: "Why then would he slowly suffocate my child with his
machine and lie in order to cover up what he had done? How many other
babies like… have been subjected wrongly to this torture and when will he
realize that it is the case of the pot calling the kettle black and that he has no
rights criticizing Mums for how they look after their children when he
experiments on hundreds at a time in the name of science even with
knowledge of adverse effects. Have the courts seen the photographs of the
tiny babies who were strangled by the neck seal on his gentle form of
In April / May 1997, Mr Brian Morgan, and Mr and Mrs X wrote similar letters
to the Editor of „Pediatrics‟ criticising my work in North Staffordshire
In June 1997, Mr and Mrs X wrote to Keele University, Mr Brian Morgan's
letter to Pediatrics was mentioned, the response of the editor to Mr Brian
Morgan was quoted and Mr Brian Morgan was described as a friend.
In March 1999, Mr Brian Morgan, Mrs Penny Mellor and Mrs X all wrote
electronic responses to an article in the BMJ about my humanitarian aid work
in Afghanistan 
A person calling herself Penny posted on the internet on 25 April 1999 details
of the alleged problems with Mrs X's consent form:
The previous respondent has been very circumspect with what they said, it is not an allegation
but a fact. Mrs X's (one of the mothers taking action) consent form that she was supposed to
have signed two hours after the birth of her daughter, had the name of her child on the
form…..except that her and her husband had not even chosen a name for their child at that point.
So who signed the form? Who entered her child's name? GMC responses please… Also all you
British lawyers and QC's whom I have heard look at this site and know who I am will you finally
have the ***** to help? Just in case any of you are in any doubt, The Mr and Mrs X's were due
for CVS according to an inside source. What would have happened then? Accused of putting
her on a bit of research machinery in order to damage her……are there any lengths these people
will not go to….?".
Prof Southall Page 38
A person calling himself Brian Morgan posted the following on the MAMA
website on 28th October 1999:
o The CNEP scandal emerged because the X family in Staffordshire were told by a
doctor looking at medical notes belonging to their brain damaged daughter CHILD"S
NAME that she had been in a study, as if they already knew knew about this.
o They didn't, and this led to them getting hold of further documents, one of which
purported to be a research consent form signed by Mrs X, with CHILD"S name on it,
spelled incorrectly as NAME.
o The problem for the hospital is that the parents did not decide on a name for
several days and NAME was not even thought of initially.
o The other problem is that the form needed to be signed between between 2 and 4
hours after birth - during these hours Mrs X was in recovery from anaesthesia
following a C section.
o Her signature on the form is perfectly formed - not the sedated scrawl you might
expect from somebody still out for the count.
o Hardly informed consent. And as Dr NAME the medical director admitted on TV it
was not possible anyway.
o The hospital has still to explain how her signature appears on this form.
o This is the theory though - Mrs X got her own notes from the hospital and (I can
confirm this) there are a number of consent forms for other procedures she
underwent - you can see clearly as a bell where someone has tried to alter the forms
and then tried to correct the alterations - but most interestingly - one consent form
Mrs X knows she signed a good while after CHILD"S NAME was born is missing. Work
o This scenario is duplicated to some degree or another in other cases I have
o The X family did a major amount of work on their case and on a number of other
cases that came forward on the back of articles in their local newspaper.
o This first of all resulted in a General Medical Council investigation being set up,
and then the Griffiths Inquiry set up by the NHS Executive in the West Midlands on
the insistence of the then health minister Baroness Hayman.
o Then when this was underway Penny Mellor took her concerns about false
allegations of child abuse and other very serious allegations about child protection
work of doctors at North Staffs and elsewhere to the same team.
o A number of other investigations have been set up, in all around 6.
o I haven't done more than scratch the surface of what my own research and the X
family's research into CNEP has shown.
Prof Southall Page 39
Professor Griffiths and the Panel made clear their acceptance of Mr and Mrs X's
criticisms. In the Sentinel newspaper (local to Stoke), it was stated on 10 May
2000: “The dogged determination of Mr and Mrs X drew high praise from Prof
Rod Griffiths as he delivered his stinging report on child health research in North
Staffordshire. Professor Griffiths … said: „we were impressed by their attitude‟…”
In a recent local newspaper article Mr and Mrs X signalled their intention to claim
damages from the North Staffordshire Hospital for many millions of pounds.
4.4 Other Errors in the Panel‟s Report
There are the following mistakes in the reference list: reference ii (eg in [6.2.2]) is
supposed to relate to the research ethics committee submission for the neonatal
CNEP study in neonatal respiratory failure (“CNEP study”), submitted by
Professor Southall and Dr SA Spencer. However, in the reference list, it is given
as a “Physiological evaluation of subjects treated with negative extrathoracic
pressure: researchers Dr Carl Bose and Dr Andrew Spencer.”
In addition, in para [6.3.2], the Panel report states “the project [the CNEP
study] was re-submitted to the LREC in June 1992 to include physiological
evaluation of infants treated with CNEP, and was approved”. This is
incorrect, as this was an entirely separate submission on the physiological
evaluation by Dr Bose and Dr Spencer.
In para [6.3.3], the Panel report states “the project was again submitted in
September 1992 and with a number of „minor revisions‟ the amendments
were approved”. This is also incorrect, as this too was to do with the
physiological evaluation study and not the CNEP study.
4.5 The role of the N Staffordshire Research Ethics Committee
The panel appear not to have recognised that the LREC submission for the
neonatal CNEP study, submitted by Dr Spencer on 29 November 1989, has an
erroneous paragraph 12 (Procedures applied to each patient), which is taken
from an earlier submission form concerning research into the physiological
effects of CNEP on cerebral function.
Prof Southall Page 40
5. Other Issues
5.1 The need for greater openness
"One of the points to emerge from the interviews was a strong feeling of
secrecy and concealment expressed by most of the interviewees. There
was a general view expressed that greater openness would be beneficial to
those conducting research as well as to the public". [10.5.2]
The Panel did not state from where the feeling of secrecy and concealment had
arisen. I am unaware of anything that might have been concealed by the
research team and welcome openness. However, the Panel themselves could
be accused of secrecy and a lack of openness. At no stage prior to publication
did they make it clear to the researchers what information they required. Had
they made clear what was needed, this could have been provided to them.
Moreover, the Panel did not discuss their findings with me before the Report was
published. They did not discuss the Study with our statistician. According to the
folder of information replicating what the Trust had given to the Griffiths' Panel
that I received from the Trust late in May 2000, the following data were not given
by the Trust to the Panel:
The randomization logbook
The parent information booklet
The project study book involving aspects of medical and nursing
The questionnaires and covering letter on mother to child bonding
The Trust‟s audit of these questionnaires
A sample of the research data folders
5.2 Child protection issues
"MUNCHAUSEN SYNDROME BY PROXY
In order to assist in the correct identification of children who have either
had illnesses induced or fabricated by their carer, the Review recommends
that the Department of Health should convene an expert and
multidisciplinary panel which review methods of identification, including
the use of covert video surveillance, within the framework of the
Government‟s interagency guidance for child protection “Working
Together to Safeguard Children” (1999). Guidelines would aid
Prof Southall Page 41
professionals in their identification of this type of abuse and would suggest
good practice in multi-disciplinary and multi-agency management of such
"Our examination of Professor Southall's work suggested that the
governance framework is unclear and there are substantial issues around
the use of the diagnostic technique which uses covert video surveillance.
New guidelines could help ensure children‟s welfare is safeguarded".
The implication is that earlier guidelines did not safeguard children. It appears
that the main evidence to suggest that this is the case results from those who
appear to post material regularly on the MAMA website at least two of whom
were interviewed and quoted by the Panel in their report (Penny Mellor and Lisa
The British Paediatric Association issued guidelines in 1994 .
We published our guidelines for undertaking covert video surveillance in 1996
 and these were commended by the Department of Health as being a good
example of interagency working. Since then, other areas of the UK, for example
the Northern and Yorkshire Region of the NHS Executive has already issued
official guidelines that drew on the earlier BPA and North Staffordshire
The process of covert video surveillance has also been approved in the High
"Recommendation: that the Department of Health convenes an expert
multidisciplinary panel to review the identification of situations where
children have had illnesses induced or fabricated by their carer (commonly
known as Factitious Illness by Proxy or Munchausen Syndrome by Proxy)
and issue guidelines which should include the use of covert video
surveillance within the framework of 'Working together to Safeguard
Children' (1999)". [4.5.2]
The panel seems to be unaware of or fails to acknowledge existing NHS
Executive guidelines (see above)
Prof Southall Page 42
1. NHS Executive West Midlands Regional Office. Report of the review into
the research framework in North Staffordshire. 8 May 2000
2. Samuels MP, Raine J, Wright T, Alexander JA, Lockyer K, Spencer SA,
Brookfield DSK, Modi N, Harvey D, Bose C, Southall DP. Continuous
negative extrathoracic pressure in neonatal respiratory failure. Pediatrics
3. Marcovitch H. Diagnose and be damned. BMJ 1999; 319:1376-1377.
4. Escobar GJ, Littenberg B, Petitti DB. Outcome among surviving very low
birthweight infants: a meta-analysis. Arch Dis. Child. 1991;66:204-211
5. Raine J, Cowan F, Samuels MP, Wertheim D, Southall DP. Continuous
negative extrathoracic pressure and cerebral blood flow velocity: a pilot
study. Acta Paediatrica 1994; 83:438-439.
6. Palmer K, Spencer SA, Wickramasinghe Y, Wright T, Samuels M, Rolfe P.
Negative extrathoracic pressure ventilation - evaluation of the neck seal.
Early Hum Dev 1994; 37:67-72.
7. Palmer KS, Spencer SA, Wickramasinghe YA, Wright T, Southall DP,
Rolfe P. Effects of positive and negative pressure ventilation on cerebral
blood volume of newborn infants. Acta Paediatrica 1995; 84:132-9.
8. Parkins KJ, Poets CF, O‟Brien LM, Stebbens VA, Southall DP. Effect of
exposure to 15% oxygen on breathing patterns and oxygen saturation in
infants: interventional study. Br. Med J 1998:316;887-894
9. Milner AD Effects of 15% oxygen on breathing patterns and
oxygenation in infants. Br Med J 1998:316; 873-874
10. Letters (VARIOUS AUTHORS). Hypoxic responses in infants. Br Med J
11. Pandit JJ. Ethical commentaries must be based on sound science. Br
Med J 1998:317; 1594
12. Poets CF, Parkins KJ, O‟Brien LM, Stebbens VA, Samuels MP, Southall
DP. Effects of 15% oxygen on breathing patterns and oxygenation in
infants. Proceedings of the Royal College of Paediatrics and Child Health
Annual Meeting 1997; 1:95
13. Anonymous. Making babies breathe low oxygen. Bull Medical Ethics
1998; March pp. 4-6
14. Editorial Bull. Medical Ethics 2000; January p1.
15. Hartmann H, Noyes JP, Wright T, Wheatley R, Spencer A, Boon A,
Alexander J, Samuels MP, Southall DP. Continuous negative
extrathoracic pressure ventilation in infants with bronchiolitis. Eur Respir J
16. Linney MJ, Marinaki T, Southall DP, Samuels MP. Negative pressure
ventilation in bronchiolitis. Care of the Critically Ill, 1997;13:161.
17. Shekerdemian LS, Bush A, Lincoln C, Shore DF, Petros AJ, Redington
AN. Cardiopulmonary interactions in healthy children and children after
simple cardiac surgery: the effects of positive and negative pressure
Prof Southall Page 43
ventilation. Heart 1997; 78:587-93.
18. Shekerdemian LS, Bush A, Shore DF, Lincoln C, Redington AN.
Cardiorespiratory responses to negative pressure ventilation after
tetralogy of Fallot repair: a hemodynamic tool for patients with a low-
output state. JACC 1999; 33:549-55.
19. Raine J, Redington AN, Benatar A, Samuels MP, Southall DP. Continuous
negative extrathoracic pressure and cardiac output - a pilot study. Eur J
20. Torelli L, Zoccali G, Casarin M, Dalla Zuanna F, Lieta E, Conti G.
Comparative evaluation of the haemodynamic effects of continuous
negative external pressure (CNEP) and positive end-expiratory pressure
(PEEP) in mechanically ventilated trauma patients. Intensive Care
Medicine 1995; 21:67-70.
21. Gappa M, Costeloe K, Southall DP, Rabbette PS, Stocks J. Effect of
continuous negative extrathoracic pressure on respiratory mechanics and
timing in infants recovering from neonatal respiratory distress syndrome.
Pediatr Res 1994; 36:364-72.
22. Nishino T, Isono S, Ide T. Effects of negative pressure assisted ventilation
on dyspnoeic sensation and breathing pattern. Eur Resp J 1998;12:1278-
23. Samuels MP, Southall DP. Negative extrathoracic pressure in treatment of
respiratory failure in infants and young children. BMJ 1989; 299:1253-
24. Cvetnic WG, Cunningham MD, Sills JH, Gluck L. Reintroduction of
continuous negative pressure ventilation in neonates: two-year
experience. Pediatr Pulmonol 1990; 8:245-53.
25. Corrado A, Gorini M, Ginanni R, Pelagatti C, Villella G, Buoncristiano U,
Guidi F, Pagni E, Peris A, De Paola E. Negative pressure ventilation
versus conventional mechanical ventilation in the treatment of acute
respiratory failure in COPD patients. European Respiratory Journal 1998;
26. Splaingard ML, Frates RC Jr, Jefferson LS, Rosen CL. Harrison GM.
Home negative pressure ventilation: report of 20 years of experience in
patients with neuromuscular disease. Arch Phys Med Rehab 1985;
27. Russell-Jones DL. Treacher DF. Lenicker HM. Tashanov M. Spencer
GT. Central hypoventilation in a seven year old child following pertussis
treated with negative pressure ventilation. Postgrad Med J 1989;65:768-
28. Raine J, Samuels MP, Mok Q, Shinebourne EA, Southall DP. Negative
extrathoracic pressure ventilation for phrenic nerve palsy after paediatric
cardiac surgery. Br Heart J 1992; 67:308-11.
29. Hartmann H, Samuels MP, Noyes JP, Southall DP. Negative extrathoracic
pressure ventilation in infants and young children with central
hypoventilation syndrome. Pediatr Pulmonol 1997; 23:155-157.
30. Hartman H. Jawad M, Noyes J, Samuels M PO, Southall D P. Negative
Prof Southall Page 44
extrathoracic pressure ventilation in central hypoventilation syndrome.
Arch Dis Child 1994;70:418-423.
31. Davis PJ, Field D Continuous negative extrathoracic pressure respiratory
support for neonates after congenital diaphragmatic hernia repair.
Acta Paediatr 1999:88; 1156-1167
32. De Munter C, Nadel S, Britto J, Habibi P. Non-invasive respiratory support
in severe bronchiolitis Proceedings of the Royal College of Paediatrics
and Child Health Annual Meeting 1998:2:81 G156
33. Electronic responses 12-14 August to Kabul Diary by Kamran Abbasi Br
Med J 1998:317;401-5
34. British Paediatric Association Working Party. Report on the evaluation of
suspected imposed upper airway obstruction. London: British Paediatric
35. North Staffordshire Hospital Trust, Staffordshire Social Services,
Staffordshire Police. Guidelines for the multi-agency management of
patients suspected or at risk of suffering from life-threatening abuse
resulting in cyanotic-apnoeic episodes. J Med Ethics 1996; 22:16-21.
36. Re DH (A Minor) (Care Proceedings: Evidence and Orders), Family
Division, 24 January 1994, WALL, J. Family Court Reporter. (1994) 2
Prof Southall Page 45
Evidence for and responses to a campaign directed against Professor
Southall and his work.
Persons calling themselves Penny Mellor and Brian Morgan have made
complaints about me on numerous occasions to various professional bodies
including the Trust I work for, the General Medical Council, the Royal College of
Paediatrics and Child health, social services, the police, politicians and
newspapers. They have also complained about a number of other
paediatricians, child psychiatrists, nurses, social workers, solicitors and judges
regarding the alleged incorrect diagnosis of child abuse, specifically in relation to
fabricated and induced illness. „Penny Mellor‟ and 'Brian Morgan' are involved
with a small group of individuals who are campaigning vociferously against this
diagnosis. They are targeting professionals who have been involved in such
cases and claim that professionals have abused children and their families by
falsely recognising and diagnosing induced or fabricated illness.
Persons calling themselves „Penny Mellor‟ or „Penny‟ and 'Brian Morgan' have
posted regularly on an internet bulletin board for the “Mothers Against
Munchausen Syndrome by proxy Allegations” (MAMA) website [see
www.msbp.com in the "heart to heart" section ]. In relation to me and other
practitioners they have posted highly critical comments and allegations.
Examples of the MAMA postings are given in Appendix 2.
It is unknown as to how the journalist identifying himself as Mr Brian Morgan
originally came to be in contact with Mr and Mrs X (the family at the heart of
complaints investigated by the Griffiths' Panel). There is, however, considerable
evidence that persons calling themselves 'Brian Morgan' and „Penny‟ have been
in contact with another parent. The mother in question was tracked on the
MAMA‟S website. What follows is a summary of events and some quotes from
8/5/00 Griffiths report made public and attracts considerable publicity. Report criticizes child
protection procedures in relation to fabricated and induced illness in children and
recommends a review.
24/5/00. Article in Daily Express „Experts closed ranks over medical scandal‟ by a journalist
called Brian Morgan. „Brian Morgan‟ outlines he is investigating Dr Southall‟s work at Crawley
Hospital and the Royal Brompton Hospital (amongst other hospitals).
2/6/00. A person identifying themselves as „New Mum‟ enters on-line discussion on MAMAS
website and becomes a regular contributor to the bulletin board.
2/6/00. Message from: „Brian Morgan‟. The message offers to put „New Mum‟ in contact with
other mothers in the UK.
5/6/00. Message from: „New Mum‟ for Penny
Prof Southall Page 46
5/6/00. Message from „New Mum‟. „ She says she is changing identity to „Ms R.‟ New e-mail
address given as V.V.I.P@…. Ms R says she initiates off-line e-mail and telephone
communication with Penny.
7/6/00. Message from „Ms R‟. For David Southall et al „See you all in court…. lies perverted
actions, abuse and sick minds‟.
10/6/00. Message from „Brian Morgan‟.: „Time to call authorities to account‟. Outlines that the
use of monitoring equipment to record physiological signals was a research project (as
opposed to a clinical investigation to aid the establishment of a diagnosis). „Brian Morgan‟
says „ remember that the CNEP scandal was about alleged failure to tell parents their babies
were going into a research project and you may begin to see a pattern‟.
10/6/00. Message from „Ms R.‟ Ms. R responds to „Brian Morgan‟ outlining that she was
„never asked to sign a consent form…..‟ (Ms R. describes how she had used home
monitoring equipment in 1992 as her child was having as Ms R put it „cyanotic, lifeless,
vacant etc episodes‟.).
12/6/00.Message from „Ms R‟. to „Brian Morgan‟ regarding an e-mail she has sent to him.
22/6/00. Message from Ms R. She asks „Penny‟ on two occasions „Any advice re my email?‟
Within 3 months of commencing a dialogue with persons calling themselves
'Penny' (Mellor) and 'Brian Morgan' on the MAMA website an account of Ms R's
complaints was given in the Sunday Mirror on 20th August 2000 with the
following headline " 900 Babies 'Victims of Professor's black box. Children
suffered burns in sinister experiments".
Reports of a number of other activities have been posted on the www.msbp.com
website. For examples:
"A public meeting has been called at the Port Vale public house Hallem Road Burslem for
Tuesday 31 August 1999 at 7PM . This meeting has been called by myself on behalf of mothers
falsely accused Ms Y for the CNEP and Hypoxia trials and Ms Z for the children brain damaged
at North Staffordshire hospital with no answers. We will be demanding the immediate suspension
of the paediatricians involved in CNEP and Hypoxia and the sacking of David Southall. We will
also be demanding a full and public inquiry into the practises of North Staffordshire Hospital and
child protection within both Staffordshire SS and Police".(attributed Penny 29/8/99)
22/5/00 „Penny‟ outlined that following the publication of the Griffiths' report she had widened her
campaign and named an additional 7 hospitals and 13 professionals that she was now
29/6/00 „Penny‟ outlined that as a result of her campaign she had now instigated investigations by
12 named Police forces through out the UK.
Between May and July 2000 there were approximately 25 postings on the website relating to a
nurse who had been accused of committing multiple murder in the South of England. This
included copies of confidential correspondence sent between the mother who made the original
complaint about the accused nurse and the hospital involved. In particular, one posting by „Penny‟
on 22/6/00 says: “RE: Nurse Suspended in childrens death probe.... THIS NURSE ACCUSED
ONE OF OUR MOTHERS WATCH THIS SPACE”.
Prof Southall Page 47
The British Medical Association on 11th October 1999 produced a press release
describing this campaign following the completion of an investigation by the Trust
into the first of three groups of allegations made against me by Mrs Penny Mellor
(1. about my personal conduct; 2. about my child protection work; 3. about my
research). To date, only investigations into those allegations concerning my
personal conduct have been completed.
The investigating team in the Trust found “there is no substance to the
serious allegations (1-5) made by Mrs P Mellor against Professor Southall
and the Trust”.
Finally, according to a quote from Northumbria police given in The Sentinel
newspaper, 2 December 1999, Penny Mellor is currently on a charge of
“detaining a child under 16 out of the lawful control of the person entitled to have
Appendix 2 Statements attributed to Penny or to Penny Mellor on the
" David Southall I have compared you to Mengele so sue me…. Where's my writ" ( attributed
Penny Mellor 27/10/99)
“DR NAME instigating a sexual assault, got you you ******, a five year old child ………. In the
notes boy in the notes” [attributed Penny 26/12/99]
“Food for thought, the abuse industry is alive and well thanks to all of you so called “experts”.
Just one question. JUST EXACTLY HOW DID YOU GET THAT EXPERTISE BY
PROCURING THE VICTIMS AND THEN WATCHING????” [attributed Penny 26/12/99]
On the North Wales enquiry: “? Have you [experts] ignored paedophilia and other forms child
abuse in the system in a trade for drug trailing? Or are you all involved in institutionalised
abuse too? Are you paedophiles? How is it you never “profile” care workers? How is it that
you never testify against local authorities in child abuse cases?” [attributed Penny 26/12/99]
“DR NAME, why in 1988 did you perform **** ********** on a minor without prior consent or
knowledge of the parent? Why after *********** where you performing this examination anyway
when it had been banned? How do you explain that the rods were kept in a nice little case
with YOUR INITIALS on it.............? You **** *** ******** *******. You questioned the mother in
a public packed out playroom on the children's ward and used these exact words "Who has
****** about with your daughter's *****? Because some women let their husbands or
boyfriends **** about with their kids, it adds a zest to their marriage, now tell me who has
Prof Southall Page 48
****** around with your daughters *****, because somebody has" ” (attributed Penny Mellor
"I'll tell you how bad this gets, one of the CNEP mothers who has lost two babies to those
******* at North Staffs was only a victim of Pin Down and guess what Southall and co have
had her two eldest children handed over to her batterer ex husband "because she wasn't fit
given the way she was bought up" THEY BOUGHT HER UP AND THEN THEY PUNISH
HER AGAIN AND AGAIN THEY ABUSED HER IN CARE THEY KILLED TWO OF HER
CHILDREN THEY TRIED TO MAKE OUT THAT SHE HAD SOMETHING TO DO WITH IT
TO SHUT HER UP THE THEY STEAL TWO MORE OF HER CHILDREN AND PUT THEM
WITH AN ABUSER?????????? Wait until Tuesday night when the whole world finds out
what they have been doing.
It makes me physically ill and so so sad. There are not enough tears to show my sorrow for
these children and their parents. For the first time in my life I have contemplated bringing
back capital punishment just for the accusers. How they have corrupted my soul...for that they
will all pay". (attributed Penny 29/8/99)
"THE DOCTORS GUILTY OF FALSE ALLEGATIONS, SUPPORTERS AND CO-CONSPIRITORS OF
DAVID SOUTHALL AND MARTIN SAMUELS
NAME OF PROFESSOR OF PAEDIATRICS retired NAME OF CITY, NAME OF DOCTOR,
NAME OF HOSPITAL Oxford, NAME OF DOCTOR, NAME OF HOSPITAL, LONDON, NAME
OF DOCTOR, NAME OF HOSPITAL, LONDON, NAME OF DOCTOR formerly of Basildon,
NAME OF DOCTOR Cardiff, NAME OF DOCTOR Cardiff, NAME OF DOCTOR Singleton, NAME
OF DOCTOR Bridgend, NAME OF DOCTOR Sheffield, NAME OF DOCTOR, NAME OF
DOCTOR, NAME OF DOCTOR Singleton, NAME OF DOCTOR retired, NAME OF DOCTOR
Telford, NAME, MP, Justice NAME, Family High Court, Justice NAME, Family High Court, Justice
NAME, Family High Court, Justice NAME, Family High Court, NAME OF SOCIAL WORKER,
Stoke-on-Trent, NAME OF POLICE OFFICER, Staffs Police, NAME OF POLICE OFFICER, Staffs
Police" (attributed Penny 7/5/00).
Prof Southall Page 49
E OF POLICE OFFICER, Staffs
Police" (attributed Penny 7/5/00).
Prof Southall Page 49