Findings And Recommendations Of The MOH Committee Of Inquiry (NNI)

08 Apr 2003

1 INTRODUCTION

1.1 The National Neuroscience Institute was involved in carrying out a study on the Haplotype Structure and Single Nucleotide Polymorphism (SNP) Frequencies in Candidate Genes in Neurological Disease and Response. Concerns about research ethical lapses in the conduct of the study were surfaced, starting in November 02, to neurologists in SGH by a number of their patients who had been recruited into this study without their doctors' knowledge. (To understand the organisation of neuroscience services in the public sector, please refer to Annex A.)

2 NATURE OF ALLEGATIONS

2.1 The allegations made against the Study were:

a. patient recruitment without the knowledge of the patients' attending physicians;

b. failure to obtain informed consent from Parkinson's Disease (PD) patients particularly in regard to challenging them with L dopa and video taping them for the Parkinson's Disease arm of the Study

2.2 Responding to these allegations, the Ministry of Health (MOH) appointed a Committee of Inquiry ('the Committee') on 21 January 03 to identify lapses in the ethics approval process and subsequent conduct of the study, and to make recommendations to address the identified lapses. The members and terms of reference of the Committee are at Annex B.

2.3 The Committee took statements from the applicant and co-applicants of the study proposal, research assistants, research nurses, interpreters, doctors whose patients participated in the study, some of the study participants, Chairmen of hospital ethics committees, chairmen of medical boards and others from the hospitals from where medical records were obtained. It also examined some participants' records, as well as interviewed the key applicants and some of the key co-applicants, Chairmen of hospitals' ethics committees and some of the doctors whose patients participated in the Study. The Committee also subsequently reviewed additional information which it had not previously had access to in the form of the NNI Inquiry's detailed review of the patient folders and medical records of 321 patients who had been recruited into the PD study.

3 Regulation of Clinical Research

3.1 For clinical drug trials (i.e. clinical trials which require human subjects to take medication for the purposes of assessing the effects or investigating new uses or new dosages), a Clinical Trial Certificate (CTC) is required under the Medicines (Clinical Trials) Regulations and issued by the Health Sciences Authority after the study protocol has been reviewed and approved by the Medical Clinical Research Committee .

3.2 Clinical drug trials are required by the MCRC to comply with the Singapore Guideline for Good Clinical Practice (SGGCP), a set of guidelines issued by MOH to ensure that the conduct of clinical trials meets internationally acceptable ethical and scientific standards.

3.3 For other forms of clinical research, MOH had, in 1998 informed the hospitals that the SGGCP would still be the relevant standard to follow.

Review of Medical Research Ethics at the Hospital Level

3.4 A central requirement of both the SGGCP and NMEC guidelines and the Helsinki Declaration is that all clinical research have to be reviewed by the Hospital Ethics Committee (HEC) [sometimes also known as Institutional Review Board (IRB)] before commencement. HECs consist of hospital medical and nursing staff, and lay persons, and are tasked with ensuring that research on hospital patients is carried out ethically.

3.5 The HECs are guided in their work by the requirements stipulated in the SGGCP (for clinical drug trials), and by guidelines issued by the National Medical Ethics Committee (NMEC).

4 THE STUDY

4.1 The aim of the Study was to study the genetic structure of patients with epilepsy, Parkinson's Disease, stroke and neuroleptic-induced tardive dyskinesia, to determine if specific genes increased the likelihood of normal persons developing these diseases and determine the responsiveness of individuals to certain medications. The target was to recruit 1,500 subjects for each of the conditions: epilepsy, Parkinson's Disease and stroke, and 1000 for tardive dyskinesia.

4.2 The Principal Applicant listed in the Study proposal to BMRC was Prof Simon Shorvon, Director of the NNI. The co-applicants were: Prof David Goldstein, Prof Nicholas Wood, A/Prof Lee Wei Ling, Dr Yee Woon Chee, Dr Lai Poh San and Prof Kua Ee Heok (see Annex C). Dr Viswanathan Ramachandran was Project Manager, while Dr Chantal Depondt was a Clinical Research Fellow at NNI. Their duties and those of the other team members are at Annex D. The key researchers in the PD study were Prof Shorvon, Prof Burgunder, Prof Woods and Dr Ramachandran. The Biomedical Research Council (BMRC) awarded the NNI a Core Competence Grant totalling S$10.8M over five years after scientific review by international and local experts.

4.3 Between 29 October 01 and 18 July 02, applications for ethics approval for the Study were made by Prof Shorvon to the Tan Tock Seng Hospital Ethics Committee (TTSHEC), Singapore General Hospital Ethics Committee (SGHEC), National University Hospital Institutional Review Board (NUHIRB), KK Women's and Children's Hospital Ethics Committee (KKHEC) and Institute of Mental Health Research Ethics Committee (IMHREC). The application that was submitted was essentially the same as the grant proposal which had been submitted to the BMRC for funding approval.

4.4 The applications did not provide detailed information on how study participants would be recruited. The applications and the patient consent form submitted for ethics approval were for the taking of blood from patients but did not include information on the dopa challenge procedure and video-taping. Approval was given by TTSHEC and SGHEC to carry out the Study, and by the NUHIRB to refer its patients to NNI for assessment as to their eligibility to participate in the Study. The IMHREC gave approval for the tardive dyskinesia arm of the Study to be conducted on its patients.

Patient Recruitment

4.5 In May 02, the recruitment of PD patients commenced through the Specialist Outpatient Clinic of a Consultant Neurologist in NNI (main campus). If patients whom he asked were willing to participate in the Study, they would be assessed by the researchers for recruitment into the PD study.

4.6 Around June 02, the researchers were concerned that this method of patient recruitment was too slow. It was also felt that the confirmation of the diagnosis of PD might not be objective enough since only one researcher was assessing the clinical condition. It was then decided that the procedures of 'on-off' observation and videotaping be undertaken to address these concerns.

4.7 The researchers decided to use listings of hospital patients from the pharmacies of TTSH, SGH and NUH to identify and contact potential Study subjects. After having obtained these lists, Dr Ramachandran then requested for access to medical records of these patients from TTSH, SGH and NUH. Access was granted by TTSH and SGH. NUH denied access by Dr Ramachandran as patients' consent had to be first obtained.

Dopa Challenge

4.8 PD patients who were identified this way were sent letters of invitation signed by Prof Shorvon, to participate in the Study. They were then called up to ask if they would participate in the Study. Before the appointed date of their participation, patients who were suffering from PD were asked by the research assistants to omit their evening dose of anti-Parkinsonian medication, and delay their morning dose until they were seen by Dr Ramachandran.

4.9 Patients who were visited by Dr Ramachandran would undergo the dopa challenge, whereby they were first examined in the 'off' state, given dopa and then examined again. For many of these participants, it was learnt that video-taping was also carried out.

4.10 The concerns with these procedures were:

a. the procedures had not been included in the applications to the hospitals' ethics committees;

b. no written informed consent was obtained from patients for the dopa challenge as well as for video-taping;

c. the dosages of medication which participants had been prescribed with were modified, raising concerns that this may jeopardise their safety and well-being.

5 ISSUES AND RECOMMENDATIONS

Issues

5.1 The Committee has identified four key areas of ethical concern as follows:

Ethics approval process

5.2 As Singapore promotes the life-sciences and biomedical research work increases, the well-being of the population and safety of patients are paramount and should not be compromised. The study of diseases and their genetic determinants most often require drawing of study subjects from the hospital population. Thus, it is important, for the ethical protection of these subjects, that the hospital system for reviewing the ethical aspects of research be stringent, effective and transparent.

5.3 In the case of the Study, the research protocol that had been submitted to HECs was the same document used to secure funding approval - it did not contain sufficient details on key areas of study methodology, such as how potential subjects were to have been identified, that 'on-off' observation and videotaping would be carried out

5.4 HECs had approved the Study without seeking further or sufficient clarifications on the study methodology and clinical procedures; Notwithstanding this, the Committee is of the view that unless sufficient information is furnished to HECs, it is not possible for HECs to anticipate or second-guess what a particular research project is contemplating in regard to the treatment of human subjects, if these are not expressly stated in the application form.

5.5 The Committee is of the view that it is the responsibility of the applicant to include all the necessary and important details in the application for ethics approval. In particular, the applicant has a duty to ensure that the application contains sufficient details with regard to specific intervention in patients' treatment or medications in the course of the research (in this case, 'on-off' observation and videotaping). In addition, the consent form and information sheet submitted to the HECs must contain these facts. It is not acceptable if such information was not included in the submission to the HECs if these procedures had been planned from the beginning.

5.6 As the modifications to the PD study protocol ('on-off' observation and videotaping) were said to be developed in mid-2002, the Principal Investigator (PI) should have gone back to the HECs for ethics approval of these substantial modifications to the study methodology, before proceeding with them. Failure to do so is a breach of protocol.

5.7 The Committee also has concerns that the HECs in hospitals are having to cope with increasing numbers of research applications that stretch resources and which may affect the quality of ethics review of research protocols.

5.8 The standards required of HECs, in the review of clinical trials, are set by the 'Declaration of Helsinki', the 'NMEC's Ethical Guidelines On Research Involving Human Subjects' and the 'Singapore Guideline for Good Clinical Practice'. Apart from the Declaration of Helsinki and guidelines issued by the NMEC in 1997, there are no other local guidelines that deal with non drug trials.

5.9 MOH had previously circulated the latter two sets of guidelines to hospitals for reference. However, the Committee notes that some of the HECs had expedited the ethics approval for the Study on the basis that it was not a clinical drug trial.

Medical confidentiality

5.10 The application to the HECs did not provide details on the patient recruitment methodology. The research team had also contacted the majority of patients without going through their attending physicians

5.11 Based on the available information, the Committee finds that the research team had, in their applications for access to patients' medical information and records, complied with the hospitals' existing rules and procedures.

5.12 The custodians of patient medical information and procedures for the release of patient information for research purposes were not clearly defined in some hospitals.

5.13 Within institutions, HECs had not been given the authority over all approval processes for the release of patient information for research purposes.

5.14 The Committee accepts that, in the case of some large international epidemiological studies drawn to the attention of the Committee, it had not been the practice to recruit study subjects through the attending doctors. The Committee feels that, where the research entails more than just the drawing of blood samples, due consideration must be given to the possibility of the research procedure affecting the patient's well-being (including the patient's emotional state of mind and the patient's confidence in the health care being given to him). The Committee therefore takes the view that the PIs and researchers should have informed the attending neurologists before contacting the patients to participate in the Study, on the basis that researchers could not be expected to have the full picture of the patients' condition, allergies, co-morbid conditions, etc., that their attending doctors would have, given that such a full picture may not be available in the patients' medical records.

Conduct of the Study

5.15 The 'on-off' observation and videotaping procedures had not been mentioned in the draft consent forms given to the HECs for their approval, nor in the actual consent forms used by the Study subjects. Consequently, only verbal consent and not written consent for 'on-off' observation and videotaping had been obtained from subjects.

5.16 Study participants had not been given a copy of the signed consent form for their retention and reference.

5.17 There had been modification of the treatment regime that patients had been prescribed in subjects in the PD study (delaying the subject's usual dose of levo-dopa, and, in some cases, substituting a normal-release equivalent for the patient's usual slow-release formulation); The Committee is of the view that changes in the type of medication or treatment regime may have the potential to expose patients to a risk of adverse effects, however small or negligible. This is especially so for the elderly patients with Parkinson's Disease who had to travel to NNI to participate in the Study, without having taking their first dose of medication. The Committee accepts the view that even a change in the timing of the patient's medication constitutes an alteration of the prescribed treatment regime of the patient. It is precisely these types of situations that the HECs should guard against and there may be a necessity for research to be monitored to ensure that there is compliance to the approved protocol.

5.18 It is unacceptable that clinically significant dosage modifications were made without consultation with the patients' attending physicians. If it had been the intention to make significant dose modifications, then consultation with the patients' attending physicians would be important to avoid jeopardising the patients' safety and well-being.

5.19 A proportion of Parkinson's Disease patients would have cognitive impairment and depressive symptoms. With such research subjects, it is not surprising that some did not remember signing a consent form, and were unable to give a full account of what had transpired. With vulnerable subjects, the consent form needs to be more explicit and yet simple. A witness or relative needs to be present to help the subject.

5.20 The Committee finds that the conduct of the Study could have been improved upon especially with regards to communication between the investigators, and with the various hospitals' neurologists,

5.21 Changes had occurred as the Study progressed and these protocol deviations ' such as changes to the PI list, modification of the treatment regime and the use of videotaping ' had not been reported back to the HECs.

Administration and organisation of the Study

5.22 The Study was a large study that involved 7 PIs,o 4 disease conditions, 4 study sites and potentially 5,500 patients. The nomenclature and definition of the different roles in the research team and the precise responsibilities of individual members had not been formalised or recorded in writing for the reference and guidance of all involved in a study as large as this.

5.23 The Committee understands that, for the Study, there had been no regular scheduled co-ordination meetings to discuss problems and the progress of the Study. The decision to have these meetings was made only in November 02, some months after the Study had already commenced.

Recommendations

5.24 In keeping with its focus which is mainly on the systems and process issues, the Committee makes the following recommendations to strengthen the ethical approval and oversight of clinical research studies.

Ethics Approval Process

Recommendation 1:

Every institution, whether a hospital or otherwise, that conducts medical research should have a robust and effective HEC. For small institutions where it is impractical for them to have their own ethics committees, clear arrangements should be made with other larger institutions for the ethics review of research projects involving patients.These HECs must be supported by an adequate number of full-time administrative staff. Where possible, the ethics review of research protocols should be supported by a full administrative office operating within the hospital. In order for HECs to function efficiently and effectively, members of HECs should also be properly oriented to the role and responsibilities of HECs and given the appropriate training where required.

Recommendation 2:

An HEC should not be expected to accept a workload that compromises the quality of ethics review. Where this is likely, the hospital / institution should establish additional HECs or make formal arrangements for other HECs to provide an opinion.l HECs could also consider appointing Scientific and Serious Adverse Events subcommittees to assist in these aspects of their work, thereby lightening the workload of the main committee.

Recommendation 3:

HECs should have formal face-to-face meetings, and proper minutes of the discussions should be kept. Decisions should not be routinely made via circulation of documents. If an application describes a study that is considered by the Chairman to be of low risk to patients and he exempts it from detailed review or gives preliminary approval for the study (as per HEC's Standard Operating Procedures), the application should still be tabled for the HEC's next regular meeting.

Recommendation 4:

HECs should ensure that a study is scientifically sound (i.e. having undergone a proper evaluation) and be responsible for reviewing the patient recruitment methodology (with particular regard to safeguarding patient confidentiality), the procedures which the patients are subject to, the consent form and the patient information sheet. If information is felt to be lacking, it is the HEC's responsibility to ensure that this is rectified to the HEC's satisfaction before allowing the study to commence.One way of ensuring compliance by investigators is to have a detailed application form that itemises the information the HEC requires, such as the recruitment methodology, any modification of patients' medication, etc.PIs should not submit to HECs the same protocol used for grant application to funding agencies, as the latter tends to focus mainly on the scientific aspect, while HECs need to review the ethical aspects of the study.

Recommendation 5:

Steps should be taken to remove the ambiguity with regard to the application of the guidelines, especially for non-clinical drug trials. Apart from NMEC guidelines, there is a lack of guidelines that deal with these trials. We recommend that this be remedied.

Medical Confidentiality

Recommendation 6:

Hospitals / institutions should take steps to determine the custodiansn responsible for patient medical information and to establish a system through which the custodians would inform the attending doctors before the release of their patients' medical information.In situations where the PI is also the administrative owner of the physical medical records, procedures should be established to address potential conflicts of interest.Hospitals/institutions should also take steps to lay down formal procedures for the approval of release of all kinds of patient medical information.

Recommendation 7:

The Committee recognises that the issue on the custody of medical records is becoming increasingly complex with the formation of national specialty centres. It recommends that further discussions be held to remove of ambiguity with regard to the custody of medical records.

Recommendation 8:

Hospitals should establish a clear system of governance on access to patient medical information, with HECs having the responsibility for the oversight where the access is for research purposes.

Conduct of Study

Recommendation 9:

The Committee strongly recommends that PIs and researchers inform the attending physicians of the participation of study subjects, especially in studies where there is a considerable level of clinical interaction with the subjects, since the welfare of participants is paramount. As a general guide, the Committee recommends the following:

For studies where there is access to patient medical records with any level of clinical interaction with patients, PIs and researchers should inform the attending physicians.

For studies where there is access to patient medical records ' with minimal level of interaction with patients ' for the purpose of obtaining more information, Pis and researchers should be encouraged to inform the attending physicians, and HECs should consider whether this should be made a condition of ethics approval.

For studies where there is access to patient medical records without any contact with patients at all, PIs and researchers need not inform the attending physicians.

Recommendation 10:

PIs should ensure that consent forms are designed such that all the necessary information for informed consent is included, as specified in the Singapore Guideline on Good Clinical Practice, yet are simple to understand. A copy of the signed consent form should also be given to the study subject for his retention.

Recommendation 11:

Details on modification(s) in treatment regime, however minor, and on procedure(s) being done to study participants should be reflected in the study protocol and consent form submitted by PIs to HECs for approval.

Recommendation 12:

When an investigator submits and signs a protocol, it is an agreement to carry out the study according to the submitted protocol. Any amendment or deviation from the protocol submitted (other than typographical corrections) should be reported back to HEC. HECs should state this as a condition when granting approval for a study.

Administration and Organisation of the Study

Recommendation 13:

For a study of this magnitude and complexity, it is desirable that the process for incorporating members into the research team, the nomenclature and definition of the different roles, and the specific roles and responsibilities of each individual in the team should be formalised.

Recommendation 14:

Research teams (i.e. Pis and researchers) should have formal face-to-face meetings regularly and proper minutes of the discussions should be kept. Decisions should not be routinely made through the circulation of documents.

Recommendation 15:

All researchers should be currently informed about the local standards / guidelines and regulations concerning clinical research. Documented certification in continuing medical education in the ethics and methodology of research should be made a requirement for all PIs and researchers who deal with human subjects.

annexA(1).pdf
Annex A - Organisation of Neuroscience Services in Singapore

annexB(1).pdf
Annex B - Composition of the Committee

annexC(1).pdf
Annex C - List of Applicants for Core Competence Grant

annexD(1).pdf
Annex D - List of Research Personnel in the Study