Cross district comparison of applications to research ethics committees

BMJ 1995; 311 doi: (Published 09 September 1995) Cite this as: BMJ 1995;311:660
  1. Paul Garfield, research associatea
  1. a Greenwood Institute of Child Health, Leicester LE30QU
  • Accepted 7 August 1995

Meade has criticised the current need to obtain approval for multicentred research from many different ethics committees.1 Variability between committees has led to research having to be modified or even abandoned in some districts but not others.2 I report the results of applying to 13 ethics committees with in a region for approvalof a study of young people.

Subjects, methods, and results

A regional study of the mental health needs of young offenders was proposed in which adolescents aged 13 to 17 were to be identified by health, social, and juvenile justice agencies. They were to be studied by means of a social and psychiatric interview, an educational assessment, an interview with staff, a postal questionnaire from parents, and a study of notes. The regional health authority identified 14 ethics committees. Approval for the study was obtained from one committee in advance. Applications were then sent to the remainder, informing them that this was a cross district study and that one committee had already given approval.

Characteristics of application forms and responses from 12 committees

View this table:

One committee did not respond to the initial inquiry. Two neighbouring committees had reciprocal arrangements for approval from each other. Eleven committees provided application forms, which were all different. One committee provided a form specifically for non-clinical studies. None provided a form on disk. Information requested varied between committees (table). Two forms asked if the study was multicentred, seemingly without this making a difference. One committee's guidelines stated that non-therapeutic research was generally unacceptable in children. One indicated that consent to a study using only interviews or questionnaires might not be required in writing, while another advised that consent for postal questionnaires should be obtained in person.

Two committees lost the original application. Two committees invited the researchers to the committee meeting: in one case four days' notice was given. One committee accepted its neighbour's decision. The table shows the responses of the remaining committee. The median time from application to the committee meeting was 4.0 weeks and from the meeting to a reply 3 days.

Eight committees had a median of three queries. Most of the queries concerned how consent was to be obtained and recorded, although three committees were concerned about confidentiality. The median number of changes requested by the eight committees was 1.5. One committee gave full approval but later decided the original meeting had been inquorate and requested changes. One committee approved the study subject to the advice that “proper legal consent was obtained” without explanation. The median time from application to final approval was 9.0 weeks.


Only two out of 13 committees had reciprocal arrangements for the approval of multicentre research, despite the recommendations of the Department of Health.3 The information requested by each committee on their forms varied widely, and the time needed to complete the different applications was considerable. Ethics committees should be encouraged to accept a national standard application form, available on disk.

Most ethics committees replied quickly after the first committee meeting, but delay was considerable when clarification was required. The ethical issues regarding consent were particularly difficult in thisstudy because of the young people's circumstances; many would have severe family difficulties and be living away from home. Given the diversity of inconsistent and sometimes vague recommendations about consent from professional bodies and the Department of Health, however, the ethics committees are bound to spend considerable time in raising queries, which are also likely to be time consuming for researchers. The training of ethics committee members is important and might produce greater consistency across districts, but it requires adequate funding.

Ethics committees have heavy workloads,4 and they should be encouraged to accept the approval of an application that has been carefully considered by one committee. Such an initiative has been introduced in the Northern region (B Sutherland, personal communication) and might be considered elsewhere. More widely, my experience adds weight to the case for a national ethics committee for multicentred studies.5


  • Funding Wellcome Trust.

  • Conflict of interest None.


  1. 1.
  2. 2.
  3. 3.
  4. 4.
  5. 5.
View Abstract