Saurab,

Ethical review of research relates primarily to the conduct of that research, the well being of research participants, research subjects (animals, things of intrinsic value and things of value to people) and third parties (things that could be affected by results of research). In the case of your specific question the precise answers will depend upon the ethics interpretations and regulations of your particular destination journal and any funders of your research. You are recommended to check out the ethics requirements of target journals at an early stage in the research to ensure that their requirements are addressed.

Normally for University research there is an expectation that the research that underpins any paper will have undergone an independent and thorough review resulting in a favourable opinion typically via an internal University research ethics committee. It is important that the reviewers are independent of the research being reviewed and that the committee can demonstrate some lay person involvement that is independent of the University. Now, to try and answer your questions from my own perspective:

1) No, in my opinion the title of the protocol on the ethical approval letter and article published do not have to be identical, but you will need to be able to show that you have a favorable opinion for the research upon which the paper is based. If for some reason the title of the paper has departed significantly from the protocol title then you could go back to your ethics review committee and request an amendment to the new title. Be careful to ensure that it remains clear that the protocol was originally reviewed before the research was actually undertaken;

2) My thinking in point number 1 is applicable also to the possibility of producing multiple papers from the original protocol - it is the underpinning research that needs to be subject to review and not each individual paper. Again, if differences in relevant titles create problems you can of course request an an amended letter of opinion from your research ethics committee. If however you intended to blend previous research with some entirely new elements you would need to consider requesting an entirely new review. Again, it would be important to ensure that it is clear that each element of research was reviewed favourably before its commencement;

3) It is the research and not the authors that are subject to review so if the research remains the same but some of the authors alter that should not matter. The only issue is that a protocol will normally identify a principal investigator who typically is responsible for preparing the submission for ethics review and responding to any reviewer questions. That person ideally should remain as one of the authors of papers drawn directly from that protocol research. They do not have to be the lead author on the paper.

I hope that I've added a little clarity or a least some sensible (to me!) considerations. As ever, ethics involves opinions and discussions of what is right and wrong so I too would be interested to learn the thoughts of others on these issues.

Best wishes

Dear Malcolm,

Nice well-thought out elaboration of the current perspective on the ethics review. Our IRB also does the same things as you said.

I think the answer given by Malcolm addresses the issues properly.

The only thing that I would add to Malcolm's very good response is with respect to Number 2 - multiple articles are fine if they are all original and distinct. Publication ethics requires avoidance of salami-slicing and redundant publications.

Hi Saurab,

I think that you will have to yourself determine whether the description you gave the IRB of the nature of your research remains sufficient, accurate, and truthful whenever you use the research, especially if human subjects are involved. 

For instance, let's assume you were initially interviewing college seniors to ascertain the extent to which they use social media as the primary means of communicating with their parents.  Your declared objective was to establish the tendency of that cohort group to rely on, say Twitter posts, rather than actually picking up the phone and dialing mom or dad.  So, in conducting your interviews with these subjects, you told them (as you had stated in writing to the IRB) that the purpose of the research was to publish an academic piece on how their generation uses social media in lieu of actual direct contact with their parents.  And, let us also assume that being an honest and ethical researcher you did indeed write up that research and publish it in a well-respected sociological journal.

After basking in the glow of positive critique of your learned article in the sociological journal, you realize that since you had personally conducted those interviews and thus had knowledge of the race and apparent ethnicity of the interview subjects, you could utilize your personal observations to write another article under the title, "Black and Hispanic Students Rarely Utilize Social Media as a Means of Staying in Contact with Their Parents."  You write the article and it is published in the popular press (and moreover picked up by UPI so widely distributed).  At the end of your article, you included the probability of statistical error based upon the size of your sample; but that is the extent of your 'scientific objectivity'. 

In such a case, you would have used the same research for an entirely different purpose -- a purpose that had not been contemplated when you originally solicited students for your in situ research. You can understand that the Black and Hispanic students you interviewed would be upset when they read your article (which might even be picked up as a UPI piece and reprinted by their school newspaper). 

Moreover, assuming you teach at the same university attended by the students you interviewed and thus cleared your work with the IRB of your university, members of the IRB would spit every time they walk by you on campus.  And, clearly, given the hypothesized scenario, you would deserve their disdain as you should have gone back to the IRB with a completely different protocol for your research and obtained approval.

Additionally, if this happened in the litigious U.S., you would be subjected to lawsuits by our aggressive Plaintiff's Bar who would have no difficulty finding numerous grounds upon which to file a class action lawsuit on behalf of the Black and Hispanic students at the hypothetical university.

Gwen (not a Certified Trial Lawyer)

I agree with Gwendolyn that research involving participants should not normally be applied towards purposes that were not intended or advertised when the data was collected (strongly justified studies that need to use deception as integral to the study design are perhaps an exception - but even then most instances can be addressed by careful post-experiment debriefing).

The critical issue is the nature and content of the consenting process. If a very tightly defined consent is sought then the data collected will be limited in application to the specific purpose agreed in the consent. If there is an anticipation that other related analyes and studies could also usefully apply that data, then researchers should include broader purposes within their consent schedules. That can significantly increase the usefulness and value of the data collected. It can, however complicate the consenting procedure so that participants may be faced with a series of consent statements. Consents for wider application of data collected can be made optional so that participants could alternatively opt for the tighter consent statement if they were unsure about wider application.

If consent statements concerning application of the data become too wide, then they risk becoming meaningless and no longer in the best interests of the participants. That is one reason why consenting arrangements and wording of consent statements are normally examined very carefully during reviews of protocols by research ethics committees.

A possible alternative where very tight consents attach to data that has high value for other applications is to attempt to revisit participants to obtain new consents to cover the intended new use. That can be very difficult since it is considered good practice to de-identify data before storing it (even if the participants are known the specific data cannot be linked to named individuals as a basic security measure).

Many thanks for the interesting discussion.

Absolutely brilliant, Malcolm!  Have you ever considered law school?

Gwen

Dear All, 

Thank you all for a very comprehensive comments,examples and argument on the ethical issues. All of these are helpful in understanding the ethical issues all round the world. 

I would like to Ask, Raj Dhungana if Nepal has any specific ethical considerations to these questions which has not been answered by Malcolm and Gwendolyn?