Chairing your first trial steering committee

I recently was given the opportunity to chair my first trial steering committee so now I’m an expert in it. Read on if you too want to be a pre-eminent expert in chairing trial steering committees.

I will be presenting my thoughts in a list format. Revolutionary, I appreciate, but bear with.

  1. Read the papers. Like *really* read them. You will want to do this anyway, because
    Bill gates
    This guy is obviously on his way to chair his first TSC

    you are nervous, but I am here to tell you your instinct is correct. Read what is there. Read what is not there. Give yourself time to go through it carefully. It is stressful (I imagine) to have to rush this. Don’t do it on the way to the meeting unless you have loads of time and reliable internet access. You will likely need to google stuff. Read the charter, read the data management plan, read it all. If you can, read it twice. There is an extra responsibility on the chair to be informed about the trial, so take it seriously. Once you have read it all, collate your notes and group your thoughts. It is important for you have clear in your mind what the broad issues are without getting bogged down in details (e.g. it may be that you want terminology standardized across all of the documents without listing all the places it is discordant). You can get into the detail time permitting, but it is likely you will be sending on specific written comments (particularly for the protocol) later.

  2. I found these papers particularly useful in defining the role and expectations of a TSC. “Exploring the role and function of trial steering committees: results of an expert panel meeting” & “What are the roles and valued attributes of a Trial Steering Committee?” & “MRC guidelines for the management of global health trials“. I then opened the meeting with a description of how I interpreted these recommendations and the order of importance I place them in. For the record, for me the broad TSC roles are:
    1. Participant safety (encompassing the health and wellbeing and rights of the participants)
    2. Offering advice- the TSC is a repository of dispassionate expertise which is made available to the study team.
    3. Endorsing TMG decisions- the TMG are much closer to the study than the TSC so I think it would be unusual for the TSC to be regularly overruling the TMG. Obviously if there is something you don’t agree with you need to say, but this is the exception rather than rule in my experience.
    4. Making decisions- in some cases the TMG will refer areas of genuine uncertainty up to the TSC for a final judgement. Be prepared to make these calls and provide a rationale for your decisions. We decided to amend our charter to remove a section which said that when a consensus could not be reached that the chairperson would take the final decision. We replaced this with a voting system, where independent member votes outweighed trial team votes. This made me happier as chair.
    5. Representing the funder- perhaps it is my inexperience, but for me pre-empting the imaginary mind of a faceless conglomerate is a little too abstract, so I interpret this to mean representing the interests of science and society (equally nebulous I grant you, but makes me sound more noble so I’m sticking with it).
  3. Man taking minutes.jpg
    This guy’s taking minutes before the meeting’s even started. This is not necessary.

    One of the most important things (that I nearly forgot) is the minutes. Minute-taking in general is one of the most underrated research activities, but for a committee that meet as irregularly as a TSC they are vital. So, first thing (probably even before introductions) is to sort out whose role it is to take minutes. Ideally, as chair, this isn’t you- you have enough to be worrying about. Secondly, the minutes must be released promptly (we agreed they would be sent to me within a fortnight of the meeting). As chair, you are independent of the study team. This is good for providing objective oversight, but bad for developing a deep understanding of the study. You aren’t involved in it that way, so you need to work hard to remember what the study is about, what are the features of the study that make it unique or tricky. Timely minutes are really important for you to be able to remember what was discussed. Similarly, when those timely minutes are produced you should be equally prompt at reviewing them and sending back for circulation to the wider TSC.

  4. Conch_drawing.jpg
    Don’t hog the conch

    One of the tips that the aforementioned papers recommended was providing your opinion as chair after everyone else does. The chair does hold a certain sway (even a newbie chair) so be conscious of that and try to get a feel for the room before weighing in with your own thoughts. As chair, one of your primary functions is eliciting consensus. This is much easier to do if you are informed about where everyone is starting from.

  5. There is lots of talk about the TSC being a critical friend to the study. I think this is a pretty good starting point, and in particular that the noun in this phrase is friend. We are not friendly critics, but critical (if necessary) friends. This is part of the reason why a chair should have experience of being a study team member at a TSC before becoming chair of a TSC. It is important to have a detailed understanding of how exposing it feels presenting your work to an entirely new committee with a whole different set of viewpoints. I think it helps to remember that ultimately we are all on the same side and all trying to ensure that the best job is done on answering the research question.
  6. One thing that I have found interesting as a statistician chair is that I have to internally moderate my statistical critique. Previously, as an independent statistician I felt that it was clear that I would be championing the importance of the quantitative aspects and focusing mostly on ensuring their rigour. My viewpoint would then be taken into consideration but the rest of the TSC and a preferred way forward agreed. Trials are complicated though and in some settings the statistical view doesn’t override everything else. So as both independent statistician and chair I felt I needed to be careful about my statistical views being expressed as chairperson’s views. This was a subtle distinction, which I think I only really appreciated about half way through the meeting.
  7. Timekeeping. This I pretty much failed at. The first meeting is so difficult to plan. It terrible clockis really unclear beforehand how communicative and forthcoming your TSC members and CIs will be. It is impossible to know which of the discussion points will become thorny and for how long. My first TSC as chair overran by about 90 minutes because there was so much to discuss and everyone was generous with their contributions, which is a great. However, we had to finish up the final parts of the meeting without two of the independent members, which is obviously not ideal. I know that in future I would ask the chief investigator to take a guess at how long each point should take, just so that I have a closer handle on how much I am overrunning. As a guide, i reckon that three hours is a minimum for an initial TSC meeting.
  8. TLAs. Three letter acronyms. Acronyms everywhereThere should be PPI (see!) representation on the steering committee and so it is important to try to cut down on jargon as much as possible. Even experienced triallists use different terminology (DMEC, DMC and IDMC all mean the same thing). Try to remember to speak plainly and clearly for everyone’s benefit and encourage everyone on the committee to pipe up if something hasn’t been introduced properly.


These were my reflections on chairing a TSC for the first time, but I would love to hear from others of all experience levels. Feel free to comment below.


Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )


Connecting to %s