Do Ethics Committees really need to meet? Rethinking Asynchronous vs Face-to-Face Review

An ethics committee meets once a month. On the agenda are 18 applications, ranging from lower-risk survey designs to higher risk health intervention studies.

Time is limited. Discussion may be uneven. Some applications are waved through quickly. Others are debated at length.

A week later, a similar study is submitted to one the committee has just seen, but must now wait three weeks for the next meeting.

In another system, the same application is reviewed within days through an online, asynchronous process. No meeting, but still a panel is convened and robust discussion had – just through an online process.

So which approach is more robust, and which is simply more familiar?

The Tradition

Ethics committees have traditionally relied on face-to-face meetings to review research applications. The logic is intuitive: bring experts together, discuss the issues, and reach a collective decision.

But as research volume increases and timelines tighten, this model is being challenged.

Increasingly, committees are adopting asynchronous (or “rolling”) review processes, where applications are assessed independently and decisions are made without real-time meetings.

So which approach is better? The answer is less straightforward than it seems.

The Assumption: meetings equal rigour

Face-to-face review is categorically assumed to be the gold standard.

It allows:

• Real-time discussion

• Clarification of any concerns

• Collective deliberation  

It also allows for rapport to be built with its members and their secretariat. Plus, it comes with coffee and a sausage roll. If you are not affiliated with the institution, you may receive compensation for your travel and your time.

This gold-standard assumption deserves scrutiny.

There is limited evidence that meeting-based review consistently produces better ethical outcomes. Indeed, when scoping the literature for this blog-post, we see some discussion about the shift from in-person to over Zoom during the covid-pandemic, but less on the removal or the requirement of that face-to-face element.

The Reality: group dynamics matter & a strong chair is vital

Face-to-face committees are influenced by:

• Personalities of members

• Time constraints for deliberation to be undertaken (one day or afternoon a month)

• Limited availability for membership attendance (particularly in peak-business hours)

• The possibility for group-think consensus pressures

They are also influenced by how a committee manages its review process. It may be the assumption that members should read every application, but is this the case in reality? It is likely that two members of the committee are ‘assigned’ several applications for review, and those two members lead the discussion for those applications. For all others, they might take a backseat. To deal with all these issues, a committee needs a strong chair who holds the members accountable.

Asynchronous review, by contrast, requires:

• Independent assessment in a flexible working environment (in a reviewers own time)

• Written justification for decisions noted in systems

• Selection of reviewing panel based on expertise for application

• Structured input from all reviewers (in other words, the full panel reviews the application)

Speed vs Scrutiny: a false trade-off

A common concern is that asynchronous review prioritises speed over rigour. In practice, this is not the case. The removal of the face-to-face does not remove the job of the reviewer to review an application.

For many well-written studies:

• Ethical issues are clear

• Risks are well-managed

• Decisions do not require extended face to face debate

In these contexts, delaying review to fit a meeting schedule does not enhance quality, it simply slows the process.

Where face-to-face still matters

This is not an argument to eliminate meetings. There should always be scope for a face-to-face meeting, if required or requested by the researcher, the Chair or any member of the reviewing panel. We are just challenging the normative approach to review, saying that a meeting is not always required.

Face-to-face review remains valuable where:

• Ethical issues are complex or contested

• There is disagreement among reviewers

• Additional context or clarification is required

In these cases, discussion can deepen understanding and support more robust decisions. It also provides a learning opportunity for the committee members who may be encountering these particular issues for the first time.

A more pragmatic model

Rather than choosing one approach over the other, many committees are moving towards hybrid models. Worldwide and in New Zealand, many committees have processes for sorting low-risk research, and the chair deals with these applications, sometimes in tandem with a couple of committee members working via email.

The Collective is one such agency that uses a hybrid-model. We are led by our seasoned ethics committee members and Chairs to know when a face-to-face must be implemented, and the NEAC Standards to guide our understanding of 'high-risk' projects which require a committee meeting.

Moreover, The Collective reimburses its members for their expertise. Irrespective of what they determine the outcome of the review to be, reviewers are reimbursed for review (simply put: pay is per review), at their own determination and not around a monthly meeting table.

To ensure our committee members have the opportunity to build rapport with each other, The Collective has monthly training sessions and drop-ins which members can attend if and when they can.

Why this matters

The structure of ethics review affects research timelines, researcher experience, committee workload and ultimately, participant protection. A process that is too slow can discourage good research. A process that is too superficial can miss important risks.

The goal is not to preserve tradition, but to apply the right level of scrutiny, in the right way, at the right time.

Ethics review is not defined by whether a committee meets. It is defined by the quality of its judgement. We argue that this can be achieved through more than one model.