RBQM Ops 2022 - Do you trust data enough to make a decision?

Do you trust data enough to make a decision?

SUMMARY KEYWORDS
data, key risk indicators, risk, site, signals, adverse event, qm, proportionate, reviewing, decisions, primary endpoint, risk assessment, subject, visits, flagging, rbqm, indicator, information, study, rates

DUNCAN HALL:

The next subject is a very important one. And we know that because in talking to many organisations, we know that some people just don't trust the data enough that they're seeing in front of them to make decisions, and so therefore, they tend to revert back to old methods such as SDV. So the question of, do you trust the data enough to make a decision is a very pertinent question. It's on that topic then that I'm going to be introducing my colleague, Kirsten McAulay, who's the RBQM Operations Manager here at TRI. In that role, Kirsten talks to many of our major customers. And that enables her to bring her 25 years of technical experience in helping to move them in the right direction and to make the right decisions based on their data. So thank you, Kirsten, over to you.

KIRSTEN McAULAY:

Good morning. Good afternoon. Good evening. Hopefully that covers everyone. And thanks for the introduction, Duncan. I agree it's an important question. And one, which to some extent, can block us from even getting started with RBQM. So how do we trust the data?

So, to get started, then I believe having confidence to trust the data comes back to what we are trying to achieve with RBQM. So, looking at this question from an RBQM perspective, what I mean by that is taking into account what is going to impact the safety of subjects, integrity of the data, and the reliability of the study results.

To do that, I'm going to focus on the absence of errors that matter, identifying sites or subjects which represent a potential risk to quality and taking a holistic view, so really being able to review signals and trends in the study level, country level, site or subject. I'm not setting out to try and find every single error or possible trend that the data could show, I'm really focusing on what matters for this specific trial.

So, what do I need to consider then to set the RBQM up for success. So RBQM doesn't start with the production of data, they need to start thinking about RBQM right at the start leveraging our initial protocol risk assessment. So illustrated on this slide is the risk assessment process as mapped out in ICH GCP. So, starting with the identification of critical data and process, the identification of risks, and then risk evaluation, we know what's important to the trial. And we can then consider what risk controls can be put in place, and will be proportionate to the level of risk and the risks that have been identified.

For example, in my study risk control strategy, I'm going to give consideration to those risks, which result in errors that matter. So those risks that really could make a difference to the data that's being collected. So that could mean errors that will impact the results and statistical analysis, or the regulatory acceptance of the trial, or even subjects. I will make sure my control strategies are documented, including in the monitoring plans.