Keep it Clean: Three Examples of SQM Driving Quality Data

Keep it Clean: Three Examples of SQM Driving Quality Data

Earlier this week I discussed how to design Study Quality Metrics (SQM).  Now, let’s take a look at three examples of SQM reports that helped identify and mitigate potential problems.

Example 1

SQM: Consistent adherence to key inclusion and exclusion criteria

Rationale: If these criteria are not met, the enrolled population may no longer represent the target study population. This can potentially lead to a diminished treatment effect if the enrolled subjects have a different severity or type of disease than intended.

Data: The chart below shows that while Site 5 is falling slightly short of overall study median for this metric, an even larger percentage of subjects at Site 3 do not appear to meet the key eligibility criteria, and Site 3 has dropped below the minimum standard of 75%.

Corrective action: Review the inclusion/exclusion criteria with Site 3 personnel, stressing the importance of enrolling a representative study population.

Example 2

SQM: Consistent number of adverse events reported

Rationale: Under- or over-reporting of adverse events (AE) can result in an inaccurate safety profile for the study.

Data: The SQM report suggests that Site 2 is reporting fewer adverse events per subject than other sites in the study, and this difference has crossed the minimum standard for under reporting.

Corrective action: Review of the guidelines for AE reporting and Case Report Form completion with the site.

Example 3

SQM: Site visit variability

Rationale: In a study with a time-to-event endpoint, events are often only recorded at clinic visits. Imbalances in the visit schedule for subjects in the active and control arms can impact the reported time of the event. This may result in a scheduling bias towards longer event times in one arm of the trial.

Data: The chart below shows that Site 5 has slightly low percentage of visits within 1 week of expected timing in comparison to all sites combined. More importantly, Site 1 has dropped below the minimum standard set for the study.

Corrective action: Identify whether the scheduling discrepancies at Site 1 relate to drug intolerability or simple scheduling problems. The solution could be as simple as providing a transportation service for subjects enrolled at that site.

An SQM reporting strategy allows for timely reaction to any developing data problems and leads to cleaner data, more credible results, and a greater likelihood for a successful study. Creation and review of SQM reports provide useful ongoing feedback to study teams regarding data quality and the ability to meet critical study objectives.

Contact us to discuss how SQM can ensure quality data within your clinical trials.

Subscribe to the UBC blog to stay up to date on clinical trial data management strategies.