From its infancy until now, clinical trials have made major progress in terms of technology. eClinical emerged and quickly became a budding business. Yet software and services continued to develop to further ease the trial process, of which clinical trial analytics (CTA) is the to-go management method.
It is impossible to leave out eClinical services when it comes to smoothening the trial process. Yet challenges remain. With independent software, say clinical data management system (CDMS) or electronic data capture (EDC) and clinical trial management system (CTMS), data are recorded separately. In other words, data are scattered everywhere and hence it is difficult and painful to cross-query.
That’s where CTA comes into play. By combining data from different sources and extracting them into a visual form, this data-driven approach enables trial investigators to track metrics and key performance indicators (KPIs) so that they can more effectively organize the studies. These include study metrics, site metrics, quality metrics, and data metrics. By getting hold of the trends and patterns of meta-data, researchers can improve the quality and efficiency of the trials.
The design of the trial is the first thing to which conductors have to pay attention. Normally, based on the expected operational and statistical outcomes, trials will be designed theory-wise. However, such approach may not be feasible in reality, because emerging problems can seriously affect the procedure and can even put the trial on hold.
Luckily, there are lessons to be learned from the past. Analysis of historical data can give researchers a big picture of what can possibly happen to one trial design. On a higher level, CTA allows simulation of different designs, so that conductors can have a rough estimation of the duration and overall cost of the trial in advance with minimum, maximum, mean values, and standard deviations. As a result, data-driven designs contain much higher success possibilities and lower risks.
Patient Recruitment and Retention
Having been involved in a clinical trial at least once, anyone has to agree that patient recruitment and retention is the most aching sore throughout the whole trial process. Not only is it the most costly, but it also the major cause of delay in trials. Yet on the bright side, CTA can be of great help in alleviating the problems.
While scouting patient availability, researchers can make use of existing database from various sources such as the Food and Drug Administration (FDA), pharmacy and hospitals to locate eligible patients. Medication restrictions, as well as inclusion and exclusion criteria, can narrow down the range. The results are shown in the form of a heat map, in which the hottest contain the largest pool of potential research subjects.
On the other hand, applying patient retention process reduces the patient drop-out rate. This works exactly like the customer retention strategy. Just as businesses collect KPIs to gain customer insights and understand their frequent behaviors in order to boost sales, researchers also base on those metrics of patients so that they can encourage them to appear at the next visit. Since it obviously costs much less to retain a patient than to recruit a new one, this data-based strategy has proved itself highly effective in shortening study duration and maintaining trial quality.
Though much less discussed, clinical supply management also poses not so small a problem to trial managers. Failure to properly allocate drugs to sites can result in re-distribution to other sites and waste of unused drugs, which lengthen study time and increase costs. The issue magnifies in adaptive clinical trial design, where the dose of drugs may subject to alteration.
CTA enables researchers and trial monitors to give more accurate forecasts of the amount of drug needed by each site. Depending on the current patient recruitment situation and historical data, drug quantity can be better calculated so that wasteful situation can be minimized.
Abandoning sites due to failure to recruit patients is a common problem clinical trials have to deal with. 80% of sites reported to recruit one or fewer subject in Phase II and III between 2008 and 2009. The opening and closing of a site impose a great financial burden on trials, while at the same time prolong it.
It is wiser, then, to utilize CTA while choosing sites. The patient heat map can be laid on FDA’s databases of investigators for patient clusters to pop up. Meanwhile, such sites’ historical data can be used to evaluate their potential and compatibility with the purposes of the trials. By delving into the site metrics, trial conductors stand a higher chance of choosing optimal sites that guarantee adequate and high-quality data.
Back in the days, clinical research associates (CRAs) made equally regular visits to every site for check-ups and tackling issues. This approach appeared too rigid to be effective because each site behaved differently. Sites that ran smoothly didn’t need too much attention, whereas sites that struggled required more frequent investigations. Hence, this traditional way is unable to optimize resources.
Two trends are emerging with regards to data-driven site monitoring. One is utilizing CTA to predict patient enrollment. Monitors anticipate a full-day workload so that CRAs can arrive at that day. The other one is risk-based monitoring, which basically means that CRAs focus on the ones with the greatest risk of data quality problems. These sites will receive the most frequent visits while sites with smoother workflow are investigated much less. FDA also encourages a risk-based approach in order to prevent data misconduct as well as cleaner and more trusted trial results.
It is important to note that CTA is not a substitute for CTMS; it is rather an add-on. By combining data from different sources and extracting them into readable and actionable reports, CTA enables smarter decisions that lead to a more effective and efficient study. Data-driven clinical trials are no doubt the to-go way of the future.