In today’s environment, it is almost certain that the pharmaceutical industry will continue the downsizing of drug development operations. In doing this, they will place an ever-greater reliance on strategic CRO alliances and collaborative technologies. As sponsors increase their reliance on CROs, the need for better collaboration and more transparency will also increase.
In managing a study, CROs must use a clinical trial management system (CTMS) to collect the necessary operational data. In some cases, they will use the sponsor’s system. But most of the time the CRO will use their own system. The CRO will give periodic updates to their sponsor-clients in the form of weekly/monthly updates, spreadsheets, reports, etc. Many CROs even go so far as to create portals for sponsor personnel to access at any time. All of this is helpful. But typically it’s not enough to make the sponsor truly happy.
Study managers on the sponsor side might be perfectly fine with this arrangement – after all, they are getting what they need (more or less) to oversee their study. Others at the sponsor might not be so thrilled. Anyone who needs to keep an eye on multiple studies can find this method of information sharing to be chaotic. This is especially true because most companies use not one but many different CROs, each with their own portal and varying sets of reports.
This being the case, those looking for updates on multiple studies can become quickly overwhelmed. In addition, the sponsor often wants this operational data in-house for use in trending analyses, populating information in an investigator database, and the like. Data integration from the CRO sounds like a smart and easy decision, but does it ever happen on a consistent basis?
Can Data Integration be Easy?
In my experience, no. There are always lofty ambitions followed by lowered expectations followed by a resigned acceptance that not all the CTMS-ish data is going to be aggregated in-house after all. Even when the sponsor and CRO are using the same system, APIs need to be set up for each study. This requires a good bit of work up front (though subsequent connections can be far less painful). If managed from a high level, where a standard set of data to be transferred back to the sponsor is consistently defined and managed for all studies across all CROs, it could work. But this is rarely done. And the wheels tend to come off the wagon in rapid fashion.
Alternatively, you can extract data to spreadsheets and then import it into the sponsor’s system. There are issues with this as well:
- inconsistent data requests
- lack of follow through in loading the data
- difficulty in accessing (or mapping to) the sponsor systems’ back-end data tables
So what’s the solution? Or perhaps more accurately, what are the right questions to ask?
Do You Need a CTMS?
How about this: Do you really need a CTMS? If all you are looking for is a way to do aggregated reporting (i.e. across studies) and/or trending analyses, why would you need a CTMS in the first place (especially since your good friend the CRO has one)?
If you don’t plan on using your CTMS to enter data (beyond basic study metadata), using a SQL database and/or a platform such as Microsoft SharePoint is an excellent (and cheaper) alternative.
A CTMS is simply a front-end data entry tool to populate information into a set of tables. If you don’t plan on using your CTMS to enter data (beyond basic study metadata), using a SQL database and/or a platform such as Microsoft SharePoint is an excellent (and cheaper) alternative.
Granted, you may have studies that you do in-house where you might still need a CTMS for tracking purposes. It is possible however to build a “CTMS Lite” using SharePoint or other tools which may (or may not) suit your needs in this capacity. This factor aside, by stripping away your CTMS you will have far more control over the back-end data tables. And when coupled with the proper reporting and/or analytics tool, you can view data on your portfolio of studies nine ways to Sunday.
There are still the business-side issues to take care of – figuring out what data you want, working with your CROs to get consistent extracts in the proper format, and making sure everything actually happens. But by taking CTMS out your side of the equation, you may find that those lofty ambitions are a little more within reach.