Outsource, outsource, outsource. It seems to be a mantra that has hit even clinical trials these days. The list of companies with significant data management outsourcing seems to keep growing. Wyeth, Eli-Lilly, AstraZeneca, Pfizer, to name a few. How is it that this situation has arisen? From a variety of factors.

When I first joined the ranks of data management, more years ago than I care to count, the staff was almost exclusively young recent college grads, everything was done on paper, and what standards existed came from the simple fact that it is easier to copy and modify from a prior study than to build everything from scratch.

Broadening Your View

All of that has changed. Data management organizations have matured, in both senses of the word. The average age of the staff is older, and the processes have matured as well. Technology and standardization have allowed much of data management to be viewed as a commodity. This commoditization has made outsourcing possible—and financial pressures have made it desirable.

So what is the future of internal data management departments? Are they to be completely replaced by outsourcing? Are they to be transformed into a small unit that simply oversees the outsourced work? Or are there other opportunities? It is my opinion that the present situation should be viewed as an opportunity.

But a shift in thinking is necessary first. Data managers need to stop thinking of the data management department as a cost center that simply manages the arriving data. Instead, data managers need to look for opportunities to expand into providing value-added services. By doing so, we can direct the evolution of the internal data management units to strengthen their value proposition and thus maintain and increase their relevance to the enterprise.

What’s the Mission?

To change our mindset, we need to step back and consider the business we are in. Perhaps an analogy will help. What are customers willing to pay biotechnology and pharmaceutical companies for? Do they want white, pink, or blue pills? Do they want injections, inhalers, pumps? The answer is no, no, and heck no. I don’t want to take any pills. Do you? I hate taking medicine.

What I want—what we all want—is good health. That is what I’m willing to pay for. Cure my infection, stop my headache—and I’ll gladly pay. If you can’t cure me, then at least alleviate my symptoms. For this I’ll still pay. If all else fails, then at least give me hope—hope that my cancer will go into remission, hope for more time. Yes, even for that, I’ll pray while I pay.

Bring It Together

What has this got to do with data management? It represents the identification of areas of opportunity, areas where data management can add value. Data management organizations can combine data from various sources to deliver meaningful information that the company needs as badly as a patient wants to get rid of a headache. Here are some ideas: Focus on cost containment. Combine data from various sources to improve operational efficiency. Seek out and identify non-traditional data sources that can be summarized, related to clinical trial data, and delivered as information.

Note here that I deliberately prefer the term “information.” Data is a commodity. It is the transformation of data into information and presenting that information in a manner that supports decision-making that holds value.

There is a wealth of information hidden within the data. But we need to know how to get it. A great example to illustrate this point comes from a clinical trial management system. Take a very simple question: how many sites do we have in the U.S.? Sounds like a simple question, doesn’t it? But it hides many nuances of database design and the need to know what is in the database to know how to refine the question to get the desired answer.

Data vs. Information

Are we asking for all sites that we’ve had in the U.S., or just active trials? If active trials, are we only interested in sites that have actually enrolled subjects? Do we want to know all sites in the U.S., or just sites that are managed by the U.S. medical group? Do we want to include or exclude sites from investigator-led trials? From post-marketing trials? And so on.

Another example: What is the average weight of subjects in this trial? Are we including subjects who dropped out? Do we care if patients were on the test product or in the control group? Do we want to keep in or exclude repeated vital sign profiles? Does the average weight really mean anything, or do we really want the body mass index?

Experienced clinicians will know how to phrase a request to get the most out of a database. What about others in the organization? Providing services to these others is an opportunity.  As I said, there is a wealth of information hidden in the data, especially once we start to combine it with other sources. Data managers should not limit their view of the world to the traditional clinical trial databases. They have established skills in working with databases, in understanding data models, in knowing how to retrieve data and turn it into information.

Outside the Firewall

Data managers need to think about how to serve that information to various user groups. Too often, data management groups focus on a narrow internal customer base that consists primarily of biostatisticians and, to a lesser degree, study managers and medical reviewers. Associated with this is a narrow approach to managing the data. At some companies, data management is merely acquiring, cleaning, coding, and delivering data. It is time to move beyond these basic activities and move into more active management of such a valuable asset. In managing data and information, as in managing any asset, we need to look for ways to maximize their value.

So data managers should expand their perception of the potential customer base. These might include study management, portfolio management, resource management, marketing, health economics, accounts payable (for investigator payments), safety/pharmacovigilance, and drug supply. (Many readers can think of ways to easily lengthen that list.) Next, consider new ways of utilizing the existing data to present information. Consider increased use of trend analysis for forecasting enrollment rates and drug supply needs. Look beyond the data from just the current trial and take advantage of historical data from related trials.

Looking Ahead

Once a data manager has expanded his or her view of the customer base, they will find that there are other data sources that can be woven together with proprietary data. Data managers are well equipped to bring in this data and deliver it in a meaningful way. Consider using commercial databases of health care records and comparing various profiles between these databases and those available internally such as from clinical trials and from safety databases.

Some of these activities might step across current department boundaries. So be it. If this is the case, concentrate on offering to partner and provide services. For providing these services to maximize the value of the data, transforming it into meaningful information, is where the future of clinical data management is headed.

welcomes feedback about data management and other topics. He has more than 25 years of experience in the pharmaceutical industry, including data management, programming and IS support. He has personally witnessed the evolution from a pure paper process to the development of e-clinical processes.