It is estimated that there will be approximately 25,000 petabytes of healthcare data available by the year 2020. Contrast that to the estimated 5,000 petabytes of healthcare data available in 2012. This developing access to healthcare data will, no doubt, lead to advances in science and improve health outcomes, but doing so depends upon continued advances in the management, organization and integration of data so that it can yield comprehensive insights.
Here are three reasons why today’s wealth of healthcare data is complicating the business planning process for life sciences companies.
1. It’s a chimera
There’s no lion's head, goat's body or serpent's tail here, but business planning in the life sciences industry is challenged by disparate data sources. With advancements in science, the treatment of cancer has grown to increasingly involve oral medications whose multiple distribution channels contribute to the fragmentation of information related to cancer drug therapy. To predict market opportunity, life sciences companies have grown accustomed to integrating conflicting findings from disparate data sources, and, too often, just like the mythological creature, the result is malformed.
The vagaries of complex modeling and multiple levels of assumptions used to concoct disparate data as a single, coherent data set introduce uncertainty that is difficult and, probably, impossible to quantify. In addition, basing critical, multi-million-dollar investments on data chimeras is not only risky, it’s expensive.
2. The window of insight is too narrow.
In the life sciences industry, single sources of data—ordered treatment, administered treatment, dispensed drugs, billed claims or paid claims—provide narrow windows of insight into drug markets. Too often the view through each of these windows differs. An informative panorama reconciles the conflicting “facts” reported by data reflecting different stages and sites of care in the patient’s treatment history, making credible panoramas hard to find.
3. You could get misleading results.
The data that drives drug innovation is almost exclusively samples of opportunity. Typical samples of opportunity in the life sciences industry include electronic medical record (EMR) data from a healthcare provider or EMR vendor, claims data from an insurer, drug wholesale data and pharmacy prescription data. This data is most often packaged and sold as unprojected sample distributions and may not represent a specific population, although it may be used as if it does. Treating a limited, raw sample as a microcosm of a population can yield the types of misleading results that undercut business planning.
To better understand the market and confidently make strategic business decisions, life sciences companies need a 360-degree view of patient data, dispensing data and claims data—potentially from a nationally representative distribution of sites of care—built at the physician, practice and patient level and blended into one common language.