A patient’s journey as they navigate and choose from the many treatments and options available to them is highly unique. The sheer number of variables can be overwhelming. Aggregate patient journey information is locked in APLD (anonymous patient level data) sources like claims, prescription, and EMR data sets. Logically sifting through multiple multi-billion row data sets looking for actionable insights is important for today’s pharmaceutical manufacturers. It requires the most leading edge possible data infrastructures to handle at scale.
The goal of analytics
Using analytics to find trends in healthcare is nothing new. Predictions of disease prevalence and potential outbreaks have saved countless lives and allowed for epidemics preparations and adequate quantities of necessary pharmaceuticals to be available. In other instances, these analytics have helped to pinpoint at-risk groups or track disease progressions.
Analytics are also used to track commonly prescribed drugs for a variety of illnesses, giving us information on the efficacy of each of those medications for that application. We are also able to look at the symptoms patients are presenting with prior to being prescribed a given medication.
Pharmaceutical companies are also interested in ascertaining what symptoms may be leading to any inappropriate prescriptions. This could be a game changer in the battle against antibiotic resistance. It is also important to find insights on the frequency of incorrect prescriptions based on a particular disease.
Data allows healthcare professionals to determine how patients across different demographics are treated. For instance, many free or low-income care clinics may routinely prescribe certain drugs for financial reasons. The efficacy of these drugs can be ascertained in relation to the diseases they’re being used to treat and the groups to whom they’re being given. Efficacy, side effects, hospitalization rates, and performance versus competitive offerings can have massive impact on formulary coverage and top line revenues if properly utilized.
The trouble, of course, is getting to the point where these insights are easy to attain.
The challenges that come with examining longitudinal patient data
The true challenge in pharmaceutical analytics is in drilling down to longitudinal patient data. Longitudinal data is information that tracks the same information across the same group of subjects over a long period of time. Pharmaceutical companies use this patient-specific data for business insights.
Getting use out of longitudinal patient data can be the difference between unrealized sales potential and choosing the correct path towards profitability. However, because there are over 280 million active patients in the U.S., this information can be difficult to put into perspective for many pharmaceutical companies. This is especially true for organizations with small teams or those using older analytics solutions.
Longitudinal patient data can contain tens of billions of rows of information comprised of doctors visits, blood work, etc. Historical data complicates this even further. More time coverage compounds the data set size.
It’s integral that you are able to query this information and determine how factors are affecting the overall repeatable patterns for patient journey mapping.
What are the best means for gaining an overall view of the bigger picture? What makes patients similar in their treatment journeys and what sets them apart from one another? Are there certain trends in their healthcare choices that seem to be invariable from one patient to the next?
Finding the answers to these questions requires you look at the longitudinal patient data, but extracting this information is incredibly difficult. Part of the problem is the continued use of cumbersome and outdated analytical tools.
What to look for in your pharmaceutical analytics solution
Traditionally, pharmaceutical analytics uses R, SAS, or MatLab for their statistical analysis. These platforms operate on relational databases or distributed environments.
Using time-series databases is a much faster and more cost-effective way to retrieve information when you are ready to query it. This alternative method increases your time to insights because it is better suited to handle information with billions of rows, which is common in the pharmaceutical industry and particularly in longitudinal patient data.
Aside from faster querying of longitudinal patient data, users should also have the ability to change their query parameters ad hoc and isolate specific drug regimens for more granular analysis.
Analysts should also look for tools that allow for advanced filtering, implementation of sophisticated business rules (ideal for concomitant drug regimens), and data extraction functions that supports additional reporting. These features will give pharmaceutical companies a more adaptable approach to look at longitudinal data, meaning they are able to approach the same datasets from multiple angles without using a large amount of resources.
Addressing the common patient journey challenges is a necessity in pharmaceuticals and healthcare to continue advancing patient care. The pharmaceutical industry needs to utilize the best data science tools in healthcare to improve their understanding of the patient journey. At RxDataScience, we provide tools for faster insights, advanced querying and filter functions, and dynamic visualizations for large datasets. Learn more about RxDataScience’s products and services.