Tracking Drug Supply Chain with Healthcare Business Intelligence
 |
Focus on: Why Cloud for Pharma |
Passed in 2013, the Drug Supply Chain Security Act (DSCSA) gives the Federal Drug Administration (FDA) more control over the manufacturing of drugs in the United States. Starting in 2015, with full compliance by 2023, most pharmaceutical drugs now must be traced, verified, and serialized for the entire supply chain, from manufacturer to pharmacy.
This provides improved healthcare business intelligence and brings confidence and security to patients by ensuring that their medication is the correct prescription, is still within its shelf life, and not from a black market.
Whereas consumer confidence is of great value to the companies in this supply chain, it also brings complexities on compliance. Whether your company is in manufacturing, wholesaling, repackaging, a dispensary or pharmacy, the DSCSA is primarily a compliance project given to Policy Operations or IT departments. This leaves them needing to find a solution or vendor that can perform to legal compliance, but also have a workable end-user experience. But, what if it could be much more than that?
With data starting at manufacturers and growing through each section of the supply chain, the amount of data gathered from compliance with DSCSA is immense. But, if used with machine learning, AI, and predictive analytics this data can help companies go from big data exhaust, to big data learning. It is all in how you utilize your back-end computations.
For example, Microsoft helped a Children’s Hospital take weather and other data, combined with sentiment analysis from social media sites to accurately predict asthma outbreaks in the hospital region. The information gathered from all that data gave the hospital the knowledge to know when to reach out to their patients with awareness and effective home treatment campaigns. This care coordination helped lower incidences of their main reason for emergency room visits, treatment for asthma.
At another health care provider, Microsoft helped build a system that tracks a patient’s symptoms at home—via wireless health monitoring devices that the patient uses, combined with other feedback to accurately predict the probability of the patient returning to the hospital. This novel use of medical data and clinical analytics not only allows real time changes to a patient’s wellness plan, but also saves the hospital money. For some of these patients it is adding years to their lives and rewriting the statistics on recovery for certain conditions.
In both cases, it is about following the data; using and grooming figures and other unstructured data that might not seem to be relevant at first glance. This innovative use of data is providing huge insights that allow health care providers to not only improve the way they work, but completely rethink how they operate.
Imagine quickly tracking a recall down to the patient level. This and many other opportunities are made possible by having the end-to-end pharmaceutical data. Other examples are:
Better forecasting across the supply chain.
- Manufacturers can ramp their productions up or down based on the data, in real time. Using predictive analytics, everything from purchasing raw materials to staffing needs, can be streamlined to reduce costs.
- Wholesalers and repackagers can begin to predict buying patterns, creating a proactive edge in providing what the dispensaries and pharmacies need, exactly when they need it. They can create customized bundled offerings that help to ensure a consistent pipeline and potentially increase margins.
- Dispensaries can tighten down on waste by ordering to modeling patterns based on healthcare business intelligence that allow them to meet patient needs.
Health Care Advantages
What if a health care provider could track the usage amongst non-typical users of Metronidazole within a specific region, which when correlated with other data, gave an indication of the rise of a drug resistant bacterium? That would be a big deal, especially when you realize this alert could be done in real time by a machine learning neural network. Now what if this system could also go across other data sources (linkages to other drugs, anti-diarrheal medicines, hospitalizations for non-specified GI complaints, etc.) to look for correlations? It could more accurately target a population trend. This could even be correlated to the efficacy of that drug in treating the symptoms.
This illustrates but one of multiple use cases that have a similar theme, they allow the health care provider, public health authorities, and other interested parties in getting in front of health events within a population. For the health care community, this type of information is key, allowing for near real-time visibility into health events and even prediction of how big the event is likely to be, including where and how it will spread.
Effective Authenticity and Efficient Recalls
Patients will be comforted in knowing that the medications they are taking are made by official manufacturers and that the efficacy of the drug hasn’t expired. There are also significant advantages in the event of a drug recall. The recall would be down to the patient level and not just the lot level. This is a big advantage in helping contain the damaging impacts of a drug recall while ensuring full coverage.
Bottom Line: Follow the Data to Lower Costs and Greater Revenue
The advanced analytics models stated above lead to lowering costs and driving revenue across the supply chain, while providing a higher quality of life for the patients being served. Companies in the drug supply chain should be looking not only at the trace and track software but should also be looking very closely at choosing the right computational backend as well. As the healthcare business intelligence grows, so does its potential, creating a situation where choosing the right back-end can become as important as the front-end software decision.
There are many computational powerhouses out there for you to choose from. Here are some tips:
Look for a cloud data provider over on-premise data storage. This will be more cost effective with the exponential growth of your database, while offering predictive analytics engines as an embedded feature.
Look for one that can handle the artificial intelligence of the data, with predictive modeling and machine learning capabilities.
Part of the equation is being able to read the data effectively, so having a robust data visualization layer is especially important. These tools have replaced having to look at numbers in spreadsheets with visual representations of data which help present key healthcare business intelligence metrics more easily.
Perhaps the most important question to ask of the computational back-end providers is breadth.
You are choosing a system that must carry you forward for the next decade and beyond. Look for the provider that has leadership not just in one or two areas, but who has the strongest overall breadth and depth within the competitive landscape of the digital transformation space. Each provider will have core competencies that are within their wheelhouse, but do they provide a set of capabilities that will grow and evolve to meet the needs of companies along the supply chain as their needs change?
Finding the right provider to meet your need for a robust computational backend could be the most important decision in your selection of a platform. It needs to meet the necessary DSCSA compliance while creating a powerhouse of cost reduction and revenue from the data generated. Those who choose the best platform will have a competitive edge over the companies that do not. Ensure your company has the tools it needs to be at the forefront of this digital transformation of health information management and the drug supply chain.
Learn more:
About Eric Henze
Eric Henze works for Microsoft as an executive consultant to Chief Technology Officers, CIO’s, CMO’s, VP, and Senior Director’s on enterprise wide digital transformational improvements. Working within the largest payors and providers in the healthcare industry, current focus areas are in predictive analytics for end of life and reduction of readmission, IoT for hospitals, along with deep learning and artificial intelligence within nephrology.
