McKesson : Optimizing Data Analytics for Oncology Practice Success

Community-based oncology practices must excel in the value-based care models that will dominate the future health care delivery system. Doing so will require practices to use the big data generated by robust advanced analytics systems to make the most clinically sound and cost-effective treatment decisions for their patients.

McKesson spoke with Dan Lodder, vice president and general manager for technology solutions at McKesson Specialty Health, about how oncology practices can close the gap between the big data capabilities they have now and the capabilities they'll need in a quickly approaching future.

What aspects of an oncology practice's business operations can be most influenced by advanced data analytics?

Lodder: The three biggest areas are revenue cycle management, supply chain and business process improvement. Analytics can help improve revenue cycle management effectiveness by illustrating what is being billed and collected. The practice can compare its performance with industry benchmarks to address problem areas impacting cash flow. Supply chain is supported by robust inventory management. Are the right specialty drugs ordered at the right time in the right amount and at the right price? Supply chain analytics can provide answers that lead to greater efficiency and less waste. The same is true for process improvement and resource optimization. Analytics can track and report productivity measures like number of patients seen by individual staff members and time spent with each patient. A practice can understand how its human capital is being utilized, whether it's being utilized optimally and where improvements can be made.

What aspects of a practice's clinical services can be most influenced by advanced data analytics?

Lodder: One big opportunity is using technology to convert unstructured data into actionable data physicians can use at the point of care to make better clinical decisions for patients. A natural language processing tool can comb through unstructured data in physician notes, pathology reports, laboratory test results, attached faxes and other documents scanned into the EHR. By structuring that information and making it accessible to the physician, the physician may see that the patient fits the profile for a specific clinical trial. Further, the newly structured data can be fed into clinical decision support modules to help the physician select the most important treatment regimen for the patients.

How can advanced data analytics help oncology practices with value-based reimbursement, which marries clinical with business outcomes?

Lodder: The most obvious way is by collecting, aggregating and reporting quality measures to payers. Whether it's Medicare with its Oncology Care Model or a commercial payer with a pay-for-performance model, practices are required to collect quality data, use that data to measure how they're performing against specific metrics and report that information to payers. Practices need technology to do that effectively and efficiently with minimal disruption to the practice and to patients.

Value-based care is about changing behaviors that lead to better outcomes at less cost. How do advanced data analytics enable that?

Lodder: Analytics provide key insights to help improve practice productivity and profitability. Incorporating information from analytics to improve practice workflow is where we see practices benefit the most. That's the real power of analytics-having the right information at the point of care and using the information to make the best decisions possible. Data and analytics should be more than just tools that create reports physicians read at the end of the week or the month or the quarter. Diagnostic test results, genetic test results, specific clinical factors captured from unstructured data, treatment options and information about clinical trials all should be available to oncologists when patients are sitting right next to them. Having all that information a day or a week later can be too late.

To what extent have oncologists embraced using advanced data analytics and big data to improve their performance?

Lodder: They've come a long way in a short time. A few years ago, it was, 'Hey, I went to medical school. I know all this stuff.' Now it's, 'Hey, thank you for giving me information to help select the right treatment regimen. Now tell me what lab test I should be ordering and what clinical trial my patient could be eligible for.' Oncologists have realized two things. One, there's no way they can keep up with the flood of new medical research and treatment options, and they need technology to help them do that. Two, advanced data analytics don't force them to do anything. The data supports them by making recommendations, but oncologists always can go 'off pathway' when it's best for patients.

What specific types of technology does an oncology practice need to become a big data-driven practice?

Lodder: It all starts with an EHR system, which serves as the technical and clinical backbone for the oncology practice. Other components include a clinical decision support tool, inventory management software, a patient portal and an analytics and reporting application. Those are the five basic pieces. And if a practice doesn't want to expend the capital to own all those pieces outright and carry ongoing labor expenses, it can license any or all of them from an outside vendor and pay a monthly subscription fee.

Do you have any advice for oncology practices that want to become a big data-driven practice?

Lodder: They should always, always identify a big data champion. That could be a doctor. That could be a nurse. It should be someone who's influential in the practice, who's comfortable with technology and who's a true believer in the power of technology to improve clinical and financial results. Without this person, you're not going to get buy in from the other clinicians. It's also important for the practice to have a strong commitment to data quality. Clean and accurate data entry is critical to the quality of the analytic results.