Motivating Supply Chain to embrace data science for clinical decision support

Jan. 25, 2017

Barely two decades into the 21st century, Supply Chain executives now find themselves regularly partnering with physicians in ways that were standard for top-tier professionals back in the 1990s, and predicted for the rest as part of the profession’s future growth.

Sequestration in the basement or merely buying stuff for clinicians represent the days of yore. Today, a growing number of Supply Chain pros serve as business consultants for doctors who want to concentrate more on patients — and rightly so — and focus less on economics directly when “functional equivalents” can offer advice and recommendations with the data to back them up.

“With the collaborative relationships with physician providers and educated vendor partnerships, the supply chain needs to leverage the depth and breadth of the health system to deliver quality care, innovation and next generation solutions while never forgetting that the patient comes first,” said Stacy Brethauer, MD, Associate Professor of Surgery, Cleveland Clinic Lerner College of Medicine and Medical Director, Supply Chain Management, Cleveland Clinic, and Medical Director, Excelerate.

Brethauer called for the creation of a “dynamic supply chain that delivers higher levels of performance and implements systems of control to manage waste, cost and expiry across a complex environment.” Among the needs for such a categorized supply chain are integration of reimbursement, supply cost, supply utilization, clinical factors and other resource utilization data by physician and MS-DRG, he added.

“Variability within a specific service line, particularly when it comes to high-cost items and implants, will drive up costs for the enterprise,” Brethauer said. “Supply chain can show the clinical leadership where the variability is and the cost impact of reducing it. By engaging physician champions to drive the process, the data becomes actionable and meaningful to the other physicians in the service line rather than a directive from non-clinicians.”

It’s what, who you know

Tim Lantz, Senior Vice President, Analytics and DataNext, Sentry Data Systems, Deerfield Beach, FL, fully supports Supply Chain pros educating physicians on the business of healthcare and working together with them toward a common goal.

“It has been my experience over my 15-year healthcare career that physicians are committed to providing the best possible care to their patients and supply chain professionals are committed to providing the best possible products at the lowest price to their clinicians,” Lantz said. “Fundamentally, these commitments are perfectly aligned.

“Supply chains are optimally positioned to help physicians understand the relationship between the supplies and drugs they choose for their patients, and the associated outcomes that the patients and the organization experiences,” Lantz continued. “As a starting point, we typically recommend tying cost, utilization and reimbursement information to some of the more fundamental clinical and operational measures such as [length of stay], hospitalization rate, readmission rate and infection rate.”

David Levine, MD, FACEP, Senior Vice President, Advanced Analytics & Informatics/ Medical Director, Center for Advanced Analytics & Informatics, Vizient Inc., acknowledged the variety of data elements Supply Chain should be collecting related to the supplies that clinicians request. They include length of stay, mortality and safety outcomes, such as post-operative infection and bleeding, according to Levine. Other important measures are patient satisfaction metrics, especially patient-reported outcomes and pain scores, such as functional status or the ability to ambulate without pain for a knee implant, he said. “It is also important to look at revisits, including readmissions, repeat ER visits and observation stays,” he added. “Utilization of resources, including imaging, is also relevant if a particular supply is related to a change in the need for imaging, similar for pharmaceutical utilization.”

Carola Endicott, Vice President, Services and Operations for Cardinal Health Inventory Management Solutions, Cardinal Health, Dublin, OH, cautioned supply chain analysts against burdening physicians with how much stock to keep on hand, such as optimal PAR levels because physicians aren’t concerned with those details. Instead, nurses and technologists tend to huddle on that topic for physicians who just want access to their products.

“So, what do physicians want to know?” Endicott asked. “One item to consider when discussing clinical data and supply chain improvements with physicians is comparative cost data. While this can be sensitive, it is some of the most compelling data available for driving change. Providing physicians with product pricing information before, during and after a procedure can reveal the impact one product choice can have to significantly impact total cost of care delivery.”

But that’s not enough, Endicott insisted. Supply Chain should match that data against comparative outcomes by requiring supply chain analysts to work with the [electronic medical record] analysts to dig deeper into how one product compares to another when used in the same case type to determine if there are any differences in outcomes, she noted. “While supply chain may not traditionally have the expertise to answer questions like this, these studies will both enhance physician understanding of their product choices and cement the partnership between physicians and supply chain,” she added.

Supply Chain should be capturing clinical data elements or attributes like HCPCS codes, GTINs for UDIs, serial numbers, lot numbers and cost, argued Michael DeLuca, Executive Vice President, Technology and Client Services, Prodigo Solutions, Cranberry Township, PA. “But those data elements are only valuable so long as they are effectively integrated into the downstream clinical information systems and data registries so meaningful action can be taken,” he said. “To do that, you have to bridge the legacy gap between clinical IT and supply chain. For years, supply chain has been a low priority for clinical IT. I’m seeing evidence that clinical is now coming to supply chain, asking for help getting access to the right clinical data elements that will help drive value-based care.”

A package deal

Packaging and presentation are key, according to Beth Meyers, RN, PhD (c), CPHIMS, Chief Nurse Executive, Analytics Strategy Director, Infor Healthcare.

“Physicians are experts, and as experts, they tend to rely on their personal experience when making decisions,” Meyers said. “So the most important consideration when preparing any type of data for a physician audience is to be sure to present it in a way that matches their experience. This experiential view and the context of the particular decision in turn, drive requirements for particular data elements.”

For example, a surgeon may not think about his or her total supply expense by vendor, Meyers explained. He or she thinks about the supplies used for each patient.

“Patient groupings must be logical for the particular supply use, which doesn’t always correlate well to easily accessible clinical data elements such as diagnosis related group,” Meyers noted. “Instead, Supply Chain analysts should gather the most granular detail available, in this case, the procedure name, which better reflects the surgical technique used. Having the correct detail in the clinical data elements is crucial to accurately reflect the way a physician naturally thinks about patient groups and supply use.”

Supply Chain also must gather data on clinically and functionally equivalent supplies from a variety of sources, including supply vendors, group purchasing organizations and third-party analytics firms, and make sure that data are validated by the physicians receiving the analysis, according to Meyers. “Positioning the validation upfront adds the necessary credibility to the resulting analysis by helping physicians think about their experience using similar products,” she added.

One crucial element involves clinical outcome data, which have experienced ups and downs, and require physician expertise to evaluate, validate and support and minimize conflicts, Meyers indicated.

“Clinical quality groups have been expanding outcome data elements rapidly over the last 10 years,” she said. “There are many options available to track infection rates, length of stay, readmission and patient satisfaction. Unfortunately, these don’t have universal support from the physician community and sometimes directly conflict with each other. For instance, is it better to have a shorter length of stay with a higher readmission rate or keep patients an extra night during an initial hospitalization in order to keep them from coming back later? The jury is still out in many cases.

“So, again, it’s best to work with the particular group of physicians to define the correct outcome measures for each particular project,” she continued. “For hernia surgery, as an example, data may need to be gathered from the physician office to find the patients’ pain and mobility scores three and six months after surgery in order to see the true surgical outcome. Even longer time frames may be needed in some cases — this data will need to be gathered from published research studies instead of directly from local outcomes.”

Using clinical data also drives the tackling of physician preference items, which can account for up to 60 percent of medical/surgical supply expenses, according to David Hargraves, Group Vice President, Strategic Sourcing, Premier Inc., Pittsburgh, PA.

“Oftentimes, physicians are influenced to select a PPI due to several factors, including information from suppliers on evidence and clinical outcomes that are not often aligned with supply chain best practices,” Hargraves said. “In theory, it may be easy to think about cutting costs by simply getting frontline physicians to change their habits, but it’s easier said than done.”

That motivated Premier to launch a comparative effectiveness initiative called the Partnership for Advancement of Comparative Effectiveness Review (PACER), which unites “diverse health systems to collaborate on clinical evidence, price and usage data, as well as proven best practices when it comes to reducing PPIs, with a focus not only on reducing costs within supply chain but also improving care,” Hargraves said.

“These health systems are seeing significant success in driving behavior change among frontline providers by including physicians throughout the decision-making process on both products and practices,” he continued. “This process provides them insight into how supply chain decisions are made, how to reduce unnecessary variation, improve outcomes and often expedite savings by partnering with vendors offering the highest-quality, most cost-effective products.” Seven health systems applied the PACER principles to cochlear implants, recruiting administrators and clinicians to participate around clinical operations, performance improvement and supply chain, leading to a GPO contract and projected average annual savings of about $670,000, according to Hargraves.

Supply Chain must incorporate the financial perspective and should not neglect the payer perspective, according to Kathy Schwartz, Solutions Owner and Associate Vice President, Strategic Messaging, Craneware.

“Data governance is a cross collaborative effort,” Schwartz said. “Supply Chain data initiatives should coincide with strategic energies that support standardization from a financial perspective while supporting a provider’s ability to collect and analyze their data. The claims data is utilized by CMS for quality reporting indicators and ensuring expert evaluation of the data to determine how to deliver quality care with better outcomes for all. Claims data offers the most structured layout for analytics and should represent the most accurate data or else clinical efficacies and care treatments could be at risk. What often gets reported on claims is not what gets documented in the EHR. This is largely due to the parallelism in data and processes that exist between these data sets.”

With the recent passage of the 21st Century Cures Act, Schwartz noted that there will be a more rapid influx of FDA-cleared implants, devices and drugs coming to the market. “In order to have meaningful discussions with care providers, a source of truth from item to charge must exist so the full patient experience is captured,” she added.