HIStalk Interviews Scott Weingarten, MD, Chief Clinical and Innovation Officer, Premier

Scott Weingarten, MD, MPH is chief clinical and innovation officer, Premier; professor of medicine and consultant to the CEO, Cedars-Sinai; and health sciences clinical professor at the David Geffen School of Medicine at UCLA.

Tell me about yourself and your work.

I’m an internist by background. I have been associated with Cedars-Sinai for many years. I have a passion for improving the quality of patient care. I have been focused on clinical decision support and information technology as a means to an end, with that end being better patient care. I started a company out of Cedars called Stanson Health. It was acquired by Premier about two years ago, and I’m now chief clinical and innovation officer at Premier.

How is Premier, along with its companies Contigo Health and Stanson Health, addressing clinical variation and waste?

Premier has rich data and analytics assets that can identify opportunities to improve outcomes of care, mortality, morbidity, and cost of care. The question then becomes, after you’ve found an opportunity, how do you realize that opportunity and demonstrate improvement?

One of the most effective, scalable, and sustainable strategies is providing context-specific information to healthcare providers that are integrated into the workflow, offering suggestions to the doctor or other healthcare provider to inform care that, in some cases, will change care for the better to be more consistent with the evidence. That’s a way to close the loop in not only identifying opportunities for improvement, but implementing those improvements and being able to measure their impact or being able to measure the actual improvement.

After the acquisition of Stanson by Premier, Contigo was formed. Contigo, as a company that is part of Premier, works with employers who have a vested interest in improving the quality and reducing the cost of care for their employees.

Is information sharing among peer health systems, as is done through Premier and vendor-sponsored programs such as Epic Health Research Network, having a significant impact on patient care?

Yes. Data is critically important to understanding the gaps between current care and optimal care. In addition to data, it’s what you do about the opportunity you recognize. How do you bring about those improvements and be able to demonstrate to your satisfaction and everyone’s satisfaction that care has improved, and significantly improved?

So sharing of data is critically important, but perhaps equally or more important is being able to change care. When I say change care, it’s the interactions among doctors, nurses, pharmacists, and the patient that lead to better care with lower mortality, morbidity, and cost.

Sharing of data is the beginning. It’s necessary, but far from sufficient. Sharing of data alone will not bring about improvement.

Is it a positive development that COVID has created an urgent demand for data in the form of anecdotal findings, observational studies, and pre-print research results that would have been delayed for years otherwise to meet formal research standards?

With a caveat. The Institute of Medicine – now the National Academy of Medicine – shared years ago that there was a 17-year delay between the discovery of important data that could save lives  and the time when that knowledge is translated into practice. That’s bad. Seventeen years is far too long, and efforts were made to shorten that gap significantly to save lives and improve care. I applaud that effort to disseminate valid scientific evidence as quickly as possible.

Pre-print publications, as we’ve seen with COVID-19, can be very helpful.  But it’s also important that the information be rigorously reviewed for scientific validity, because invalid information that is disseminated and translated into practice can be potentially dangerous. We want to shorten the gap for scientifically valid, scientifically credible information, that gap between when the discovery is made and when all patients benefit from that information. But we want to remain responsible that the information that’s disseminated proves to be scientifically valid.

Clinical decision support and evidenced-based medicine are sometimes seen by physicians as intrusive, where they trust their personal experience and practices more than the results of someone’s study. Are we seeing any new capabilities or use cases that would lead to their wider acceptance?

I believe so. The key is providing physicians, nurses, pharmacists, other healthcare providers with information that is integrated in the workflow that they find helpful, that will help them take better care of patients rather than being annoying. One of the best ways to do that is to suppress the information when it just validates what the physician or healthcare provider was going to do anyway. Only provide that information when it’s incremental to what the healthcare provider is already doing or informing them when whatever they are doing is in conflict with the evidence.

There may be many good reasons that it’s in conflict. Maybe they offered the patient a treatment where the evidence has shown that the treatment can be effective, but the patient, for whatever reason — sometimes a very good reason — refuses to take the treatment. But you always want to make sure that the care is informed by the latest scientific evidence.

Another development to improve the precision and value of the information is to examine the free text information in the electronic health record. Not only discrete data elements — such as demographic information, medications, or laboratory values — but the notes that the provider has written in the electronic health record. To be able to read, interpret, and contextualize the notes to further guide the clinical decision support that can be potentially be most effective for an individual patient. It’s really a type of precision medicine, where to the best of your ability, you get to know the patient based on what is recorded in the electronic health record and tailor the guidance for that specific patient when the evidence suggests there is a testing strategy or treatment that would be best for the patient.

Have EHRs gotten better at surfacing information that tells the key story of the patient and the clinicians who have treated them so that a quick glance at the electronic chart provides the most situationally relevant information?

Electronic health record vendors have worked very hard at solving this problem. They’ve certainly heard from healthcare providers that this is an important issue that needs to be solved, and that it contributes to burnout. But I have heard that in the United States, notes of healthcare providers are much longer than those of our colleagues in other countries in Europe and so on. If that’s true, then we may be inadvertently contributing to this issue.

The question is, how can we — in addition to the electronic health record vendor — help solve this problem? Can we have shorter notes, where the high-value information, the clinically important information, is still available to other healthcare providers, and potentially the patient if OpenNotes or other strategies are used to enable patients to retrieve the information in the notes? How do we, together with the electronic health records vendors, make the notes more concise, easy to read, and easy to interpret in a short period of time?

What is the status of large health systems, such as Cedars-Sinai, getting involved with health IT accelerators, health IT investment, and acquiring commercial businesses?

I think you’ll see some health systems, not all health systems, having a greater interest in accelerators, venture capital funds, and even creating companies or spin-outs for a variety of reasons. Out of Cedars-Sinai, my colleagues and I created two companies, the order set company Zynx and my current company Stanson. We were able to commercialize the IP, which was largely related to clinical decision support, and sell both of those companies. 

You will see this trend continue with some health systems, in particular, with academic medical centers. They are in the business of creating a new knowledge and discovery and disseminating that information to improve patient care, not only at their own organization, but across the country and potentially globally.

In addition to publishing the results in peer-reviewed journals, a way to increase the impact to a greater extent is to commercialize or productize that IP so that it can be used across the country and around the world. Many health systems will say that is consistent with their mission, including academic medical centers.

The second thing you will see is that patient care revenue is increasing very slowly. In many cases, wages are increasing faster than patient care revenue, so some health systems are looking to diversify their sources of revenue. You hear about health systems thinking about creating startups, creating accelerators, and having venture funds.

What technology and data needs have been exposed by COVID that will accelerate future development and adoption?

The American Recovery and Reinvestment Act was a very large subsidy of electronic health record purchases, installments, and implementations in physician offices and hospitals. However, with that investment, there was a very small investment in comparison in public health infrastructure. We’re finding that public health information technology infrastructure has lagged significantly behind, and we are reading almost daily of the consequences of not having state-of-the-art information technology for our public health professionals across the country during a pandemic.

COVID-19 will change healthcare and the way it is practiced for the foreseeable future in many ways. We’re going to see investments made to upgrade to 2020 standards the public health information technology infrastructure to benefit from the information in the electronic health records, so that we are ahead of the curve and ready for the next pandemic, for bio-terrorism surveillance, and for understanding the next carbon monoxide poisoning or diarrheal disease outbreak.

What personal characteristics and practices allow you to be involved in so many things simultaneously?

I have a certain intellectual curiosity that causes me to do a number of different things. Some might question how well I do any of those things. But the underlying theme is that I have a passion for improving care. My mission is to make a contribution to improving patient care. I’ve set out to fulfill that mission through a variety of things — direct patient care, academics and teaching, implementing quality improvement and value improvement strategies across a health system, and creating businesses that hopefully will enable health systems across the country and beyond to improve care. That’s the underlying theme of all of my attempts to make a difference.

Do you have any final thoughts?

As someone who has been in clinical decision support for over two decades, I’m now quite bullish on the field. I feel like the advances in technology, electronic health records, natural language processing, machine learning and AI, and speech recognition will enable transformation and significant improvements in the field of clinical decision support. I’m quite hopeful and optimistic that we’re going to see greater improvements in patient care from clinical decision support in the future than we have in the past.