Clinical decision support and meaningful use: Perspective from Zynx Health CEO Dr. Scott Weingarten (transcript)

This is the transcript of my recent podcast interview with Zynx Health CEO, Dr. Scott Weingarten.

David Williams: This is David E. Williams, co-founder of MedPharma Partners and author of the Health Business Blog.  I’m speaking today with Dr. Scott Weingarten, CEO of Zynx Health.  Scott, how are you today?

Dr. Scott Weingarten: I’m fine.  How are you David?

Williams: Good, thanks.  I know a lot of people are thinking about meaningful use. Definitions are being finalized, the qualifying period is coming up, and people are trying to figure out how to meet the different parts of it.

Give me a sense about clinical decision support and how it fits in to meaningful use from your perspective.

Weingarten: There is a lot involved with meaningful use.  A number of aspects relate to clinical decision support and a number do not.  I can focus on those aspects that are likely to be in the final definition of meaningful use that may relate to clinical decision support.  We’re in the comment period and are waiting for the final definition of meaningful use, so with anything I say there is a possibility that it could change. However I believe that clinical decision support will continue to be prominently featured in some form in the final definition of meaningful use.

The reason for my confidence is that the scientific literature overwhelmingly supports the benefits of clinical decision support when included in an electronic health record or computerized physician order entry system. Those benefits include reductions in mortality, reductions in morbidity and reductions in health care costs.

The data suggest that the benefits of installing electronic health records without clinical decision support are not that impressive.  One example that I shared at the HIMSS conference was a study that was published in the Archives of Internal Medicine by Doctor Linder and Dr. Bates. They found that when physicians installed an electronic health record without clinical decision support, their patients had only minimal improvements in care compared to patients whose physicians had not installed electronic health records.

They evaluated 17 ambulatory quality indicators, such as appropriate use of mammography. They concluded that for 14 of the indicators there was no difference between those physicians who had an electronic health record and those who did not.  Two out of the 17 quality indicators showed improved performance with an electronic health record, so that’s not overwhelmingly positive, and one of the 17 actually showed worsened care with an electronic health care record as compared to no electronic health record. The authors concluded that, as implemented, electronic health records were not associated with better quality ambulatory care. They summarized by saying in selecting an electronic health record, physician practices should carefully consider the inclusion of clinical decision support to facilitate quality of care.

There are many other studies that demonstrate that by providing clinical decision support, you improve the quality, safety and cost of care. They also show that when you do not provide clinical decision support, often organizations are disappointed with the clinical benefits and the financial benefits of installing electronic health records.

Because the literature is so compelling I would be very surprised if the final definition of meaningful use did not include the need to provide clinical decision support. The HITECH Act is not about just installing software but it’s also about providing benefits to the American people. That means better quality health care at an affordable cost.

Williams: What you are describing makes a lot of sense. The evidence is in the literature and certainly the intent of spending $50 billion is to achieve improvement for patients, not just wire up offices.  It also fits the intuitive sense that people have from using an electronic health record without decision support, so I can see how it’s going to be pretty central to meaningful use.

In terms of how it will be implemented, I assume it’s not just all or nothing.  Will people start with a few measures?  How will it get phased in over time?

Weingarten: I think we are likely to see clinical decision support rules with an evidence grade.  The data is overwhelming that they can improve care.  These clinical decision support rules are different from the way many of us have thought about clinical decision support roles in the past, when we had thought mainly about drug databases –rules that prevent drug interactions, drug/food interactions and drug allergies. Those types of rules are unlikely to satisfy the final definition of clinical decision support rules with an evidence grade.

We’re more likely to see under-use rules such as providing an ACE inhibitor for patients with chronic heart failure. I think medication safety in the elderly will be included. Some of the criteria that have been used include the NQF and NCQA criteria, which I think will prove to be important.

There are a number of quality measures that will need to be reported to demonstrate meaningful use. I think most organizations that report quality measures will wonder what will happen with the information they report.  Could the information one day be publicly reported? Will they be responsible for demonstrating excellence of care and also improvement of quality indicators?

So they’re also thinking about clinical decision support to demonstrate excellence in clinical care represented by the quality metric.  There is clinical decision support related to reducing preventable hospital readmissions.  There is clinical decision support related to reducing inappropriate imaging procedures.  There are evidence based order sets, which could very well continue to be a care goal especially with the literature so overwhelmingly positive about the impact of evidence based order sets on reducing mortality, morbidity and health care costs.

So those would be some of the examples of clinical decision support that I believe have a strong probability of making it into the final definition of meaningful use. But of course we’ll have to wait and review the final definition when it is available.

Williams: Will there be different decision support rules for meaningful use between the inpatient and ambulatory settings? Obviously there are differences in the patients.

Weingarten: There are a number of differences related to the overall definition of meaningful use, but as it relates to clinical decision support, we’re likely to see a number of similarities. However, the types of rules will be different. For example, one rule in the inpatient or hospital setting might be for an acute myocardial infarction or acute coronary syndrome. Rules might include providing aspirin or beta blockers or ACE inhibitors or ARB’s for appropriate patients. Clearly that would be appropriate and potentially beneficial in the hospital setting.

In the ambulatory setting you’re more likely to see preventive care rules such as mammography, pap smears, cholesterol levels, and disease management-type clinical decisions, such as the appropriate care of patients with asthma or diabetes. There are going to be content differences in the rules depending on whether we’re talking about a hospital or a physician’s office.

Williams: Based on what you’ve said, it’s becoming clear that clinical decision support is really needed.  So are EHR vendors simply incorporating clinical decision support directly into their products?

Weingarten: In many cases they are not. It depends on the type of clinical decision support. The electronic health record suppliers are very good at developing software, and there are a number of requirements to meet the overall definition of meaningful use that in some cases will take additional development work.  They are often proficient in working with drug database vendors to offer the drug allergy and drug/food rules that I alluded to, but very few of the electronic health record vendors create their own evidence based order sets, create their own clinical decision support rules with an evidence grade, have clinical decision support to optimize performance on the reported quality measures, have clinical decision support to help reduce inappropriate or preventable readmissions or to reduce inappropriate imaging procedures.

Decision support is more of a clinical competency, often requiring teams of doctors, nurses, pharmacists and editors. It often requires a different skill set and a different competency than many of the electronic health record companies currently have.

Williams: Tell me how Zynx Health fits in.

Weingarten: Zynx Health is a clinical decision support company, which has been around 14 years developing evidence based clinical decision support.  Many physicians, nurses, pharmacists, and editors are on staff day in and day out to evaluate the literature. They identify changes, including new publications, new regulatory measures, new pay-for-performance measures, and new guidelines. They grade the literature, they summarize it, and they synthesize it in a manner that can be consumed for clinical decision support.

We also work with virtually all of the major electronic health record vendors, both in the inpatient setting and the physician office setting. In addition, our clients have customized over 100,000 order sets and plans of care with Zynx-derived material. Many of our order sets and plans of care have undergone 10 or 20 iterations based on the experience of our clients to try to optimize their use with electronic health records or CPOE systems.

Williams: I’ve been speaking today with Dr. Scott Weingarten, CEO of Zynx Health.  We’ve been talking about clinical decision support and its application to meaningful use.  Scott, thanks for your time.

Weingarten: Thank you very much David.

April 2, 2010

One thought on “Clinical decision support and meaningful use: Perspective from Zynx Health CEO Dr. Scott Weingarten (transcript)”

Leave a Reply

Your email address will not be published. Required fields are marked *