OR WAIT null SECS
Your guide to cutting down the number of clicks in your EHR.
The administrative burden that electronic health records (EHRs) place on providers factors into physician burnout. But these systems can be optimized to reduce the amount of work for physicians. Improving EHR systems can help reduce burnout and increase patient safety and provider satisfaction.
Michael Hogarth, M.D., is chief clinical research information officer of the University of California, San Diego Health. His role is all about EHR optimization and helping oversee data extraction for research and analysis.
Hogarth received his medical degree from the University of Texas Southwestern Medical School and completed his residency in internal medicine at UC Davis. He was a professor in the in the internal medicine and pathology and laboratory medicine departments at UC Davis for 20 years. He also serves as the chief medical informatics officer of Athena Breast Health Network.
I spoke with Hogarth about physician burnout, how he’s optimizing the EHR, and the development of a data warehouse at UC San Diego Health.
Editor’s note: This interview has been lightly edited for length, clarity and style.
Michael Hogarth: There are two parts of my job. One is to provide IT infrastructure for clinical trials. The other part is to oversee data extraction from the EHR for the purpose of research and analysis. This is also done for quality improvement.
Value-based reimbursement seems to be coming in with accountable care organizations, Medicaid doing the Prime program. Being able to query populations accurately for particular metrics is important, and that’s a significant crossover. We need to make sure we are providing accurate information and that the quality of data is good. That’s often a challenge.
A colleague of mine refers to data in the EHR as “dirta.” It’s data, but they are not clean. So you have to make inferences and map them to a common coding system so that everyone is speaking the same language.
Michael Hogarth: Optimizing the EHR is reducing the number of clicks people have to do. That’s a very frequent complaint for nurses and providers. It’s important to facilitate workflows by limiting the number of clicks to boost efficiency.
At UC San Diego Health, we have a team that observes the clinic and works with them to optimize their workflow and highlight things that can be done in the EHR to make their jobs easier. That’s fairly common nowadays.
Most healthcare organizations that have EHRs should really think about optimizing their systems.
Just to give you an idea, there was a paper a few years ago that mentioned that emergency medicine physicians are making about 1,000 clicks in the EHR per shift. If you think about 1,000 clicks per shift, that’s a lot.
Michael Hogarth: I think it would mitigate one factor in physician burnout. It’s a complex and multifaceted issue that ranges from compensation to work-life balance to EHRs being efficient.
Most EHRs are not smart and predictive in understanding where you are in the workflow and where you’re heading. Then you end up doing the same thing over and over again. The EHR could say, “You should discharge the patient. Do you want to do this?” Instead, you find yourself clicking quite a bit. Nobody has really looked at trying to make a system that predicts the next move.
Navigation of EHRs can be a challenge for complicated cases where you have many notes, and a lot of useful information to take care of the patient is in those clinical notes. Some EHRs have search engines or keyboard search engines, but they are not very good.
We did a study about four years ago in breast cancer to look at where data were being sourced. We itemized about 80 elements important for clinical care, research and the registry and then looked at pain points. The main pain point was information finding. We tend to document for billing purposes extensively, and we copy things into the note that are already somewhere else in the EHR, rather than just referencing them.
Optimizing information and finding and reducing click burden would be huge productivity boosts for providers who use EHRs.
Michael Hogarth: It’s hard. You look at the workflow. There are things the EHR can be configured to bypass. We also see people practicing where they’re still doing old maneuvers and not relying on EHR systems. It’s observation and advice. People who have done that have seen a significant improvement in productivity.
At UC San Diego Health, we launched the Home for Dinner Program intended to target clinics where providers are spending a lot of time on the EHR and are noted to be documenting late into the evening. The team goes to figure out how to make things better. The same program exists at the University of Colorado. They’ve also been successful in reducing click-rate and spending less time on the EHR.
There are things that have been put onto the providers in the EHR system process that weren’t there on paper. Not only did we go to the system, but we added things like medication reconciliation. And billing-type documentation has become more burdensome.
It’s interesting because the largest healthcare payer in the U.S. is Centers for Medicare and Medicaid. Yet its burdensome documentation requirements are what makes us very inefficient. Because we are inefficient, we see fewer patients and it costs more money. It’s interesting how that entity hasn’t come to make itself more efficient with documentation and pursuing that as a strategy to reduce costs.
Some places implemented scribes to do the clicks to let physicians focus on the patient.
Michael Hogarth: I think it’s good, but I would love to know whether the documentation requirements that it imposed over the years have improved the quality of care. I haven’t seen data that shows that. What’s the point? Is it reduced costs? Is it improved quality? That would be the proof, but where is it? I haven’t seen it and am not aware of the requirements resulting in decreased costs or improved quality.
There are other maneuvers that can be done to get us to both of those things.
Michael Hogarth: I think you have someone with a role in the health system information IT organization. You don’t want to have a separate ecosystem, which often happens. Health systems often view research as another mission. There are specialized computing systems that happen in that venue. Clinical trial management system is one, biospecimen management system — there’s a variety of specialized systems. But running systems, data centers, databases and securing sensitive data is all the same in both systems. So you should really avoid having two organizations do that. You should have one organization doing the core fundamental foundation and then a specialized team.
Separate organizations are counterproductive because they can duplicate costs. Sometimes you don’t have a good depth of skill on the research IT side because all of them are focused on the clinical mission. Integration of the two is key, and that’s happening here at our institute.
We can often optimize the EHR for research or quality improvement purposes. A good example would be data capture. Right now, we export data manually. There are ways emerging that would allow you to identify the data you use in the EHR and have them pulled by an electronic data capture system called REDCap.
Michael Hogarth: It started with the Athena Breast Health Network. Laura Esserman, M.D., wanted to frame breast health and understand the genesis of breast cancer. It involved the five University of California medical centers. I was the informatics person behind Athena Breast Health Network. We needed data, and I realized my colleagues were standing up research data warehouses, so we decided to do this in a more synchronized fashion. At the end of the day, we can all exchange data, if we wanted to do that, for research purposes.
We implanted a distributed data network where we each implemented the i2b2 data warehouse, which is an open-source system from Harvard. There was a query mechanism for querying all the sites at once, so we implemented that for people to do patient counts based on particular filtering. There wasn’t patient-level data; it was all aggregate data. We wanted to demonstrate that we could connect the systems. It took a while to get everyone aligned.
It really set the stage for what we have now in the UC data warehouse, which does not implement i2b2. It implements an open standard called Observational Medical Outcomes Partnership (OMOP). I think it’s become one of the largest active data warehouses in the country. We transitioned to OMOP, which has a lot more data than UC ReX did.
Michael Hogarth: We have a mobile app that helps with patient engagement. We also partnered with Apple. If you have Apple HealthKit, you can put data on it. We are constantly optimizing best practice alerts and making things safer and understanding what we can do better from that perspective. We are constantly doing that with population health. We have population health analytics with 70 registries and dashboards to understand what the practice is doing from a safety and economic standpoint. We are leveraging the EHR — and it’s almost like EHR 2.0.
Get the best insights in digital health directly to your inbox.
Hear from More Executive Voices