HIStalk Interviews Ritu Agarwal, Director, Center for Health Information and Decision Systems

Ritu Agarwal, PhD is Professor and Robert H. Smith Dean’s Chair of Information Systems at the Smith School of Business of the University of Maryland, College Park, MD. She is also the founder and director of the Center for Health Information and Decision Systems (CHIDS), a research center within the business school.

Give me some background about yourself and about your organization.

I’m a professor of information systems at the Smith School of Business at the University of Maryland. I’m also the director of the Center for Health Information and Decision Systems. I established the Center in 2005 before health IT became trendy.

The mission of the Center is to investigate how technology can be used to transform healthcare fundamentally. We’ve been involved in doing research for the last six years or so on this topic. We work with a variety of partner organizations from the business sector, the government sector, and not-for-profits.

I saw on your Web site that you are working on a number of projects. What are the top two or three?

In our portfolio, we have a couple of projects around health information exchanges, which I think are extremely exciting and important. One is a project with the District of Columbia Regional Health Information Organization. We spent about a year doing an assessment study, which involved a wide range of data collection from different stakeholders. Based on that, we developed a generalizable model that can be used to assess any health information exchange.

We’re currently still engaged with the DC RHIO in helping them evaluate the usability of the technology and the value that it’s generating for all stakeholders. This is going to be an ongoing effort as more and more people join the collaborative and more hospitals and clinics come online.

The second project, which is just getting kicked off, is part of the Office of the National Coordinator’s Health Information Exchange Challenge Grant. We’re working with the Chesapeake Regional Information Systems for our Patients. That is the Maryland health information exchange. They’re in the process of rolling out an intervention which involves direct integration between acute care in long term facilities for exchange of continuity of care documents as well as advance directives. 

We are responsible for the research around the assessment of this particular intervention. We’re comparing how quickly the information can be transmitted, whether it’s reducing hospital readmissions and a host of other outcomes across the set of intervention hospitals compared with a pilot group of hospitals. I think both of these projects are going to provide some important insight into how health information exchanges can be used to deliver more value into the healthcare system.

We’re also working with the eHealth Initiative. They collect data from health information exchanges every year as part of their annual survey. We’re doing some econometric analysis to understand what the predictors of health information exchange sustainability and operational maturity are. We’re looking at financial break-even and looking at what specific aspects of the business model and the revenue structure help predict whether the exchange will be sustainable or not.

Are people showing interest in your findings? How would you intend those findings to be used?

Yes, absolutely. I think they have a lot of implication for how health information exchanges are going to structure their business models in the future. The grant money is going to run out — it’s not infinite. 

Clearly there are examples of health information exchanges that have managed to attain some level of sustainability. The Delaware case is one example. Vermont is another example. They have specific revenue structures and business models that provide some kind of value to all the participants that motivates them to join the exchange. Certainly this is going to be an important aspect in the future.

In terms interest in our findings, I would certainly think so. We made a presentation at HIMSS last month on the DC RHIO evaluation. There was a lot of interests in that. Several people have requested a copy of the report. We’re just in the process of working with the HIE to put out a policy brief on some of our findings around their data, and I think there will be significant interest in that as well.

Another project listed was AHRQ-funded research on EHR usability. What thoughts do you have about that in terms of EHR adoption?

I have interacted with a lot of doctors in the last five or six years around this whole notion of EHR usability. I’ve also seen so many of the products that are out there in use and my own research in the past in usability and other domains. I’ve done research around Web site usability for the retailing industry, for example.

Suffice it to say that usability is probably one of the most important factors that drives any individual’s adoption, especially when you think about how these products are going to be used. Many of them might be used while the doctor is actually interacting with the patient. The last thing you want is the workflow to be awkward or in any sense disruptive in the doctor-patient relationship or engagement.

The answer to your question, “Is usability important?” is a resounding yes. One of the things that we are doing in this project is developing a very simple usability toolkit that physicians can use in an ambulatory setting in the physician offices to figure out whether their EHR is working for them or not. If it isn’t, what specifically they might be able to do in terms of either changing their workflow or making some modifications to the EHR. 

I think it’s going to have a big impact. ONC has significant interest in looking at the usability of EHR products. That’s going to become an important criteria in their certification processes as well. It’s not just the functionality, because all these products are loaded with lots and lots of functions. They probably have a least 80 or 90% overlap in functions, but there’s a lot of variation in usability.

Usability as a condition of “do you want to buy this product” is one thing, but what about usability in the context of “are patients safe based on sound usability principles?”

Both adoption as well as safety are the two important outcomes. I’d say safety trumps adoption. Clearly if the physician is not able to interpret the information that’s coming out of the EHR, or if the EHR is awkward to use in an emergency situation when it’s absolutely imperative to get to the correct information, then the patient safety compromise is completely unacceptable.

But even if the EHR was being used more in a non-real time fashion, just simply to record data after the interaction with the patient is over — even then, usability becomes a concern. It has an implication for how much time the clinician, whether it’s a physician or nurse assistant, spends in updating and accessing information. It’s supposed to make them more effective as well as more efficient.

Are you studying anything related to using government incentives to encourage providers to adopt technology they didn’t want and how that might impact their chance of success?

One of the studies that we’ve done has been around this whole notion of physician identity and how that’s changing as a result of technological innovations and the ARRA mandates and pressure from the government and other important agencies. It is eventually in the interest of the entire system and all the stakeholders if physicians willingly adopt this technology, rather than believe that it’s being something that’s being imposed on them. There has been lots of prior research documenting misuse, ineffective use, sabotaging of this technology when individuals perceive that it’s not their volition or choice to use it. 

The important thing is in the messaging and marketing around these technologies. There has to be a very clear articulation of value to everybody who’s required to use it.

One of the things that we have not been able to do very compellingly yet, which we’re trying to do, is to be able to walk into a physician’s office and say, “Look, here are some reasons why this technology is going to make your life better. It’s going to improve your effectiveness. It’s going to help you take care of your patients better. It’s going to help you improve patient safety. It’s going to help you improve effectiveness,” 

In other words, there’s not enough evidence yet around the value of electronic health records and such technologies. But one recent study that came out of the Office of National Coordinator which was published last month in Health Affairs seems to suggest that now the evidence base has started growing. I think now we have a better story to tell.

That study had some problems, being a meta analysis written by folks who clearly had a bias. And hospitals, where employed physicians were already mandated to use electronic systems, haven’t seen the kind of numbers they hoped on raising quality or lowering cost. Is an interest of  yours proving the value of these systems?

We have a lot of interest in proving the value of these systems. As with every other research organization, we are limited by data availability. We have started on some specific granular studies around individual systems in hospital as well as physician practice settings.

For example, we did a study at Children’s National Medical Center, early findings from which were also presented at HIMSS, with a group of pediatric physicians looking at the readability of clinical documentation system and how much that improves readability over just regular handwritten notes. So you know, those are more micro-level studies. We have several of those ongoing. 

But we also have some studies at the hospital level, where we’re using some of the HIMSS data and combining that with quality measures to try and establish if there is a relationship between different types of information technology investments that the hospital makes and different measures of quality. But it’s going to take a few years before, as a community, there’s enough understanding and data for these affects to start appearing.

I should also point out very quickly that we had similar issues around information technology in general at the turn of the century. There was a very famous economist who said, “You see computers everywhere except in the productivity numbers.” It took a while before there was enough macro-level data to be able to establish that causal link. I think we’re getting there, but I’m not going to say in the next one year we’ll have the definitive answer on health IT value.

That makes it tough to sell a small physician practice since it involves a leap of faith.

Many of their concerns can be allayed with the appropriate kind of assistance and help. There is a learning curve, but they’re not horrendously difficult. Sometimes you get overwhelmed with the complexity of an EHR system, but I think there’s ways to help doctors assimilate it into their workflow.

Part of it is that there has to be a clear understanding of how both the technology and the workflow need to evolve to fit each other. What ends up happening is that the doc sees the technology and then says, “OK, here’s how I do my business. Here’s how I do my all my clinical work and administrative work.” That’s almost like a square peg in a round hole.

You’ve done some work with personal health records. What’s your feeling on where those are and where they’re going?

My own personal opinion is that this next generation of healthcare consumers that’s going to enter the system in the next decade or so … it’s almost a cliché now, it’s a very highly technologically savvy group.

I think personal health records have a big role to play in how people take the control of their own health and wellness and well-being. I personally believe that personal health records or some equivalent is going to be a significant application in the next five to seven years. The question remains is, how should these applications be designed so that they have the same level of exponential growth in adoption as in something like a Facebook?

Has anybody studied what it would take to motivate consumers to use personal health records? They don’t seem very interested.

One of the ARHQ-funded projects that we’re currently working on is learning best practices and principles from the design of other consumer products that can be applied to health IT. We’ve identified 24 highly successful products in other domains. We’ve been examining their development methods and processes that have been used in their construction, their fee, what are some principles and best practices that could applied to consumer health IT as well.

I’m also currently involved in a project with the Air Force medical system and personal health record to users at one of the major Air Force bases, 40,000 users. What they’re discovering in the early stages of the research is that the consumers love it. They love it, they are delighted with the idea that have access to their personal information, that they can update their medications and allergies and everything else. That product is slowly being extended with different kinds of devices to help them monitor their blood pressure if they are hypertensive and various other services depending on their disease condition. I see a growth in personal health record type of technology — consumer health IT in general.

If you could work on any healthcare IT project that would have wide impact on both cost and outcomes, what work would you undertake?

I think I would love to study the comparative effectiveness of health IT interventions. There are resources and funding for that available, but I think a better understanding of how these health IT interventions are assisting people with managing their disease conditions as compared with traditional therapeutic regimens.

Let me just give you an example. We all know that social networks and social influence plays a major role in how people take care of themselves. The moment you use health information technology — or any information technology, for that matter — to connect up people in social networks, suddenly you have the exponential effect of a lot more influence on the focal person. It would be fascinating to study how those types of interactions, social interactions, coupled with health IT stack up in terms of critical effectiveness and cost of care, as compared with just traditional therapeutic regimens of, “Take this prescription for 20 days.”

A lot of interesting work and has been coming out of Kaiser since they have the captive audience of users. I would think that’s a pretty rich mine of data to look at if you could get at it.

Absolutely, yes, that is an amazing repository they have. We’ve had some conversations with Kaiser in this regard, but we’re not quite there yet in having access to the data.

Do you have any concluding thoughts?

I’m quite a passionate believer in the importance of health information technology interventions. I think they can help healthcare achieve many of the goals that they’re all trying to achieve of being safer and more cost effective. I also think that the system has a major problem with the incentive alignment now. Health information technology can have an impact only when it’s coupled with other complimentary changes at the system level — some alignment of incentives around payment reform, some around insurance reform. That has to take place for health IT also be influential.