Friday, February 13, 2009

Which Comes First — EHR or Quality Improvement?

Many people seem to believe that the only thing standing between us and a completely transformed healthcare system that has higher quality and lower cost is the lack of Electronic Health Record (EHR) systems in every physician’s office.

That’s a little like saying that the only reason the country is in a recession is that every American doesn’t have a Blackberry to improve their productivity.

There’s no question that many aspects of care coordination and quality improvement are very difficult for healthcare providers to deliver without appropriate health IT support. It’s inefficient and impractical to flip through paper patient charts to find out which patients are due for an immunization or a diabetic checkup, when a simple query of an electronic database could provide the answer quickly and easily. It’s inefficient and problematic for hospitals and primary care practices to be faxing each other admission and discharge information in order to coordinate care transitions when the information could be electronically transmitted and stored in a common electronic health record.

But merely having an EHR doesn’t guarantee that providers will, in fact, deliver better care to diabetics or more effectively coordinate hospital care and discharges. The physician practice and/or hospital must still redesign the actual processes of care to achieve those goals. The EHR can make that possible, or at least much easier, but only if the EHR is designed in a way that actually supports the improved care processes.

And therein lies the rub — it’s hard for today’s EHR systems to be designed to support improved care processes, when the improved care processes don’t exist. Indeed, it’s likely that, if anything, today’s EHRs will best match the way providers work today, rather than the way we want them to work in the future.

For example, most studies have found that the key computer support for improved management of chronic disease patients, preventive care, etc. is a patient registry. Yet most commercial EHR systems do not have, or do not come with, a registry component. Similarly, it’s not surprising that the only research showing an impact of EHRs on quality is from healthcare systems which developed EHRs in-house, since it’s more likely that an EHR will match care delivery processes if it’s developed in cooperation with the practitioners who will use it, rather than independently by an IT company.

Does that mean we should not be pushing aggressively for development and implementation of EHRs? No, but it means that we shouldn’t be pushing for EHRs in isolation – they should be developed and implemented as integral parts of quality improvement initiatives, ideally at a regional level, rather than provider by provider.

Comments

Can you describe what exactly is a patient “registry”. Having worked some years ago in TB control at a local health department we did have access to the TB registry of patients – and this was before EHRs were being installed in community and private physicians’ and OPD/hospital systems’ offices – but I am forgetting the registry’s main purposes, who created and maintained it (well, the health department, but I mean did it demand a lot of back stage staffing??). In this case, the registry was comprised of all reported cases of TB disease and infection and the health commissioner signed off and closed the case once TB staff verified patient’s having completed treatment.

Also, how does a registry differ from an EHR? Is it merely derived from a query of the EHR database?

Comment by Jean — March 27, 2010 @ 9:44 pm

A registry is a list of patients and selected information about those patients that is used for one or both of two purposes: (1) to identify groups of patients that have not received something they should have received, and (2) to analyze reasons why certain patients have better outcomes than others. So, for example, if I have a diabetic registry, I would enter the names of all of my patients with diabetes into the registry and record anytime they come in for an HbA1c test. Every month, I could query the registry to find out which patients had not had their HBA1c checked recently and call them to get them into the office, and I could also pull the list of those who hadn’t had their HBA1c checked to see if there were any common elements among them.
The information that goes in the registry is a subset of the information that would go in the EHR. The problem with many EHRs is that they don’t give you the ability to do queries like those defined above. You may only be able to find out which patients haven’t had their HBA1c checked by tediously going through each patient’s record. The EHR may remind you when the patient comes in that they should have their HBA1c checked, but it may not remind you to have them come in in the first place. Moreover, an EHR is overkill if what you really want to do is track a specific, small set of information about a specific set of patients.

Comment by Harold Miller — March 28, 2010 @ 3:08 pm

aha, the capacity of an EHR to do a query. Now I understand. Thanks so much for clarifying.
(Recent registered user of CHQPR website. Learned about your org./website while working on a paper in my MSN program last semester whose topic was, loosely, Disease Mgm’t. (Nurse Led) in CHF and its potential impact on reducing readmissions amongst Medicare benefs. Got the idea from a May 2009 article in the NYTimes on a hospital-based program in the mid-west that was reducing admissions but might be abandoned due to the cost of operating it. I think that you/CHQPR and “ACOs” were quoted by the writer. I really appreciate the clarity of your mission and now that I have looked through the tabs on the website I see that CHQPR is partnered with NRHI and their focused collabs. around ‘true’ or truer payment reform.)

Comment by Jean Sale-Shaw — March 31, 2010 @ 12:58 pm

RSS feed for comments on this post.

TrackBack URL