Those who follow health IT for a long time can easily oscillate between overenthusiasm and despair. Electronic records will bring us into the 21st century! No, electronic records just introduce complexity and frustration! Big data will find new cures! No, our data’s no good!
Indeed, a vast gulf looms between the demands that health reformers make on information technology and the actual status of that technology. But if we direct a steady vision at what’s available to us and what it provides, we can plan a path to the future.
This is the goal of a report I recently wrote for O’Reilly Media: The Information Technology Fix for Health: Barriers and Pathways to the Use of Information Technology for Better Health Care. As part of a comprehensive overview, it dissects the issues on some topics that often appear on this blog:
Patient empowerment. After looking at the various contortions hospitals go through to provide portals and pump up patients’ interest in following treatment regimes, I conclude that the best way to get patients involved in their care is to leave their data in their own hands.
But wresting data out of doctors’ grip will be heavy exercise. Well aware that previous attempts at giving patients control over data (Google Health and Microsoft HealthVault) have shriveled up, and that new efforts by Box and Apple seem to be taking the same path, I suggest a way forward by encouraging people to collect health data that will hopefully become indispensable to doctors.
What’s wrong with current EHRs? We know that doctors grab any opportunity handed them to complain about their EHRs. Even more distressing, the research bears out their pique; my report cites examples from the medical literature finding only scattered benefits from EHRs. Sometimes their opacity and awkward interfaces contribute to horrific medical errors.
One might think that nobody is actually getting what they want from their EHR, but in fact plenty of providers are quietly enjoying their records–success has a lot to do with their preparation and whether they take the extra effort to make effective use of data gathered by the EHRs.
New interfaces such as tablets, convenient storage in the cloud, and agile programming may be producing a new crop of EHRs that will meet the needs of more clinicians. But open source software would lead to the most widespread advances, enabling more customization and a better response to bug reports.
The viability of ACOs. Accountable care, pretty much a synonym for the notion of pay-for-value, is on the agendas of nearly all payers, from CMS on down. It certainly makes sense to combine data and keep close tabs on people as they move from one institution to another. But it’s really a job to be done on a national level, or at least a regional one. Can a loose collection of hospitals and related institutions muster the data and the resources to analyze patient data, created viable health information exchanges, and perform data analysis? I don’t think the current crop of ACOs will meet their goals, but they’ll provide valuable insights while they try.
Can standards such as ICD-10 improve the data we collect? What about the promise of new standards, such as FHIR? I’m a big believer in standards, but I’ve seen enough of them fail to know they must be simple, lithe, and unambiguous.
That doesn’t characterize ICD-10 to be sure. Perhaps it does pretty well in the unambiguous department. But like most classifications, it’s a weak representation of the real world: a crude hierarchy trying to reflect many vectors of interlocking effects–for instance, the various complications associated with diabetes. And although ICD-10 may lead to more precise records, the cost of conversion is so burdensome that the American Medical Association has asked the government to just let doctors spend their money on more pressing needs. The conversion has also been ruthlessly criticized on the EMR & EHR site.
FHIR is a radical change of direction for the HL7 standards body. For the first time, a standard is being built from the ground up to be web-friendly as well as sleek. It currently looks like a replacement for C-CDA, so I hope it is extended to hold patient-generated data. What we don’t need is another hundred vendors going off to create divergent formats.
For real innovation, we should look to the open SMART Platform. Its cleverness is that it functions as a one-way valve channeling data from silo’d EHRs at health providers to patient-controlled sites.
We need to know what current systems are capable of contributing to innovative health solutions, and when to enhance what we have versus seeking a totally disruptive solution. I look forward to more discussion of these trends. Comment on this article, write your own articles on the topics in the report, and if you like, comment to me privately by writing to the infofix alias @ the oreilly.com domain.