Juergen Fritsch

The unfulfilled promises of health information technology

February 27, 2013 6:00 AM EST

A 2005 RAND report predicted that widespread use of electronic health records technology would save the US healthcare system at least $81 billion per year. At the time, the vendor-funded report helped drive substantial growth in the electronic health records industry and probably contributed to the federal government making billions of dollars of incentive payments available to physicians and hospitals to adopt and meaningfully use electronic health record (EHR) systems via the Health Information Technology for Economic and Clinical Health (HITECH) Act.

Realizing that the cost savings and improvements in healthcare delivery are nowhere near what was optimistically predicted in 2005, RAND recently commissioned a new study to take a fresh new look at the state of health information technology.  The new study paints a very different picture and received broad coverage by mainstream news outlets, including “In Second Look, Few Savings From Digital Health Records” by the New York Times in January.

To put it bluntly, the authors of the new report essentially admit that the original RAND study was dead wrong. Healthcare spending has risen by $800 billion since the first report was published, and while much of that is due to an aging population and the increase in overall medical services, there is scant evidence of cost savings due to electronic health records. In addition, there is increasing concern that electronic records have actually made it easier for providers to over-bill for certain services.

Some of the key points made in the new RAND study:

  • EHR technology has been adopted too hastily, before it was ready for prime time. Many of the EHR systems deployed in hospitals today were originally developed many years ago and still have major usability issues that have been shown to adversely affect physician productivity.
  • Lack of interoperability. Existing EHR systems make it hard or sometimes even virtually impossible to share medical information across systems, despite the availability of mature standards governing the exchange of medical information. Proprietary data models and silos still dominate the field.
  • Lack of adoption. Despite the payout of $6.5 billion of Meaningful Use incentives, adoption of EHR systems is still lower than 40%. On top of that, most of the facilities that have adopted EHR systems are larger, urban and academic hospitals - leaving a significant gap around the smaller, rural and non-teaching hospitals.
  • Healthcare professionals are not embracing the technology.  The government incentives are encouraging providers to 'just buy what's out there anyway' instead of demanding more usable systems. They tend to get frustrated with healthcare IT in general rather than pushing towards a next generation of electronic record systems.
  • Patients are not embracing the technology either. Despite the availability of patient portals, few of them are signing up to access their records, and most of those never actually look at their data.

The study puts a lot of the blame on HITECH - which only was signed into law in 2009 - and on the usability shortcomings of the prevailing group of EHR systems dominating today's market. Even the former National Coordinator for Health Information, David Brailer, is quoted as saying HITECH was a "colossal strategic error". But that is a little unfair, as John Halamka, CIO at BIDMC and Harvard Medical School , argued in a recent blog post, and it misses an important point, as Farzad Mostashari, the current National Coordinator explained recently. While technology plays an important role, the key to reducing health care costs is to put reimbursement changes in place that reward better care coordination and that create incentives to improve patient outcomes, rather than simply reimbursing for services provided.

That in turn will drive innovation towards a new generation of electronic health record systems - one that in combination with the policy changes will be more likely to fulfill at least some of the promises outlined in the original RAND report. What those systems will need to look like and be capable of doing is going to be the topic of my next post on this topic. Stay tuned!