Electronic health records: health-policy dreams versus reality

Much hope and consternation surrounds the topic of electronic health records (EHR, or EMR, for electronic medical records). Justmeans regulars may recall coverage of the $billions in incentive for implementation in Obama's health reform, uses for genetic research, telemonitoring, and e-prescribing. Naturally, some skepticism about their health benefits remains.

Two interesting new studies muddy the waters further. It's frequently argued that EHR improve the practice of medicine within individual hospitals or practices, but troubles arise in networking disparate sites of care into a truly integrated community- (or nation-) wide network. Findings from the Morgan Institute for Health Policy at Mass General Hospital suggest that we may need to rethink even that basic premise.  Researchers analyzed 3,000 acute care member hospitals of the American Hospital Association (about 3/5 of all hospitals in the US) and compared quality measures with the degree of EHR integration. Quality measures were standard Medicare process measures related to heart attack, congestive heart failure and pneumonia, and Medicare Provider Analysis and Review File efficiency measures. For EHR integration they compared hospitals that were well integrated, meaning they had fully implemented computerized systems for 24 of 32 basic care functions in all units of the hospital, against those with less integration (10 functions in major units), and those with no EHR.

The results? EHR provided some spotty, minor improvement in measures of surgical infection and length of stay, but overall there was no indication that EHR improved major measures of hospital quality such as mortality. In a nutshell: Even fully wired hospitals did not see significant levels of quality improvement from their EHR.

A second study in the same issue of Health Affairs looks at EHR from a different angel. Researchers at the Group Health Research Institute in Seattle report on efforts to involve patients in EHR, with apparent great success. Since the ultimate goal should be the improvement of population health through the improvement of individual health, it makes sense to allow patients access to their own records. Since 2003, Group Health Cooperative patients have been able to access their medical records via a secure internet portal, and review test results, schedule appointments, request medication refills, and email their provider. Early adoption was slow, but Group Health now has nearly 60% of their patients registered with their system. How did they get so much participation? A truly novel approach in medicine: They asked patients how they would want to interact with their own medical information, and what types of functionality would be useful to them (the patient.) Every two years a random sampling of patients has received an extensive survey whose results have helped to guide development, improvement and ultimately the current level of success.

This second study also raises a possible shortcoming of the first: Perhaps a key utility of EHR is patient empowerment, something that is currently not considered important enough by most researchers and policy makers to warrant its own measures and incentives, and that receives relatively little attention among serious researchers.