EHRs Lower Mortality Rates, But Effects Take Time

Electronic health records (EHRs) are improving patient mortality rates, but the benefits of the software systems take time to appear, according to a new study.

A team of researchers from the University of Michigan, Harvard University and the University of California, San Francisco, sought to gain a better understanding of the real-world effects of EHR implementation. Thus far, reports on the links between EHR use and level of care have been mixed, the authors wrote. In search of a more definitive answer, they looked at five years’ worth of Medicare data, analyzing EHR adoption and 30-day mortality rates.

>> READ: How EHRs and Behavioral Interventions Can Fight Antibiotic-Resistant Superbugs

They found that when a hospital introduced its baseline EHR system, 30-day mortality rates increased slightly, up 0.11% per EHR function used at baseline. However, as time went on mortality rates began to drop, at a rate of 0.09% per year, per function. As new functions were added during the study period, mortality rates dropped even more, by 0.21% per year, per function.

Corresponding author Julia Adler-Milstein, Ph.D., an associate professor of medicine and director of the Clinical Informatics and Improvement Research Center at UCSF’s School of Medicine, told Healthcare Analytics News™ that EHRs’ increasing effects over time was in line with other research.

“We did expect that, by measuring EHR adoption in a more nuanced way, particularly in a way that allowed for a ‘maturation’ effect (i.e., additional benefits from baseline technology over time), we would see a stronger relationship — largely because studies of technology in other contexts have found such a maturation effect,” she said.

Adler-Milstein said the study data didn’t allow a clear elucidation of why exactly maturation seems to improve the effects of EHR systems.

“In terms of the benefits from new functions, it could be because more sophisticated functions tend to be adopted later and so that ‘greater sophistication’ is driving the benefits we see from new function adoption,” she said. “Or it could be that, by the time providers get to adopting new functions later in time, they are just better at knowing how to realize benefits from IT because they have greater experience (i.e., better implementation process).”

Notably, the effects of EHRs were most pronounced among small hospitals and teaching hospitals, the researchers found.

The research backs up the notion that a national effort to implement EHRs is medically worthwhile, but it also implies that healthcare organizations and public health officials should be patient when waiting to see the fruits of their labor.

Longer term, EHRs could have broader effects on hospital-level mortality rates as more providers use the data generated by EHRs to fuel population health management programs, clinical decision support software and other tools.

Adler-Milstein said it may be a few years before those kinds of effects can be accurately measured, and she noted that doing so won’t be straightforward.

“We'd need to think about what outcomes best capture those measures — for example, should we look at mortality at the community level as a function of how many health systems have population analytics capabilities?” she said.

That’s a different level of analysis, Adler-Milstein said, but it will be feasible to do so in the coming years.

The study is titled, “Electronic Health Records Associated With Lower Hospital Mortality After Systems Have Time To Mature.” It was published this month in Health Affairs.

Get the best insights in healthcare analytics directly to your inbox.

Related

How 1 Small Change Reduced Post-Op Opioids at 3 Hospitals

Rage Against the Machines

Instinct Outweighs Data When ICU Physicians Order Diagnostic Tests