Unintended Consequences of Nationwide Electronic Health Record Adoption: Challenges and Opportunities in the Post-Meaningful Use Era

This follow-on article moves the discussion forward by adding further dimension to the issue of unintended consequences from the perspective of the United States health system.

Unintended Consequence 1: Failed Expectations

Recent systematic reviews have found that most HIT evaluations published before MU reported predominantly positive outcomes. These outcomes served as the foundation for the MU program and have produced a hype around HIT. Such a hype led to a nationwide adoption of commercial EHRs with high expectations for improving the US health care cost and quality. However, after 4 years of nationwide EHR adoption, health care in the United States is still the most expensive and lags behind in some quality outcomes when compared with other developed countries, which indicates that the expected benefits of a digital health system have not yet materialized. As the adoption of commercial EHRs increased, new, unanticipated modalities of problems emerged. The first systematic review of HIT impact published after MU continued to find mostly positive results; however, it also reported that 19% of the studies found no significant HIT impact, and the lack of negative outcomes is likely explained by publication bias.

The same systematic reviews that have reported positive findings have also reported several mixed results, which leaves unanswered questions as to the impact of HIT on quality, productivity, and safety. Furthermore, studies from other industries demonstrate that IT adoption rarely produces positive results if not accompanied by complementary factors or investments. Several internal and external factors have been identified as potentially affecting care outcomes during HIT interventions, which suggests that previous studies may have been subjected to similar context-dependent factors, as they are common to HIT interventions. Pre-MU studies are being criticized for relying on weak research designs such as short-term pretest-posttests and for the use of a small set of nonconsensus measurements. The latter is an important barrier to the reproducibility of studies and to the comparison of outcomes across studies, which prevents more comprehensive assessments of HIT impact and produces questions regarding the strength of the evidence supporting HIT effectiveness. The lack of consistent evidence resulting from the use of poorly designed studies indicates that what others have called positive outcomes are in fact  putative outcomes. It has been estimated that without improved research methods, around 100 hypotheses per year will continue to be tested without providing any valuable knowledge.

With insufficient evidence to support the hype around HIT and generalizable effects of HIT across care outcomes, settings, and EHR systems, an important question remains unanswered: was the over 20-billion-dollar investment in HIT from the America Recovery and Reinvestment Act (ARRA) worth it?