Adoption of computerized clinical decision support functionalities and the quality of hospital care
Experience the future of the Tulane University Digital Library.
We invite you to visit the new Digital Collections website.
Description
The Health Information Technology for Economic and Clinical Health (HITECH) Act of 2009 has triggered the wide adoption of electronic health records (EHRs) among eligible hospitals nationally. With the increased adoption of EHRs, hospitals’ efficiency improved. However, the evidence of improvements in healthcare quality after the adoption of EHRs was mixed. This work assesses whether computerized clinical decision support (CCDS) functionalities, which are a part of EHRs, affected inpatient health care outcomes among acute stroke patients to illustrate whether CCDS may improve the quality of hospital care for acute health conditions. Previous works, limited to randomized clinical trials, did not evaluate how CCDS adoption in hospitals impacted the quality of acute stroke care at the national level. In answering this research question, it is assumed that improvements in hospital care quality were warranted by the adoption and use of CCDS functionalities, which contributed to resolving the complexity of clinical decisions for care providers. The association between the adoption of CCDS and acute stroke quality measures in the 2013 and 2017 sample of the U.S. non-federal acute care hospitals was performed using rich administrative data on the key characteristics, including the adoption of CCDS adoption (American Hospital Association surveys), inpatient quality outcome measures (Hospital Compare) employing multivariable ordinary least-squares regression analyses with an instrumental variable controlling for time-invariant hospital characteristics (fixed effects). It has been shown that the adoption of CCDS had a clinically significant positive effect on the 30-day all-cause acute stroke mortality rate (-0.015, p=0.69), and a statistically significant negative effect on the 30-day acute stroke readmission rate (0.068, p=0.01). Even though the results were inconclusive, this work informs further research on the adoption and implementation of CCDS, including those that employ machine learning and artificial intelligence, by expanding the framework for assessing health IT adoption using administrative data and opening avenues for enhancing measures of care quality.