HealthAffairs: Hospitals could play waiting game for ROI on health IT
Jeffrey S. McCullough, PhD, assistant professor in the division of health policy and management at the University of Minnesota School of Public Health in Minneapolis, and colleagues measured changes in the quality of care following adoption of EHRs among a national sample of 3,401 nonfederal, acute-care U.S. hospitals from 2004 to 2007.
Multivariate regression techniques measured the effect of health IT on six quality measures independently across academic and nonacademic hospitals: ACE inhibitor/ARB use, smoking cessation advice (for heart failure patients), pheumococcal vaccination, blood culture precedes antibiotic, smoking cessation advice (for pneumonia patients) and most appropriate antibiotic for pneumonia patients.
“Academic hospitals have been the leading adopters of health IT and the setting for much of the health IT value literature. Academic hospitals also differ from other hospitals in their case-mix and organization. Consequently, it is crucial that we understand whether and how the academic context influences health IT value,” wrote the authors.
Average quality was higher for hospitals with EHRs and CPOE for all sampled hospitals, the report stated. “However this difference was statistically significant only for pneumococcal vaccine administration and use of the most appropriate antibiotic for pneumonia,” McCullough and colleagues reported.
For academic hospitals, the difference in quality following IT adoption was larger than it was for hospitals on average, the study found. Significant improvements were found for pneumococcal vaccine administration (6.1 percentage points) and the most appropriate initial antibiotic use for pneumonia patients (3.7 percentage points).
“The effects of health IT for academic hospitals were about threefold larger than the effects for hospitals on average,” the report found. “Health IT is correlated with quality improvement in nonacademic hospitals. Although positive for each measure, the differences are small and not statistically significant.”
“If context dependence is caused by differences in technology and implementation, then the average U.S. hospital will generate large returns only when it fully implements comprehensive EHR systems,” the authors concluded. “Under these conditions, stringent ‘meaningful use’ criteria…will be essential to generating widespread returns from health IT. If the meaningful use is set low, we should not expect substantive returns from widespread health IT adoption.”