Association between Electronic Prescribing among Ambulatory Care Providers and Adverse Drug Event Hospitalizations in Older Adults

Grishma Patel Bhavsar, University of South Carolina - Columbia


Purpose: This dissertation research sought to determine whether the proportion of physicians using electronic prescribing (e-prescribing) was associated with the hospitalization rate for adverse drug events (ADEs) among patients aged 65 and older in 2011. Additionally, we sought to determine whether increases in the proportion of eprescribing physicians in a county were associated with decreases in the hospitalization rate for ADE among older adults.

Methods: Two study designs were used, a cross-sectional study using 2011 data and a prepost- study using 2008 and 2011 data. Data from the 2008 and 2011 State Inpatient Databases, the Office of the National Coordinator Health IT Dashboard, and the Area Health Resource File were gathered for six states: Arizona, Florida, Maryland, Michigan, New Jersey, and Washington. ADE hospitalization rates were calculated for adults 65 years and older. The independent variable, the rate of e-prescribing, was an ecological measure for both analyses. Multivariable linear regression examined county rates of ADE hospitalization in 2011, multivariable logistic regression examined the odds that a discharge would have been ADE associated versus other causes in 2011, and negative binomial regression was used to model the ADE hospitalization rate among older adults in 2011 based on the ADE hospitalization rate in 2008, the change in e-prescribing rates, and county characteristics.

Results: Results indicated that county e-prescribing rates were not significantly associated with county ADE hospitalization rates among older adults (p=0.4705). Further, after adjusting for patient, provider, health infrastructure, and community factors, the county eprescribing rate was not a significant factor in determining the odds of an ADE hospitalization. Change in e-prescribing rates was not significantly associated with the change in ADE hospitalization rates; no other county characteristics were found to be significant factors.

Conclusion: Though the adoption of e-prescribing has continued to increase throughout the U.S., our findings indicate that population-level benefits, such as decreased ADE hospitalization among older adults, have yet to be seen. It may be too early to detect population-level changes due to low levels of implementation of health information technologies, such as e-prescribing. Researchers and policy makers must continue to monitor the population impact that the implementation of HITs is having on the health of the nation.