EHR Use Tied to Lower Adverse Event Rates

Tara Haelle

February 19, 2016

Patients admitted for pneumonia, cardiovascular complaints, or surgery conditions were less likely to experience an adverse event in the hospital if the hospital health records were fully electronic, according to a retrospective analysis published online February 6 in the Journal of Patient Safety. The reduced odds of adverse events varied, however, according to type of event and medical condition.

"Electronic health records have been used for electronic surveillance to automatically identify patients at risk and to create checklists and dashboards to monitor unit-wide central line-associated bloodstream infections," explain Michael F. Furukawa, PhD, from the Agency for Healthcare Research and Quality, and colleagues.

"Computerized physician order entry and [clinical decision support] have been shown to reduce catheter-associated urinary tract infections through electronic reminders and alerts that facilitate improved catheter management and to reduce the risks of falls and pressure ulcers through electronic integration of protocols and risk assessments," the authors write. "Beyond the direct effects on safety, [electronic health records] can also indirectly reduce overall adverse event rates by enabling quality improvement initiatives."

Dr Furukawa's team reviewed 2012 and 2013 patient discharge data from the Medicare Patient Safety Monitoring System to look for any relationship between adoption of electronic health records and adverse events among those patients.

The 45,235 patients in the sample, all adults aged 18 years and older, were admitted for pneumonia, any condition that required surgery, and acute cardiovascular disease, including acute myocardial infarction and congestive heart failure. The authors compared the 5876 (13.0%) patients with fully electronic records with those who had partially electronic or nonelectronic records.

The researchers looked specifically at the occurrences of 21 in-hospital adverse events that depended on patients' condition and treatment. They calculated the rates as the number of adverse events that occurred over a denominator of the total possible adverse events for which the patient was at risk. All patients were at risk for in-hospital falls and hospital-acquired pressure ulcers.

The adverse events fell into four broad categories: hospital-acquired infections, adverse drug events, general events, and postprocedural events. The investigators calculated that there were 347,281 patient exposures to potential adverse events across the sample, with each patient at risk for an average of 7.8 adverse events.

Overall 7820 adverse events occurred, which translated to an occurrence rate of 2.25%. Adverse drug events accounted for the highest proportion of adverse events, ranging from 5.9% to 6.8% of exposures. Hospital-acquired infections, meanwhile, occurred at a rate of just 0.7% to 0.8%, despite having the greatest number of exposures.

Patients with fully electronic records had 17% to 30% lower odds of any adverse event compared with the other patients, after adjustment for both patient and hospital characteristics. Patient confounders included age, sex, race, smoking status, and common clinical comorbidities, including congestive heart failure history, obesity, coronary artery disease, renal disease, cerebrovascular disease, chronic obstructive pulmonary disease, cancer, and diabetes. Hospital characteristics included size, ownership type, teaching status, geography (urban or rural), case mix (based on Medicare and Medicaid shares of discharges), and nurse-to-patient ratio.

Patients with pneumonia benefitted the most from electronic records, with 35% reduced odds of adverse drug events (P < .001), 34% reduced odds of hospital-acquired infections (P = .030), and 25% reduced odds of general events (P = .003).

Patients admitted for conditions requiring surgery had 36% lower odds of hospital-acquired infections (P = .013) if they had fully electronic records.

Similarly, acute cardiovascular disease patients with fully electronic records had 31% lower odds of postprocedural events (P = .004) and 21% lower odds of occurrence of general events (P = .034).

No reduction in odds for adverse drug events occurred for patients hospitalized for cardiovascular disease or surgery conditions.

"Consistent with expectation, [electronic health record] adoption had limited association with the occurrence of postprocedural events," the authors write. "The finding of lower postprocedural events for patients exposed to [electronic health records] and hospitalized for acute cardiovascular disease might in part reflect [clinical decision support] alerts to order prophylaxis to prevent venous thromboembolism."

The research was not supported by external funding. The authors have disclosed no relevant financial relationships.

J Patient Saf. Published online February 6, 2016. Abstract


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.
Post as: