Patient safety will improve with new tools for better data collection

As many as a third of patients admitted to the United States’ best hospitals have experienced harm as part of their care. That astounding 2005 statistic is often cited to underscore the alarming need for improvement in health care. More recent studies have found that patient safety problems may be responsible for hundreds of thousands of deaths per year.

Now, a paper co-authored by a UB faculty member and published recently in the Journal of Biomedical Informatics is the first to explore how a new federal reporting requirement that’s part of the Affordable Care Act may play a key role in improving patient safety.

The new requirement, which takes effect on Jan. 1, 2017, states that most Medicare and Medicaid hospitals participating in Affordable Care Act exchanges must be part of a Patient Safety Organization. Those organizations, in turn, must report adverse patient events in a uniform way, using standard forms called the Common Formats for Patient Safety data collection and reporting.

Tracking patient-safety trends, as well as adverse events, is extremely difficult when data-collection methods and definitions vary from organization to organization.

Based on clinical guidelines and expert opinion from the Agency for Healthcare Research and Quality (AHRQ) and the National Quality Forum’s expert panel, the Common Formats require that organizations now use the same definitions to report patient-safety problems and events. Similar kinds of requirements have been successfully applied to the aviation and nuclear power industries, the paper reported.

While this requires additional data entry by patient-safety officers at hospitals — usually nurse managers who already may be burdened with data reporting — these steps are critical to making health care safer, according to the paper. Such steps also can provide hospitals with better legal protection in the event that a liability issue arises.

“We believe this is worth the effort as data collection of outcomes and root cause analysis — the factors that make errors more likely — are the best way for us to determine the important interventions needed to improve patient safety,” says Peter L. Elkin, first author and professor and chair of the Department of Biomedical Informatics in the Jacobs School of Medicine and Biomedical Sciences.

The standardization of such tools has another advantage, he continues, since they are easily incorporated into electronic systems, such as electronic health records. “These tools can be deployed at the point of care and can assist us in finding systematic error in the clinical practice and in resolving incongruities toward a safer patient care environment for our patients,” he says.

The expectation is that the Common Formats will help improve the health care system as a whole by creating what is termed in the paper “a patient safety life cycle of standardization” so that problems are detected, analyzed, studied and ultimately used to improve the entire system.

Elkin adds that the new reporting requirements are an excellent example of translational medicine because they take research results from disciplines like biomedical informatics and apply them to the real-world health care system.

“The Common Formats are helping to systematize the practice of medicine by providing the tools, workflow and knowledge at the point of care to assist clinicians,” he says. “This helps decrease both errors of commission and of omission.”

Since its inception, Elkin has been a member of the expert panel of the National Quality Forum, which is charged with helping the health care system become better and more affordable through the development of ways to measure and assess different health care variables. The Common Formats were created by the AHRQ and approved by the NQF’s expert panel. 

Elkin’s co-authors are Henry C. Johnson of Henry Johnson Healthcare Consulting; Michael R. Callahan of KattenMuchinRosenman; and David C. Classen of the University of Utah School of Medicine and Pascal Metrics PSO.