Want to see how health IT causes harm? Scot Silverstein, MD, a professor of health informatics at Drexel University in Philadelphia, has been chronicling problems and failures since 1998, and has posted dozens of examples on his Drexel homepage (cci.drexel.edu/faculty/ssilverstein) and on his "Health Care Renewal" blog (hcrenewal.blogspot.com).

Silverstein, a former hospital chief medical information officer who has seen numerous EHR failures firsthand, is clear where he stands on the state of health IT.

"This site takes a patient- and human rights-focused view on healthcare information technology, an experimental technology that in the U.S. is largely unregulated," Silverstein explains on his compendium, which he calls "Contemporary Issues in Medical Informatics: Good Health IT, Bad Health IT, and Common Examples of Healthcare IT Difficulties."

The Boston Globe quoted Silverstein in its recent in-depth look at EHR-related medical harm ("Hazards tied to medical records rush," July 20).

"The more complex the case, the more confusing the picture that electronic health record screens will present," he told the Globe. "All it takes is one mistake in a complex and frail patient to seriously harm them or kill them."

Silverstein was not available for an interview in time for this issue's deadline, but in an e-mail, he referred to a post he wrote on his "Health Care Renewal" blog in April, shortly after the Food and Drug Administration, the Office of the National Coordinator for Health IT and the Federal Communications Commission released a report recommending continued voluntary, confidential error reporting. 

This report, mandated by the 2012 Food and Drug Administration Safety Innovation Act, or FDASIA, said that "health management health IT products," including EHR, CPOE, medication management and clinical decision support systems, pose "sufficiently low risk" to avoid the need for premarket regulatory clearance.

That conclusion incensed health IT safety watchdogs like Silverstein.

"The FDA statement is particularly reckless, perhaps criminally so," he wrote, calling the decision "an academic idealist's and industry's dream" and "a patient's nightmare."

Last year, in an open letter to David Bates, MD, chairman of the Health IT Policy Committee's FDASIA workgroup, Silverstein called the committee's proposed Health IT Safety Center "a toothless organization without true regulatory authority." He said the recommendation to exempt EHRs from premarket testing was "quite frankly, negligent."

(For their part, the American Hospital Association and the American Medical Association issued statements in favor of this proposed collaboration between government and vendors, even though the AHA was largely critical of the FDASIA report as a whole.)

Silverstein listed 11 points to argue his case, including: a leaked 2010 internal FDA memo that showed the agency was aware of at least six health IT-related deaths; a much-publicized 2011 Institute of Medicine report calling for a new health IT safety watchdog agency at the federal level; a history of the FDA recalling at least four health IT products since 2011 and March 2014 letter from CMS stating that the agency that administers meaningful use does "not have any information that supports or refutes claims that a broader adoption of EHRs can save lives."

Another longtime critic of health IT design, University of Pennsylvania sociologist Ross Koppel, calls the FDA plan a gift to vendors "to make sure they would remain unregulated."

Koppel derides voluntary reporting systems, which he blames for allowing General Motors to guard evidence of faulty ignition switches in millions of vehicles until at least 13 people had died.

"GM effectively created voluntary reporting, and we see where that went," Koppel says. (Silverstein, too, has cited the delayed GM recalls in his writings.)

"This kind of error reporting is pretty much meaningless," Koppel says, leading to what he calls a "false sense of security."