HIMSS: IOM report on health IT, patient safety explicated
LAS VEGAS—The latest report from the Institute of Medicine (IOM) is the 2011, health IT version of the 1999 IOM medical errors report called "To Err Is Human," said David Classen, MD, associate professor of medicine at the University of Utah in Salt Lake City and CMIO of Pascal Metrics in Washington, D.C. Classen discussed the report, "Health IT and Patient Safety: Building Safer Systems for Better Care," during a presentation at the 2012 Healthcare Information and Management Systems Society (HIMSS) conference.
When the report was issued, many claimed the error rate was inflated, Classen said. However, he cited a study that found that 15,000 Medicare beneficiaries per month experienced adverse events contributing to death. “That’s just in the Medicare population,” he pointed out. “Clearly, the landscape has changed.”
Patient safety remains a huge problem in the U.S., he said. As a member of the IOM’s Committee on Patient Safety and Health IT, he said the group “thought there is an enormous opportunity going forward to find ways to improve the safety of patients with health IT.” For one thing, voluntary reporting is not a good way to measure patient safety. “We must view patient safety as part of a sociotechnical system rather than in individual silos. The technical aspects are merely part and parcel of a complex system and when it fails, it does not fail in isolation.”
Key findings of the IOM report included the idea that health IT could improve patient safety in some areas, particularly medication safety. “Safer implementation and use begins by viewing health IT as part of a larger system,” Classen said. The most important reason for the difficulty in estimating patient safety problems is poor measurement systems. Plus, legal barriers and vendor contract provisions prohibit the sharing of patient safety information about vendor products.
Patient engagement is very important in this effort, Classen said. “Health IT could enable safer care by enabling patients and families to participate in their care.”
The committee’s recommendations allow for the Secretary of the Department of Health and Human Services to create a plan to implement those recommendations within one year. "The first thing we focused on was getting the relevant federal funding agencies to expand their funding for safe practices," said Classen. "Another focus was developing patient safety measures. We couldn’t find measures in those areas and we felt they needed to be created. Also, oversight of health IT should include accrediting organizations and there should be funding and research for how EHRs could improve the measurement of safety. There needs to be a great deal of focus on promoting post deployment safety testing of EHRs for patient safety risks. Post deployment testing standards should be more broadly applied. That could be a common focus of organizations, not only to test systems initially but in an ongoing fashion as well."
In other industries, usually the software is tested routinely. That is not always the norm in healthcare. Going forward, an appropriate example would be measuring an EHR after deployment for medication safety and providing each system with an overall score. That score could be broken down into various categories, such as drug-drug interactions.
Simulation has not proved to be successful, he said, because it picked up just 53 percent of the medication errors that would have killed patients. All the simulation testing in the world probably had little to do with actual operation of these systems. "Initially, we thought that certification bodies could handle all safety problems. We found most of the action was in post deployment."
Given the fact that comparative performance of vendor products is hard to get shaped, the recommendations will probably have a great impact on vendors, Classen asserted. Vendor contracts include many provisions that prevent the sharing of this type of performance testing. A big part of the IOM report was to change the patient safety landscape so adverse experiences could be shared. Comparative user experiences could be made more widely available. Vendors should create risk management processes.
Reporting of adverse events should be mandatory for vendors and voluntary for users, the committee recommended. Patient safety organizations (PSOs) offer one of the best frameworks for reporting these problems, Classen said, because it’s a reasonable forum for users to report through. Because there are so many patient safety incidents, PSOs could have a body that handles, analyzes and investigates those reports.
“Healthcare is a complex, high-risk industry,” he said. “Aviation and nuclear, two other high-risk industries, all have reporting bodies that can investigate and analyze trends. Healthcare doesn’t even though it is a bigger and potentially riskier industry.”
The committee felt that if the first recommendations in the report do not lead to progress then the second round of recommendations should have an option to move ahead with federal regulation of health IT. The goal, however, is for the first recommendations to be implemented before that step is taken. The FDA probably needs to create a new conceptual framework, Classen said, because the current medical device framework is not applicable to health IT.
The recommendations have a big impact on health IT systems because managing the complex lifecycle of health IT is a challenge for organizations.
In an editorial about the report, published in the British Medical Journal, Christopher A. Longhurst, MD, CMIO of Lucile Packard Children’s Hospital in Palo Alto, Calif., and Howard M. Landa, MD, CMIO of Alameda County Medical Center in Oakland, Calif., said that rigorous and transparent evaluation of software is crucial, but regulation may stifle innovation (BMJ 2012;344:e1096). “Policymakers everywhere should be aware that aggressive regulation of health IT systems may bring unintended consequences, although it is crucial to rigorously evaluate and openly share lessons learnt in this evolving field,” the editorial read.
When the report was issued, many claimed the error rate was inflated, Classen said. However, he cited a study that found that 15,000 Medicare beneficiaries per month experienced adverse events contributing to death. “That’s just in the Medicare population,” he pointed out. “Clearly, the landscape has changed.”
Patient safety remains a huge problem in the U.S., he said. As a member of the IOM’s Committee on Patient Safety and Health IT, he said the group “thought there is an enormous opportunity going forward to find ways to improve the safety of patients with health IT.” For one thing, voluntary reporting is not a good way to measure patient safety. “We must view patient safety as part of a sociotechnical system rather than in individual silos. The technical aspects are merely part and parcel of a complex system and when it fails, it does not fail in isolation.”
Key findings of the IOM report included the idea that health IT could improve patient safety in some areas, particularly medication safety. “Safer implementation and use begins by viewing health IT as part of a larger system,” Classen said. The most important reason for the difficulty in estimating patient safety problems is poor measurement systems. Plus, legal barriers and vendor contract provisions prohibit the sharing of patient safety information about vendor products.
Patient engagement is very important in this effort, Classen said. “Health IT could enable safer care by enabling patients and families to participate in their care.”
The committee’s recommendations allow for the Secretary of the Department of Health and Human Services to create a plan to implement those recommendations within one year. "The first thing we focused on was getting the relevant federal funding agencies to expand their funding for safe practices," said Classen. "Another focus was developing patient safety measures. We couldn’t find measures in those areas and we felt they needed to be created. Also, oversight of health IT should include accrediting organizations and there should be funding and research for how EHRs could improve the measurement of safety. There needs to be a great deal of focus on promoting post deployment safety testing of EHRs for patient safety risks. Post deployment testing standards should be more broadly applied. That could be a common focus of organizations, not only to test systems initially but in an ongoing fashion as well."
In other industries, usually the software is tested routinely. That is not always the norm in healthcare. Going forward, an appropriate example would be measuring an EHR after deployment for medication safety and providing each system with an overall score. That score could be broken down into various categories, such as drug-drug interactions.
Simulation has not proved to be successful, he said, because it picked up just 53 percent of the medication errors that would have killed patients. All the simulation testing in the world probably had little to do with actual operation of these systems. "Initially, we thought that certification bodies could handle all safety problems. We found most of the action was in post deployment."
Given the fact that comparative performance of vendor products is hard to get shaped, the recommendations will probably have a great impact on vendors, Classen asserted. Vendor contracts include many provisions that prevent the sharing of this type of performance testing. A big part of the IOM report was to change the patient safety landscape so adverse experiences could be shared. Comparative user experiences could be made more widely available. Vendors should create risk management processes.
Reporting of adverse events should be mandatory for vendors and voluntary for users, the committee recommended. Patient safety organizations (PSOs) offer one of the best frameworks for reporting these problems, Classen said, because it’s a reasonable forum for users to report through. Because there are so many patient safety incidents, PSOs could have a body that handles, analyzes and investigates those reports.
“Healthcare is a complex, high-risk industry,” he said. “Aviation and nuclear, two other high-risk industries, all have reporting bodies that can investigate and analyze trends. Healthcare doesn’t even though it is a bigger and potentially riskier industry.”
The committee felt that if the first recommendations in the report do not lead to progress then the second round of recommendations should have an option to move ahead with federal regulation of health IT. The goal, however, is for the first recommendations to be implemented before that step is taken. The FDA probably needs to create a new conceptual framework, Classen said, because the current medical device framework is not applicable to health IT.
The recommendations have a big impact on health IT systems because managing the complex lifecycle of health IT is a challenge for organizations.
In an editorial about the report, published in the British Medical Journal, Christopher A. Longhurst, MD, CMIO of Lucile Packard Children’s Hospital in Palo Alto, Calif., and Howard M. Landa, MD, CMIO of Alameda County Medical Center in Oakland, Calif., said that rigorous and transparent evaluation of software is crucial, but regulation may stifle innovation (BMJ 2012;344:e1096). “Policymakers everywhere should be aware that aggressive regulation of health IT systems may bring unintended consequences, although it is crucial to rigorously evaluate and openly share lessons learnt in this evolving field,” the editorial read.