Quality & Safety

The prospect of hackers gaining access to remote access to networked IT systems and connected medical devices raises concerns about serious safety risks.

This case study shows how Hardin Memorial Hospital lowered that rate from 5.6 to 2.2 percent – and improved its HCAHPS scores thanks to better communication between doctors and patients.

This case study looks at a health information network that has gone from zero uniquely identified patients to 7.6 million – and nearly all major health systems statewide are set to participate in its Common Key Service.

In a bipartisan new report on artificial intelligence, Rep. Will Hurd, R-Texas, chair of the House Oversight and Government Reform Subcommittee on IT, and Ranking Member Rep. Robin Kelly, D-Illinois, put forth funding and policy recommendations for addressing AI's impact on the public and private sectors – including healthcare. WHY IT MATTERS As artificial intelligence continues to evolve at an exponential pace and find its way into all manner of technology, it's already having profound effects on all industries and many important facets of society. Healthcare, of course, is no different, with AI making itself felt in fundamental ways – from newly-approved tools to detect diabetic retinopathy, to applications for new techniques for pop health and precision medicine. This new report explores the ways AI could impact four areas, in healthcare and elsewhere: workforce, privacy, data integrity and cybersecurity. Specifically: Workforce implications could include loss of jobs, so the report suggests government agencies boost efforts to improve training and reskilling so workers to be "more competitive in an AI-driven economy." Privacy concerns include the fact that AI depends on algorithms that depend on troves personal data. The feds should review privacy laws and perhaps update them for the era of AI. Biased data poses a big problem for AI's efficacy and, sometimes, safety. Agencies "should ensure the algorithms supporting these systems are accountable and inspectable." Malicious use of AI is an emerging concern, with its potential to vastly increase the scope and risk of cyberattacks. Government should take more proactive steps to know how AI "could be used to harm individuals and society, and prepare for how to mitigate these harms." WHAT IS THE IMPACT The House report about AI is not strictly about healthcare, but each of those four concerns have potential direct impacts across the industry, from lost jobs for radiologists to patient safety concerns. The study offers an exploration about those challenges, but also the opportunities presented by AI, and gives recommendations for how the government can combat them and capitalize on them, respectively. Increased funding for various government agencies will be a big way of accomplishing both goals. The United States "has traditionally led the world in the development and application of AI-driven technologies," they said. "This is due in part to the government’s prior commitment to investing heavily in research and development that has, in turn, helped support AI’s growth and development. In 2015, for example, the United States led the world in total gross domestic R&D expenditures, spending $497 billion. But U.S. leadership in AI is "no longer guaranteed," said Hurd and Kelly. "To maintain American leadership there is a need for increased funding for R&D at agencies like the National Science Foundation, National Institutes of Health, Defense Advanced Research Project Agency, Intelligence Advanced Research Project Agency, National Institute of Standards and Technology, Department of Homeland Security, and National Aeronautics and Space Administration." ON THE RECORD "AI has the potential to disrupt every sector of society in both anticipated and unanticipated ways," according to the report. "In light of that potential for disruption, it’s critical that the federal government address the different challenges posed by AI, including its current and future applications." THE BIGGER TREND Government agencies are increasingly cognizant of the rise of AI, in healthcare and beyond, and their need to get in front of it to properly harness its evolution, deployment and safety. For all the risks it carries, there are also huge benefits to be had when it's properly deployed. That's why, among other examples, the U.S. Food and Drug Administration early this year voiced its qualified support for AI in healthcare, and promised to refine its regulatory approach for the new era, with initiatives such an FDA-supported tech incubator for AI innovation and a new machine learning partnership with Harvard. "We’re implementing a new approach to the review of artificial intelligence," said FDA Commissioner Scott Gottlieb, MD. "AI holds enormous promise for the future of medicine, and we’re actively developing a new regulatory framework to promote innovation in this space and support the use of AI-based technologies." Twitter: @MikeMiliardHITN Email the writer: mike.miliard@himssmedia.com

Data-driven partnerships with skilled nursing facilities helped the Ohio health system improve care outcomes and lower costs.

Hospital chief medical information officers are almost unanimous in recognizing the patient safety improvements that have been brought about by better medication management processes, according to a new poll from their U.S. trade group, the Association of Medical Directors of Information Systems. The results come amid the ongoing opioid epidemic in America. Technology vendors are trying to help in the battle. IT-savvy hospitals such as Intermountain, which has tweaked its Cerner EHR to reduce such prescriptions in the first place, and Ochsner, which has integrated an opioid monitoring tool into its Epic system, are taking matters into their own hands. More and more states, meanwhile, are passing e-prescribing laws and standing up prescription drug monitoring programs. Even with all that going on, there is a lot that can be done to further boost hospital safety initiatives, said AMDIS members, and that depends largely on bigger and better investments in health IT systems for inpatient settings. Nearly all of the physician informaticists surveyed, in a poll sponsored by e-prescribing and med management developer DrFirst, agreed that patient safety issues are less likely to occur today, as compared to five years ago. In large part, that's due to better medication administration processes enabled by technology. Some 82 percent of the CMIOs polled say such initiatives have led to fewer adverse drug events, for instance. That said, many still thought there was big room for improvement in medication management processes, and saw broader tech deployments to be a way to further safeguard their patents. Only half of the AMDIS members polled said they were satisfied their hospitals' processes, and 12 percent said they were dissatisfied. The CMIOs pointed to gaps such as incomplete patient medication histories (80 percent) and misaligned medication reconciliation and care transition cycles that lead to misinformed decisions by care teams (75 percent), according to DrFirst. The AMDIS survey also noted several other areas where bolstered technology infrastructure could improve medication safety surveillance. For instance, 91 percent of CMIOs said one of their biggest challenges was a lack of visibility into their patients' medication adherence: Pharmacies know when a prescription has been filled, but hospitals are often left in the dark. And 85 percent of the clinical informaticists polled pointed to the fact that patients aren't often active participants in the med rec process as a big culprit when when it comes to spotty medication history at their hospitals, suggesting a bigger role for more integrated patient engagement technologies. Meanwhile, large majorities of CMIOs pointed to challenges with interdepartmental workflow variations, and challenges with process buy-in, compliance and ownership as issues that could adversely impact safety. Nearly all of the AMDIS members surveyed said a holistic focus on the entire medication management process was key to building on safety improvements. At the same time, they said reducing order entry and data validation burdens for pharmacy and clinical staff will enhance patient safety and process efficiencies, according to DrFirst. "While the industry has clearly made significant strides to improve medication management processes, CMIOs remain troubled by a number of gaps that compromise patient safety and quality outcomes," said the company's president, G. Cameron Deemer, in a statement. Opioid risks top concern The survey also found a lot of concern among informaticists specifically related to opioid related monitoring gaps. Nearly two-thirds (65 percent) of CMIOs called for more tightly integrated clinician workflows to enable better coordinating the entire medication management process – including electronic prescribing of controlled substances and access to state prescription drug monitoring programs. While PDMPs have their limitations, being able to better discern patients opioid histories and EPCS data could help stem the spread of addiction and avoid harmful drug combinations. And almost half (41 percent) of AMDIS members said they had concerns about hospitals' opioid readiness, given the difficulty to get discern so-called drug shoppers from patients with genuine pain management needs. As we noted earlier this week, several technology vendors are exploring, from different angles, the potential of blockchain to help solve that challenge. Twitter: @MikeMiliardHITN Email the writer: mike.miliard@himssmedia.com

Add this to blockchain’s growing list of innovative use cases: tracking prescription and population health data.

Absent more rigorous federal regs, Pew Charitable Trusts, MedStar Health and AMA offer model test cases to help vendors and providers detect potentially dangerous usability risks.

Studies supported the groundbreaking IDx artificial intelligence tool, which can detect diabetic retinopathy with medical imaging – but the investigation of its validity wasn't above critique.

Several news stories have questioned what reported struggles with the cognitive platform mean for the larger idea of AI in healthcare.