The revelation that medical error is the third leading cause of death in the United States sent unsettling reverberations through the healthcare industry last week, but the news is likely only the tip of the iceberg and much more must be done to address this growing health issue.
Researchers Martin Makary and Michael Daniel recently published findings in BMJ revealed that deaths caused by medical error numbered more than 250,000 annually — making medical error a leading cause of death only behind heart disease (611k) and cancer (585k) — and are likely higher as a result of system-wide under-reporting.
"To achieve more reliable healthcare systems, the science of improving safety should benefit from sharing data nationally and internationally, in the same way as clinicians share research and innovation about coronary artery disease, melanoma, and influenza," they wrote.
Sound scientific methods, beginning with an assessment of the problem, are critical to approaching any health threat to patients," Makary & Daniel continued. "The problem of medical error should not be exempt from this scientific approach. More appropriate recognition of the role of medical error in patient death could heighten awareness and guide both collaborations and capital investments in research and prevention."
Despite its prevalence, medical-error-related death is neither listed on death certificates nor listed as a disease. The lack of accurate information on deaths caused by medical error, moreover, warrants deeper analysis and reporting of available data on the subject with members of the healthcare community.
"Although we cannot eliminate human error, we can better measure the problem to design safer systems mitigating its frequency, visibility, and consequences," wrote the authors, who identified three steps for developing strategies for reducing these types of deaths.
First, medical errors must be more visible. Second, remedies must be available to safeguard patients from medical errors. Third, best practices must be shared to make medical errors less frequent.
What is a medical error?
The question appears on the surface to be straightforward, but that isn't necessarily the case.
"One of the problems in this patient safety area is the under-reporting of errors because there is a fear of reporting them," Mick Murray, PharmD, MPH, tells EHRIntelligence.com. "Any project you wind up seeing errors reported is typically the tip of the iceberg. Medical errors are better characterized in the inpatient setting than the outpatient setting, mainly because there are more people there to detect them."
Among his many roles, Murray serves as the head of the Executive Director of Regenstrief Center for Healthcare Effectiveness Research as well as Director of the Regenstrief Institute Data Core. (His academic positions are more numerous.)
For Murray, an important first step in addressing deaths caused by medical errors begins with determining how prevalent they are, particularly those stemming from medication administration errors — and time is of the essence.
"A big area where they oftentimes go undetected is the outpatient area," he observes. "There’s a projection that medications for outpatients are going to swell from four billion prescriptions a year to six billion within the next several years. And a lot of those medications are going to become more and more potent. Somewhere along the line we need to get health professionals and patients on the same page in terms of better prescribing, better dispensing, and better management of medications in the home."
His Regenstrief colleague Paul Dexter, MD, is in agreement. A hospitalist by training, Dexter also serves as the CMIO of Eskenazi Health, Associate Professor of Clinical Medicine, Indiana University School of Medicine, and Research Scientist at Regenstrief.
"Every error that happens needs to be in a somewhat judgment-free environment so that it can be recognized and people aren’t trying to hide it so that everything possible can be learned from that error and next time a patient comes in the same mistake won’t be done and technologies would be used as necessary to provide a reminder about how to do the right thing at the right time," he asserts.
Along with that understanding must come the recognition that not all medical errors are created equal.
"Not everything that happens is an error," Dexter maintains. "The first time a patient gets a medication there are risks of allergies and very rarely life-threatening allergies. There’s simply insufficient data to have been able to predict that. Sometimes it’s bad luck," Murray's Regenstrief Paul Dexter, MD, tells EHRIntelligence.com.
In his capacity as a research scientist, Dexter studies the role of EHR use and functionality in mitigating instances of medical error, but he is quick to admit that technology is not an end in itself.
"You won’t solve medical errors by just throwing technologies at it," he maintains. "But you also won’t solve medical errors by pure human processes. If you had teams of physicians being all able to check each other, I’m sure you could reduce the rate of error by 10 percent. But we already have a physician shortage. We’ll never get where we want to without a mixture of processes — and recognizing these errors and being willing to talk about them. It isn’t just one."
That approach is shared by his colleague Murray.
"One thing we have to get over in terms of the people issue is freedom to report errors. Otherwise, we’re never going to get a good handle of these things. When they do get reported, there is a lot of private-eye work done to detect what went wrong. That’s exactly how we learn from it," he adds.
What is a the role of health IT in preventing medical error?
While the solution to reducing medical errors goes well beyond technology, the latter will in fact play a critical role.
It is only a matter of time, according to Murray. "The IT, the technology approach is really the great hope. Technology is going to progressively get smart enough to detect these things before they happen. We’re just not there yet," he remarks.
At present, health IT slots in as a safety net.
"In the vast majority of cases, you have well-meaning, well-intentioned clinicians who are trying to do right thing but who may be scattered in terms of their focus," says Dexter. "That’s where you want to have a safety net so that technologies, personnel, and processes double-check and triple-check decisions and hopefully reduce the rate by orders of magnitude."
EHR technology is already proving beneficial in detecting and avoiding medical errors in practice. One such example is computerized physician order entry (CPOE), a health IT feature in place in many inpatient settings.
"Medication errors are frequent and common, and order entry systems where they enter dose and the medication can do a lot of crosschecking and decrease errors," says Dexter. "That’s one of the most established of areas or domains in technology improving safety. At an early stage, it was shown that order entry systems can decrease medication errors by somewhere in the range of half — through legibility, preventing errors of accidentally being off on one’s dosage by an order of magnitude."
"Medication errors are frequent and common, and order entry systems where they enter dose and the medication can do a lot of crosschecking and decrease errors."
Similarly, Dexter sees a valuable role for health information exchange to provide a longitudinal record of a patient that enables providers to view a patient's allergies and history of illness in a single location.
"These electronic health records at best have the possibility of improving communication among clinicians and patients, consultants, and all that. That’s what we need to get to, but we’re a long ways away from it," he states.
More down the road, diagnostic error technology, remote monitoring, and telemedicine are likely to feature prominently in avoiding potential medical errors.
A time for clinical decision support to shine?
Clinical decision support has received criticism for being "cookbook medicine." In truth, CDS systems and services can prevent harmful reactions to prescribed treatments. The challenge is presenting the right type of information at the right time to the right person.
That's one of several lessons learned by Peter Edelstein, MD, in his role as Chief Medical Officer of Elsevier Clinical Solutions.
"Who in their right mind believes spending 16 to 18 percent of GDP but having preventable medical errors be in the third leading cause of death is a good equation?" he asks rhetorically. "When we talk about these single source of truth ideas, we’re talking about this big repository of current and credible evidence-based content. But how that is presented to different providers is enormously different."
For CDS technology to pay off, it must be consistent for all and at the same time tailored to the end user.
"Physicians, whether they’re specialists or generalists, we tend to want rapid access and we like things presented in a certain type of format," Edelstein explains. "We learned it in med school. We learned it internships and residencies. We process information very quickly looking at certain types of charts, tables, and synoptic content. That’s true whether your ER doc or an OB, whether I’m doing surgery in Vermont or doing dermatology in California."
That isn't necessarily the case for nurses and other clinicians.
"It’s the way physicians quickly understand, evaluate, and incorporate information — entirely different than the way that nurses incorporate information," he elaborates. "It may even be the same information, but it’s presented with much more narrative, much more holistically, much more in the context of patient care, not disease management. Pharmacists? Different information. Different presentation. And patients of course see the same information entirely different based on educational level, language, culture, socioeconomics."
"These electronic health records at best have the possibility of improving communication among clinicians and patients, consultants, and all that. That’s what we need to get to, but we’re a long ways away from it."
In other words, CDS technology is only valuable insofar as all end-users are on the same page, following the same information — hence, the single source of truth.
"It is extremely important to format and deliver it in the way appropriate for each one of those four groups. At the same time, it has to say the same thing fundamentally. That’s the single source of truth. It can’t have competing, conflicting instructions or we have our current disaster," Edelstein claims.
Extending the value of CDS will require its use first among the right user base before rolling out its use to the general provider population.
"Physicians, nurses, pharmacists, and therapists are much more homogeneous than the general population," asserts Edelstein. "It makes sense to start to learn how to use CDS with a small population, the traditional providers who have a fair amount in common and try to drop those numbers. Then when we really understand how to do things, we just have to slam it in to the general population."
Investments in this and other health IT should help chip away at the number of deaths caused by medical errors over the years to come. However, their impact will remain uncertain until the number of medical errors reported is accurate.
"Regulation and policy is probably going to be one important area. But we’re still going to have to do a better job in the non-punishment of reporting an error," says Murray.
Because to err is human, these errors will persist. Having a trustworthy baseline will give the industry a much-needed target for mitigating them.