It has been over 10 years since the publication of the Institute of Medicine's study on preventable medical errors: To Err Is Human: Building a Safer Health System
The first story, above, indicates errors are still a major problem:
While patient safety in US hospitals is improving, “medical mistakes still occur at an alarming rate,” according to the sixth annual HealthGrades study of patient safety in American hospitals, released today.
Between 2005 and 2007, medical errors cost Medicare over $6.9 billion and were responsible for more than 92,000 potentially preventable in-hospital deaths among Medicare beneficiaries, report Dr. Rick May and others at HealthGrades, a healthcare ratings organization in Golden, Colorado.
Keep in mind these are just Medicare patients. What percentage of all patients are on Medicare? I don't have that number handy… would really be interested to extrapolate 30,000+ Medicare patient deaths to the whole patient population.
Some more numbers:
More than 913,000 total “patient safety events” occurred, representing 2.3 percent of the nearly 38 million Medicare hospital admissions.Patients who suffered one of these mistakes had a one-in-ten chance of dying, the report indicates.
The AHRQ says there are about 29 million hospitals admissions each year in the U.S. At 0.23%, that would be a total of 66,700 deaths per year.
Are these incidents random events? It seems not — it behooves us, as patients, to find out which hospitals have better quality. You can do so at the HealthGrades website, but alas they sell the detailed reports (and no need for disclosure here, I don't take a cut for referring you to them). It seems difficult to, as a patient, draw valid conclusions from the data they show, but maybe that's the subject of another post (or for another blogger).
The investigators observed that, on average, Medicare patients treated at award-winning hospitals were 43 percent less likely to experience a medical error compared with those at bottom-ranking hospitals. “This finding of better performance was consistent across all 12 patient safety indicators studied,” the authors write.
At least things aren't consistently bad. So what are the “award winning” hospitals doing differently? It's got to be a matter of PROCESS, not people. I can't imagine the award-winning hospitals are hiring people who are smarter or more careful.
Errors with the highest occurrence rates were “failure to rescue,” defined as death among surgical inpatients with serious treatable complications; bed sores; postoperative respiratory failure; and serious postoperative infections.
There are some problems inherent in extrapolating the Medicare numbers, since those patients are older and, presumably, more likely to get bed sores. But still, this is a serious problem.
This related article also caught my attention:
Nearly 90 major medical mistakes logged at Utah hospitals in 2008 – Salt Lake Tribune
This number was only the events that caused death or serious harm in Utah, not all errors.
Despite a years-long effort to cut down on one type of medical mistake – surgical errors – they remain Utah's top problem. There were 45 surgical errors last year, such as performing the wrong surgery on the wrong body part. One example: A gastrointestinal tube that was guided into a lung instead of the stomach.
“We're struggling,” said Iona Thraen, who reviews the mistakes as director of patient safety for the state health department.
Why are they struggling? It's not just Utah… why do the experts (I'm not pointing at myself as the expert) say many or most of these are preventable?
The standard practice is for hospital staff to manually count the sponges before and after surgery to ensure they are removed and confirm the removal with an X-ray, said Thraen. When reviewing the cases when sponges were left inside patients, the staff members are usually certain they counted and re-counted the material, she noted.
OK, you might wonder — how hard is it to have some standardized work that says you count and even re-count? Does the phrase “usually certain” set off red flags?
Counting — this is a form of visual inspection, done by a person — it's going to be prone to error. We're human. We make mistakes.
But do we have to make as many? The sidebar at the bottom of the article highlights a situation that might not be unique or rare:
During a routine inspection of McKay-Dee Hospital in Ogden last year, state health department surveyors cited the facility for compromising patient safety because surgical staff didn't count instruments before and after surgery. They did count sponges and needles.
Inspectors were told that staff only counted instruments during open-heart surgery and that surgeons were “reluctant to allow staff to perform instrument counts” because “it added more time to the surgical procedure,” according to the inspection report obtained by The Salt Lake Tribune.
Whoa. So when things like this are happening, as fancy RFID technology the only solution? The most cost effective solution?
How do you win the cultural battle that patient safety, not speed, comes first? Can you eliminate waste and take time out of the procedure rather than eliminating this step that impacts patient safety?
Seems like this, and many other medical mistakes, are cultural and social problems more than they are technical problems.
So I'll leave it on that. The LEI's John Shook has a blog post about looking beyond the technical for root causes of problems that fall in the social realm. Seems to fit here.
Comments?
Subscribe via RSS | Lean Blog Main Page | Podcast | Twitter @markgraban
What do you think? Please scroll down (or click) to post a comment. Or please share the post with your thoughts on LinkedIn – and follow me or connect with me there.
Did you like this post? Make sure you don't miss a post or podcast — Subscribe to get notified about posts via email daily or weekly.
Check out my latest book, The Mistakes That Make Us: Cultivating a Culture of Learning and Innovation:
I am not too familiar with this subject (healthcare errors) but in manufacutring this would be related to compliants and risk management. The obvious risk of death being signficant. Wouldn't FMEA be a tool you would use to evaluate proces or euqipment design potential opportunities for mistakes, errors, failures and the associates risk. Then I guess the harder part of improving those weakneses to reduce the risk.
Where humans are involved you can not elminate human error but improvements can and certainly have been made. The hardest part in lean or continuous improvement is learning to "see" the waste. I guess I wonder if the can and what tools are they using to do that.
The same principles apply in Healthcare, Tim, only they are often a little more difficult to see, but with training and experience there is no reason improvement shouldn't happen. Marks's article, with reference to John Shook, illustrates the cultural issues very clearly, and this only supports the notion that using only lean tools without cultural change will fail (the "counting" is just level 2 error-proofing at best). So they know they are supposed to count… This is effective inasmuch as they know why, what is at stake, and to the extent the Director audits the process, reinforces it, and removes barriers to it being omitted.
Mark:
Sitting here waiting for a family member to come back from the operating room (minor surgery = surgery on somebody else!) your topic of the day naturally led me to try to divine where this particular hospital stands. It is not easy at the "HealthGrades" website! Not only do they require you to chase down one rabbit trail at a time, by procedure and region, to see how an individual hospital is rated, but the ratings are for complications and other outcomes – NOT explicitly "errors" – so they are only really useful as relative measures. You can't really expect a "zero complication" environment – this (hospital care) is rework, after all, fixing problems! A complication in (or after) the O.R. is not necessarily an error – some people are really very sick! It would be great to see comprehensive error reporting. Maybe I just missed it.
Any how, that said, the "BEST" results (relative or not) I found on "HealthGrades" web site for this hospital in almost all the categories I tracked down reflected my observations in the gemba here: visual cues, standardized communications at the pre-op level (with patients and with family), and flexible scheduling as opportunities for accelerated movement arose. Checklists in use! White boards (that weren't here last time I passed through) with obviously prototyped labelings, speak of an environment of continuous improvment (and reminded me of what I see and strive for in my own plant).
So, my shallow investigation into a totally unscientific sample of ONE, says: This stuff works. Who is surprised? It is said that when you walk into a lean plant, you can tell – you can see what is supposed to be happening, and where the problems are even if you don't know the business or the industry. My experience in health care gemba is the same.
-Anonymouse today, due to reference to family member in hospital!
Mark,
There is no doubt error and unexpected or undesired outcomes happen in healthcare. A very real problem in healthcare is knowing what constitutes quality.
For instance, many of the Medicare quality indicators do not really reflect on quality. Moreover, rating services, like Healthgrades, are far too crude to provide helpful measurement, and don't really contribute to understanding what quality actually is.
Healthgrades gets its data from the individual hospital. Its revenue is derived from the hospital paying Healthgrades if the hospital wants to use their ratings for marketing.
In one instance I am intimately familiar with, Healthgrades presented a hospital with one of its few 5 star ratings in a particular state, only to have the hospital lose its Medicare and Medicaid funding one month later for serious and unresolved quality problems!
How can this be? It's hard to responsibly and effectively measure what we don't truly understand. In lieu of this, we create measurements that do more harm than good by misleading us into thinking we are doing quality, creating rather than reducing waste. Metrics must be clear and relevant. So far, medicine has neither. And this is a real problem
Mark – I heard similar criticism of HealthGrades last week. That seems very shady when it's "pay for play" or "pay for rate."
It's interesting to see how efforts like the Wisconsin Collaborative for Healthcare Quality play out:
http://www.wchq.org/index.php
I don't blame Healthgrades or Leapfrog or Medicare or any of a number of other groups trying to begin an effort to rate healthcare. We must, however, be very mindful of their limitations before we quote them as being definitive. They, too, must practice continuous improvement to someday zero in on what really matters.
Doctors get bashed over not having done this work, but it just reflects how difficult and complex it is. Doctors know that the history of medicine contains too many examples of a great idea that just didn't work out when tested against reality.
Therein lies the potential for Lean
Mark
The recent US News and World hospital ratings are a perfect example. Look at their initial screening parameters, which are based on being a teaching hospital, number of beds and number of nurses, none of which have any relationship to quality. Where is their methodology to justify these criteria? And yet, most people will not even note this. They will just accept that there is merit to these ratings.