Wednesday, July 28, 2004
The HealthGrades report used 16 of 20 'Patient Safety Indicators' developed by the federal Agency for Healthcare Research and Quality to screen hospital administrative data for safety-related incidents from 2000 through 2002.
In all, the report looked at 37 million Medicare patient records, representing about 45 percent of all hospital admissions in the United States -- not including obstetrics patients.
...There were about 1.14 million 'safety-related incidents' associated with 323,993 deaths in hospitals during the period reviewed by HealthGrades, which is based in Denver. Eighty-one percent of those deaths were directly attributable to the incident.
And one in every four Medicare patients who experienced an incident died, the report found.
'Failure to rescue' (which refers to failure to diagnose and treat conditions that develop in a hospital), bedsores, and postoperative sepsis accounted for almost 60 percent of all ' safety-related incidents,' according to the report.
The report's authors said these errors accounted for $8.54 billion in excess costs to the Medicare system over the three years studied. If that number were extrapolated to the entire United States, it would mean an extra $19 billion was spent and more than 575,000 preventable deaths occurred from 2000 to 2002, the authors concluded.
That sound horrible. Maybe we should fire all the doctors and close all the hospitals. Then no one will be harmed and we'll save billions of dollars.
But wait, the "study" leaves more questions than answers, as one of the authors of the report that started all of this, To Err is Human, by the Institute of Medicine, points out:
'Medicare patients have a higher adverse event rate because they have a lot more treatments, they're sicker, they have multiple diseases, so the mortality rate, the error rate, all these things are higher,' said Dr. Lucian Leape, adjunct professor of health policy at the Harvard School of Public Health, and one of the authors of the IOM report.
Furthermore, he added, 'failure to rescue' is not normally used in calculating deaths from medical errors; it's not an accepted standard."
The problem is, these folks act as if we already live in a FuturePundit world where no one ages, no one gets sick, and no one dies. We have not reached that state, and until we do, people will continue to get sick and continue to die. Not every rescue attempt can be successful. And many times, especially in the very old whose bodies are worn out, our rescue attempts are like spitting in the wind.
So why would this HealthGrades company go to such lengths to exaggerate the dangers of modern medicine? Because, they make their money by giving error-reduction advice. Got to drum up business. And a perception of a problem is sometimes better than an actual problem at stirring up the market. Trouble is, unless they have the waters of the fountain of youth in their vaults, they can't correct these "failure to rescue" events, either.
UPDATE: A reader notes another problem with the study:
Not only are all your points well-taken but the software they used was designed to identify POTENTIAL preventable deaths, and these people treat them all as preventible.
posted by Sydney on 7/28/2004 07:58:00 AM 0 comments