How to misinterpret statistics to make your organization sound important

Stroke associations use this when they report on stroke dropping from 3rd to 5th leading cause of death in the US. They don't report on disability rates in the same time period because nothing has changed. I bet a lot of the drop in death rates is because younger adults are getting strokes and are more likely to survive.
http://medcitynews.com/2016/05/misinterpret-statistics-make-organization-sound-important/?
Post a comment /

31 Shares

• 
  
• 
  
• 
  
• 
  

/ May 27, 2016 at 9:05 PM

Brace yourself for upcoming news reports on how a new, federally funded initiative from a pharmacists’ group could save 98,000 lives a year. Those reports will be wrong, because a publicist for that group misunderstood statistics and news outlets will still take him at his word, because, well, nobody can be troubled to check facts anymore.
Also, the publicist really needs to be fired.
Friday, I received a pitch that started:

Since 2000, roughly 98,000 people die annually from medication errors and are the 3rd leading cause of death in the U.S. This is staggering when compared to some of the most widely reported causes of death: AIDS (13,712); car accidents (43,510); and Breast Cancer (41,555).

In addition to the poor English (people are the 3rd leading cause of death?), there’s a glaring mistake right in the first sentence, and at least one other mistake that might not be so obvious.
No, 98,000 people do not die each year from medication errors in this country. That 98,000 figure is the high end of the estimate of annual deaths from adverse events in hospitals, as reported in the landmark 1999 Institute of Medicine report, “To Err Is Human: Building A Safer Health System.” The number of deaths attributed to medications was closer to 7,000.
By the way, the report was based on retrospective data, so claiming that these deaths started in 2000 is disingenuous.
As for the “3rd leading cause of death,” that also refers to inpatient hospital deaths attributed to all medical errors, not necessarily medication errors. That statistic comes from one of two places: A 2013 study in the Journal of Patient Safety or a BMJ study published just a few weeks ago. (We discuss those here.)
The 2013 report put the range of preventable deaths at 210,000-440,000, while the new study said there were at least 250,000 preventable hospital deaths per year. Extrapolating from the IOM statistics, you might be able to argue that 15,000-30,000 people die annually from medication errors, which is terrible, but still far less than 98,000.
Still, thanks to this misleading pitch, someone is going to report that the American Society of Health System Pharmacists is trying to save nearly 100,000 lives a year.
The ASHP recently won a three-year contract from the Food and Drug Administration to implement a plan called Standardize 4 Safety. That program will develop standard concentrations for IV and oral liquid medications, in hopes of cutting medication errors.
Wonderful. This needs to happen. Just don’t tell me that this is going to prevent 98,000 deaths annually, and don’t let me read that number in other publications that don’t know the truth.