What Gives?

“Get your facts first, then you can distort them as you please.”

Mark Twain

I was listening to a travel podcast recently about President Trump’s decision to immediately stop visits to Cuba by Americans. The podcast hosts were very upset and one said that the immediate impact would be on 800,000 passengers who had booked a cruise which included Cuba as part of the itinerary. That caught my notice.

I am not interested in the politics of the decision. I am interested in the cavalier use of statistics to make a point. There is no way that 800,000 cruise ship passengers will be affected by this decision. Given the number of ships calling on Cuba this year and their size, that is a dubious number.

We see this same careless use of statistics in healthcare where the stakes for being wrong far outweighs that of a disappointed passenger. Careless use of statistics means resources may be wrongly deployed to everyone’s detriment.

For instance, take the Institute of Medicine’s 1999 report, “To Err Is Human”. The report was a shocker with its estimate that between 44,000 and 98,000 people a year died in hospitals as the result of medical errors. Even one such death is one too many.

That report resulted in a great deal of concern as it should have. Hospitals and medical professionals made changes in care including things as simple as being more diligent about hand washing between caring for patients. That effort continued as incentives both positive and negative were created by the government and others. No one I knew in healthcare was happy about the idea of patients needlessly suffering or dying.

Fast forward now fourteen years to 2013 when the Journal of Patient Safety published an article estimating that there were between 210,000 and 440,000 hospital deaths due to medical errors. All the efforts to decrease the number of preventable deaths noted by the Institute of Medicine in 1999 seemed to have had no impact at all. It was discouraging. There was a question which should have been asked: What gives?

Two years later in 2015 the Leap Frog organization, which attempts to give hospitals letter grades for quality using a problematic methodology, estimated that there were 205,000 preventable deaths in hospitals. That was followed three years later by another estimate by Leap Frog of 160,000 preventable deaths. Again, I have to ask the question, what gives?

The common thread to all these statistics is this word: “estimate”. That is a word which needs clarification by the people issuing these reports. The words “estimate” and “estimated” could simply be another way of saying we don’t know.

Estimates are not facts; they are guesses. In baseball, no one says that Buster Posey’s batting average is estimated as being .294. Should we not expect a more rigorous committment to facts and statistics in healthcare when dealing with important matters?

When you have a range from 44,000 to 440,000 deaths, you have a lack of preciseness which is troubling. Furthermore, when the estimates increase significantly after a decade of process improvements directed at reducing such deaths, you have to wonder whether the report issuers have their own error problems. Or maybe they are just lousy estimators.

I certainly knew during my career as a healthcare system CEO and a board member of a risk retention company providing professional liability insurance of medical errors which led to needless suffering and/or deaths. There is no defense for any such occurrence and every and all effort should be made to eliminate the errors. There were too many such errors.

We have a problem but we do not know exactly how big a problem it is. We need more than just estimates if we are going to be able to measure how effective we are in reducing preventable deaths to zero. Of equal concern is we don’t know how many faulty “estimates” are being used in other areas of healthcare to drive decision making in important areas. Perhaps there is a need for a report on “To Estimate Is Human”.