By Dr. Howard Brody 

The estimates of how many patients die in American hospitals due to potentially preventable errors keeps climbing — despite more than a decade of serious efforts to lower these rates.

Is it possible that we are ignoring a simple answer to this problem?

Back in 1999, an Institute of Medicine report made headlines by calculating that as many as 98,000 Americans die of errors annually in hospitals.

Just recently, Dr. John T. James of Patient Safety America, based in Houston, used newer methods and argued that the true death rate is nearer to 400,000, making medical errors the third largest cause of death in the U.S.

This is discouraging not only because a single death of a patient, that could have been prevented had the staff been more careful in some way, is itself a tragedy.

It’s discouraging because as anyone who works in hospitals knows, the entire system has been on red alert since 1999 to find ways to reduce and prevent errors.

Yet the harder we all try, the more the numbers keep climbing.

We now get into territory where I have no evidence at all to back me up and am wandering off into pure speculation.

I also have distinct biases from my training in the field of family medicine, where we tend to believe that the simplest approach is often the best approach.

So, in a purely speculative vein, I wonder if we have missed the boat because we ignore the root cause of so many errors, the ever-increasing complexity of health care.

Whenever we trip over our complicated technology and find that it is causing harm to patients, we decide that the best fix is — more technology.

The new technology is well intended and in its own way sounds completely logical.

But by increasing the total technological complexity of the whole system, what if the end result is more errors, not fewer?

An example is provided by electronic medical records.

These gizmos were introduced with great enthusiasm because it made sense that a computer could avoid a lot of human error — such as the fabled illegible handwritten prescription.

The trouble is, of course, that humans have to use the computers. Some of these records are so complicated that they create new ways to make errors for every old way they avoid.

When we add more technological complexity, we imagine we’re making health care better — and we usually are, but only to a marginal degree.

We never ask whether by increasing the sheer complexity of the entire system, we may be undermining any hoped-for improvement.

What if we save one extra life out of 10,000 patients with our fancy new technology; but the more complex health system kills 5 out of 10,000 extra patients with medical errors?

So I have a very controversial proposal. Let’s try an experiment where smart people review a lot of complicated procedures we now do in hospitals, look carefully at the evidence of what benefit they provide, and figure out ways to simplify the processes as much as possible.

Then let’s try it and see if the error rate goes down.

Dr. Howard Brody is director of the Institute for the Medical Humanities at the University of Texas Medical Branch at Galveston.