Inside Medicine: Accuracy is elusive when it comes to providing a prognosis.

Published: Wednesday, Jan. 22, 2014 - 5:21 pm

“So, Doctor, how long do I have before I die from this disease?” Not an unreasonable question from a person just diagnosed with cancer or serious heart disease. There are many reasons a reasonable person might want to know the answer, including being able to plan the rest of his or her life, to assist in deciding on treatment options, for estate planning and to engage in psychological preparation.

Despite advances in understanding the mechanisms of many diseases and in understanding disease patterns, doctors still do poorly when providing a person an accurate prognosis. Doctors are wrong more often than they are right. There are several reasons for doctors’ inaccuracy, including their poor understandings of probability, the uniqueness of each individual person, and even the doctor-patient relationship. Experts who care for terminally ill people overestimate the chance of an individual surviving a set period of time often by as much as five-fold compared to what the data would predict.

It turns out that the closer the relationship between doctor and patient, the more likely the chance the doctor’s thinking will be clouded by what psychologists call an “optimism bias.” Doctors and patients both hope for the best outcomes and this often leads to a collusion in hopeful thinking whereby the doctor assumes the best for the patient and provides an overly optimistic prediction.

Take as an example a person who the doctor believes has a 33 percent of dying in the next three months. What this means is that over the next three months, if we took 100 people with the same disease, 33 of them would be expected to die by the end of that period and 67 would be alive.

There are several problems with this type of thinking. First, each of those 100 people are not, and cannot, be the same either in terms of their underlying health or in the biology of the disease, Also, each person’s human behaviors (sleep, smoking, weight, diet, exercise, etc.) and attitudes differ greatly. Secondly, the doctor has no sorting hat that tells him to which group his/her patient belongs – the 33 percent or the 67 percent group.

Having hospitals, the NIH or CDC collect and analyze more scientific data is unlikely to improve our prognosis accuracy. Diseases are often sneaky and unpredictable. Still, not knowing the prognosis makes it difficult for the person, his or her family and the health care team.

When there is uncertainty doctors often feel obligated to collect more data, request more doctor appointments and seek second, third, fourth opinions. All this medical attention requires sick people to spend more time in and around hospitals thus denying them a chance to enjoy the present (unless of course they really enjoy being around the medical team).

Often people with bad diseases become so consumed trying to predict events in the future that they fail to enjoy today and may miss important opportunities and activities.

There is another side of this. My end-of-life colleagues point out that living in the present can be scary for a sick person, and many people feel a driving need to be told an upbeat prognosis that can provide them with hope and a feeling that better times may lie ahead.

If one believes in transparency, it is important for doctors to engage in discussions around uncertainty and limitations in predicting the future while also providing opportunities for hope – hope for a longer life and hope for the highest quality of existing life.


Michael Wilkes, M.D., is a professor of medicine at the University of California, Davis. Identifying characteristics of patients mentioned in his column are changed to protect their confidentiality. Reach him at drwilkes@sacbee.com.



Sacramento Bee Job listing powered by Careerbuilder.com
Quick Job Search
Sacramento Bee Jobs »
Buy
Used Cars
Dealer and private-party ads
Make:

Model:

Price Range:
to
Search within:
miles of ZIP

Advanced Search | 1982 & Older

TODAY'S CIRCULARS