As a cardiology fellow, I remember reading a 1995 study1 testing the ability of primary care clinicians to predict cardiovascular risk. The conclusion was that clinicians did a poor job estimating absolute risk, although they made better estimates of the relative risk between patients. Now in 2018, a new study2 tested clinician’s ability to estimate absolute risks of a woman being diagnosed with breast cancer, having breast cancer saved by mammogram screening, and the side effects associated with screening. Again, clinicians were unprepared with accurate numbers of side effects. In the intervening 23 years, other studies have produced similar findings. While at first glance it may seem that these studies are uncovering an important deficit in medical care, I believe we are testing the wrong skills.

My son recently took his first high school exams and I was struck that across math and science classes, students were not tested on memorization. They could take in a 3×5 card with formulas and standard constants, and then they were tested on their abilities to understand the mechanics of reaching the answer. Similarly, as I directed a second-year medical school course for 13 years, we focused on testing knowledge of the principles of cardiac physiology and not just recall of facts.

The important skills for clinicians should not be memorization of numbers but rather interpreting those numbers in ways that help patients learn the more complex concepts involved, and thus make more thoughtful and informed decisions. For example, fully discussing mammography screening includes addressing over-diagnosis, a difficult concept that goes by different names and can be confused with false-positives. For a representative 55-year-old white woman with a family history of breast cancer and a breast density of B, we should not expect clinicians to retain that in 10 years of annual screening, over-diagnosis occurs in 5 out of 1,000 patients. Instead, the more important skill is teaching a patient that over-diagnosis is based on population studies that pushed us all to recognize that there are cancers out there that we don’t want to find. Maybe one is barely growing at all, or the patient’s immune system will take care of it, or the patient will die from other issues before the cancer is noticed. In all these cases, finding that cancer and all the accompanying anxiety, life changes, medications, and procedures is, in hindsight, unnecessary. Teaching the patient a number is an exercise in fact recall. But teaching our patients the complex ideas behind the numbers is the mastery of the experienced clinician.

This is why HealthDecision exists. Our tools gather data, run calculations, show guidelines, and lay out treatment options generally in the first 60 seconds of use. We offer the correct 3×5 card of fact recall and the time we save the clinician up front can then be spent engaging the patient about complex topics. Using our tools allows clinicians to focus their valuable mental horsepower on the patient’s understanding of the issues behind the numbers which makes the visit more meaningful for both patient and clinician. Getting accurate numbers is necessary and being aware of national guidelines is a key step, but that’s not where the important job of helping patients ends. It is where it begins.


  1. Grover S, Lowensteyn I, Esrey K, Steinert Y, Joseph L, Abrahamowicz, M. Do doctors accurately assess coronary risk in their patients? Preliminary results of the coronary health assessment study. BMJ (Clinical Research Ed) 1995;310(6985):975–978.
  2. Martinez KA, Deshpande A, Ruff AL, Bolen SD, Teng K, Rothberg MB. Are Providers Prepared to Engage Younger Women in Shared Decision-Making for Mammography? Journal of Women’s Health 2018;27(1):24–31.