- inputs - data that might affect our health
- process - an action we can take to change our outcome
- decision - a question that nudges us to make deliberate choices and
- end result - the consequence of the decision process
What troubled me, however, was the implicit message that simply taking tests, like genetic testing for a fee, or getting a wealth of medical data and then using a decision tree could result in patients choosing the right care for themselves. The internet has allowed individuals to no longer rely on experts to book plane tickets or manage our investments. The author seems to suggest that in healthcare we've reached the time that patients can empower themselves in a similar manner.
[We’re at a moment] when more data than ever lies within ready reach. Whether it’s personal genomics services like 23andMe or screening tests or self-tracking iPhone apps, each of us can draw on a wealth of personalized data sources that turn generic medical advice into customized health equations. And this is always-on data: Instead of checking in on our health episodically — when we visit the doctor or get lab test results — we can now tap into a constant stream of information and opportunity. We can minimize our uncertainty and maximize our control. We can build ever more sophisticated, and useful, decision trees.
If it was really that simple. There are significant nuances between the theoretical and practical, especially when it comes to genetic testing, which in many instances has not yet been proven to be an accurate predictor about one's future health. To understand the nuances take the simple routine cholesterol test. You already know to eat healthy, exercise, and maintain a healthy weight to achieve a lower cholesterol. Your question is whether your total cholesterol of 280 and a HDL (good cholesterol) of 35 is a problem? Do you need to take a cholesterol lowering medication?
The answer depends.
If the patient is a 40 year old man, who smokes, and otherwise healthy with a blood pressure of 120, then his risk of a heart attack is 21 percent over the next decade.
If the patient is the same man, but a non-smoker, his risk now is 5 percent over the same time of ten years.
If he is a non-smoker, but taking medication to maintain his blood pressure at 120, his risk for heart attack becomes slightly higher at 6 percent.
For the first example, the first thing the person should do is to quit smoking. The risk of heart attack drops by 75 percent. If he refuses, then his doctor should recommend starting a cholesterol lowering medication as well as suggesting taking an aspirin daily. In the last two examples, the risk is small enough that diet and exercise alone are adequate. Would a flowsheet have captured this difference in outcome?
While a decision tree and flowchart can be helpful and it is likely doctors are using a similar algorithm in our heads, the issue is who is interpreting the information. A flowchart for a critical health care decision may get the patient an end result. In situations where newer tests and technologies are involved, a review with a doctor who has the experience and expertise, will provide the framework for a candid discussion and a great result.
If this article demonstrated anything then it is that we as doctors have consistently failed to take the medical information available, interpret, translate, and then communicate clearly the risks, benefits, and choices personalized for an individual person. We can and must do better.