More power of personal feedback

Now that I’ve dumped on emerging behavioral feedback technologies, perhaps I should share a personal success story, in which measurement technology played a key role.

Ten years ago, a routine test revealed that my cholesterol was 280 mg/dl, and even higher in a confirmation test. That’s not instant death, but it’s bad. NIH calls <200 desirable, and many argue for even lower levels.

This was a surprise, because I was getting a fair amount of exercise and eating healthier than the typical American diet. I suspect that their must be some genetic component.

Without any discussion, my doctor handed me a prescription for Lipitor. Now, I liked that doctor, and I know he was smart because we’d just had an interesting conversation about wavelet analysis of time series data in biomedical research. But I think he was operating under the assumption that there was no potential for improvement from behavior change. This idea seems to grip much of the medical profession, and creates nasty self-fulfilling prophecy and eroding goals dynamics.

I decided that I didn’t want to take statins for the rest of my (hopefully long) life, so with the aid of spousal prodding and planning, I eliminated all cholesterol and saturated fats (essentially all animal products) from my diet. I was quickly below 200, and then made more gradual progress to a range of about 160 to 180.

Interestingly, since then I’ve also cut out a lot of carbohydrates, because the rest of my family is gluten intolerant, which takes the fun out of bread and pasta. My cholesterol is now lower than ever, 149 at last check, in spite of adding eggs, a big dietary cholesterol source, back into my diet.

While my wife deserves most of the credit for my success, I think technology played a key role as well. Early on, I bought a home cholesterol test meter (a Bioscanner 2000, predecessor to the CardioChek that I now have). The meter allowed me to close the loop between behavior and outcome without the long delay and expense involved with a trip to the doctor. That obviously had a practical benefit, but it was also very motivating.

Interestingly, home cholesterol tests are comparatively rare, even though heart disease is the leading cause of death in the US. That might have something to do with the fact that testing involves stabbing and blood, and insurers won’t cover it. It might also have something to do with bad medical advice. For example:

First, the most readily available (and affordable) tests only measure total cholesterol. … [I measure total, HDL, and triglycerides, from which LDL can be inferred, and it’s a lot cheaper than a heart attack.]

Second, even if you get a sophisticated cholesterol test, a doctor needs to review your results … [Bah. Heart surgery is complicated, but cholesterol testing is simple. You don’t need a doctor to do it any more than you need an automotive engineer to watch your speedometer.]

Third, and perhaps most important, blood cholesterol — unlike blood pressure — doesn’t change on a day-to-day or even week-to-week basis. Doctors recommend that healthy adults get cholesterol tests every five years … [The time constants are actually fairly short – you can see triglycerides go up in response to a glass of wine on a time scale under an hour, and see a response to diet change in weeks.]

Bottom line: At-home cholesterol tests may satisfy your curiosity, but they don’t provide enough information to be truly helpful.

The implicit assumption above is that people are using home tests for screening, which probably is overkill. There’s also a hint that people are ignorant and helpless without a doctor. But the real niche for measurement is not screening, it’s control. The food pyramid and medical advice didn’t help me to discover a better way to eat; closed-loop feedback did. 5-year or even annual doctor visits are inadequate for control, and frequent visits would be far too expensive.