I was recently reading one of Castle Grok's posts where he mentions he used to eat SAD and felt very sickly all the time, but when he went to the doctor, the doc said he had a nice low trig score and all healthy metrics and was pronounced to be 'as healthy as a horse.' But he knew he was not because he felt sick all the time. Because of that, he no longer trusts doctors to be the judge of his health.
Then there is the man who ate nothing but pototoes for 60 days. The end result was the doctors felt his blood metrics were much improved: http://20potatoesaday.com/index.html Is such a diet truly healthier? If you go just by the doctor's metrics, the answer is yes.
Then there is the guy who decided to eat 2/3rds of his caloric intake in the form of twinkies! Yes, he too lost weight, his triglycerides went down, 'bad' cholesterol went down and 'good' cholesterol went up. Was this guy healthier?
How reliable are the doctor's health metrics when it comes to determining health?
asked byEva (20817)
Get FREE instant access to our Paleo For Beginners Guide & 15 FREE Recipes!
on February 19, 2011
at 12:33 PM
You raise a very interesting question that I've never experienced discussed in medical school except for superficial discussions of sensitivity and specificity. However, to my knowledge, it is something that technicians, the more diagnostic specialties, and researchers in industry often grapple with.
The punch line is that these tests only show how different you are from a group of people whom the government thinks are normal. There is something called the Harvard-MGH death: the patient dies with perfect labs.
All lab values are compared to a normal range and the person establishing the test, who usually isn't the doctor, must apply his test to a reference population. One then says that your test values are abnormal if they deviate too much from the reference population.
This creates two statistical caveats.
- For simplicity most tests assume that the disease is randomly distributed among the population. This isn't in general true. (Although it makes calculating the variance of something so easy!) For example, any disease is conditionally dependent on its causative risk factors. So, if the risk factors aren't normally distributed, the disease must not be. So sometimes a test is valid only for a certain population. Often the doctors don't bother to learn that and wrongly apply it too widely.
- Signal Detection Theory quantifies how poor an aid many tests are in medical decision making (evidenced-based medicine is notoriously light on evidence). Even if the test really indicates that you have a lab value very different from everyone else's, science has often only shown that the thing you are testing for is correlated with some disease process. This makes for good tenure-getting and dinner conversation but without a causative role (thing X is necessary and sufficient for thing Y), it can be myopic to focus on lab values.
I hope that helps, Mike
on February 19, 2011
at 02:08 PM
Interesting question, Eva. I'm not so sure that it can be done justice on this forum, but I'd like to throw out a few nuggets for consideration:
The practice of medicine aims at treating disease (including injury and illness). It is focused on understanding anatomy (the structure of body parts and systems), physiology (the functioning of body parts, systems and networks), identifying pathology of those and treating them to minimize symptoms (palliation), repair injuries (surgery, invasive procedures such as ablation, dilatation of narrowed parts, and excision and removal of foreign bodies and intrusive growths (cancer, overgrowths, etc).
Physicians receive a breadth of education in the basic sciences. Social science and humanities education is more variable, and in some cases, may be missing.
Education about health and well-being is very much a secondary focus, if covered at all.
Success is measured by the absence of disease and recovery from injury/illness. Primary care is aimed at the prevention of illness.
So you can see where this is heading.
Health and well-being and their promotion are not a focus of the practice of medicine.
However, professional nursing does address these, but only in university curricula for nurses educated at a minimum of the baccalaureate degree. Unfortunately, that's only about 1/3 of all nurses practicing in the US. Nurse practitioners and advanced practice nurses are educated at the master's level or above.
An interesting item to note is that multiple studies have shown equal or better patient outcomes and satisfaction with their care when received from nurse practitioners. My guess is that this is due in part to the focus on health and patient self-management - in other words partnering with patients instead of prescribing to them.
Where would I get reliable health-focused care?
I would use a primary care physician for illness prevention/screening and as a gatekeeper referral to these services:
A certified nurse practitioner.
A certified exercise kinesiologist.
A crowd sourcing health site, such as Patients Like Me and Cure Together.
PubMed, to read and evaluate the science.
A careful selection of research-based science and health-focused bloggers who provide primary source links and references and who make their own biases clear to readers.
Systematic reading of the health focused literature.
on February 24, 2011
at 11:43 AM
Yes, I think lipid ratios, blood pressure, and bodyfat percentage are fairly good indicators of general health.
Don't forget that a low-calorie weight loss diet is a high fat diet. ;) I think those guys were much healthier for the month they were on those ridiculous diets, because they were getting half their calories from many pounds of their own fat! I feel like hell eating like that immediately (I have done it before unfortunately), because I only have around 20 lbs of body fat and can't spare much to fuel my body in the absence of healthy food and fats.
And plenty of people have good general health even eating like crap, especially young people. Life isn't fair. I still had perfect cholesterol ratios, low blood pressure, etc when I was eating high-grain - and felt terrible! Good lipid panels, low blood pressure, and low weight run in my whole family - and we don't get heart disease and diabetes. We do have tons of autoimmune issues (most of my problems) and mental illness.
on February 19, 2011
at 03:31 PM
Sometimes you can trust labs as an indicator of health problems (A1C, PSA) and sometimes you can't (total cholesterol).
The Twinkie diet guy didn't really get 2/3 of his calories from Twinkies. He ate some pretty healthy food, along with some junk food, but kept his calories low to lose weight. http://www.fathead-movie.com/index.php/2010/11/16/the-twinkie-diet/
on February 19, 2011
at 07:18 AM
The guy who did the Twinkie diet is on Facebook and he posted his body comp test results. As I recall it (and you can go look him up), he lost something like 8 pounds of lean mass and he only got his bodyfat level to just under obese for his age, gender, and activity level. It does not take much loss of muscle in terms of space taken up in your body to register significant weight loss on the scale. He did it all wrong and he is emphatically not "healthier" as a result. One of the parameters of his weight loss that he claimed was a sign of improved health was lower cholesterol. I'm sure you've heard by now that lower total cholesterol is associated with increased all-source mortality, including deaths from accidents. Apparently, low cholesterol makes you dumber and more amenable to risk-taking. Or just more clumsy. Or some combination of all three. Hope he's got a good life insurance policy.
on February 19, 2011
at 07:05 AM
If the medical school curiculums weren't bought and paid for by the drug companies, it might be different, but unless that changes, I will be staying out of the doctor's office unless I need an antibiotic or stitches or something of that nature. It's really too bad that chronic illness is so bloody lucrative.