I'm not ready to write off the value of preventive medicine just yet. Dr. Welch's point about candidates touting things that they know little about is well taken. However, I would be curious to see how he defined preventive medicine when he did his analysis (not to mention where the numbers are to back up his claim). I have a sense that his version of preventive care involves a lot of testing, and to be fair I think that's the same version that many share. However, we need to get back to the old school definition that he highlighted, in which providers have engaging discussion with patients about their health. That face to face time will be a cornerstone of effective reform.

some simple inexpensive tests can detect a disease before it becomes chronic and expensive. For example, hypertension, heart disease, and diabetes. Data have shown that the USA has the most expensive healthcare system, but ranks near the bottom as far as quality of healthcare. One thing stands out between our system and others. We base our healthcare system on procedures while the better ones base theirs on preventive medicine. Adding blood pressure screening, cholesterol testing and glycosylated hemoglobin testing will add little to our healthcare costs, but will save down the road, tens of billions on chronic diseases.

Why do you suppose the candidates are talking about preventative medicine? Because it resonates with voters. Patients are increasingly drawn to alternative medicine practitioners because they sell the idea that mainstream medicine "only treats symptoms" and doesn't treat "the whole person". I don't think patients necessarily view preventative medicine as "more tests more often". I think they want their doctors to spend more time talking with them about how to stay healthy, and I think most patients would agree that physicians should be compensated for time spent discussing healthy lifestyle choices.

Anon at 3:07: whether or not cost-savings are important depends on the subject being discussed. If it is - which procedures to recommend and to cover by insurance, then you are right. However, the moment cost-savings claims come into play, you get policy-makers and employers trying to mandate certain preventive procedures (like annual checkups) mandatory in the mistaken belief that this will save them money. Edwards, for example, wanted to mandate annual checkups. There are some employers who tried to do the same or to even penalize employees for having LDL over 130 regardless of their weight. German government wanted to penalize people who didn't want to be screened for cancer believing it would save them money if everyone is screened. Never mind the evidence that it would not.

Hence, it is important to talk about cost savings as these misguided claims can lead to policies that interfere into the decisions that should be between a doctor and a patient.

I read an article in Atlantic Monthly in 1993 on health care that made the point that early prevention and treatment does not reduce costs. It increases cost. One reason is that you extend the lives of people. It costs more to care for the old - who develop serious health problems that costs lots of money to treat. Years ago, we paid much less for health care because pneumonia killed everyone in their 40s.

Perform randomized clinical trials and measure LDL, blood pressure, diabetes, etc. (...) I don't need to learn about epidemiology when it is well know that billions could be saved as well as thousands of lives if intervention was imposed.

Last time I checked "it is well known" wasn't exactly a scientific argument. How many things exactly that were "well known" at one time or another before they were proven to be untrue?

OK, let's take statins as an example, since you mentioned LDL. Yes, treating one high-risk person with statins is cheaper than treating the heart attack in that person (unless the heart attack is immediately fatal). What about treating 50 people for 10 years to prevent a heart attack in one person? 100 people? Based on randomized trials, how many people do you need to treat with statins and for how many years (for PRIMARY PREVENTION) to prevent one heart attack? What about primary prevention in women? What exactly randomized studies say about that?

Or maybe we should do math on some of the screening tests? Ever heard about NNS? False positives? Overdiagnosis?

Can you multiply? E.g. the cost of preventive tests/treatment multiplied by the number of people by years vs the cost of one heart attack or one advanced vs early cancer? Add to it, treating side effects, if any. Or, in case of tests, evaluating false positives and, because of overdiagnosis, treating more people.

Talking about randomized studies, can you show me a randomized study that says that annual checkups in otherwise healthy symptomless individuals save money?

Now you want to impose interventions? Just because you BELIEVE that it'll save money even if you have no evidence that it will. NO, randomized studies of any of these interventions didn't include cost saving computations. As to saving lives, while this is a valid argument for recommending an intervention, since when this is a valid argument for imposing it on adults? What if tomorrow new evidence comes out that the particular intervention did more harm than good? Or do you think this never happen? There are no biases in randomized studies, all potentially lethal side effects are known? They never later shown to be wrong?

You talk about science and randomized studies, yet in the same sentence you say that you don't need to know epidemiology. Tell me, how can you read and understand the studies results if you feel you don't need to know epidemiology? Especially since you clearly don't understand either NNT or NNS? Do you understand the difference between relative risk and absolute risk? How about cost-effectiveness and cost-savings?

Pray, somebody tell me that this guy is an exception, and that doctors today actually do learn some epidemiology....

The normal retirement age must also be indexed to life expectancy and quality of life.
Life expectancy is an interesting measure since it is an average. It is affected by fewer infants dying, it is affected by people who are diagnosed with heart desease or cancer at the age of 50 dying at 60 instead of 55.

Additionally, increased life expectancy doesn't mean that one experiences mental decline or decline in physical abilities later. Their problem-solving ability, ability to learn, memory is still declining which makes competing with younger people for jobs difficult. Same about physical strength.

But most people still die between 60 and 80. Those who survive into their 80s and 90s are mostly those with "good genes" - those whose parents also lived into 80s and 90s. Most 70-year-olds are still unable to work. Even if they are able to work, most companies don't want to hire them. Even if they are still smart or strong (depending on type of job they do), most employers still don't want them. Try getting a job at the age of 60, then talk about how retirement age shall be indexed to life expectancy.