Sarah Palin: bat****, or howling-at-the-moon bat****?

[Straybow]

****, it even has its own Wikipedia entry

http://en.wikipedia.org/wiki/Lead_time_bias

Suppose you are correct in positing that lead time bias is at work, can you demonstrate that lead time bias is responsible for all the difference in death rate?

Note also, lead time bias can only change the statistics for the period of time the patient is still alive. Once the patient dies, the lead time no longer matters. Unless the study is conducted over a period of time that is not long compared to the lead time differential, there is little effect on the result.

[KrazyHorse]

Suppose you are correct in positing that lead time bias is at work, can you demonstrate that lead time bias is responsible for all the difference in death rate?

I do not claim that lead time bias is the SOLE factor at work. But for any disease for which there are large differences in screening rates (as there are between the US as other OECD countries) it is COMPLETELY UNREASONABLE to post the raw "survival" figures without noting it. In fact, I DO believe that the US does a better job of treating prostate cancer than do other First World countries. As I've previously posted, this paper:

http://repository.upenn.edu/cgi/view...working_papers shows a significant difference in the paths of mortality rates for prostate cancer between the US and the rest of the OECD. Prior to 1990, the US and the rest of the OECD had similar mortality rates from prostate cancer (with the US being ~5% higher). Since then, there have been significant advances in the treatment of prostate cancer. Nowadays, the mortality rate (deaths per population) of the US from prostate cancer is ~20% lower than in the rest of the OECD. It's reasonable to assign this change to more aggressive screening and treatment in the US, and this is a REAL CHANGE, not simply a statistical artifact. The **** statistic you keep posting is not.

Note also, lead time bias can only change the statistics for the period of time the patient is still alive. Once the patient dies, the lead time no longer matters. Unless the study is conducted over a period of time that is not long compared to the lead time differential, there is little effect on the result.

I have no idea what the **** this is supposed to mean.

[notyoueither]

NYE, the point is a bit subtle but important.

When a business purchases its employee's health insurance for him then that health insurance has some value to the employee. When a business (or employee) pays the taxes that arise from his salary, the employee sees absolutely no value from that. If the employee values his health insurance at the same price that the business paid for it (which is at least true to a first approximation) then there is no "wedge" between the the two; it's just compensation in a different form. When the employee values the taxes paid at 0 then there is a tax "wedge"; the business is paying more to hire the employee than the employee is seeing in total compensation.

Would you think that wages have to increase due to this perception of lack of value? Is there any data to suggest they do?

[notyoueither]

It would be interesting as an aside to see the differences in wages/salaries between US and Canada for companies with significant operations on both sides of the border.

[KrazyHorse]

Would you think that wages have to increase due to this perception of lack of value?

It's not a perception; it's a truth. If I personally pay the government 1000$ extra in taxes then the value to me personally is minuscule (since my 1000$ extra is diluted among the entire population).

Is there any data to suggest they do?

This is literally what the entire field of tax incidence is based around, NYE.

[KrazyHorse]

It would be interesting as an aside to see the differences in wages/salaries between US and Canada for companies with significant operations on both sides of the border.

While you might find such a comparison interesting it would have little to nothing to do with the issue at hand.

[notyoueither]

Yes, but our 'extra taxes' pay for our entire system (or most of it). In the US you are still paying taxes for government healthcare (roughly GDP per person equal to our entire system) as well as paying for your own insurance for the rest of the system.

I'm not buying that Canada's system is an export killer, while the US system is not. Quite the reverse is my intuition, since many companies locate here for our educated, medicated workforce, and they do not have the hassles of the American system.

Is this where I ?

[KrazyHorse]

NYE, what I'm laughing at is your apparent contention that comparing wages between two different countries would tell you anything about the effect of taxes on wages, ceteris parebus. There are a lot more differences between the two countries than a couple of percentage points in the overall tax burden.

I never said that Canadian medicare -> export killer. As you then went on to point out, the American Medicare + Medicaid programs are basically as expensive per person as is the entire Canadian system, and these are basically paid out of taxes. Add in a few more things like defense spending etc. and Americans are just about as profligate with gov't expenditure as we are.

[notyoueither]

OK. I was not contending 'that comparing wages between two different countries would tell you anything about the effect of taxes on wages.'

I said as an aside, I would be interested in data on that subject.

I'd also be interested in data that taxes put pressure on wages that healthcare premiums, employment insurance premiums, CPP or Social Security premiums, etc, do not.

[KrazyHorse]

CPP and SS premiums DO put upward pressure on wages, but the problem is not as extreme as for taxes going to health care.

The reason is that SS and CPP contributions lead to a benefit which is related at least partially to the contributions. So, the more CPP I pay during my life the higher my retirement benefits are. This has some value to me, so I include it in my calculation of my total compensation.

[notyoueither]

Hmmm.

I get what you're saying, I'm just not buying in yet.

Can you direct to a good source that supports what you are claiming?

[KrazyHorse]

Like I said, the keyword is "tax incidence". I can't think of a single source that presents a bunch of basic material in the same place, so if you're interested I would suggest googling about for a bit.

[Straybow]

Note also, lead time bias can only change the statistics for the period of time the patient is still alive. Once the patient dies, the lead time no longer matters. Unless the study is conducted over a period of time that is not long compared to the lead time differential, there is little effect on the result.

I have no idea what the **** this is supposed to mean.

Maybe you should try to figure it out. It is written in Canadian-compatible English, with well-formed sentence structure, a premise, supporting content, and a conclusion.

[KrazyHorse]

The problem is semantic, not syntactic

[KrazyHorse]

Finally hunted down what that statistic is. It's the mortality rate divided by the incidence rate. Crudely, it can be thought of as one over the "average" survival time with the disease from diagnosis to death (there are a couple of problems with this view, but it's not terrible as a heuristic).

You're a ****ing idiot, Straybow.



The "length of the study" has absolutely nothing to do with anything. Lead time bias is hugely important to this statistic. And saying that "the death rate for prostate cancer in the US is 19% while in the UK it's 57%" is simply idiotic.

****.