I'm not sure where exactly you got the idea that the 'criteria were listed'. Bascially, all I could find about the initial screening of articles was "reviewers independently evaluated titles"... To me, that sounds a little vague.
The section starting with:
"Eligibility criteria
We included published and unpublished prospective or retrospective observational studies comparing health outcomes (mortality or morbidity) in Canada and the United States for patients of any age with the same diagnosis."
followed by seven paragraphs describing the process is the part I was referring to. Further details can be obtained from the corresponding author. I can go over them with you, if you're interested. It may need a new thread, but I'm always interested in explaining the process (this is my area of research, but I recognize that my fascination may not be shared by others
Correction... there were 2 people involved at each step. Given the political stance of the lead author of this study (and the nature of this particular peer reviewed publication) I do have to question who exactly was doing the initial screening.
I did say that wrong. There were multiple people involved in evaluating each study, as different pairs of reviewers were employed at each step for any particular study.
Why are you interested in who did the initial screening. The initial step is undertaken in order to exclude all the studies that have nothing to do with this particular research. Anything that could possibly be relevant is passed on for review - i.e. it's not the point at which any real judgments are being made.
I use MRIs because they are a good example of the problems associated with an all-user-pays system (at least here in Canada)... its infrastructure that COULD be improved by allowing free market participation, but is not because of government rules. That also includes CAT scans, PET scans, and any other diagonstic tools you can think of.
I understand. But the information you gave doesn't tell us whether or not diagnoses are delayed. Diagnostic tools aren't only used for initial diagnosis, and an average waiting time does not tell us what we really want to know, since we are interested in whether the wait corresponds to the urgency. For example, while my patient who needs a routine CT waits for 3 months, my patient that needs a CT done within a day or two gets it within a day or two.
We're skeptics... we should recognize that even if we set out with the best of intentions to remove any bias its just not possible. You can't have a double blind study with people who cannot effectively do any sort of random selection.
Random sampling is not an issue since the studies represent the entire population (all studies that looked at a particular issue) not a sample.
I'm not saying that all bias can be removed. A good researcher takes into consideration sources of bias and talks about the steps taken to reduce or eliminate those sources.
Just out of curiosity... did you actually read the results in any detail?
According to their study, they found 14 studies that showed Canada's health care system was better, vs. 5 in the U.S.. Yet if you look at the studies that showed Canada's system was better:
- 5 involved kidney disease. Basically, they were measuring the same thing 5 times. Even if the Canadian system IS better at handling Renal failure, including 5 different studies is a little misleading
But that would be poor research practice. I thought we were supposed to be opposed to that.
- Multiple Studies on cystic fybrosis are also included (2), as are multiple studies (3) covering breast and prostate cancer. (Ironically this conflicts with other studies they've done that how the U.S. system is better at handling certain cancers).
So now you want to arbitrarily include information that supports your particular bias?
- They've also included a study on AIDS survival rates... however, such results likely depend more on the affordability of drugs rather than the overall health care system, so personally I think that study should be discounted
Oh good. Even more cherry-picking after the fact.
- Something else I noticed... many of the pro-U.S. studies involved Arthritis, Cataracts, and Heart disease. These are (as far as I know) more common than AIDS and Cystic Fibrosis. So, why is a study of a disease which affects few people given as much weight as a study that affects a lot of people?
Once the 'duplicates' are removed (multiple studies supporting the Canadian system) the Canadian system doesn't exactly look as good.
Lets face it, your 'article' does not so much prove the superiority of the American system, but in the ability of researchers to research articles.
I agree that the article does not prove the superiority of the American system (or the Canadian for that matter). I agree with the researchers that a reasonable conclusion is that neither system can claim hegemony in terms of quality of care. However, you demonstrated that while the researchers stuck to the principles of good research practices, you would not have been interested in doing so, had it been your choice. I find that I cannot go along with that.
Linda
