Directly measuring quality is very difficult, and it wouldn't surprise me if not a single one of those 206 charts directly measured quality. I'm not sure there is even one example of a way to measure it, that could actually be reliably measured at a reasonable cost and difficulty for a whole nations health care. Individual doctors, or even hospitals might be rated with some measure of reliability, but to measure all health care quality in the country you would have to make almost as big of effort as you do to actually provide the health care in the first place. Well I suppose measuring 10 percent of the health care chose at random could in theory be affordable, and would map well to the quality of the system as a whole, but no one is going to want to pay for that much expert observation and analysis.
Measuring proxies for quality is still hard but doable. A few things it addressed, like waiting times*, are proxies for equality. Not necessarily great proxies, but they do tell you something. Another it measured was survival rates where the US did well. Again not really a measure of health care quality, but a useful enough proxy to include as long as its realized that it isn't a direct measure of health care quality. One measure might be to try to take life expectancy (or some form of quality adjusted life expectancy) and then adjust for a huge number of factors, but it gets pretty hard to be sure your accurate on all those adjustments, or to be sure your not missing some important adjustment. Adjusting life expectancy for murder rates and fatal vehicle accidents puts the US in number one by some measurements, but I'm not going to claim its the perfect way to measure things just because it supports what I think.
Most of the charts there are about health outcomes. Health outcomes are rather distant and tenuous proxies for health care quality. Edit - Actually a lot are also gross inputs. MRI's or doctors per person etc. Again not really a measurement of quality.
Continuing on to the 2nd page at your link
"Quality Indicators in Select OECD Countries, 2009" is actually about outcomes not quality. Perhaps outcomes with a closer connection to quality then randomly selected outcomes, but its not a measurement of quality, certainly not directly.
"Supply and Utilization of Doctors and Hospital in Select OECD Countries, 2009" - Well now its not about outcomes, but its about inputs in gross terms, not a measure of the quality of those inputs. I guess to get quality you have to have a certain level of gross inputs, but it seems that all of the countries listed have a sufficient level here. There doesn't seem to be any correlation between these measures of inputs and any attempt to measure quality that I've seen.
"Volume of Knee and Hip Replacements, 2009" - Measuring how many procedures are done, not how well they are done or how many are done relative to need. Also even if you where to take this as a measure of quality, it would be a confusion one. For example the US is 2nd in knee replacements but only a bit above average in hip replacements. If your going to call it quality that's still a noticeably above average result for the US, but I'm not going to be dishonest to support my opinion. I care more about the truth, then potentially seeming to have won some argument on some internet discussion forum.
"Magnetic Resonance Imaging (MRI) Units per Million Population, 2009"
Another gross input measurement. Again not totally unrelated to quality, you need resources to get good results, and extra resources can decrease waiting times even if they don't produce better end results. But still its not a direct measurement of quality or a good indirect measurement except perhaps if some of the listed countries had so few resources in these areas as to be very problematic. The US is number two in this measurement at 25.9 units per million population. The UK brings up the rear at 5.6 per million. The larger number in the US is likely to reduce American waiting times compared to the UK, but one could easily question if spending that much more on these units is cost effective.
"Number of Practicing Physicians per 1,000 Population, 2009"
Another gross input measure. The US is 2nd to last, but then Japan, with some very high life expectancies and often considered a quality health care system is last.
"Rated Quality of Care in Past Year as Excellent or Very Good by Medical Home"
Well that is not directly a measure of quality, its a measure of opinions of quality. Probably more relation to actual quality then most of the other measurements, but expectations can differ, problems with cost or billing could cause someone to rate high quality care as low quality, etc. Still better than most of the data at your link in terms of measuring quality, but that's not saying much. The UK scores number one in both measurements here, but I'd hardly say it has the highest quality. The US scores third in one measurement and middle of the pack in the other.
"Patient Engagement in Care Management for Chronic Condition by Medical Home"
I'm not sure what they mean by "Medical Home", perhaps what would commonly be called a nursing home in the US, but I'm not sure. Engagement might be a positive, and might contribute to quality, but wouldn't directly be a measure of quality (or even of results). US is third for all that it matters.
"Medical Medication or Lab Test Errors in Past Two Years by Medical Home"
This might be a decent if limited measure of quality. Although different systems might define or measure error differently. The figures are for reported medical mistakes, but if what gets counted as a medical mistake and/or what gets reported as a medical mistake differs based on country then this data could be useless. For what its worth the US is below average, but New Zealand and Norway are worse and Canada almost as bad in this measurement. A little more clustering here than in the other measurements. The UK does the best which doesn't fit with what I've heard about there system.
I went up to page 10, selecting some that may be halfway decent proxies for quality (no really great ones). I'll look in to them in my next reply. |
