What are we controlling for?

[Richard_Marken]

[From Rick Marken (940903.1100)]

Obviously, I got the wrong date on my previous post. It was posted on
940902, not 940828 of whatever was there.

Martin Taylor (940901 13:40) --

You might like, then, to find a good PCT-theoretic reason why among the
nations with roughly equivalent living standards, the US has by far the
most expensive health care system and one of the worst health-care
records.

Bill Leach (940902.05:57 EST) --

I might but then it would be necessary for me to believe that this
assertion is true. Given even that I believed that it is true (and I
don't), there is still the matter of why.

Givem Bill Powers' recent caveat about the quality of social data, I
suppose one has to respect this conclusion. I am going to try to check
the World Almanac today for infant mortality rates in differnt countries.
I don't have one at home so if anyone can get this data I would appreciate
it. Would the World Almanac qualify as a data source that does not have a
vested interest in the health care debate? Does infant mortality count
with you (Bill L.) as a reasonable measure of the quality of a nation's
health care? If not, why not?

Does anyone know of an unbiased, archival source of information about
per capita health care costs? I have seen such numbers but they probably
were presented by advocates of a single payer systems.

Suppose that we could get data from what we could all agree was an unbiased
source (or, better, a source that was biased in the US health care direction)
that showed that countries with single payer type health care systems payed
less per capita for health care and had lower infant mortality rates than
the US. Would this incline you (Bill L.) to at least be willing to TRY a single
payer system here? My guess is "no" because (based on what you say about it)
it seems to me that you are controlling for something other than "best
health care for all". I think you are controlling for a perception of
the government being out of it; as long as that perception is under
control (which it seems to be now) then I don't think it would matter if
you found that our infant mortality rate is comparable to that in
Zambia. It may be the shittiest health care system in the world, but
at least the govenment isn't involved.

By the way, if it turns out that our health care system is the cheapest,
with the lowest infant motality rate of any developed nation, then I'll
fight the "government takeover of health care system" right along with you.
I'm controlling for good health care for everyone at the lowest cost
possible cost. I am not controlling for the means used to achieve it. If
"free enterprise" is the best route, then great. It would be a huge surprise
if it was -- but that would be fine. It would be nice for the Republicans
to be right ONCE;-)

Best

Rick

[Bruce_Buchanan1]

Rick Marken (940903.1100) writes:

Givem Bill Powers' recent caveat about the quality of social data, I
suppose one has to respect the conclusion. [i.e. that the assertion that the

US has by far the most expensive health care system and one of the worst
health-care records, may be open to legitimate doubt.]

Does anyone know of an unbiased, archival source of information about
per capita health care costs?

Any university department of health care studies has many sources, as well
as the background knowledge required to interprete the data, which should
indeed be compared and assessed as to reliability. But I would like to
comment rather on the assumptions which may be involved in this discussion
to date.

Governments, which are accountable for obtaining sound data on which to
base their decisions (and which are the sources for such compilations as
the World Almanac in any case), are reasonably sound sources of data. In
the Western World, with university departments of health care
administration and policy think tanks second-guessing at every turn, it
would be political stupidity for any government not to try very hard to be
competent. Granted, mistakes can be made, but the consequences in terms of
ineffective or bad policy, even if not discovered by opposition critics,
are best avoided.

I do not have the precise figures in hand, but for over a decade it was
part of my job as a senior policy advisor to the Ontario Minister of Health
to arrange for briefings of U.S. government visitors - federal and state
officials and Senators and the like - who wished to meet with senior
Ontario and Canadian policy makers and well as Ministers of Health and
Social Services. I can assure you that the provision of data which would
stand up to careful scrutiny was part of the expectations. And while
accurate data does require attention, it is not easily second-guessed by
outsiders. A lot of expert resources are required for credibility.

For ten years before that I was the Ministry official responsible for
manpower planning e.g. how many doctors of various specialties to we need
to train? I can tell you that the information on which such decisions are
based is politically extremely sensitive. In Ontario we had committees and
working groups which included representatives of all the stakeholders
(doctors, medical schools, licensing bodies, public interest reps, etc.)
who oversaw and appraised the information being obtained by the information
specialists and consultants. A lot of university income monies as well as
many professional careers are at stake. It does remain a fact that the key
information for decision-making is seldom readily available and must be
obtained in timely ways by special inquiry.

My memory is that in Ontario health care costs in the early '90's were
somewhere about 8% of the gross provincial product (GPP) whereas in the
U.S. the costs were 12% of GNP or more. The exact amounts are well studied
and known to the parties most concerned. On many indices of population
health (e.g. infant mortality, overall mortality) the U.S. record is indeed
surprisingly poor by developed world standards. (I think Scientific
American and well as The Economist have had authoritative articles on these
topics.)

So much for content of this debate. Now for something on PCT and process!

Rick says further:

I'm controlling for good health care for everyone at the lowest
possible cost. I am not controlling for the means used to achieve it.

The trouble with most general statements of this kind is that they are
sufficiently imprecise that no one can resonably disagree. However, in the
event and when the decisions must be made, people who think they share the
same general principles very often disagree. This must mean, I think, that
a genuine (rather than merely supposed) controlling variable must have
clear operational impact. The meaning will come out in the trade-offs -
e.g.are routine physicals for employment purposes to be included as part of
"good health care" to be funded by the public purse? - and innumerable such
questions. Who is included in "everyone", really? Is 94% of the population
enough? Where does one draw the line in deciding upon the lowest possible
costs? Indeed along what kinds of continua or ranges of services will
lines be drawn? Should the key decisions be made by those most affected?
And who are they?

In addition, while it is hardly conceivable that the ends do not control
for at least some possible means which may be used there are undoubtedly
many constraints on means that will appear to those who are asked to
undertake the project.

The point of such questions here is to illustrate the need to recognize
that vaguely stated general principles are just that. They may lead to more
useful formulations in terms of controlling variables, but they have a long
way to go.
The situation may be analogous to the history of Management Science, where
operational research methods were preceded for many decades by statements
of general managment principles. And there is always the Book of Proverbs.
The trouble is that general principles can be found to contradict almost
any proposal.

That is surely enough for a while!

Cheers!

Bruce B.

[Bill_Leach1]

<[Bill Leach 940904.10:37 EST(EDT)]

[Rick Marken (940903.1100)]

Does anyone know of an unbiased, archival source of information about
per capita health care costs? I have seen such numbers but they probably
were presented by advocates of a single payer systems.

No, I don't think that there really is such a thing, but at least this is
a useful approach.

Infant mortality...

I am not sure why this particular figure is THE measure of health system
care quality. I question this not to be contrary but rather because I
really don't seen where the measure of this one parameter is necessarily
dependent upon "the quality of the health care system" in a manner that
all (most) other measurements would be. That is, are there factors that
influence this measure that are independent of both "the quality of
health care" and other standards by which health care can be measured?

Selection of this particular parameter also bothers me a little for two
other reasons. The first is that I am lead to believe that SIDS is a
"significant" factor in infant mortality and it is presently a complete
mystery. Indications are however, that it is probably not affected by
"health care quality". If the SIDS rate is affected by some other
factors then differences in infant mortality rates could not be used to
judge health care quality.

Second reason should have no real basis for influence here.

it seems to me that you are controlling for something other than "best
health care for all". I think you are controlling for a perception of
the government being out of it; as long as that perception is under
control (which it seems to be now) then I don't think it would matter if

I am not sure what to say to this. It appears to me to be a "personal
attack" veiled in PCT terminology. Or maybe just an attempt to "push my
buttons". I suppose that this also could have been written just due to
some high level of frustration at perceiving that you have not changed my
perceptions. I believe that I "understand" you well enough that I
believe that you do not, at least consciously, control for "control of
others". As to the last matter, I doubt that you can have any idea of
how much you have influenced my thinking on such subjects as we have been
trying to discuss.

A serious problem with any such discussions (that Bill pointed out rather
nicely) is that all such "complex" matters are not binary in nature (that
is there is not an absolute right and an absolute wrong).

I also sense, rightly or wrongly a strong bias in your postings. I am
one of those that believes that in any compromise between "True and
False", True looses. However, I also believe that "True" seldom is! I
tend to agree that in, at least most, disagreements there is "truth" in
both (or many) sides.

My initial enthusiam for Rand waned as I began to recognize a measure of
"truth" in some of the objections to her philosophy. Bill P. mentioned
"the rational man" concept upon which her entire ediface was constructed
and pointed out some of the fallacy. However, it seems to me that any
even nominally democratic system has to suffer the same problem.

-bill