`Unnecessary Medical Care' Hoaxsters
Shoot Selves in the Foot
by Ned Rosinsky, M.D.
[PDF version of this article, including graphs]
Oct. 6—Leading staff from Dr. John Wennberg's Dartmouth Institute for Health Policy and Clinical Practice, the source of the fraud circulated for the past six months by President Obama, that nearly one-third of Medicare expenditures is unnecessary medical care, have essentially admitted their lies. This pulls the rug out from under the argument made by all the "expert" proponents of the Obama "reform," namely, that approximately 30% of medical expenses can be cut, if "overuse" Medicare payment in certain areas of the country, is cut back to the rate in other regions.
The admission appeared in The New England Journal of Medicine of Sept. 24, 2009, in an article entitled "Getting Past Denial: The High Cost of Health Care in the United States," authored by Dartmouth researchers including Elliott Fisher, MD.
While the study reported on in the article nowhere admits that the methodology used is fraudulent, its new data document that allegedly unjustified regional variations in health-care spending total 9.5% of costs, not 30%, as previously claimed.
A Closer Look at the Graph
The fraud used by the Dartmouth Group to reassert their argument that "regional disparities" (they mean overspending) account for the major differences between the highest and lowest per-capita areas, in terms of Medicare spending, appears in a bar diagram purporting to show the "proportion of higher regional Medicare spending attributable to differences in race, income, health factors, and regional factors." To devise the graph, the authors divided the study geographic areas into five equally populated quintiles, and arranged them by annual per-capita Medicare spending. The chart gives the visual impression that the highest quintile is many times the height of the lowest, and the highest quintile bar is mostly marked as unnecessary health-care spending. But, on closer inspection, the chart does not show all five quintiles, but only four (numbers 2-5), and depicts only the differences in spending above the amount in the first, missing quintile.
Why was the first quintile not shown? Possibly because, when it is included, it totally changes the visual effect, demonstrating that regional differences amount to only 9.5% of the total cost, not the much higher percentage which the study's graph implies. The text of the article focuses, not on the total spending differences among the quintiles, but only on the supposedly small relative size of the differences due to poverty and disease severity.
You can do the calculation, by looking at the bar heights.
The four bars shown are approximately $500, $800, $1,500, and $3,300. If the highest bar is 50% of the height of the hidden first quintile (which the authors state), then the first quintile must be $6,600. Thus the actual heights of the five quintiles are $6,600, $7,100, $7,400, $8,100, and $9,900, and the total of the five bars, to be used in the calculation below, is $39,100. Each of the four bars shown in the figure is subdivided into the portion of the cost due to race, health factors, income, and a leftover segment termed "regional factors." It is the regional factors that the article says represent "unnecessary" health care.
By visual inspection, the regional factors are at most $100, $500, $800 and $2,300, totaling $3,700. These numbers can be used to find the overall proportion of total costs that the authors think is unnecessary. Thus, the study finds that $3,700 out of a total of $39,100 is based on regional variation that is not accounted for by disease severity or patient income. This is 9.5% of the total cost.
EIR will be publishing a further review of the Dartmouth fraud in a forthcoming issue.
