Problematic assumptions in study of costs of care

By Claudia Chaufan, M.D.
Health Affairs, Letters, June 2012

Tomas Philipson and coauthors assert that the substantially higher costs of cancer care in the United States versus Europe are “worth it” (Apr 2012). The problem of lead-time bias aside, their assertion relies on additional problematic assumptions.

One is the use of the Surveillance, Epidemiology, and End Results (SEER) database, which, as the authors admit, represented “about 14 percent of the U.S. population during the time period studied.” In contrast, the EUROCARE (European Cancer Registry on Survival and Care of Cancer Patients) databases, used for European countries, represent “national registries [including] the full population of each country” except France and Germany. Thus, the question arises: Which 14 percent of the U.S. population do the SEER data include? Are they people who have good coverage; the underinsured[1], thousands of whom go bankrupt because they cannot pay their medical bills[2]; or some of the nearly fifty million Americans who lack even modest coverage and die for that reason?[3] And how many in the latter groups might have died of cancer, as noted by the American Cancer Society[4], and thus were excluded from the authors’ study? It appears that the SEER population is better off, better educated, and more urban than the overall U.S. population, so naturally the majority of SEER registries underrepresent the U.S. cancer mortality experience.[5]

An article in a prestigious journal featuring complex statistics leads most readers to conclude that its results must be true. Yet a cursory examination of a few basic assumptions should warn against hasty conclusions.

Editor’s Note: Tomas Philipson and coauthors’ rebuttal to reader responses is available online on Health Affairs Blog: http://healthaffairs.org/blog/2012/05/14/when-epidemiology-goesastray-valuing-cancer-care-in-theunited- states-and-europe/

Claudia Chaufan, M.D., Ph.D., teaches at the University of California, San Francisco.

NOTES

1 Bodenheimer T. Underinsurance in America. N Engl J Med. 1992;327(4):274–8.

2 Himmelstein DU, Thorne D, Warren E, Woolhandler S. Medical bankruptcy in the United States, 2007: results of a national study. Am J Med. 2009;122(8):741–6.

3 Wilper AP, Woolhandler S, Lasser KE, McCormick D, Bor DH, Himmelstein DU. Health insurance and mortality in US adults. Am J Public Health. 2009;99(12):2289–95.

4 American Cancer Society. Report links health insurance status with cancer care [Internet].
Atlanta (GA): ACS; 2007 Dec 20 [cited 2012 Apr 19]. Available from: http://www.cancer.org/Cancer/news/News/report-links-healthinsurance-status-with-cancer-care

5 Merrill RM, Capocaccia R, Feuer EJ, Mariotto A. Cancer prevalence estimates based on tumour registry data in the Surveillance, Epidemiology, and End Results (SEER) Program. Int J Epidemiol. 2000;29(2):197–207.

doi: 10.1377/hlthaff.2012.0484

http://content.healthaffairs.org/content/31/6/1369.3.full.pdf+html