Miranda Mugford, "A Comparison of Reported
coverage necessarily the best or most efficient use
Differences in Definitions of Vital Events and
of resources for improving health. Helen Levy and
Statistics," World Health Statistics Quarterly 36
David Meltzer, "What Do We Really Know about
(1983), cited in Nicholas Eberstadt, The Tyranny of
Whether Health Insurance Affects Health,"
Numbers: Measurements and Misrule (Washington:
Economic Research Initiative on the Uninsured
AEI Press, 1995), p. 50. Some, but not all, coun-
Working Paper no. 6, December 2001. Moreover,
tries are beginning to standardize figures, and
in many cases, expanding insurance coverage will
future data may be more reliable.
exacerbate the problems of third-party payment.
25. Anthony DePalma, "SiCKO, Castro, and the
14. Cited in Tamar Nordenberg, "Make No Mis-
120 Year Club," New York Times, May 27, 2007.
take: Medical Errors Can Be Serious," FDA Con-
sumer Magazine 34, no. 5 (2000).
26. Arduino Verdecchia et al., "Recent Cancer Survival
in Europe: A 200002 Period Analysis of EUROCARE-
15. Elizabeth McGlynn et al., "The Quality of
4 Data," The Lancet Oncology 8, no. 9 (2007): 78496,
Health Care Delivered to Adults in the United
http://www.thelancet.com/journals/lanonc/article/
States," New England Journal of Medicine 348 (2003):
PIIS1470204507702462/abstract; Nicole Martin, "UK
263545.
Cancer Survival Rate Lowest in Europe," Daily
Telegraph, August 24, 2007. Of course, some argue that
16. Rita Mangione-Smith et al., "The Quality of
these figures are skewed by aggressive U.S. testing and
Ambulatory Care Delivered to Children in the
diagnostic procedures. In the United States, doctors
United States," New England Journal of Medicine
catch many cancers that would go undetected in other
357 (2007): 151523.
countries. These cancers may be small or slow growing
and might not kill the person. That these cancers are
17. See, for example, http://www.michaelmoore.
diagnosed in the United States but not in other coun-
com/sicko/health-care-proposal/ and http://www.
tries makes our survival rate look higher. Jonathan
pnhp.org.
Cohn, "What Jacques Chirac Could Teach Us about
Health Care," New Republic, April 10, 2007. That theo-
18. World Health Organization, "The World Health
ry is worth considering; increased screening likely does
Report 2000" (Geneva: WHO 2000); SiCKO, Dog Eat
affect the figures for slow growing cancers such as
Dog Films, 2007. For a critical look at the WHO
prostate cancer (the source of much controversy since
study, see Glen Whitman, "WHO Do You Think
Rudy Giuliani raised the issue in his campaign). "Rudy
You're Fooling: The World Health Organization's
Wrong on Cancer Survival Chances," Washington Post,
Problematic Ranking of Health Care Systems," Cato
October 31, 2007; David Gratzer, "Rudy Is Right in
Institute Briefing Paper, February 28, 2008; Twila
Data Duel about Cancer," Investors Business Daily,
Brase, "WHO's Hidden Agenda," Ideas on Liberty,
November 6, 2007.
Foundation for Economic Education, October 2000.
As Robert Ohsfeldt and John Schneider con-
cede in their book, The Business of Health, "[Many]
19. World Health Organization.
cancer survival rate estimates . . . do not adjust for
cancer stage at diagnosis. This could result in sur-
20. Ibid.
vivor time bias--those with cancers detected at an
earlier stage would exhibit longer post diagnosis
21. OECD, "Health at a Glance: OECD Indicators,
survival times, even for cancers that are essentially
2005" (Paris: OECD Publishing, 2005), p.11.
untreatable." Robert Ohsfeldt and John Schneider,
The Business of Health (Washington: AEI, 2007), pp.
22. Census Bureau, 2000 Census (Washington: U.S.
2324. However, survivor time bias is not as big an
Census Bureau).
issue for cancers that have faster metastasizing
times or that strike younger patients. As Ohsfeldt
23. Robert L. Ohsfeldt and John E. Schneider, The
and Schneider go on to note:
Business of Health: The Role of Competition, Markets,
and Regulation (Washington: American Enterprise
Survivor time bias, however, should not be
Institute Press, 2006).
a significant concern for cancers that
respond well to treatment if detected early.
24. In Austria and Germany, fetal weight must be
For such cancers, early detection makes a
at least 500 grams (1 pound) to count as a live
substantive contribution to survival time--
birth; in other parts of Europe, such as
the longer survival time associated with
Switzerland, the fetus must be at least 30 cen-
early detection thus is not a spurious effect
timeters (12 inches) long. In Belgium and France,
of early detection. An example is thyroid
births at less than 26 weeks gestation are regis-
cancer. In the United States, virtually all
tered as lifeless. And some countries don't reliably
females with thyroid cancer survive for at
register babies who die within the first 24 hours
least five years. The lower survival rates for
after birth. For a full discussion of the issue, see
37