FOUR YEARS AGO, in the presidential campaign, in major medical journals and medical associations, in the media, and in civic groups, our nation was engaged in a great debate about the best way to provide health insurance coverage to all Americans. By contrast, these debates have been conspicuous spicuous their absence in this election year, The health system reform debate was marked by some controversy about whether there was a health insurance crisis at all.(1) On one side were those who said that most of the uninsured could get care when they needed it, a view that was expressed in a commentary in the Wall Street Journal in 1994 that noted that ''these [uninsured] citizens are not denied health care'' and only 1 in 5 uninsured (about 3% of the population) cannot obtain affordable insurance.(2) On the other side were those who claimed that many of the uninsured faced major barriers to needed health care serviccs and experienced health and economic consequences because of these barriers While the political urgency of these concerns has subsided, we know that gaps in insurance coverage remain for a sizable number of Americans-recent estimates indicate that approximately 40 million people in the United States were uninsured in 1994 (oral communication, B. Katherine Swartz, PhD, April 23, 1996, regarding unpublished data from analyses of March 1995 Current Population Survey); another 29 million are underinsured.(3) Reports of public opinion surveys hare noted that support for reform was driven in large part by anxieties about access to and cost of medical care among both insured and uninsured persons.(1-6) Prior research has documented decreased. access to health care services, and increased burdens of economic hardship, ill health, and mortality that the uninsured and underinsured experience.(7-9). In them midst of the heal system reform debate, media reports about this research were frequently augmented with vignettes illustrating the consequences of access and financial problems for individuals and families, While vignettes can add a valuable qualitative dimension to quantitative estimated of problem, they are not randomly selected and may illustrate extreme, rather than average, consequences. Verbatim responses are common in the development of sample surveys, but in larger-scale surveys, structured and coded responses are preferred to minimize cost and increase analytic power. Now that the politically charged health system reform debate has subsided, we return to the basic questions that marked that discussion. What problems do uninsured and insured people have in getting and paying for medical care in today's changing health care system? How man Americans are affected? How severe are their problems and their consequences? We sought to answer these questions by using a combination of quantitative measures of access to health care and personal qualitative accounts in a recent survey of Americans to determine who has experienced problems getting medical care and paying for it.
