Journal of General Internal Medicine

, Volume 28, Issue 9, pp 1122–1124 | Cite as

Life or Debt: Underinsurance in America

  • Steffie Woolhandler
  • David U. HimmelsteinEmail author

Life or debt.

Millions of our patients face that choice, including many with insurance.

Health reform has focused on America’s 50 million uninsured. But the predicament of the underinsured is also dire, and they will find less solace in the Affordable Care Act (ACA).

In this issue of JGIM, Magge et al.1 cast welcome light on the plight of insured, low-income (0–125 % of poverty) families. More than a third of them met criteria for “underinsurance”; 31.5 % devoted more than 5 % of their meager incomes to medical expenses, while many skipped or delayed needed care or medications because of costs.

Not surprisingly, Medicaid enrollees fared somewhat better than those with private coverage. Medicaid has generally been more comprehensive than private insurance, with minimal cost-sharing. However, Medicaid’s low fees have caused many physicians and hospitals to shun Medicaid, compromising enrollees’ ability to get appointments—a problem that wouldn’t show up in Magge’s analysis.

While among low-income insured individuals whites were at higher risk of underinsurance, a much higher share of all Blacks and Hispanics are uninsured or low-income. Hence, the low-income uninsured and underinsured account for a larger proportion of the total Black and Hispanic populations.

Magge’s research extends previous findings indicating a steady erosion of the financial protection offered by health insurance. Farley’s analysis of the 1977 National Medical Expenditure Survey (NMES) found that 12.6 % of individuals with private coverage had a 1 % annual probability of incurring out-of-pocket medical expenses exceeding 10 % of family income (one of several alternative definitions of underinsurance that she explored).2 Using this same definition, underinsurance had increased to 29 million persons, 18.5 % of those with private coverage by 1994.3

The NMES’ successor—the Medical Expenditure Panel Survey (MEPS)—has not released the insurance benefit schedules needed to replicate Farley’s definition. But more recent studies indicate that the ranks of the underinsured continue to grow.

Between 1996 and 2003, among individuals with employer-based coverage, the share with health expenditures (including premiums) exceeding 10 % of family income increased from 14.2 % to 18.2 %.4 The burden was especially heavy on the poor (among whom 33.3 % spent > 10 % of income); on those in fair or poor health (32.3 %); and on those with chronic conditions such as diabetes (39.1 %), hypertension (30.9 %) or a mental disorder (29.2 %).4

Using an alternative definition—inflation-adjusted out-of-pocket spending > $5,000 (excluding premiums)—underinsurance among households headed by a working-age adult with full-year coverage increased from 2.6 % to 4.5 % between 1999 and 2006. Among households that included someone with a hospitalization, underinsurance rose from 7.2 % to 11.6 %.5

A series of surveys of non-elderly adults by the Commonwealth Fund estimated underinsurance at 9 % in 2003, increasing to 16 % in 2010;6 the proportion spending > 10 % of income on out-of-pocket costs and premiums rose from 21 % in 2001 to 32 % in 2010.

Striking evidence of widespread underinsurance also comes from the bankruptcy courts. Nearly 1.2 million families seek bankruptcy protection annually; medical bills or illness contributed to 62 % of filings in 2007—a 49.6 % increase since 2001.7 Sixty percent of the medically bankrupt had private coverage at the onset of the bankrupting illness; only 22 % were uninsured.7

Several studies have shown that skimpy insurance menaces more than just financial health. In the Rand Health Insurance Experiment, the only randomized trial of cost sharing, high deductibles didn’t harm affluent, healthy patients, but increased the risk of dying by 21 % among lower income, sicker participants.8 That study almost certainly understates the hazard of underinsurance, because it excluded the poorest and sickest individuals (i.e. those most likely to be harmed). Moreover, it predated widespread adoption of several life-prolonging therapies such as beta-blockers, ACE inhibitors, and statins, whose use is decreased by copayments.

In a large national survey in 2007, 29 % of individuals with high-deductible plans vs. 16 % with low deductibles reported delaying or avoiding care due to cost.9 Disturbingly, in a study of patients hospitalized with acute myocardial infarction, underinsurance predicted pre-hospital delays (OR 1.21 compared to the well-insured).10

Many hope that the ACA will fix both uninsurance and underinsurance. Once fully implemented, it will expand coverage by about 26 million, eliminate lifetime benefit caps which have ensnared a few thousand families annually, and ban pre-existing condition exclusions.

But, paradoxically, the ACA may actually increase the number of underinsured. About 40 % of those gaining coverage will get Medicaid. As Magge shows, many current Medicaid enrollees are woefully underinsured. Moreover, CMS looks set to allow state Medicaid programs to demand copayments and deductibles, even from the poorest of the poor. Several states have already reduced benefits, cut provider payments, and narrowed provider networks.11 Hence, underinsurance among Medicaid recipients will probably increase. More ominously, the White House is encouraging state officials to use federal Medicaid expansion funds to purchase private insurance,12 a shift likely to raise both taxpayers’ costs and poor patients’ copayments.

The new private coverage offered to near-poor and middle income individuals through insurance exchanges will also leave many underinsured. Bronze plans—the minimum coverage mandated by the ACA—will cover only 60 % of average medical expenses; silver plans will cover 70 %. That’s far worse than the roughly 80 % coverage under today’s average job-based policy—equivalent to the ACA’s Gold plans. (A complex system of sliding-scale discounts on copays and deductibles available to some of those with incomes 138 %–250 % of poverty will offset some, but not all, of the near-poor’s cost-sharing.)

In concrete terms, a 56-year-old making $45,900 (399 % of poverty, and hence eligible for premium subsidies) will pay an estimated $4,361 in premiums for individual Bronze coverage, and up to $4,167 in additional deductibles and copayments for covered services.13 At 401 % of poverty ($46,100) subsidies disappear; the mandatory premium would be $10,585, with out-of-pocket costs for covered services capped at $6,250. In effect, the federal government has lent its imprimatur to skimpy plans (long-promoted by private insurers) that offer scant protection from pauperization.

Little wonder that expanded coverage under the Massachusetts reform (where Medicaid has remained comprehensive, and the Bronze plans’ actuarial value is 70 % vs. the ACA’s 60 %) yielded no reduction in medical bankruptcies.14

Unfortunately, both Massachusetts and the ACA eschewed the social insurance approach which makes care free at the time of use, puts the burden of health costs on those most able to pay—the healthy and wealthy—and relies on readily enforced global budgets for cost control. Instead, they embraced market-based policies that demand far more (percentage wise) from the middle class than the rich, and compound the misfortune of illness with financial penalties. Such policies conflate patients seeking care with price-sensitive consumers whose voracious appetites for excessive services must be curbed.

International evidence indicates that cost-sharing is neither necessary nor particularly effective for cost control; the U.S. has high cost-sharing and the highest costs. Canada, which outlawed copayments and deductibles in 1981, has seen both faster health improvement and slower cost growth.15 Canadian provinces control costs by tax-based funding; global hospital budgeting; binding, negotiated physician fee schedules; and a simple unified single-payer structure that minimizes administrative burdens and costs. Scotland, which has eschewed market-based policies and patient payments—even going so far as to abolish parking fees—has costs about half those in the U.S. Scots view patients as owners of their health care system, not its customers.

Magge’s sobering data remind us that wish-it-would-work health reforms such as the ACA won’t end the unnecessary suffering that fragmented, market-oriented health financing inflicts on patients. Only thoroughgoing, evidence-based reform will do that.


  1. 1.
    Magge H, Cabral HJ, Kazis LE, Sommers BD. Prevalence and predictors of underinsurance among low-income adults. J Gen Intern Med. 2013. doi:  10.1007/s11606-013-2354-z.
  2. 2.
    Farley PJ. Who are the underinsured? Milbank Q. 1985;63:476–503.Google Scholar
  3. 3.
    Farley PJ, Short P, Banthin JS. New Estimates of the underinsured younger than 65 years. JAMA. 1995;274:1302–1306.Google Scholar
  4. 4.
    Banthin JS, Bernard DM. Changes in financial burden for health care: national estimates for the population younger than 65 years, 1996 to 2003. JAMA. 2006;296:2712–2719.PubMedCrossRefGoogle Scholar
  5. 5.
    Bernstein D. Statistical note #1. Changes in out-of-pocket expenses by insurance category (1999 to 2006). Social Science Research Network, 2008. Available at: (Accessed March 5, 2013).
  6. 6.
    Collins SR, Doty MM, Robertson R, Garber T. Help on the horizon: Findings from the Commonwealth Fund Biennial Health Insurance Survey of 2010. The Commonwealth Fund, Washington, DC, March 2011. Available at: (accessed March 5, 2013).
  7. 7.
    Himmelstein DU, Thorne D, Warren E, Woolhandler S. Medical bankruptcy in the United States, 2007: results of a national study. Am J Med. 2009;122:741–746.PubMedCrossRefGoogle Scholar
  8. 8.
    Brook RH, Ware JE Jr, Rogers WH, et al. Does free care improve adults’ health? — Results from a randomized controlled trial. N Engl J Med. 1983;309:1426–1434.PubMedCrossRefGoogle Scholar
  9. 9.
    Fronstin P, Collins SR. Findings from the 2007 EBRI/Commonwealth Fund Consumerism in Health Survey. EBRI Issue Brief No. 315, March, 2008. Available at: (accessed March 5, 2013).
  10. 10.
    Smolderen KG, Spertus JA, Nallamothu BK, et al. Health care insurance, financial concerns in accessing care, and delays to hospital presentation in acute myocardial infarction. JAMA. 2010;303:1392–1400.PubMedCrossRefGoogle Scholar
  11. 11.
    Galewitz P, Fleming M. 13 states aim to limit Medicaid. USA Today July 24, 2012. Available at: (accessed March 9, 2013).
  12. 12.
    Pear R. Expanding Medicaid with private insurance. New York Times March 22, 2013:A14–A15.Google Scholar
  13. 13.
    The Henry J. Kaiser Family Foundation. Health reform subsidy calculator. Available at: (accessed March 7, 2013).
  14. 14.
    Himmelstein DU, Thorne D, Woolhandler S. Medical bankruptcy in Massachusetts: has health reform made a difference? Am J Med. 2011;124:224–228.PubMedCrossRefGoogle Scholar
  15. 15.
    Himmelstein DU, Woolhandler S. Cost control in a parallel universe: medicare spending in the United States and Canada. Arch Intern Med. 2012;172:1764–1766.PubMedGoogle Scholar

Copyright information

© Society of General Internal Medicine 2013

Authors and Affiliations

  1. 1.The City University of New York School of Public Health at Hunter CollegeNew YorkUSA

Personalised recommendations