The Affordable Care Act (ACA) was. The Share Of People With High Medical Costs Increased Prior To Implementation Of The Affordable Care Act

Transcription

1 By Peter J. Cunningham The Share Of People With High Medical Costs Increased Prior To Implementation Of The Affordable Care Act Health reform is in part a response to steady increases in the number of uninsured people, rising health care costs, and stagnant family incomes. Between 2001 and 2006 the percentage of nonelderly Americans living in families with a high burden of out-of-pocket medical expenses those spending more than 10 percent of their family income on health care increased from 14.4 percent to 19.2 percent. 2,3 This increase was driven both by a 33 percent increase in spending on premiums (after inflation) and by an 18 percent increase in out-of-pocket spending on services which included copayments, deductibles, and noncovered services. The increases in the costs of health insurance and health care have moderated in recent years. 4 However, many families without access to employer-sponsored coverage continued to face a choice between being uninsured and having greatly reduced access to medical care or enrolling in coverage that puts a strain on family fidoi: /hlthaff HEALTH AFFAIRS 34, NO. 1 (2015): Project HOPE The People-to-People Health Foundation, Inc. ABSTRACT The percentage of Americans with high medical cost burdens those who spend more than 10 percent of their family income on out-of-pocket expenses for health care increased to 19.2 percent in 2011, after having stabilized at 18.2 percent during the Great Recession of The increase was driven primarily by growth in premium expenses in for people with employer-sponsored coverage. Out-of-pocket spending on health services, especially for prescription drugs, continued to decrease between and Medical cost burdens were highest for income groups most likely to benefit from the Affordable Care Act s coverage expansions, including people with private insurance coverage. Those who purchased nongroup coverage before the implementation of the health insurance Marketplaces in 2014 spent an especially high proportion of their income on health care, and over half of these people will qualify for premium subsidies in the Marketplaces. Federal subsidies will substantially reduce medical cost burdens for many people who do not obtain health insurance through their employers. Peter J. Cunningham (pjcunningham@vcu.edu) is a professor in the Department of Healthcare Policy and Research, School of Medicine, Virginia Commonwealth University, in Richmond. The Affordable Care Act (ACA) was designed to increase access to affordable health insurance and health care for people who are uninsured. In addition, many individuals and families who relied on nongroup coverage before 2014 will be able to find more affordable options because of the larger and more varied risk pools for health plans available through the federal and state Marketplaces, premium subsidies for lower-income people, and restrictions on underwriting practices that had allowed insurance companies to charge much higher premiums (or deny coverage altogether) to people considered to be high health risks. 1 Such provisions could help narrow the gap in affordability between people with employersponsored coverage who in most cases will be required by the ACA to retain their coverage and people buying insurance in the nongroup market. January :1 Health Affairs 117

2 nances. In addition, the lingering effects of the recession and the modest economic recovery have led to stagnant or decreasing family incomes, which in turn have negated any benefit to families resulting from the slowing growth in health care costs. 5 This article updates previous analyses published in Health Affairs of the medical cost burdens for people younger than age sixty-five. 2,3 This analysis includes estimates for the most recent years for which data were available and compares them to earlier trends. The time period included the official end of the Great Recession, in June However, it was characterized by slow economic growth, still-high unemployment, a continued decrease in inflation-adjusted family income, and slow increases in health care costs. Health insurance coverage, family income, and health care costs have changed little since 2011, the year before the major coverage expansions of the ACA were implemented. Thus, the estimates likely reflect medical cost burdens for American families just prior to the implementation of the ACA s major coverage expansions. 4,5 This article also provides a more detailed analysis of medical cost burdens for people with nongroup coverage before the opening of the federal and state health insurance Marketplaces. In the early months of the implementation of the Marketplaces, media reports suggested that many people with individual policies faced higher costs in the Marketplaces, which led to concerns that the reform would not accomplish the primary objective for which it was named. 6 In addition, a 2014 report by McKinsey and Company indicated that most new enrollees in plans offered in the Marketplaces previously had coverage through the nongroup market. 7 The results show that people with nongroup coverage before the ACA were incurring high medical cost burdens about half were spending more than 10 percent of their family income on health insurance and health care and that new coverage options in the Marketplaces will allow them to lower their costs dramatically. Study Data And Methods The analysis was based on data from the Medical Expenditure Panel Surveys (MEPS), sponsored by the Agency for Healthcare Research and Quality. The survey s Household Component collects detailed information on health care expenditures and use of services, insurance coverage, sources of payment, health status, employment, and other sociodemographic characteristics of individuals and families. The Census Bureau s Current Population Survey (CPS) has also begun collecting data on outof-pocket expenditures. However, changes in spending and medical cost burdens can be assessed through that survey only since 2010, when the survey first included questions on these topics. 8 Furthermore, the CPS relies entirely on respondents reports of out-of-pocket expenditures for the entire year. In contrast, household reports of use and spending in MEPS are supplemented with information on third-party payments, as well as billing codes from provider billing records. In addition, MEPS is unique in that it includes estimates of total expenditures (from all sources), as well as spending by type of service. This makes it possible to calculate the share of total spending that each person paid out of pocket, as well as breakdowns of total and out-of-pocket spending by type of service. Sample sizes in MEPS for people younger than age sixty-five are about 28,000 people for each of the survey years. Estimates are weighted to be nationally representative of the US civilian noninstitutionalized population of the same age. To account for potential underreporting or bias on key population characteristics, survey weights are poststratified to reflect the distribution of the population by income, employment status, race/ethnicity, and Metropolitan Statistical Area status. Standard errors used in tests of statistical significance take into account the complex survey design of MEPS. As in previous studies, 2,3 medical cost burden for families was defined as the ratio of total out-ofpocket spending for health care services and premiums to total family income. Both income and medical cost burden were measured at the family level, although population estimates were reported at the individual level, and each individual was assigned the family burden measure. People in families that spent more than 10 percent of family income on health care were defined as people with high burdens. Although the 10 percent threshold is somewhat arbitrary for defining high burden, it was used to be consistent with prior studies. Zero or negative values for family income were recoded to $100 for the purposes of computing burden levels. Premiums included payments for private policies reported in MEPS as well as simulated Medicare Part B premiums for people enrolled in Medicare and not eligible for Medicaid premium support. All premiums were prorated to account for the duration of coverage during the year. Individuals who did not have coverage for all of a given calendar year were classified in this study as uninsured. The remaining individuals were assigned to one of three mutually exclusive in- 118 Health Affairs January :1

3 surance categories based on the type of coverage they held for the longest period of time during the year (measured by number of months). The three categories were private employment-sponsored insurance, private nongroup insurance, and public insurance. Trends in out-of-pocket medical spending are reported for the period For the years before 2010, three-year averages are reported (for , , and ) to reduce year-to-year fluctuations in estimates and to facilitate the interpretation of longer-term trends. The groupings correspond to major periods of change reported previously in Health Affairs. 2,3 Single-year results are reported in the online Appendix. 9 Study Results Increase In High Medical Cost Burdens In 2011, 19.2 percent of Americans younger than sixty-five lived in families with high burdens of out-of-pocket medical spending an increase from the 18.2 percent seen in that category during the period (Exhibit 1). Family outof-pocket burdens stabilized during the later Exhibit 1 Components Of Family Out-Of-Pocket (OOP) Burdens For People Younger Than Age 65, By Insurance Status, Years Total Population (1,000s) Family income ($) a Family OOP spending on services ($) Family OOP spending on premiums ($) b Total family OOP burden ($) c ,414 68,920** 1,360** 1,589** 2,948* 15.7** ,827 68,100** 1,471** 1,856** 3, ** ,524 65,900 1,336** 1,914** 3, * ,416 64,800 1,250 2,107 3, ,559 63,900 1,193 2,069 3, Employer-sponsored insurance ,713 84,700 1,489** 2,003** 3,491* 13.7** ,666 85,300 1,659** 2,441** 4, ** ,627 84,300 1,568** 2,592** 4, ** ,502 85,000 1,543 2,885 4, ,853 83,800 1,449 2,838 4, Nongroup private insurance ,965 74,600 2,004 4,358** 6,362** 41.6** ,780 74,300 2,310 4,939** 7,249** ,812 73,200 2,142 4,854** 6,996** ,479 73,700 2,062 5,866 7, ,600 74,500 2,209 5,799 8, Public insurance ,898 26, ** 330* 1,239** 20.1** ,749 27, ** 342* 1,246** 18.5** ,454 27, ** 382 1,056* ,211 26, ,418 26, Uninsured ,839 36,500 1,060** 226** 1,287** ,630 36,200 1,095** 271 1,366** ,631 35, ** 270 1, ,225 33, , ,689 33, , People in families with high OOP burden (%) d SOURCE Author s analysis of data from the Medical Expenditure Panel Surveys, Household Component, NOTES Insurance status is based on monthly insurance indicators and reflects coverage for the entire year. People with multiple coverage sources were assigned the coverage with the longest duration. For , , and , three-year averages are reported. Single-year results can be found in online Appendix Exhibit A1 (see Note 9 in text). Income and expenditure amounts are averages in 2011 dollars, based on the Consumer Price Index for urban areas. Significance refers to difference with a Family income is income of all family members, before taxes. The actual family income reported by survey respondents was used to compute financial burden measures. For this exhibit, estimates of average family income were rounded to the nearest hundred. b Premiums are enrollee payments for private coverageandmedicarepartbpremiums.premiumsforpeoplewithpublicinsuranceornocoverageareprivatecoveragehelpforpart of the year, private coverage for other family members, or both. c BurdenisOOPspendingoncareandpremiums.Careandpremium spending may not sum to total burden because of rounding. d High burden is combined OOP expenses for services and premiums that are greater than 10 percent of family income. *p < 0:10 **p < 0:05 January :1 Health Affairs 119

4 14% Of uninsured Among uninsured people, 14 percent were in families with high out-ofpocket spending burden in period. However, the increase since 2009 is part of a longer-term trend in decreasing affordability of medical care for Americans relative to their incomes. In the period , 15.7 percent of Americans were in families with high out-ofpocket burdens. Private Coverage The increase in high outof-pocket burdens for families was concentrated primarily among people with employer-sponsored insurance. Among people in that category, the percentage with high burdens increased from 17.9 percent in to 19.9 percent in 2011 (Exhibit 1). Increases in spending on premiums and medical services both contributed to increases in outof-pocket burdens in previous years. However, the more recent increase was attributable primarily to increases in family spending on premiums. For people with employer-sponsored insurance, the annual average spending on premiums rose from $2,592 in to $2,838 in 2011 (Exhibit 1). According to findings from the 2013 Employer Health Benefits Survey, 10 this likely reflects an increase in the cost of health insurance coverage during this period, instead of an increase in employees share of the total premium. In contrast, family incomes were unchanged, and family out-of-pocket spending on health services decreased during this period for people with employer-sponsored coverage. Family medical cost burdens were highest for people with nongroup private insurance. In 2011, 48.0 percent of people with this type of coverage experienced high out-of-pocket burdens a percentage largely unchanged from (Exhibit 1). High burdens for people with nongroup coverage reflected their higher spending on premiums ($5,799 versus $2,838 for people with employer-sponsored coverage), higher out-of-pocket spending on services ($2,209 versus $1,449), and lower family incomes. People with nongroup coverage had much higher out-of-pocket spending before the implementation of the ACA, compared to people with employer-sponsored coverage. This difference reflects the fact that nongroup policies were not subsidized by employers, that premiums were based in large part on individual health risk (rather than that of the group), that there were fewer restrictions on medical underwriting practices by insurance plans, and that deductibles and copays for services were typically much higher. Public Coverage Although the Medicaid program includes little or no cost sharing for premiums and services, people with public coverage may nevertheless be in families with high out-of-pocket burdens for several reasons. Some of their family members may have private insurance or be uninsured, they may live in a state that imposes some cost sharing or out-of-pocket payments on some Medicaid enrollees, they may use services not covered by some state Medicaid programs (such as oral health care), and they may have been spending down assets before qualifying for Medicaid. In any case, because people with public coverage typically have very low incomes, even small or modest out-of-pocket spending can be burdensome. Nevertheless, high family out-of-pocket burdens among nonelderly people with public coverage have been decreasing during the past decade, from 20.1 percent in to 16.0 percent in 2011 (Exhibit 1). This trend may reflect in part an increase in public coverage of children as a result of the Children s Health Insurance Program (CHIP) and expansions in Medicaid coverage for children. Children generally use fewer services and less expensive ones compared to adults. Uninsured Among uninsured people, 14 percent were in families with high out-of-pocket burdens in 2011 (Exhibit 1). These burdens for the uninsured are not as high as might be expected because of their lower access to and use of medical care. Many uninsured people forgo medical care to avoid high burdens. Out-Of-Pocket Spending Trends By Type Of Service A previous analysis showed that despite recession-related decreases in family income, the lack of an increase in high out-of-pocket burdens in the period was a result of lower out-of-pocket spending on services. 2 The same analysis also showed that lower spending on prescription drugs part of which was caused by people shifting from brand-name medications to less expensive generics accounted for virtually all of the decrease in out-of-pocket spending on services. Estimates from 2010 and 2011 show this trend to be continuing. Per person out-of-pocket spending on prescription drugs averaged $172 in 2011 a decrease from $203 in and $279 in (Exhibit 2). New generic medications became available between 2009 and 2011, including those used to treat attention deficit hyperactivity disorder, schizophrenia, epilepsy, depression, migraine headaches, and early-stage breast cancer. 11,12 Purchasers and health plans likely increased the use of valuebased tiering of drugs in health plans, which includes very low cost sharing for many generic medications used to treat common chronic conditions. The decrease in spending on prescription drugs was experienced across all insurance coverage categories (Exhibit 2). Nevertheless, out- 120 Health Affairs January :1

5 Exhibit 2 Components Of Out-Of-Pocket (OOP) Spending For People Younger Than Age 65, By Insurance Status, Years Total Total OOP spending ($) Average per person spending ($) on: Office-based physician visits Hospital care (inpatient, ED, and outpatient) Rx drugs Home health, vision, and medical equipment Dental care ** 143** 79** 254** ** ** ** 45** 165** ** 45** 156** Employer-sponsored insurance ** 66** 226** ** 180** ** ** Nongroup private insurance , ** , ** , , Public insurance ** ** 21* ** ** 23** 60** ** 30** Uninsured ** ** 353** 40* 168** ,003** * 398** 42** 159** ** 42** SOURCE Author s analysis of data from the Medical Expenditure Panel Surveys, Household Component, NOTES Average expenditures were adjusted to reflect 2011 dollars based on the Personal Health Care Expenditure Index, from the Office of the Actuary, Centers for Medicare and Medicaid Services. Spending is on health services only, not on premiums. For , , and , three-year averages are reported. Single-year results can be found in online Appendix Exhibit A2 (see Note 9 in text). Significance refers to difference with ED is emergency department. *p < 0:10 **p < 0:05 of-pocket drug spending accounts for one-third of all out-of-pocket spending on services, although this share has been decreasing during the past decade (data not shown). In contrast, out-of-pocket spending on officebased physician visits accounts for a similar share of out-of-pocket spending (about onethird), but the average annual amount spent changed little between 2004 and 2011 ($165 $175, on average; Exhibit 2). Out-of-pocket spending on hospital-based care also changed little during this period. However, this spending accounted for a much smaller share of total out-of-pocket spending (about 8 percent; data not shown). Out-of-pocket spending on dental care decreased from $173 in to $131 in 2011 (Exhibit 2). The decrease was especially large for uninsured people. This may reflect in part the expansion of dental clinics at community health centers that charge uninsured patients on a sliding scale based on their ability to pay. Between 2007 and 2012 the use of dental services at such centers increased more than 50 percent. 13 Medical Cost Burdens By Family Income Before the implementation of the ACA, family out-of-pocket burdens were highest among people with family incomes of less than 200 percent of the federal poverty level (Exhibit 3). Few people at this income level have private insurance (data not shown). However, those who do spend about half of their incomes on health care, pri- January :1 Health Affairs 121

6 Exhibit 3 Percentage Of People In Families With High Financial Burden, By Family Income And Insurance Coverage, Percent of people with high family out-of-pocket burden Family income (percent of poverty) All With private insurance With public insurance Uninsured All Up to 138% >138% 200% ** 13.4** 12.1** >200% 400% 22.0** 26.7** 8.9** 9.3** >400% 9.0** 9.5** 5.3** 4.1** SOURCE Author s analysis of data from the Medical Expenditure Panel Surveys, Household Component, NOTE High out-ofpocket burden is combined out-of-pocket expenses for services and premiums that exceed 10 percent of family income. Private insurance includes both employer-sponsored insurance and nongroup private insurance. Significance refers to difference with family income of up to 138 percent of the federal poverty level. **p < 0:05 marily because of their low income since most people at this income level have employer-sponsored coverage. Even for people with incomes of less than or equal to 138 percent of poverty with public coverage, about one-fifth have high family out-of-pocket burdens. Out-of-pocket burdens are relatively high for income groups that could benefit from premium subsidies in the new health insurance Marketplaces (such as those with family incomes of percent of poverty). Of privately insured people with family incomes of percent of poverty, 46.5 percent experienced high family burdens; the share was 26.7 percent for the privately insured with incomes of more than 200 percent to 400 percent of poverty (Exhibit 3). Out-of-pocket burdens were much lower for people who were in families with incomes of more than 400 percent of poverty and who were not eligible for public or subsidized coverage through the ACA, compared to people in lowerincome groups who were eligible for such coverage. More Affordable Options For People With Nongroup Coverage Among the privately insured, people with nongroup coverage who do not have access to employer-sponsored coverage are eligible for premium subsidies in the new Marketplaces if their family incomes are between 100 percent and 400 percent of poverty. Based on MEPS, premiums for the 7.1 million people with nongroup coverage in who did not have access to employer-sponsored coverage (through their own job or their spouse s job) averaged $6,233, which amounted to 10 percent of their family income. 14 Importantly, 53 percent of those with nongroup coverage in had family incomes of 400 percent of poverty or less (making them potentially eligible for premium subsidies in the Marketplaces), and 23 percent had family incomes of percent of poverty. People in families with incomes at that level spent an average of $5,435 in annual premiums for nongroup coverage, or 15.9 percent of their family income (Exhibit 4). If these people purchased subsidized coverage through the secondlowest-cost silver plan in the Marketplaces, their out-of-pocket premiums would decline by more than two-thirds to $1,797 on average, or 5.0 percent of their family income. The amount of the subsidy decreases as family income increases. Thus, families at higher income levels percent of poverty would see a smaller decrease in what they pay for premiums, from 11.3 percent of family income, on average, to 8.7 percent of family income for subsidized plans (Exhibit 4). Still, 47 percent of people with nongroup coverage had incomes above 400 percent of poverty and therefore would not be eligible for premium subsidies. In they spent $6,774 annually on premiums, or about $565 per month. Early information on premiums in the Marketplaces suggest that the savings will be substantial for many of these people. Based on information about qualified health plans made available by the Department of Health and Human Services in September 2013, the lowest-cost silver plan averaged about $300 per month nationally, with only three states (Alaska, Mississippi, and Wyoming) exceeding an average of $400 per month. 15 Based on MEPS data, among people with nongroup coverage who had family incomes of more than 400 percent of poverty, two-thirds paid more than $300 per month in , half paid more than $400 per month, and one-fourth paid more than $750 per month (data not shown). Switching to Marketplace plans could also result in substantial savings on out-of-pocket spending for services. The ACA imposes limits 122 Health Affairs January :1

7 Exhibit 4 Out-Of-Pocket Spending For People With Nongroup Coverage, By Family Income, Family income (percent of poverty) 138% 250% >250% 400% >400% People with nongroup coverage (millions) Premiums Annual average $5,435 $5,711 $6,744 Percent of family income 15.9% 11.3%** a Annual average after ACA subsidy b $1,797 $4,753 a Percent of family income after ACA subsidy 5.0% 8.7% a Health services c Annual average $2,200 $1,441 $1,652 Percent of total spending 33.6% 31.0% 33.9% SOURCE Author s analysis of data from the Medical Expenditure Panel Surveys, Household Component, NOTE Significance refers to difference with family income of percent of federal poverty level. a Ineligible for ACA premium subsidies. b Computed by applying the maximum percentage of income a person is required to pay for premiums if he or she obtained subsidized coverage in a silver plan to that person s actual income, as specified in the Affordable Care Act (ACA). c Reflects spending on all family members covered by the plan. **p < 0:05 on out-of-pocket spending associated with deductibles and copays for covered services and provides cost-sharing subsidies to individuals and families with incomes of percent of poverty. Silver plans sold in the Marketplaces are required to have an actuarial value of at least 70 percent (that is, the plan must pay at least 70 percent of the average total covered expenses incurred by the beneficiaries). However, the cost-sharing subsidies will increase this to as high as 94 percent for people with incomes of percent of poverty and 87 percent for those with incomes of percent of poverty. On average, people with nongroup coverage and incomes of percent of poverty (who are potentially eligible for cost-sharing subsidies) spent $2,200 out of pocket annually in for health services (Exhibit 4). Their health plans paid for about two-thirds of their total health care bills, which means that switching to Marketplace plans and obtaining costsharing subsidies could lower their health bills dramatically. Implications Growth in national health care spending has slowed in recent years, which many observers attribute to the lingering effects of the recession of and continued high unemployment. 4 However, the percentage of Americans with high out-of-pocket burdens continues to increase. The increase between and 2011 was driven primarily by growth in what Americans paid for employer-sponsored health insurance coverage and the lack of change in family income after inflation. This occurred in spite of a decrease in out-of-pocket spending on care, which was driven primarily by continued decreases in spending on prescription drugs. Despite concerns about the effect of the ACA on employment and whether small employers (those with fewer than fifty workers) will continue to offer coverage, the ACA will have little impact on people who continue to obtain health benefits from their employers. People with employer-sponsored coverage are not eligible to receive premium subsidies through the Marketplaces unless they have family incomes below 400 percent of poverty and their out-of-pocket premiums for employer coverage would exceed 9.5 percent of family income. Few people with employer-sponsored coverage will qualify for a subsidy because the affordability threshold will be based on the cost of single coverage for people who currently have family coverage. Nevertheless, if premiums continue to rise faster than incomes, more people with employer coverage may qualify for subsidies to purchase coverage through the Marketplaces. The results from this study indicate that many people with nongroup coverage before 2014 will be able to dramatically lower their health insurance and health care spending through the Marketplaces. Some critics of the ACA have contended that most of new Marketplace enrollees already had insurance instead of being uninsured. 7 As was the case with Medicaid expansions and the implementation of CHIP in the past, the concern is that federal subsidies are displacing or crowding out coverage that was previously obtained without the assistance of federal dollars. Such displacement is an inevitable part of all expansions in publicly subsidized coverage. January :1 Health Affairs 123

8 Nonetheless, the high financial burden that people with nongroup coverage incurred before the ACA and the opportunity to dramatically lower their health care spending through the ACA need to be considered. Some policy makers may argue that federal subsidies are unnecessary because many lower-income individuals and families were able to purchase coverage despite its cost. However, research has shown that high levels of family spending on health care can consume income that would otherwise have been dedicated to buying other necessities, which leads to high levels of debt and even bankruptcy, especially for lower-income families. 16 Conclusion Rising economic inequality is a growing concern among policy makers. Thus, the fact that people with employer-sponsored coverage spent half as much on health care premiums and services as people with nongroup coverage did before the ACA s major coverage expansion in 2014 should be considered by policy makers when assessing the impact of the ACA. The potential for the new Marketplaces to reduce disparities in spending between people with nongroup coverage and those with employer coverage, and the potential to alleviate some of the impact of economic inequality in the United States, should not be overlooked by policy makers as an important goal of the ACA. This study was funded by the Commonwealth Fund. Programming support was provided by Ase Sewell, of Sewell Inc. NOTES 1 Levitt L, Pollitz K, Claxton G, Damico A. How buying insurance will change under Obamacare [Internet]. Menlo Park (CA): Kaiser Family Foundation; 2013 Sep 24 [cited 2014 Nov 6]. Available from: health-reform/perspective/how-buying-insurance-will-change-underobamacare/ 2 Cunningham PJ. Despite the recession s effects on incomes and jobs, the share of people with high medical costs was mostly unchanged. Health Aff (Millwood). 2012;31(11): Cunningham PJ. The growing financial burden of health care: national and state trends, Health Aff (Millwood). 2010;29(5): Martin AB, Hartman M, Whittle L, Catlin A, National Health Expenditure Accounts Team. National health spending in 2012: rate of health spending growth remained low for the fourth consecutive year. Health Aff (Millwood). 2014;33(1): DeNavas-Walt C, Proctor BD, Smith JC. Income, poverty, and health insurance coverage in the United States: 2012 [Internet]. Washington (DC): Census Bureau; 2013 Sep. Figure 1: real median household income by race and Hispanic origin: 1967 to 2012 [cited 2014 Nov 20]. (Current Population Reports). Available from: 6 Cha AE, Sun LH. For consumers whose health premiums will go up under new law, sticker shock leads to anger. Washington Post Nov 3. 7 McKinsey and Company. Healthcare systems and services [Internet]. New York (NY): McKinsey and Company; [cited 2014 Nov 6]. Available from: service/healthcare_systems_and_ services/latest_thinking 8 Caswell KJ, O Hara B. Medical outof-pocket expenses, poverty, and the uninsured [Internet]. Washington (DC): Census Bureau; 2010 Dec 29 [cited 2014 Nov 6]. (SEHSD Working Paper No ). Available from: povmeas/methodology/ supplemental/research/caswell- OHara-SGE2011.pdf 9 To access the Appendix, click on the Appendix link in the box to the right of the article online. 10 Kaiser Family Foundation, Health Research and Educational Trust. Employer health benefits: 2013 survey [Internet]. Menlo Park (CA): KFF; 2013 Aug 20 [cited 2014 Nov 6]. Available from: kff.org/?post_type=reportsection&p= Food and Drug Administration. Generic drug roundup: December 2010 [Internet]. Silver Spring (MD): FDA; 2010 Dec [cited 2014 Nov 6]. Available from: downloads/forconsumers/ ConsumerUpdates/UCM pdf 12 Food and Drug Administration. Generic drug roundup: September 2009 [Internet]. Silver Spring (MD): FDA; [last updated 2014 Oct 14; cited 2014 Nov 6]. Available from: Consumers/ConsumerUpdates/ ucm htm 13 Health Resources and Services Administration Health center data [Internet]. Rockville (MD): HRSA; [cited 2014 Nov 6]. Available from: datacenter.aspx?year= This estimate is somewhat higher than the average premium for nongroup coverage in Exhibit 1. The difference reflects the fact that Exhibit 4 includes only people with nongroup coverage who did not have access to employer-sponsored coverage. Also, people in the family with other types of coverage are excluded from Exhibit Department of Health and Human Services, Office of the Assistant Secretary for Planning and Evaluation. Health insurance marketplace premiums for 2014 [Internet]. Washington (DC): HHS; 2013 Sep [cited 2014 Nov 6]. (ASPE Issue Brief). Available from: marketplacepremiums/ib_ marketplace_premiums.cfm 16 Cunningham PJ, Miller C, Cassil A. Living on the edge: health care expenses strain family budgets [Internet]. Washington (DC): Center for Studying Health System Change; 2008 Dec [cited 2014 Nov 6]. (HSC Research Brief No. 10). Available from: CONTENT/1034/ 124 Health Affairs January :1