|
D A T A W A T C H
T H E U N D E R I N S U R E D
14 June 2005
Insured But Not Protected:
How Many Adults Are Underinsured?
The experiences of adults with inadequate coverage
mirror those of their uninsured peers,
especially among the chronically ill.
By Cathy Schoen, Michelle M. Doty, Sara R. Collins,
and Alyssa L. Holmgren
ABSTRACT:
Health insurance is in the midst of a design shift toward greater financial risk for patients. Where medical cost exposure is high relative to income, the shift will increase the numbers of underinsured people. This study estimates that nearly sixteen million people ages 19–64 were underinsured in 2003. Underinsured adults were more likely to forgo needed care than those with more adequate coverage and had rates of financial stress similar to those of the uninsured. Including adults uninsured during the year, 35 percent (sixty-one million) were under- or uninsured. These findings highlight the need for policy attention to insurance design that considers the adequacy of coverage.
The relentless rise in health care costs and insurance premiums has spurred a move away from more-comprehensive insurance benefits for the population under age sixty-five. Trends instead point toward plans with higher deductibles, patient cost sharing, and, in some instances, a more restricted scope of benefits. Faced with the fourth consecutive year of double-digit premium inflation and the demise of managed care options, employers have sought to moderate their premium costs by offering new insurance products that shift more financial risk to workers.1 Insurance plans sold to small businesses, in particular, have moved to sharply higher deductibles; however, increases are also spreading across larger firms.2
Recent federal policies have also pushed toward higher patient cost sharing through enactment of new tax-protected health savings accounts (HSAs) available only to people with insurance policies having deductibles of at least $1,000 per person. Some proposals would further encourage such high-deductible plans.3 Although the United States already stands out among industrialized countries for the high share of medical costs its citizens pay out of pocket, trends point to still greater patient and family exposure to medical care costs in the future, reversing gains in more comprehensive coverage that occurred in the 1990s.4
To date, efforts to redesign insurance have proceeded with little regard to patients’ or families’ ability to pay or the consequences of exposure to financial risk. To the extent that patient cost sharing or benefit gaps leave insured adults without adequate financial protection in the event of illness, erosion in the quality of insurance coverage will raise the number of Americans who are underinsured. If inadequate protection erects barriers to appropriate care, market trends could undermine the central goals of health insurance: to facilitate timely access to care when needed and to protect patients from costs that would be catastrophic relative to their income.
The adverse access and health effects of increased cost sharing can be particularly acute for low-income populations, because plans rarely adjust cost exposure relative to income.5 Moreover, higher cost sharing, by design, shifts costs to sicker populations. The combination of poor health and low income increases the risk of both access barriers and financial stress. For example, a recent Canadian study of the impact of increased patient payments for prescription drugs for the elderly and welfare recipients found reductions in use of essential drugs in both populations, which, in turn, led to higher rates of serious adverse events and emergency department visits.6 A study of copayments for medications among chronically ill Americans similarly found sizable reductions in use and increased risk to health, including a 23 percent reduction in use of antidiabetic drugs among diabetics.7 Another U.S. study that followed chronically ill adults over four years found that higher cost sharing reduced care for both serious and minor symptoms.8 In general, past studies, including the RAND Health Insurance Experiment, indicate a need for caution in insurance designs that put patients at increased financial risk, particularly those who are poor and sick.9
Even though concerns that inadequate insurance can contribute to access barriers and financial hardships have long been recognized, the extent of the underinsured problem among nonelderly U.S. adults has not been assessed since the seminal work by Pamela Farley Short and Jessica Banthin in 1995.10 Their study included a methodology developed by Short that relied on detailed analyses of private insurance policies to estimate out-of-pocket costs that would be high relative to family income in the event of catastrophic illness. The paucity of research since the mid-1990s stems in part from the difficulty of gathering the type of detailed insurance and demographic information they used to count the underinsured as well as a lack of consensus on what it means to be underinsured.11 To enable periodic national updates on the number of Americans who are underinsured and monitor the impact on access, policymakers need a working definition of underinsured that can be easily applied.
The primary goal of this study is to focus policy attention on the issue of the underinsured and potential risks for access. The study also seeks to develop household survey measures that can be easily updated to monitor trends.
We define underinsured as being insured all year but without adequate financial protection. Using a 2003 cross-sectional survey of adults, we estimated the number of adults who were underinsured based on exposure to out-of-pocket costs that were high relative to incomes.12 We then examined the effects of being underinsured on access to care, with comparisons to insured adults having more adequate coverage and to the uninsured.
Study Data And Methods
Data. Study data come from the Commonwealth Fund 2003 Biennial Health Insurance Survey, a nationally representative telephone survey of 4,052 adults age nineteen and older living in the continental United States, conducted by Princeton Survey Research Associates. Interviews took place from 3 September 2003 through 4 January 2004. In this study we restricted the analysis to the sample of 3,293 people ages 19–64 who participated in the survey.
The survey consisted of twenty-five-minute telephone interviews administered in either English or Spanish. To enable more detailed analyses of adults with low incomes, the survey oversampled adults from telephone exchanges with a high density of low-income households. The final sample weights corrected for the disproportionate sample design and weighted the adult population by age, sex, race/ ethnicity, education, household size, and region, using the 2003 Annual Social and Economic Supplement of the U.S. census. The resulting sample for analysis is representative of the 172 million people ages 19–64 living in the continental United States. The overall survey response rate was 50 percent.13
Key study variables and methods. The survey included an array of questions about access and care experiences, out-of-pocket medical care costs, insurance, income, and other demographic characteristics. When surveyed, 36 percent of all respondents had health problems: They either rated their health as fair or poor or had some type of disability or one of four chronic diseases.14
Respondents reported their current insurance status and also whether they had been uninsured at any time during the past year. We used these responses to categorize adults as insured all year or uninsured during the year.15
Defining “underinsured.” The study used indicators of financial risk to define underinsured. Following the lead of earlier work on the underinsured, we assessed risk by comparing cost exposure to family income.16 Using respondents’ estimates of out-of-pocket medical care expenses, plan deductibles, and income, we classified them as underinsured if they were insured all year but reported at least one of three indicators: (1) Medical expenses amounted to 10 percent of income or more; (2) among low-income adults (below 200 percent of the federal poverty level), medical expenses amounted to at least 5 percent of income; and (3) health plan deductibles equaled or exceeded 5 percent of income.17
We selected 10 percent of income because it was the threshold most commonly used in past studies of the underinsured or analyses of catastrophic costs. We included the lower threshold, 5 percent of income, for low-income adults based on the national policy embedded in the State Children’s Health Insurance Program (SCHIP) that permitted some cost sharing for low-income families but limited total exposure to 5 percent of income. The RAND experiment also used a 5 percent threshold for low-income participants.18
The deductible indicator measures potential risk and was the only measure of plan cost sharing available in the survey. We specified the 5 percent threshold based on analysis revealing that the majority of adults with deductibles this high were in families with two or more people and that 93 percent had incomes below 200 percent of poverty (63 percent had incomes below 100 percent of poverty). Thus, deductibles exposed families to financial risks equal to this study’s other two cost-to-income thresholds.
Two of three indicators (with deductibles as the exception) relied on current-year out-of-pocket costs and medical bills. As a result, unlike methods that simulate expenses based on analysis of insurance plans, age, and health, this “snapshot” approach will miss healthier adults with inadequate coverage who had little need for medical care during the year. The methodology thus likely underestimates the number of underinsured people.
Analysis. We used this composite indicator variable to divide continuously insured adults into two groups: underinsured and not underinsured. The “not underinsured” group also includes 545 insured adults (25 percent of the insured group) who did not provide sufficient income or expense information to assess financial risk (missing numerator or denominator values).
The analysis compared the care experiences of underinsured adults with those of adults having more adequate insurance and to uninsured adults. By construction, the “underinsured” category includes only adults who were insured all year, which allowed us to associate access experiences and out-of-pocket costs during the year with the quality of insurance and avoided the chance that events occurred at a time when the person was uninsured.
Exhibits compare experiences by insurance groups and indicate where differences are significant at the 5 percent level or better, using “adequate” insurance as the referent group. We used statistical software (STATA 7.0) that took into account the sample weights and adjusted the standard errors for the stratified sampling design.
Study Results
How many insured adults are underinsured? Three of four people ages 19–64 said that they had been insured all year. When we applied the three indicators of being underinsured, 12 percent of these insured adults—nearly sixteen million—were underinsured. Exhibit 1 shows the incidence of each indicator separately as well as the cumulative effect as indicators are added sequentially following the hierarchy in the exhibit. Based on reported expenses and income, about 7 percent of continuously insured adults spent 10 percent or more of their income on family medical expenses during the year. Adding low-income adults who reached or exceeded the threshold of 5 percent of income, 11 percent of adults had expenses at or above the two income-related out-of-pocket indicators during the past year; 3 percent faced deductibles that amounted to 5 percent or more of their income. Respondents often met more than one threshold, accounting for the difference between the marginal contribution of each indictor and its incidence.
How many adults are uninsured and underinsured? About one-quarter of adults—an estimated forty-five million people—were uninsured for all or part of the year, based on the survey (Exhibit 2). When uninsured adults were added to those who were underinsured based on financial indicators, we found that an estimated sixty-one million adults, or 35 percent of the population ages 19–64, had either no insurance, sporadic coverage, or insurance that exposed them to catastrophic medical costs during 2003.
Adults with lower incomes were more likely than those with higher incomes to be uninsured and, when insured, to be underinsured (Exhibit 2). In total, seven of ten low-income adults (with household income below 200 percent of poverty) were either uninsured or underinsured during the year. Similarly, underinsured adults were disproportionately low income: 73 percent had annual incomes below 200 percent of poverty (data not shown.)
Underinsured rates were also high among adults with health problems, reflecting the use of indicators based on recent out-of-pocket expenses. Among sicker adults defined broadly, 43 percent were either uninsured or inadequately insured, compared with 31 percent of healthier adults (Exhibit 2).
African American and Hispanic adults were at high risk of being uninsured and, when insured, were somewhat more likely than white, non-Hispanic adults to have inadequate insurance. The elevated rates with gaps and inadequate coverage were related to income: 52 percent of African American and 59 percent of Hispanics adults reported incomes below 200 percent of poverty, compared with 27 percent of whites (data not shown).
Access and care experiences by insurance group. Comparisons of care, satisfaction, and confidence among the three insurance groups revealed the importance of having adequate insurance to facilitate access and more positive medical care experiences (Exhibit 3). Relative to adults with more adequate insurance, underinsured as well as uninsured adults were significantly more likely to go without care because of costs, to lack confidence that they would receive high-quality care when they needed it, and to rate care experiences negatively. More than half of the underinsured (54 percent) and uninsured (59 percent) went without at least one of four needed medical care services during the year—double the rate among those with more adequate insurance. Across the measures of cost-related access problems, underinsured adults reported forgone care at rates two to four times higher than those reported by adults classified as having more protective insurance.
Notably, despite being insured all year, underinsured adults reported negative care experiences at rates similar to those of uninsured adults. Having a regular doctor was the only measure for which underinsured adults’ responses were similar to those of adults with adequate insurance.
Underinsured adults were also significantly less confident about their ability to get care in the future and less satisfied with the quality of the care received than were those with more protective coverage. Perhaps signaling underlying anxiety about their ability to get care when needed, half of underinsured adults thought that they would have received better care if they had been covered by a different insurance plan.
Medical bill burdens by insurance group. Underinsured and uninsured adults also were at risk for facing collection agencies and enduring high levels of financial stress as a result of medical bills. Nearly half (46 percent) of underinsured adults reported that they were contacted by a collection agency during the past year about owing money for medical bills; more than one-third (35 percent) said that they had to change their way of life dramatically to pay for medical bills. The latter group includes a high proportion saying that they took on high levels of credit card debt or took out a loan or mortgage against their home to pay medical bills (data not shown). As illustrated, rates of medical bill stress among the underinsured were almost equal to those reported by the uninsured.
Insurance adequacy and access for adults with health problems. Access and care concerns were most acute among adults with health problems who were either underinsured or uninsured (Exhibit 4). Two-thirds of sicker adults who were underinsured and three-fourths of sicker adults who were uninsured went without needed care because of costs during the year. Notably, nearly half of underinsured sicker adults with chronic diseases or poor health did not adhere to medications, 38 percent did not see a doctor when sick, and one-third did not follow up on diagnostic treatments or care recommended by their doctors because of costs. Nearly half of sicker adults who were underinsured lacked confidence in access—near the levels reported by uninsured adults with health problems.
Inadequate coverage as well as being uninsured also left sicker adults grappling with medical bill burdens. Nearly three of five in each group faced collection agencies during the year. Nearly half of underinsured sicker adults said that medical expenses resulted in a major change in family life circumstances to pay bills.
These findings reflect the double jeopardy of health problems and low incomes among the uninsured and problems with policies that entail substantial cost sharing relative to income. Seventy percent of underinsured sicker adults and 65 percent of uninsured sicker adults shown in Exhibit 4 had incomes below 200 percent of poverty.
To control for the differences in income, health, age, and other demographic characteristics across insurance categories, we also conducted a series of multivariate analyses. In all regression analyses, the underinsured and the uninsured remained significantly more likely to report access concerns or negative care experiences than those with adequate coverage, after demographic differences were adjusted for.19
Inadequate insurance and other insurance characteristics. Using descriptions of insurance plans and plan experiences, the study found that being underinsured is related to underlying gaps in benefit structure, cost sharing, and other insurance limits (Exhibit 5). Compared with those having more adequate coverage, adults who were underinsured were less likely to have prescription drug coverage, dental coverage, or vision benefits and more likely to report having none of these benefits. Underinsured adults were also significantly more likely to have deductibles of $500 or more, a possible indication of overall higher cost sharing.
Experiences with plan limitations indicate generally more restrictive coverage. Underinsured adults were more likely than those with more adequate coverage to encounter limits on what their plans would pay to treat an illness or injury or to discover that their plans did not cover care they had thought would be covered. On both measures, underinsured adults reported coverage limitations at double the rate of those with more adequate coverage.
Overall, underinsured adults were more likely than those with more adequate coverage to face higher cost sharing, plan limits. and more restrictive benefits. Yet, despite more limited coverage, underinsured adults often incurred high annual premium costs: One-third had annual premium shares of $1,500 or higher, and 47 percent paid premiums that amounted to 5 percent or more of their annual incomes. These results indicate that the underinsured are coping with financial stress from both premiums and out-of-pocket costs.
Likely reflecting the benefit structure, financial exposure, and access experiences, one-third of the underinsured rated their insurance negatively, about double the rate of those with more adequate insurance (Exhibit 5). Negative plan ratings were correlated with benefit gaps as well as exposure to financial risk among both insurance groups.
Sources of insurance varied somewhat among the two insured groups. The underinsured were more likely than those with more adequate coverage to be insured by sources other than employer-based plans. Among those with job-based coverage, the distribution by firm size was similar (Exhibit 5). Within employer groups, low-wage workers were the most likely to have inadequate coverage. This pattern likely reflects both the types of policies available from lower-wage employers and the fact that any level of cost sharing will expose low-wage workers to relatively greater financial risk than that faced by their higher-income coworkers.
Discussion And Policy Implications
Based on a financial definition of inadequate insurance and recent cost experience, the study indicates that nearly sixteen million adults with full-year coverage were underinsured in 2003. Including those who lacked coverage at any time during the year, one in three U.S. adults were underinsured or uninsured.
The findings indicate that having inadequate insurance as well as being uninsured undermines access to care, satisfaction, and confidence in the quality of care obtained. Access barriers reported by the underinsured at times approach rates observed among uninsured adults.
Implications of higher cost sharing. Some policymakers and researchers have endorsed the move toward greater cost sharing, arguing that encouraging people to pay more of the health care dollar will make them more prudent consumers of health care services and help moderate health care cost inflation. Those advocating insurance redesign often envision coupling high deductibles with employer-funded or personally funded HSAs that could be used to pay front-end costs. The most recent published studies, however, indicate that only a small share of high-deductible plans included employer-funded accounts.20 HSAs also tend to favor healthier and higher-income adults who have the most to gain from tax-preferred status.21
This study indicates that without targeted protections, insurance policies that expose patients to costs that are high relative to income are likely to have a negative effect on access and adherence to recommended care. The risks are particularly high for low-income patients and those with chronic illnesses.
Past studies of the effects of cost sharing indicate that it is at best a blunt tool. Patients are nearly as likely to cut back on essential care as on more elective or discretionary services.22 The RAND experiment found that low-income families were particularly sensitive to costs and that steep reductions in physician visits undercut doctors’ ability to assess and screen as well as to treat chronic or acute disease. These results are particularly notable because RAND made an effort to protect those with lower incomes by putting a ceiling on cost exposure relative to income and having lower caps for the poor and near-poor.23
In the current marketplace, insurance policies rarely adjust patient cost sharing or out-of-pocket limits for varying incomes. Yet the effect on care patterns is likely to relate directly to incomes. A $1,000 deductible might appear to carry modest financial risk from the viewpoint of a family earning $100,000 or more per year with assets. The same policy would expose a minimum-wage family earning $10,000 a year to substantial risk, leaving them reluctant to seek care.
A clear consequence of deductibles and cost sharing will be to shift more of the costs of medical care to the sickest patients and their families. Given the concentration of health expenditures, this shift may do little to address underlying cost trends yet unduly burden families that are already under stress because of poor health.24
Shifting trends. The findings are likely to be symptoms of more to come. The survey took place at the end of 2003, at the onset of a shift toward higher deductibles and cost sharing. In the survey a small percentage of insured adults—just 6 percent—reported deductibles of $1,000 or more. Given the public policy push and market trends toward higher deductibles, it will be important to track coverage adequacy and related care patterns over time. The United States may well be on a path to where it becomes harder to distinguish the insured from the uninsured if insurance no longer provides either access or financial protection.
Study limitations and strengths. Estimates of the number of insured people whose coverage is inadequate will clearly vary depending on the method and measures used to categorize insurance. In this study, except for deductibles, we used indicators of financial risk that relied on cost experiences. A limitation of this methodology is that it misses healthier adults having inadequate policies who have used medical care less recently. Thus, the study—and any approach that relies on experiences—likely underestimates the number of underinsured adults.
Despite this limitation, the methodology offers an approach that could easily be adapted to national federal surveys such as the Medical Expenditure Panel Survey (MEPS) and the National Health Interview Survey (NHIS) to enable monitoring of trends. By relying on measures easily reported by adults in interviews, without recourse to extensive review of insurance documents, the indicators provide an efficient means of identifying the extent of financial stress and of tracking change. Although likely to undercount the underinsured, these indicators could yield a first alert to inform policy if markets move to higher front-end cost sharing or benefit limits on essential services.
Need for attention to insurance benefit design. Overall, the study points to a need for new policy and research attention to health insurance benefit design, with a focus on assessing design effects on access and financial protection relative to income. International studies repeatedly find that the United States often lags behind other countries in key measures of health status and timely access to care while leading the world in exposure to medical care costs.25 Without attention to insurance adequacy and whether patients receive effective care, an increase in the number of underinsured people could undermine health, productivity, and financial security in the future.
The authors thank the anonymous reviewers for their helpful comments. The views presented here are those of the authors and should not be attributed to the Commonwealth Fund, its directors, or its officers.
NOTES
1. J. Gabel et al., “Health Benefits in 2004: Four Years of Double-Digit Premium Increases Take Their Toll on Coverage,” Health Affairs 23, no. 5 (2004): 200–209; S.R. Collins et al., Job-based Health Insurance in the Balance: Employer Views of Coverage in the Workplace (New York: Commonwealth Fund, March 2004); and J.C. Robinson, “Reinvention of Health Insurance in the Consumer Era,” Journal of the American Medical Association 291, no. 15 (2004): 1880–1896.
2. J. Gabel and J. Pickreign, Risky Business: When Mom and Pop Buy Health Insurance for Their Employees (New York: Commonwealth Fund, April 2004).
3. The Bush administration and some Senate proposals would make premiums for high-deductible plans tax-deductible. Other proposals would forgive federal student loans if students buy health savings accounts (HSAs).
4. P. Hussey and G.F. Anderson, “A Comparison of Single- and Multi-Payer Health Insurance Systems and Options for Reform,” Health Policy 66, no. 3 (2003): 215–228; and G.F. Anderson et al., “Health Spending in the United States and the Rest of the Industrialized World,” Health Affairs (forthcoming).
5. S.R. Collins et al., The Affordability Crisis in U.S. Health Care: Findings from the Commonwealth Fund Biennial Health Insurance Survey (New York: Commonwealth Fund, March 2004).
6. R. Tamblyn et al., “Adverse Events Associated with Prescription Drug Cost-Sharing among Poor and Elderly Persons,” Journal of the American Medical Association 285, no. 4 (2001): 421–429.
7. D.P. Goldman et al., “Pharmacy Benefits and Use of Drugs by the Chronically Ill,” Journal of the American Medical Association 291, no. 19 (2004): 2344–2350.
8. M.D. Wong et al., “Effects of Cost Sharing on Care Seeking and Health Status: Results from the Medical Outcomes Study,” American Journal of Public Health 91, no. 11 (2001): 1889–1894.
9. For a review of the cost-sharing literature, see K. Davis, “Consumer-Directed Health Care: Will It Improve Health System Performance?” Health Services Research 39, no. 4, Part 2 (2004): 1219–1234.
10. P.F. Short and J.S. Banthin, “New Estimates of the Underinsured Younger than Sixty-five Years,” Journal of the American Medical Association 274, no. 16 (1995): 1302–1306; and P.J. Farley, “Who Are the Underinsured?” Milbank Quarterly 63, no. 3 (1985): 476–503.
11. Short and Banthin used detailed insurance plan descriptions and demographic information about the privately insured to estimate out-of-pocket expenses in the event of a catastrophic illness. They classified those insured all year as underinsured if out-of-pocket costs would exceed 10 percent of family income. For a discussion of concepts, see R. Bashshur, D.G. Smith, and R.A. Stiles, “Defining Underinsurance: A Conceptual Framework for Policy and Empirical Analysis,” Medical Care Review 50, no. 2 (1993): 199–218; and A.C. Monheit, “Underinsured Americans: A Review,” Annual Review of Public Health 15 (1994): 461–485.
12. Other studies of the underinsured have used access measures to indicate deficits in insurance. See R.J. Blendon et al., “Paying Medical Bills in the United States: Why Health Insurance Isn’t Enough,” Journal of the American Medical Association 271, no. 12 (1994): 949–951; J. Cook et al., “State-specific Prevalence Estimates of Uninsured and Underinsured Persons—Behavior Risk Factor Surveillance System, 1995,” Morbidity and Mortality Weekly Report 47, no. 3 (1998): 51–55; and K. Donelan, C.M. DesRoches, and C. Schoen, “Inadequate Health Insurance: Costs and Consequences,” Medscape General Medicine, 11 August 2000, www.medscape.com/viewarticle/408069 (19 November 2004).
13. Calculated according to the American Association for Public Research Definition 1.
14. Heart attack or heart disease, cancer, diabetes, or arthritis.
15. In the survey, 13 percent of respondents were uninsured part-year and another 13 percent, all year. We combined these two groups after analysis found similar responses on survey access measures.
16. Short and Banthin, “New Estimates of the Underinsured”; and Farley, “Who Are the Underinsured?”
17. The survey asked about respondents’ out-of-pocket health care spending during the year, including specific expense categories. Using reported expenses and family income, we calculated whether costs amounted to 10 percent or more of income or 5 percent or more of income for adults with incomes below 200 percent of poverty.
18. J.P. Newhouse, Free for All? Lessons from the RAND Health Insurance Experiment (Cambridge, Mass.: Harvard University Press, 1993).
19. Multivariate results predicting access problems, confidence, and satisfaction by insurance groups controlled for age, sex, health status, income, race and ethnicity, and other family characteristics. Results of this analysis are available from the authors upon request. Send e-mail to cs{at}cmwf.org.
20. J.R. Gabel et al., “Employers’ Contradictory Views about Consumer-Driven Health Care: Results from a National Survey,” Health Affairs, 21 April 2004, content.healthaffairs.org/cgi/content/abstract/hlthaff.w4.210 (5 May 2005).
21. S. Glied and D. Remler, The Effect of Health Savings Accounts on Health Insurance Coverage (New York: Commonwealth Fund, April 2005); and B. Lyke, C. Peterson, and N. Ranade, Health Savings Accounts, Pub. no. RL 32467 (Washington: Congressional Research Service, 23 March 2005).
22. Davis, “Consumer Directed Care.”
23. R.H. Brook et al., “Does Free Care Improve Adults’ Health? Results from a Randomized Controlled Trial,” New England Journal of Medicine 309, no. 23 (1983): 1426–1434; K.N. Lohr, “Use of Medical Care in the RAND Health Insurance Experiment: Diagnosis- and Service-specific Analyses in a Randomized Controlled Trial,” Medical Care 24, no. 9 Supp. (1986): S1–S87; and J.P. Newhouse, “Some Interim Results from a Controlled Trial of Cost Sharing in Health Insurance,” New England Journal of Medicine 305, no. 25 (1981): 1501–1507.
24. A.C. Monheit, “Persistence in Health Expenditures in the Short Run: Prevalence and Consequences,” Medical Care 41, no. 7 Supp. (2003): III53–III64.
25. C. Schoen et al., “Primary Care and Health System Performance: Adults’ Experiences in Five Countries,” Health Affairs, 28 October 2004, content.healthaffairs.org/cgi/content/abstract/hlthaff.w4.487 (19 November 2004); and P.S. Hussey et al., “How Does the Quality of Care Compare in Five Countries?” Health Affairs 23, no. 3 (2004): 89–99.
Cathy Schoen (cs{at}cmwf.org) is a vice president of the Commonwealth Fund in New York City. Michelle Doty is a senior analyst there; Sara Collins, a senior program officer; and Alyssa Holmgren, a program assistant.
DOI:
10.1377/hlthaff.w5.289
©2005 Project HOPE–The People-to-People Health
Foundation, Inc.
|
