Literature on the Effects of Uninsurance on Children

Studies on children were identified through systematic searches of the National Library of Medicine’s MEDLINE and the American Economic Association’s EconLit databases. The evidence base for establishing how and under what circumstances health insurance affects the health and functioning of children remains limited. The literature search identified many more studies examining the effects on children’s access to care and service use than on their health status or functioning. Studies on children were included in this review if they estimated the effects of health insurance coverage on validated access measures (i.e., having a usual source of care, having a preventive visit or any ambulatory care visit, having unmet health needs, receiving recommended immunizations, having a usual source for dental care, having received a preventive dental visit or any dental care, and having an unmet need for dental care) or health outcomes. Ultimately, 57 studies on children were selected for inclusion in this review.

Literature on the Effects of Uninsurance on Adults

Studies on adults were similarly identified through systematic searches of the National Library of Medicine’s MEDLINE database and the American Economic Association’s EconLit database. A key requisite for inclusion in the review was the demonstration of a distinct contribution to the research reviewed in the IOM’s previous report (IOM, 2002a). Potential contributions were considered in each of the following dimensions: (1) strength of study design and methodological rigor (e.g., quasi-experimental vs. observational design, inclusion of sensitivity analyses, handling of missing data); (2) quality of the data (e.g., longitudinal vs. cross-sectional, level of clinical detail, unique linkages); (3) importance of outcomes (e.g., validated measures of health vs. processes of care); and (4) external validity of results (e.g., findings generalizable to broader populations or previously unstudied diseases). Observational studies were excluded from the review unless they compared health outcomes for insured and uninsured adult subjects and investigated the sensitivity of results to statistical adjustments for observed demographic and socioeconomic characteristics. Ultimately, 42 studies on adults were selected for inclusion in this review.

Effects on Children’s Access to Health Care Services

Finding: Children benefit considerably from health insurance, as dem onstrated by evaluations of enrollment in Medicaid and the State Chil dren’s Health Insurance Program (SCHIP).

Finding: When children acquire health insurance, their access to health care services, including ambulatory care, preventive health care (e.g., immunizations), prescription medications, and dental care, improves.

Finding: When children acquire health insurance, they are much less likely to experience unmet health care needs, both when they are well or when they have special health care needs.

Although children in the United States are typically perceived as in good health relative to adults, certain conditions including asthma, diabetes, and obesity have become relatively common among children. Further, there is a population of particularly vulnerable children with special health care needs that require ongoing medical attention and other health-related services. More than 10 million children in the United States meet the federal definition of children with special health care needs—i.e., children “who have or are at increased risk for a chronic physical, development, behavioral, or emotional condition and who also require health and related services of a type or amount beyond that required by children generally” (American Academy of Pediatrics, 2008). Such children with conditions such as asthma, arthritis or other joint problems, autism, blood problems, Down syndrome, mental retardation/developmental delays, depression, diabetes, heart problems, cystic fibrosis, cerebral palsy, or muscular dystrophy require health and related services of a type or amount beyond that required by children generally.

Research linking health insurance and children’s access to care has flourished since the IOM’s last study, particularly regarding the potential benefits of enrolling in a publicly sponsored health insurance program. New evidence from well-designed studies draws on state-level expansions of the SCHIP and Medicaid programs in 14 states and for local programs in three different California counties. The 14 states—California, Colorado, Florida, Illinois, Iowa, Kansas, Louisiana, Massachusetts, Missouri, North Carolina, New Jersey, New York, Tennessee, and Texas—account for over 60 percent of the nation’s low-income children, represent all four census regions and major SCHIP program types, and vary with respect to program size and composition (Kenney, 2007).

The new evidence, discussed further below, strongly supports the finding that expansions in eligibility for Medicaid and SCHIP have produced gains in access to medical care for children targeted by the eligibility expansions, as well as the finding that positive spillover effects may arise for children who were already eligible for coverage. Well-designed evaluations of children’s participation in SCHIP, Medicaid, or county-based initiatives have generated consistent and robust evidence showing that children’s ac cess to health care services improves after children enroll in a public health insurance program. Regardless of the state or locality, with health insurance, children’s access to and use of preventive care (including immunizations), prescription medications, and dental care improves substantially. Importantly, these gains in children’s access to health services do not seem to depend on the particular type of the public program in place or on the composition of the target population for the program. Thus, for example, similarly positive and significant effects were found in states that have different SCHIP models (e.g., Medicaid expansions vs. separate non-Medicaid programs vs. combination programs).

National studies examining the general impacts of eligibility expansions under Medicaid and SCHIP find that acquiring health coverage increases children’s ambulatory care visits (Banthin and Selden, 2003; Currie et al., 2008; Davidoff et al., 2005). The study by Davidoff and colleagues also finds that eligibility expansions reduced children’s unmet health needs (Davidoff et al., 2005).

Similarly, evidence from 14 state and local studies have found that acquiring public health insurance coverage increases the likelihood that children have a usual source of care and that children receive preventive or other ambulatory visits, including immunizations and dental visits, and reduces the likelihood that children experience unmet health needs (Damiano et al., 2003; Dick et al., 2004; Eisert and Gabow, 2002; Feinberg et al., 2002; Fox et al., 2003; Howell et al., 2008a,b; Kempe et al., 2005; Kenney, 2007; Kenney et al., 2007; Moreno and Hoag, 2001; Slifkin et al., 2002; Szilagyi et al., 2004; Trenholm et al., 2005).

Improvements in Children’s Access to a Usual Source of Care. All but one study report significant improvements in having a usual source of care as a result of gaining public health coverage.² The magnitude of the impacts differs and ranges from increases of 20 to 30 percentage points or more (Howell et al., 2008b; Kenney, 2007; Kenney et al., 2007; Moreno and Hoag, 2001; Slifkin et al., 2002; Trenholm et al., 2005) to smaller increases in the likelihood of having a usual source of care (Damiano et al., 2003; Dick et al., 2004; Fox et al., 2003; Howell et al., 2008a; Szilagyi et al., 2004). Studies conducted with adolescents find that uninsured adolescents were significantly less likely to have a usual source of care and health care visits than insured adolescents and that the differences are larger than for younger children (Dick et al., 2004; Kenney, 2007; Klein et al., 2007; Probst et al., 2005; Slifkin et al., 2002). After enrolling in SCHIP, adolescents were more likely to have a range of ambulatory encounters, including preventive visits, dental visits, and specialist visits. Kenney finds significant increases for adolescents in the likelihood of having a usual source of care after enrolling in SCHIP (Kenney, 2007).

Improvements in Children’s Access to Preventive Care. Being enrolled in public coverage improves the likelihood that a child receives preventive care. The magnitude of the estimated impacts varied across the studies, though not as dramatically as the usual source of care estimates. Some studies (Fox et al., 2003; Howell et al., 2008b; Moreno and Hoag, 2001; Slifkin et al., 2002; Trenholm et al., 2005) find that coverage increases the likelihood of children’s receiving a preventive visit by over 15 percentage points;³ other studies find increases in the range of 8 to 13 percentage points (Dick et al., 2004; Eisert and Gabow, 2002; Kenney, 2007; Kenney et al., 2007; Szilagyi et al., 2004).⁴ Insurance coverage also increases the likelihood that children have an ambulatory care visit (Howell et al., 2008a,b; Kenney, 2007; Kenney et al., 2007; Szilagyi et al., 2004; Trenholm et al., 2005).

Evidence on children’s access to immunizations and dental visits further supports the positive relationship between health insurance and access to health care services. The majority of studies suggests a positive effect of insurance on immunization rates. Regardless of the immunization series examined, all but one of the studies finds that young uninsured children are less likely to be up-to-date on their immunizations than insured children, controlling for observed characteristics of the children (Allred et al., 2007; Dombkowski et al., 2004; Henderson et al., 2006; Joyce and Racine, 2005; Zhao et al., 2004).

Joyce and Racine controlled for selection into insurance by examining changes during the period of SCHIP implementation for all poor and near-poor children, compared to nonpoor children (Joyce and Racine, 2005). While immunization rates improved during the SCHIP implementation period for all children regardless of income, there was greater improvement for poor and near-poor children for the recently added varicella vaccine. No recent studies have shown that uninsured children have higher rates of preventable diseases, such as measles, compared to insured children. Still, the documented relationship between immunization and disease can be used to infer that children who lack timely immunizations are at greater risk of developing such diseases (Guerra, 2007).

Improvements in Children’s Access to Dental Care. The majority of studies on children’s access to dental care showed significant improvements; with health insurance, use of dental services increased from 16 to 40 percentage points (Damiano et al., 2003; Fox et al., 2003; Howell et al., 2008a,b; Kenney, 2007; Kenney et al., 2007; Lave et al., 2002; Lewis et al., 2007; Mofidi et al., 2002; Selden and Hudson, 2006; Trenholm et al., 2005; Wang et al., 2007). These results mirror the findings on children’s use of medical care, but the impact is sometimes greater for dental care access. Although the magnitude of effects varies somewhat depending on the study, the consistency of these findings across a range of studies—from local, to statewide, to national studies—increases their generalizability.

Reductions in Children’s Unmet Health Needs. As just discussed, well-designed evaluations of children’s participation in SCHIP, Medicaid, or county-based initiatives have generated consistent and robust evidence showing that children’s access to health care services improves after children enroll in a public health insurance program. Thus, children and adolescents who gain health insurance are much less likely to have unmet health care needs than those who are uninsured.

Numerous studies conducted in recent years demonstrate that once children acquire health insurance through a public program, they are significantly less likely to have unmet health needs (Damiano et al., 2003; Dick et al., 2004; Feinberg et al., 2002; Fox et al., 2003; Howell et al., 2008b; Kenney, 2007; Kenney et al., 2007; Slifkin et al., 2002; Szilagyi et al., 2004; Trenholm et al., 2005).

Parents with children who have acquired health insurance are much less likely to report that their child has an unmet need for prescription drugs, mental health or specialty care, vision care, and preventive care. Gaining coverage was associated with 1- to 12-percentage-point declines in children’s unmet need for prescription drugs (Damiano et al., 2003; Feinberg et al., 2002; Fox et al., 2003) and 10- to 14-percentage-point declines in unmet need for preventive care (Howell et al., 2008a; Szilagyi et al., 2004). After enrolling in SCHIP, adolescents experience declines in unmet needs for medical, specialist, hospital, and dental care, but not for mental and reproductive health care (Klein et al., 2007).

Uninsured children are generally at least twice as likely as children with dental insurance to have unmet need for dental care (Damiano et al., 2003; Davidoff et al., 2005; Feinberg et al., 2002; Fox et al., 2003; Howell et al., 2008a,b; Kenney, 2007; Kenney et al., 2007; Lave et al., 2002; McBroome et al., 2005; Mofidi et al., 2002; Szilagyi et al., 2004; Trenholm et al., 2005; Wang et al., 2007).

Nine studies of children with special health care needs indicate that uninsured children with special health care needs have higher unmet need and lower use of critical services such as genetic counseling, mental health care, and specialist services than insured children with special health care needs (Busch and Horwitz, 2004; Davidoff et al., 2005; Dick et al., 2004; Jeffrey and Newacheck, 2006; Kenney, 2007; Mayer et al., 2004; Porterfield and McBride, 2007; Wang and Watts, 2007; Yu et al., 2006). Two studies demonstrate that uninsured children with special health care needs are six to eight times more likely to have an unmet need for health care than their insured counterparts (Mayer et al., 2004; Yu et al., 2006). For children with special health care needs, being uninsured can have disastrous consequences (Box 3-3).

BOX 3-3

Ginny: A Girl with Special Health Care Needs Who Died When She Lost Her Health Insurance Coverage. Ginny was born with a congenital cardiac anomaly that was repaired successfully when she was 5 years old and funded through her Medicaid insurance. She (more...)

Two studies have shown that following enrollment in SCHIP, children with special health care needs experienced greater reductions in unmet health needs than other children in SCHIP (Davidoff et al., 2005; Kenney, 2007). Kenney’s evaluation of the effects of SCHIP in 10 states found that, after enrolling in SCHIP, children with special health care needs had significantly improved access to a wide range of health care services, including a usual source of care, preventive care, prescription medications, specialty and hospital care, and dental care (Kenney, 2007).

Effects on Adults’ Access to Health Care Services

Finding: Adults benefit substantially from health insurance for preven tive care when they are well and from early diagnosis and treatment when they are sick or injured.

Finding: Without health insurance, chronically ill adults are much more likely to delay or forgo needed health care and medications.

Finding: Without health insurance, adults are less likely to receive ef fective clinical preventive services.

Finding: The benefits of health insurance have been clearly demon strated through recent studies of the experiences of previously unin sured adults after they acquire Medicare coverage at age 65. These studies demonstrate that when previously uninsured adults gain Medi care coverage:

• Their access to physician services and hospital care, particularly for adults with cardiovascular disease or diabetes, improves.
• Their use of effective clinical preventive services increases.

Table 3-1 shows data based on tabulations from the 2005 Medical Expenditure Panel Survey (MEPS) on the prevalence of serious medical conditions among uninsured adults in the United States (AHRQ Center for Financing Access and Cost Trends, 2008). The prevalence of chronic diseases reported by uninsured adults in the United States is high. More than 40 percent of uninsured adults ages 19 to 64 reported having one or more chronic conditions, such as asthma, hypertension, depression, diabetes, chronic obstructive pulmonary disease (COPD), cancer, or heart disease in 2005. Because uninsured adults seek health care less often than insured adults, they are often unaware of underlying health problems such as hyper tension or hyperlipidemia (Ayanian et al., 2003). Nevertheless, uninsured adults report high rates of chronic disease (Table 3-1).

TABLE 3-1

Prevalence of Serious Medical Conditions Among Uninsured Adults Ages 19-64, 2005.

Data from the 2005 MEPS indicate that uninsured adults with chronic conditions are far less likely to use health care services than insured adults with the same chronic conditions. In 2005, as shown in Table 3-2, chronically ill adults who lacked health insurance had five to nine fewer health care visits per year than chronically ill adults who have health insurance. Uninsured adults with chronic illnesses were much more likely than their insured peers to go without any medical visits during the year—even when they were diagnosed with serious conditions such as asthma (23.4 of uninsured adults with no visits vs. 6.2 percent of insured adults), COPD (13.2 vs. 4.0 percent), depression (19.3 vs. 5.2 percent), diabetes (11.0 vs. 5.2 percent), heart disease (8.7 vs. 2.9 percent), or hypertension (12.7 vs. 5.3 percent).

TABLE 3-2

Comparison of the Use of Health Care Services by Insured and Uninsured Adults with Serious Medical Conditions, 2005.

Similarly, uninsured adults with asthma, cancer, COPD, diabetes, heart disease, or hypertension are at least twice as likely as their insured peers to say that they were unable to receive or had to delay receiving a needed prescription (Wilper et al., 2008).

Studies of the Effects of Health Insurance on Adults’ Access to Care. Important new research has emerged since the IOM last studied the impact of uninsurance on adults in 2002 (IOM, 2002a). These include various quasi-experimental studies that have evaluated what effects gaining Medicare coverage at age 65 has on access to clinical preventive services and general health care services, including visits to physicians and hospitalizations, among adults who previously lacked health insurance or were members of groups that were more likely to be uninsured prior to age 65 (e.g., adults of lower socioeconomic status or racial and ethnic minorities).

Improved Access to Important Clinical Preventive Services for Adults. Five recent well-designed, quasi-experimental studies have estimated the effects of health insurance coverage on adults’ use of important clinical preventive services (Busch and Duchovny, 2005; Card et al., 2004; Decker, 2005; McWilliams et al., 2003; Sudano and Baker, 2003).

In one of the quasi-experimental studies, Sudano and Baker used longitudinal survey data that provided detail on continuity of coverage and concluded that adults’ rates of mammography, Pap testing, cholesterol testing, and influenza vaccination decreased in a stepwise fashion with increasing number of episodes of uninsurance over a 4-year period (Sudano and Baker, 2003). For women who reported 0, 1, 2, and 3 episodes of uninsurance, for example, rates of mammography screening were 76.7 percent, 62.0 percent, 53.4 percent, and 34.7 percent respectively, suggesting dose-response effects of coverage on use of recommended clinical services.

In another study, Card and colleagues assessed changes in adults’ use of preventive care services that were associated with gaining Medicare eligibility at age 65 (Card et al., 2004). Using a quasi-experimental approach, the researchers used age trends in adults’ utilization of preventive care before age 65 to predict utilization of such care after age 65 and attributed any abrupt deviations from predicted trends occurring at age 65 to Medicare coverage. Several significant increases, ranging from 5 to 10 percentage points, in rates of influenza vaccination, cholesterol testing, mammography, or diagnosed hypertension occurred at age 65 for some groups of adults who were more likely to be uninsured before age 65 given their race, ethnicity, and educational attainment. However, the increases in the utilization of preventive care services were not consistently greater for all of these sociodemographic groups across all preventive services.

Decker, in a similarly designed analysis of a larger sample of over 250,000 women participating in the Behavioral Risk Factor Surveillance System surveys during 1991 to 2001, estimated rates of mammography more precisely and found increases at age 65 to vary significantly and more consistently across race, ethnicity, and educational attainment (Decker, 2005). The percentage of women reporting a mammogram in the prior 2 years increased by 2.6 and 4.8 percentage points among women with high school degrees and less than a high school education, respectively. Similarly, rates of mammography increased by 2.4 percentage points among white women, 4.4 percentage points among black women, and 7.5 percentage points among Hispanic women. In tests of differential effects, increases were significantly greater for women who were less educated or members of minority groups.

In another study, McWilliams and colleagues used longitudinal data from the Health and Retirement Study (HRS),⁵ a nationally representative longitudinal survey of adults and their spouses over the age of 50 in the continental United States. The researchers assessed the receipt of basic clinical services before and after Medicare eligibility at age 65 for adults who were continuously uninsured, intermittently uninsured, or continuously insured from age 60 to 64 (McWilliams et al., 2003). Differences in use of cholesterol testing and mammography between continuously insured and continuously uninsured adults were significantly reduced after age 65 by 17.7 and 15.3 percentage points, respectively. Differential effects of gaining coverage on service use were positive but smaller for intermittently uninsured adults, and increases in cholesterol testing after age 65 were greatest for continuously uninsured adults with hypertension or diabetes, in whom such testing to guide cardiovascular risk reduction is particularly important.

In the fifth study, Busch and Duchovny assessed the effects of state Medicaid eligibility expansions from 1995 to 2001 on Pap testing for previously uninsured mothers (Busch and Duchovny, 2005). They estimated that 29 percent of these uninsured mothers who had not been screened for cervical cancer were screened after they became eligible for Medicaid.

Improved Access to General Health Care Services for Adults. Three recent quasi-experimental analyses have examined the effects of health insurance coverage on adults’ general use of health care and contribute noteworthy findings on the effects of health insurance coverage on adults’ health outcomes (Card et al., 2004, 2008; Lichtenberg, 2002; McWilliams et al., 2007b). Sharing the same analytic strategy used in several analyses of preventive services, all three of the quasi-experimental studies of the effects of health insurance on the general use of care assessed the changes in health insurance coverage that occur after individuals reach age 65 due to nearly universal Medicare coverage among the U.S. elderly population.

In one of the three studies, Lichtenberg used cross-sectional data from the National Ambulatory Medical Care Surveys from 1973 to 1998 to examine annual per capita physician visits among adults (Lichtenberg, 2002). This study found an abrupt and persistent increase in use of physician visits among adults who were age 65 and older. A similar increase in hospital admissions among adults age 65 and older was also observed in data from the National Hospital Discharge Surveys during 1979 to 1992, but most of this surge in hospital admissions seemed to be due to the postponement of presumably elective admissions in the 2 years preceding Medicare eligibility.

In a second quasi-experimental study, Card and colleagues analyzed doctor visits reported in the National Health Interview Survey from 1997 to 2003 and hospital discharges from 1992 to 2002 in three states, disaggregated age profiles by sociodemographic predictors of coverage, and conducted more formal testing of trends (Card et al., 2004, 2008). Among individuals age 65 and older, the increase in routine doctor visits was especially pronounced among less educated adults and members of racial and ethnic minority groups who were more likely to be uninsured before age 65. Hospital admissions also increased sharply and persistently after age 65, but racial and ethnic differences in these increases varied by admission diagnosis and primary procedure performed.

McWilliams and colleagues, in a longitudinal analysis of the HRS from 1992 to 2004, found that near-elderly adults who were intermittently or persistently uninsured before reaching Medicare eligibility at age 65 reported significantly greater increases in doctor visits, hospital admissions, and total medical expenditures after age 65 than adults who had continuous private health insurance coverage before age 65 (McWilliams et al., 2007b). These investigators found that the differential increases in doctor visits, hospital admissions, and total medical expenditures were concentrated among adults who suffered from cardiovascular disease (hypertension, heart disease, or stroke) or diabetes—conditions for which there are many effective treatments to prevent costly complications. Moreover, previously uninsured Medicare beneficiaries with cardiovascular disease or diabetes reported relative increases in doctor visits (13 percent), hospitalizations (20 percent), and total medical expenditures (51 percent) compared with previously insured beneficiaries who were otherwise similar across observed characteristics at age 59 to 60 and had comparable generosity of coverage after age 65 (i.e., supplemental insurance and prescription drug coverage). These persistently elevated health care needs suggest that uninsured near-elderly adults with chronic conditions enter the Medicare program at age 65 with greater morbidity than they would if they had previously had health insurance.

In an observational study, Ward and Franks used longitudinal data on adults ages 21 to 64 from the 2000 to 2003 MEPS and found that total medical expenditures were higher for previously uninsured adults after they gained coverage but were not significantly different from expenditures among continuously insured adults (Ward and Franks, 2007). However, the sample of uninsured adults who experienced coverage gains was small (N = 385), and the changes in insurance status were voluntary and therefore could have been caused by changes in health.

Effects of Health Insurance on Children’s Health Outcomes

Finding: With health insurance, children receive more timely diagno sis of serious health conditions, experience fewer avoidable hospital izations, have improved asthma outcomes, and miss fewer days of school.

There are 13 recent studies on the health effects of health insurance coverage for children, including 5 studies that used quasi-experimental methods (Aizer, 2007; Bermudez and Baker, 2005; Cousineau et al., 2008; Currie et al., 2008; Howell et al., 2008a). These studies suggest that health insurance is beneficial for children in several ways, resulting in more timely diagnosis of serious health conditions, fewer avoidable hospitalizations, better asthma outcomes, and fewer missed school days (Aizer, 2007; Bermudez and Baker, 2005; Cousineau et al., 2008; Currie et al., 2008; Damiano et al., 2003; Fox et al., 2003; Froehlich et al., 2007; Howell and Trenholm, 2007; Howell et al., 2008a,b; Maniatis et al., 2005; Szilagyi et al., 2004, 2006).

The study by Maniatis and colleagues, for example, assessed children when first diagnosed with diabetes (Maniatis et al., 2005). They found that uninsured children were less likely to have their conditions diagnosed as early as insured children. Among the children diagnosed with diabetes, the uninsured children were more likely to present with severe and life-threatening diabetic ketoacidosis.

Three quasi-experimental studies found significant reductions in hospitalizations related to ambulatory care sensitive conditions (ACSC) for children enrolled in Medicaid or SCHIP (Aizer, 2007; Bermudez and Baker, 2005) or county-based health insurance programs (Cousineau et al., 2008). Aizer, in a quasi-experimental analysis of state hospital discharge files, Medicaid enrollment, and U.S. Census data, found that a 10-percentage-point increase in Medicaid enrollment led to about a 3-percentage-point reduction in ACSC admissions. Similarly, Szilagyi and colleagues reported improvements in asthma-related outcomes for New York children after they enrolled in SCHIP: the rate of asthma-related hospital stays for these children dropped from 11 percent to just 3 percent (Szilagyi et al., 2006). These investigators also reported significant declines in asthma-related emergency room visits among these children after they enrolled in SCHIP.

Effects of Health Insurance on Adults’ Health Outcomes

Finding: Uninsured adults who acquire Medicare coverage at age 65, particularly if they have cardiovascular disease or diabetes, experience substantially improved trends in health and functional status.

Finding: Without health insurance, adults with cardiovascular disease or cardiac risk factors are less likely to be aware of their conditions, their conditions are less likely to be well controlled, and they experience worse health outcomes.

Finding: Without health insurance, adults are more likely to be di agnosed with later-stage breast, colorectal, or other cancers that are detectable by screening or by symptom assessment by a clinician. As a consequence, when uninsured adults are diagnosed with such cancers, they are more likely to die or suffer poorer outcomes.

Finding: Without health insurance, adults with serious conditions, such as cardiovascular disease or trauma, have higher mortality.

Finding: When previously uninsured adults gain Medicare coverage:

• They experience substantially improved trends in health and functional status.
• Their risk of death when hospitalized for serious conditions declines.

Improved Health and Physical Functioning of Adults. Several quasi-experimental studies have advanced our understanding of the effects of health insurance coverage on adults’ overall health and physical functioning. Many of these studies were similar in their analytic approach, recognizing near-universal Medicare coverage after age 65 as a source of variation in insurance status that allows effects on health to be estimated more rigorously.

Card and colleagues, in a quasi-experimental analysis of cross-sectional data from the National Health Interview Survey during 1992 to 2001, found significant improvements in self-reported general health among adults in sociodemographic groups who experienced the largest gains in insurance coverage at age 65 (Card et al., 2004). The gap between more educated white adults and less educated black and Hispanic adults was narrowed after age 65 by a 12 percent relative reduction. In sensitivity analyses, Card and colleagues found no evidence for discontinuities in employment status, marital status, geographic location, or family income that might explain discontinuities in general health, suggesting these reduced disparities were due to increases in coverage after age 65 (Card et al., 2004).

In a related study, Decker and Remler used Canadian adults as international controls (Decker and Remler, 2004). Age profiles of general health status were constructed from National Health Interview Survey data for U.S. adults and from the National Population Health Survey for Canadian adults ages 55 to 74 and compared by income, country, and age (65 or older vs. under 65) in a quasi-experimental approach. Among the near-elderly age group, low-income adults in the United States were 15 percentage points more likely than high-income adults to be in fair or poor health, compared to an 8 percent absolute difference between low- and high-income adults in Canada. Among adults age 65 or older, this 7 percent international difference was reduced to 3 percent, suggesting that near-universal Medicare coverage reduced the excess risk of fair or poor health among low-income adults by 4 percent—or equivalently, that not having insurance explained more than half of the health disparity between low-income and high-income nonelderly adults in the United States.

In these studies of cross-sectional data, effects of Medicare coverage on other measures of health were not assessed, and uninsured adults, particularly those with specific conditions who might benefit most from coverage, could not be longitudinally followed as they became eligible for Medicare. Several subsequent studies used longitudinal data from the HRS to assess the effects of gaining Medicare coverage on the health of previously uninsured adults.

In an observational study, Baker and colleagues found that adults without health insurance for 1 to 2 years prior to age 65 were more likely to report a major decline in general health or a new functional limitation in their first interview after age 65 (Baker et al., 2006a). In subsequent interviews, after at least 2 years of Medicare eligibility, previously uninsured adults no longer reported significantly higher rates of these health declines.

In a quasi-experimental study, Polsky and colleagues used more recent data from the HRS to follow participants longer after age 65 and compared trajectories in self-reported general health between insured and uninsured near-elderly adults before and after age eligibility for Medicare (Polsky et al., 2006). Health declines became less frequent after age 65 for previously uninsured adults, such that the proportion of these adults reporting excellent or very good health after age 65 was nearly 8 percent higher than expected based on their trajectory before age 65. However, previously insured adults also reported a significant improvement in their health trajectory after age 65. This change in trajectory was slightly smaller, suggesting a net positive health effect for previously uninsured adults attributable to Medicare coverage, but the differential effect was not significant. No other measures of health were included in this analysis.

Using the same data and similar quasi-experimental methods, McWilliams and colleagues compared changes in health trends reported by previously insured and uninsured adults after age 65 for a more comprehensive set of six self-reported general, physical, and mental health measures and a summary measure of these component items (McWilliams et al., 2007a). In comparison with adults who were continuously insured, adults who were intermittently or persistently uninsured from age 55 to age 64 reported significantly improved health trends after age 65 for the summary health measure and several component measures. Analyses of agility and mobility scores also suggested that gaining Medicare coverage improved trends in physical functioning for previously uninsured adults.

Two recent studies employed another quasi-experimental approach (instrumental variables methods) to estimate the effects of private health insurance on self-reported general health and physical functioning for near-elderly adults using data from the HRS (Dor et al., 2006; Hadley and Waidmann, 2006). Relative to no insurance coverage, both studies found statistically significant and substantial effects of private coverage on health that were much larger than those obtained from observational studies using more basic statistical models. In a third study that used instrumental variables techniques, Pauly analyzed the 1996 MEPS to assess the effects of health insurance coverage on medical expenditures, utilization, access to care, and general health status among nonpoor young women (Pauly, 2005). Although results were not statistically significant for the health status outcome, effects on expenditures, utilization, and access to care were significant and larger in magnitude than predicted by standard statistical comparisons commonly performed in observational studies.

Worse Outcomes for Adults with Serious Chronic and Acute Conditions. Uninsurance has profound health implications for the estimated 40 percent of uninsured adults age 19 to 64 who have chronic disease (Table 3-1). As discussed below, new studies provide compelling evidence demonstrating that health insurance is the most beneficial for adults who have chronic conditions such as hypertension, diabetes, and cancer, as well as serious injury, heart attack, stroke, and other acute conditions for which effective treatments are available (Table 3-4). Uninsured adults are more likely than insured adults to be unaware that they have an asymptomatic chronic condition, such as high blood pressure or early-stage cancer. Yet such underlying disease is often responsive to timely preventive and diagnostic services and appropriate management and treatment.

TABLE 3-4

Recent Research Findings on the Harmful Effects of Uninsurance for Adults with Selected Acute Conditions and Chronic Disease.

Delayed Diagnosis and Worse Health Outcomes for Adults with Chronic Conditions Such as Cardiovascular Disease or Diabetes. Ayanian and colleagues analyzed findings from the National Health and Nutrition Examination Survey (NHANES) III, a nationally representative assessment of the health and nutrition of 10,946 insured and uninsured adults ages 25 to 64 (Ayanian et al., 2003). They found that uninsured adults were much more likely than insured adults to have undiagnosed hypertension (high blood pressure) and hypercholesterolemia (high cholesterol), largely because of their limited access to health care services. A later analysis of NHANES III data analyzed outcomes for adults with diagnosed and treated hypertension (Duru et al., 2007). This study found that in comparison with insured adults, uninsured adults had substantially poorer blood pressure control even when treated.

The health implications of underdiagnosed and poorly managed hypertension can be profound. An observational analysis of longitudinal data from the Atherosclerosis Risk in Communities Study compared cardiovascular outcomes among insured and uninsured adults ages 45 to 64 in four U.S. communities from 1987 to 2000 (Fowler-Brown et al., 2007). Uninsured adults were more likely to be unaware of clinically determined hypertension, diabetes, and hypercholesterolemia, and those with hypertension were more likely to have inadequate blood pressure control—echoing the NHANES results described above. Uninsured adults had higher adjusted risks of stroke and death by 65 percent and 26 percent, respectively.

Uninsured adults who are hospitalized for acute ischemic stroke are far more likely than privately insured adults to suffer extremely poor outcomes. A recent observational analysis of the Nationwide Inpatient Sample found that among patients hospitalized for acute ischemic stroke, uninsured patients had higher levels of neurological impairment, a 24 percent higher mortality risk, and among those with intracerebral hemorrhage, 56 percent higher mortality risk (Shen and Washington, 2007). Because there are few therapeutic interventions for acute stroke, particularly acute intracerebral hemorrhage, these findings suggest that uninsured adults suffer more severe strokes because of poorer management of cardiovascular risk factors, such as hypertension (high blood pressure) and hypercholesterolemia (high cholesterol), and fewer preventive carotid endarterectomies when indicated for asymptomatic or symptomatic carotid stenoses.

Quasi-experimental research by McWilliams and colleagues has shown that previously uninsured adults report substantial improvements in cardiovascular health after turning age 65 and acquiring Medicare coverage (McWilliams et al., 2007a). The researchers analyzed data from the HRS during 1992 to 2004. Study participants were asked to rate their general health status, changes in general health, mobility, agility, pain, and symptoms of depression. The study compared the health status reports of previously insured and uninsured adults with cardiovascular disease or diabetes, finding that, by age 70, the disparities between the two groups had dropped by 50 percent. The previously uninsured adults reported significantly improved general, physical, and mental health as well as mobility and agility. Although these outcomes were not disease-specific, self-reported physical functioning correlates strongly with clinical complications of cardiovascular disease such as angina, dyspnea, neuropathy, visual impairment, myocardial infarction, and stroke (Guralnik et al., 1993; McHorney et al., 1993). Fur thermore, previously uninsured adults also reported better outcomes related to myocardial infarctions, angina that limited activities, and hospitalization for congestive heart failure.

In addition, glycemic control was compared in two groups of diabetic adults: those under age 65 and those age 65 or older (with Medicare) (McWilliams et al., 2007a). Uninsured adults in the younger age group had significantly worse glycemic control compared to their insured counterparts. At age 65 or older, however, the outcomes were similar for the previously insured and previously uninsured adults. Finally, differential increases in doctor visits and hospital admissions after age 65 were also concentrated among adults with cardiovascular disease or diabetes (McWilliams et al., 2007b).

Thus there appear to be several ways in which acquiring Medicare coverage improves the health of uninsured adults with chronic disease, among them (1) improving access to prescription medications; (2) improving diabetes management, especially glycemic control; and (3) increasing access to outpatient and hospital care for adults with cardiovascular disease or diabetes. Access to prescription drugs appears to be especially important—previously uninsured adults who gained prescription drug coverage experienced the greatest health improvement. In sum, this body of observational and quasi-experimental research suggests that gaining health insurance coverage improves the health of previously uninsured adults with cardiovascular disease or diabetes, as improved access to care, greater use of effective procedures and medications, and better management of these conditions alleviates symptoms, maintains functioning, and prevents or postpones complications.

Delayed Diagnosis and Worse Health Outcomes for Adults with Cancer. A large body of observational research completed before 2002 demonstrated that uninsured adults with cancer are diagnosed at more advanced stages of disease, have poorer outcomes, and die sooner, even after adjusting for stage of disease (Box 3-4) (IOM, 2002a). These findings suggest deficits in screening and diagnostic services, as well as in curative and palliative procedures and therapies for uninsured adults with prevalent cancers, among them breast, cervical, colorectal, and prostate cancer and melanoma.

BOX 3-4

Sheila: Unable to Obtain Breast Cancer Treatment After She Lost Her Health Insurance. Sheila had been diagnosed with breast cancer and initially treated while her husband still received health insurance. But for 2 years after her mastectomy, she went (more...)

Two recent observational studies used data from the U.S. National Cancer Database to generalize these associations to the national population, as most prior studies of insurance coverage and cancer outcomes relied on state or regional cancer registries (Halpern et al., 2008; Ward et al., 2008). One study found that uninsured patients with cancers diagnosed between 1998 and 2004 were more likely than privately insured patients to be diagnosed at advanced stages of disease, particularly those with cancers that can be detected early by screening (e.g., breast or colorectal cancer) or by symptom assessment (e.g., melanoma or bladder cancer) (Halpern et al., 2008). In contrast, no significant differences in stage at diagnosis were found between insured and uninsured adults with cancers that typically become clinically apparent only at late stages and for which there are no effective screening tests (e.g., ovarian or pancreatic cancer).

Another study found that after adjusting for cancer stage, 5-year survival rates for uninsured adults were significantly lower than for privately insured adults diagnosed with breast or colorectal cancer—two prevalent cancers for which there are not only effective screening tests, but also treatments demonstrated to improve survival (Ward et al., 2008). Similar associations between insurance status, stage at diagnosis, and survival have also been demonstrated for laryngeal and oropharyngeal cancer in several recent observational studies using these data (Chen and Halpern, 2007; Chen et al., 2007a,b).

Increased Risk of Death Among Adults Hospitalized with Serious Condi tions. Since 2002, three quasi-experimental studies have more rigorously assessed the effects of insurance coverage on mortality among adults with a variety of acute conditions, such as acute myocardial infarction, COPD or asthma exacerbation, hip fracture, respiratory failure, severe trauma, and stroke (Card et al., 2007; Doyle, 2005; Volpp et al., 2003).

Volpp and colleagues assessed mortality rates for insured and uninsured patients hospitalized for acute myocardial infarction in New Jersey before and after state market reforms in 1994 reduced subsidies for hospital care for the uninsured and changed the hospital payment system to price competition from one in which reimbursement was based on costs (Volpp et al., 2003). The research team performed difference-in-differences analyses of New Jersey hospital discharge data, in which discharge data from New York and the Nationwide Inpatient Sample were used to control for secular trends. While not directly addressing the effects of losing or gaining health insurance, this study sheds light on what happens to uninsured patients hospitalized with cardiovascular disease when changes in reimbursement policy restrict hospitals’ ability to recoup the costs of providing care to uninsured adults.

Volpp and his colleagues found no significant changes in mortality from acute myocardial infarction for insured patients in New Jersey in comparison to patients in New York or the nation prior to 1994. In contrast, the absolute mortality rate increased differentially after the reform by 3.7 percent to 5.2 percent for uninsured patients in New Jersey compared to uninsured patients in New York, representing a 41 percent to 57 percent relative increase over their baseline death rate of 9.1 percent before the reform. Concurrent relative decreases in rates of cardiac procedures were also observed for these uninsured patients. These findings provide strong evidence that lack of health insurance coverage exposes uninsured patients with acute myocardial infarction to poorer quality of care and higher mortality risks when providers are reimbursed less for uncompensated care or are unable to use profits from insured patients to cover the costs.

In a similarly designed quasi-experimental analysis of discharge data in New York and New Jersey, Volpp and colleagues found relative increases in mortality for uninsured New Jersey patients admitted for congestive heart failure and stroke when compared to uninsured New York patients with these conditions (Volpp et al., 2005). On the other hand, they found mortality trends during the 1990 to 1996 period to be similar in New Jersey and New York for hospitalized patients with hip fracture, gastrointestinal bleeding, pulmonary embolism, or pneumonia, regardless of the patients’ insurance status.

Card and colleagues, in another quasi-experimental study of state discharge data, assessed the effects of near-universal Medicare coverage after age 65 on mortality among acutely ill patients in California who were hospitalized between 1992 and 2002 (Card et al., 2007). To avoid a nonrepresentative sample of uninsured adults under age 65, the analysis focused on serious acute conditions or acute exacerbations of chronic conditions, including acute myocardial infarction, stroke, respiratory failure, COPD or asthma exacerbation, hip fracture, and seizure. A quasi-experimental analy sis of mortality rates identified an abrupt decrease of 1 percentage point in 7-day mortality at age 65, suggesting that Medicare coverage reduced the overall death rate for these acutely ill patients by 20 percent. The mortality effect persisted for at least 2 years after admission, suggesting a lasting impact of increased use of beneficial procedures and medications. The decline in mortality was too large to be explained by changes in cross-sectional rates of uninsurance from age 64 to 65, suggesting that near-elderly adults who have limited public or private coverage may also benefit from Medicare coverage.

Doyle conducted an observational study of adults injured in severe automobile accidents in Wisconsin (Doyle, 2005). By focusing on seriously injured drivers who were unable to participate in their initial treatment decisions, this analysis successfully addressed any differences in care-seeking behaviors between insured and uninsured adults that might bias the findings. The study found that, compared to privately insured patients, the uninsured crash victims received 20 percent less care (i.e., especially more costly procedures and services) and had a substantially higher mortality rate—an increase of 1.5 percentage points above the mean rate of 3.8 percent. The study controlled for an array of potentially confounding factors and included sensitivity analyses to test the robustness of the results.

These studies of hospitalized patients suggest that health insurance coverage may not only affect patients’ demand for health care services, but also provider behaviors in delivering care. Therefore, coverage expansion may not only improve outcomes for acutely ill patients by reducing delays before needed care, but also by allowing providers to offer effective but costly procedures and treatments at more equitable rates.

In another observational study, Hadley assessed insured and uninsured adults medical care and health status after unintentional injuries or newly diagnosed chronic conditions (Hadley, 2007). Using data from MEPS during 1997 to 2004, he found that, compared with insured adults, the uninsured adults received significantly fewer health care services, were less likely to fully recover, and more likely to report subsequent declines in health status.

Higher Mortality Rates Overall. Prior to 2002, two observational studies provided evidence that uninsured adults die at younger ages than their privately insured counterparts (IOM, 2002a). Of these two studies, the analysis that adjusted for more demographic, socioeconomic, and health characteristics estimated that the relative risk of death over 13 to 17 years was 25 percent greater for adults who were uninsured at baseline than for adults who were privately insured (Franks et al., 1993). Two subsequent observational analyses of data from the HRS estimated this increased relative risk to be 35 percent to 43 percent for uninsured near-elderly adults after controlling for even more predictors of mortality (Baker et al., 2006a; McWilliams et al., 2004). The association between uninsurance at baseline and subsequently higher mortality risk was particularly strong among near-elderly adults who were white, had low incomes, or had diabetes, hypertension, or heart disease (McWilliams et al., 2004). A sensitivity analysis demonstrated that the explanatory effect of an unmeasured predictor would have to be greater than the impact of smoking on mortality differences between insured and uninsured adults in the study for the increased risk among the uninsured to lose statistical significance. Another related study also found lack of health insurance was associated with major health declines, but not an increased risk of death within 2 years, suggesting that premature death is likely to be a long-term rather than short-term consequence of uninsurance among near-elderly adults (Baker et al., 2006b).

Because mortality generally represents a longer-term outcome for all but the severely or acutely ill, quasi-experimental analyses designed to identify abrupt discontinuities in mortality rates or even linear trends in mortality may not be suitable for estimating the effects of health insurance on mortality in the general population. Lichtenberg used life tables produced by the Social Security Administration and found a dramatic drop in the growth rate in annual probabilities of death for adults beginning at age 65 (Lichtenberg, 2002). However, in a subsequent analysis of National Center for Health Statistics Multiple Cause of Death files, Card and colleagues found no evidence of a deceleration in mortality rates at age 65 (Card et al., 2004). Similar assessments of the introduction of Medicare in 1965 found no discernable impact on mortality for beneficiaries (Card et al., 2004; Finkelstein and McKnight, 2005), although many subsequent medical advances have improved the effectiveness of health care for elderly adults in the United States.

Because of the often delayed effects of health services on survival, these null findings from some types of quasi-experimental studies should be interpreted with caution and are not necessarily inconsistent with positive findings from the observational comparisons described above. Indeed, in an instrumental variables analysis of longitudinal data, Hadley and Waidmann estimated that with universal health coverage, the absolute death rate for nonelderly adults could decrease from 6.7 percent to 3.9 percent (Hadley and Waidmann, 2006).