Mixed Signals: Trends in Americans' Access to Medical Care, 2007-2010

Mixed Signals: Trends in Americans' Access to Medical Care, 2007-2010

Tracking Report No. 25
August 2011
Ellyn R. Boukus, Peter J. Cunningham

Likely as a consequence of the deep economic downturn and the resulting drop in healthcare utilization, the share and total number of Americans who reported forgoing or postponing necessary medical care fell modestly between 2007 and 2010, based on results from the Center for Studying Health System Change's (HSC) nationally representative 2010 Health Tracking Household Survey. Even as the uninsured population grew, just over one in six Americans -- roughly 52 million individuals -- indicated they did not receive or delayed required medical care in 2010, a decrease from one in five -- about 58.6 million people -- in 2007. This improvement was largely attributable to fewer access difficulties among insured individuals, likely reflecting recession-driven reductions in the use of medical services. Nonetheless, the disparity in access between those with and without insurance widened in 2010 relative to 2007, particularly for lower-income populations and those coping with health conditions. Among those who experienced difficulties obtaining care, the expense of treatment became an even larger obstacle than in earlier years. At the same time, fewer individuals faced health system-related barriers such as securing timely physician appointments, possibly a reflection of additional capacity freed up by diminished demand.

Fewer Access Problems, More Uninsured

The count and proportion of uninsured Americans surged between 2007 and 2010 -- climbing from 42.8 million to 51.7 million -- as the recession eroded employer-based health coverage. Ordinarily, such a steep rise in uninsurance would coincide with worsening access to care overall, given that uninsured individuals experience access barriers at much higher rates than their insured counterparts. Yet in 2010, roughly 17 percent of the U.S. population -- approximately one in six people -- said they had not received or had postponed needed medical services in the preceding 12 months, a decline from the 20 percent -- or one in five -- who reported similar problems in 2007.

In 2010, about 19.7 million people went without needed care and approximately 32.3 million postponed seeking treatment, totaling 52 million individuals with access problems, according to HSC's nationally representative 2010 Health Tracking Household Survey. The share of Americans reporting unmet needs fell by 1.3 percentage points between 2007 and 2010 (from 7.8% to 6.5%). Adjusted for population growth, this translates to roughly 4 million fewer people going without necessary care in 2010 than would have occurred had the unmet-need rate held steady.

Reductions in unmet needs among the insured population were primarily responsible for the overall decrease and more than compensated for the growth in uninsured people with unmet medical needs. The percentage of insured Americans reporting an unmet medical need dropped by more than a quarter -- from 6.2 percent to 4.5 percent -- between 2007 and 2010. Concurrently, the rate of unmet need among the uninsured edged down slightly from 17.5 percent to 16.6 percent, though this change was not statistically significant. While unmet need did not worsen for the uninsured between 2007 and 2010, the gap between insured and uninsured populations in terms of unmet need grew wider in 2010 compared to 2007.

Additionally, fewer individuals delayed necessary medical care in 2010 relative to 2007. Among insured Americans, the share reporting delayed care dropped from 10.7 percent in 2007 to 9.6 percent in 2010. Uninsured individuals experienced an even steeper decline: the proportion postponing care fell 4.1 percentage points, from 20.5 percent to 16.4 percent.

Notwithstanding the recent improvement, the number and proportion of Americans struggling to obtain needed medical care was still elevated in 2010 compared to 2003, owing to the sharp increase in access problems between 2003 and 2007 for both insured and uninsured populations.

Declining Unmet Needs, Decreasing Demand

The drop in unmet needs among insured individuals most likely reflects both reduced demand for healthcare services and an easing of system capacity pressures. During the economic downturn, many Americans cut back on healthcare utilization, including physician visits, prescription medications, and medical procedures. Health Tracking Household Survey data reveal a 4 percent decrease in physician visits per person between 2007 and 2010, following an 11 percent rise between 2003 and 2007.

The recent decline in utilization reflects the growing uninsured population -- who visit physicians far less often than insured individuals -- a sharp reduction in utilization among the uninsured between 2007 and 2010, and stable utilization among the insured between 2007 and 2010 after an increase from 2003 to 2007. Lower demand for services means fewer opportunities for people to encounter access barriers. Simultaneously, freed-up capacity for those who do seek care makes them less likely to face obstacles such as lengthy appointment queues or extended in-office wait times. This reduced strain on system capacity may be especially important in explaining the decline in delayed care between 2007 and 2010.

Lower-Income and Sicker People Face More Access Problems

Unmet medical needs were more prevalent among individuals with low or moderate incomes than among higher earners, regardless of whether they had insurance. Overall in 2010, people earning below 200 percent of the federal poverty level -- $44,100 for a family of four -- were 3.1 times as likely to report an unmet need compared with those at or above 400 percent of poverty (9.3% versus 3%). This gap had widened since 2007, when lower-income people were only 2.2 times as likely as higher earners to forgo care. Between 2007 and 2010, unmet needs fell for high-income individuals -- from 5 percent to 3 percent -- but remained essentially flat for those in the low- and moderate-income brackets.

Within each income tier, uninsured people were approximately three times as likely as those with insurance to report going without necessary care. Additional analysis indicates that the access disparity between insured and uninsured individuals grew wider in 2010 for lower-income people but not for those with moderate or higher incomes.

Individuals who described their health as fair or poor were also substantially more likely to report forgoing needed medical care compared with those in good, very good, or excellent health (16.9% versus 4.6%). This is partly because people with greater health needs use more care and therefore encounter more chances to face access barriers. Between 2007 and 2010, unmet needs dropped 1.6 percentage points -- from 6.2 percent to 4.6 percent -- among the healthier group and held steady for those in fair or poor health.

However, the access trends for insured and uninsured people diverged markedly among those in fair or poor health. Unmet needs grew for uninsured individuals in fair or poor health while declining for insured individuals, though these shifts were not statistically significant owing to small sample sizes for persons in fair or poor health. Still, the access gap between insured and uninsured individuals in fair or poor health expanded, with unmet need 2.5 times higher among the uninsured than the insured in 2010, compared to 1.7 times higher in 2007.

Children's Access Stable

Children continued to experience the fewest unmet medical needs among all age groups, primarily because most children are healthy and have comparatively few healthcare requirements, and also because a larger proportion of children are covered by insurance. Following a dip in access during the mid-2000s, children's access improved slightly between 2007 and 2010. In 2010, approximately 3 percent of children did not receive needed medical care, and another 2 percent experienced delayed care -- the latter figure being nearly half of the 2007 level.

Children in families earning 400 percent of the poverty level or more -- $88,200 for a family of four -- had the fewest access problems, with just over 1 percent reporting an unmet need. As with the general population, declines in delayed care were most pronounced among children in families earning less than 200 percent of poverty. Only 2 percent of children in this income range delayed care in 2010, a decrease from 5.5 percent in 2007.

Cost Concerns Rise, System Barriers Down

Among the roughly 52 million Americans who reported an access problem, cost remained the primary obstacle to care and became an even greater deterrent over time. In 2010, 75 percent of people who went without or delayed needed treatment cited concerns about cost, an increase of 6 percentage points since 2007.

While uninsured people continued to voice the greatest concern about expense -- with more than 95 percent pointing to cost as a barrier -- worries about care costs also escalated among insured individuals facing access problems. Two-thirds of these insured individuals cited cost concerns in 2010, up from 61 percent in 2007. Even as growth in overall U.S. health spending moderated in recent years, expenditures continued to outpace income growth, and many employers raised patient cost-sharing requirements and scaled back the comprehensiveness of coverage in response to the weak economy. These factors, along with higher unemployment, may have fueled growing cost worries among insured individuals.

As cost concerns grew more prominent, health system-related barriers became less significant for both insured and uninsured individuals. More than half of insured people with an access problem cited a health system barrier in 2010, a decline of 11 percentage points from 2007. Roughly one-quarter of uninsured people encountered a system-related barrier in 2010, down 16.2 percentage points from 2007. Among all individuals citing a health-system obstacle, the most notable declines were: inability to secure a timely appointment (10.2 percentage point decrease); excessive travel time to reach the provider (6.0 percentage point decrease); inability to visit the provider during office hours (5.7 percentage point decrease); and difficulty getting through on the telephone (5.5 percentage point decrease).

The reduction in system-related barriers between 2007 and 2010 contrasted sharply with the 2003 to 2007 period, when such barriers rose amid significantly higher demand for care and the resulting severe capacity strain. In contrast, the recent decrease in system-related barriers likely reflects freed-up resources resulting from lower demand for medical services. Moreover, as financial pressures intensify during economic hardship and elevated unemployment, individuals who decide against seeking care because of cost are prevented from encountering additional insurance- or system-related barriers. For instance, a patient who opts not to call for an appointment will not have the opportunity to report difficulty finding a doctor, reaching the office by phone, or scheduling a convenient appointment.

Implications

Following a steep deterioration in access to medical care between 2003 and 2007, Americans' access stabilized in 2010 for both children and adults. However, the overall trend masks the reality that the access disparity between insured and uninsured populations grew between 2007 and 2010, especially among low-income individuals and those with health conditions.

Under the 2010 Patient Protection and Affordable Care Act (PPACA), Medicaid eligibility was set to expand to cover individuals earning up to 138 percent of poverty, and subsidies would be available to those earning under 400 percent of poverty to purchase coverage through health benefit exchanges beginning in 2014. This study's findings demonstrate that these provisions indeed target populations that disproportionately experienced access difficulties in 2010. Uninsurance rates were highest among low and moderate earners, and among the uninsured, more than 86 percent had incomes below 400 percent of poverty. Furthermore, the proportion of uninsured individuals experiencing access problems was greatest for those earning less than 400 percent of poverty, with 30 to 33 percent reporting an unmet need or delayed care. Even among uninsured individuals in the highest income bracket, nearly one-quarter reported an access problem, indicating that such difficulties were not confined to the poorest uninsured.

Because uninsured and low-income individuals are more likely to seek low- or no-cost care, rising unmet needs among these groups may point to strain on safety net resources. Recent HSC research shows that patient demand for safety net services surged during the economic downturn as more people lost insurance or enrolled in Medicaid. Yet to a large degree, community health centers and safety net hospitals were able to absorb this additional demand thanks to federal stimulus funding. Increases to the federal medical assistance percentage (FMAP) for state Medicaid programs, enhanced Medicaid disproportionate share hospital (DSH) payments, maintenance-of-effort requirements for states receiving federal stimulus funding, and expanded grants to federally qualified health centers helped ease pressures on local and state budgets that might otherwise have caused more drastic safety net reductions. Without these measures, it is possible that access would have worsened more significantly for vulnerable populations. With enhanced federal Medicaid matching funds expiring on June 30, 2011, and ongoing state and local budget deficits, the loss of federal support may trigger an increase in access problems.

While access problems diminished between 2007 and 2010, this was most likely driven by an overall reduction in demand for health services that loosened health system capacity constraints for patients seeking care. During a period of economic hardship for many, patients may have become more cost-aware consumers by cutting back on care, a trend noted by some providers. This could be a positive development to the extent patients reduced wasteful or duplicative care, but a far greater concern if they also deferred necessary tests, procedures, and treatments. Sustained declines in demand could exert downward pressure on health care costs and help moderate the cost trajectory, but ideally such reductions would be concentrated among services of limited value to health outcomes.

Much of the decrease in demand likely represents a temporary response to strained finances during the recession. As the economy recovers, demand for care will grow. Furthermore, as the PPACA coverage expansions are implemented in 2014 and many people gain insurance, system-related access problems may intensify as pent-up demand from newly insured individuals is unleashed. To ensure that needs are met, it is critical for providers and policymakers to anticipate and take appropriate steps to prevent potential system-related barriers, particularly in areas with provider shortages and where coverage increases will be greatest.

Notes

1. Estimates are based on the 2007 and 2010 HSC Health Tracking Household Surveys and are broadly consistent with other national surveys, including the Current Population Survey and the National Health Interview Survey.

2. This finding is based on a weighted logistic regression analysis of the likelihood of having unmet medical needs that combined the 2007 and 2010 surveys and controlled for age, gender, family income, education, race/ethnicity and health status. Indicators for insurance status (uninsured set to 1) and survey year (2010 set to 1) were also included. An interaction term for insurance status and survey year was positive and statistically significant at the .05 level, indicating that unmet need for the uninsured relative to the insured population was higher in 2010 than in 2007.

3. Cunningham, Peter J., and Laurie E. Felland, Falling Behind: Americans' Access to Medical Care Deteriorates, 2003-2007, Tracking Report No. 19, Center for Studying Health System Change, Washington, D.C. (June 2008).

4. Johnson, Avery, Jonathan D. Rockoff and Anna Wilde Mathews, "Americans Cut Back on Visits to Doctor," The Wall Street Journal (July 29, 2010).

5. Results were nearly identical when defining the lowest income group as individuals earning less than 138 percent of poverty, which aligns with federal guidelines governing Medicaid eligibility under the 2010 Patient Protection and Affordable Care Act. According to the law, all adults younger than 65 will be eligible for Medicaid if their income does not exceed 133 percent of poverty, after a deduction of 5 percent based on the upper income limit for that group, effectively making 138 percent of the federal poverty level the cutoff for Medicaid eligibility.

6. These findings were based on the same analysis as described in Note 2, subsetting the samples by income level.

7. Martin, Anne, et al., "Recession Contributes to Slowest Annual Rate of Increase In Health Spending In Five Decades," Health Affairs, Vol. 30, No. 1 (January 2011).

8. Kaiser Family Foundation/Health Research and Educational Trust, Employer Health Benefits 2010 Annual Survey (September 2010).

9. Felland, Laurie E., Joy M. Grossman and Ha T. Tu, Key Findings from HSC's 2010 Site Visits: Health Care Markets Weather Economic Downturn, Brace for Health Reform, Issue Brief No. 135, Center for Studying Health System Change, Washington, D.C. (May 2011).

10. Johnson, Rockoff and Mathews (2010); and Abelson, Reed, "Health Insurers Making Record Profits as Many Postpone Care," The New York Times (May 13, 2011).

Data Source and Funding Acknowledgement

This Tracking Report presents findings from the HSC 2007 and 2010 Health Tracking Household Surveys and the 2003 Community Tracking Study Household Survey. All three telephone surveys use nationally representative samples of the civilian, noninstitutionalized population. For the first time, the 2010 survey incorporated a cell phone sample because of declining percentages of households with landline phones. Sample sizes include approximately 47,000 people for the 2003 survey, about 18,000 for the 2007 survey, and roughly 17,000 for the 2010 survey. Response rates were 57 percent in 2003, 43 percent in 2007, and a combined 35 percent in 2010 (46% for the landline sample and 29% for the cell phone sample). Population weights adjust for probability of selection and differences in nonresponse based on age, sex, race or ethnicity, and education. The weights also adjust for the increased probability of selection in cases of households using both landline and cell phones. Although all three surveys are nationally representative, the sample for the 2003 survey was largely clustered in 60 representative communities, while the 2007 and 2010 surveys were based on a stratified random sample of the nation. Standard errors account for the complex sample design of the surveys. Questionnaire design, survey administration, and the question wording of all measures in this study were similar across the three surveys.

Estimates of unmet need and delayed care were based on the following two questions: (1) "During the past 12 months, was there any time when you didn't get the medical care you needed?" and (2) "Was there any time during the past 12 months when you put off or postponed getting medical care that you thought you needed?" For those reporting either an unmet need or delayed care, follow-up questions were asked to determine why. Responses included worry about cost, problems with health insurance, problems with availability of medical providers, and personal reasons such as lack of time or procrastination. This Tracking Report includes only responses where at least one of the reasons related to cost, health insurance, or the health care system; responses involving only personal reasons were not counted as unmet need or delayed care. Insurance status reflects coverage on the day of the interview and includes employer-sponsored and individually purchased private insurance, Medicare, Medicaid, the Children's Health Insurance Program (CHIP), other state programs, TRICARE and other military insurance programs, and the Indian Health Service.

Funding Acknowledgement

This research was funded by the Robert Wood Johnson Foundation (RWJF). The HSC 2007 and 2010 Health Tracking Household Surveys and the HSC 2003 Community Tracking Study Household Survey used for the analysis also were funded by RWJF.

Sources and Further Reading

Robert Wood Johnson Foundation (RWJF) -- Funded the HSC Health Tracking Household Surveys (2003, 2007, and 2010) that provided the primary data for this analysis of access trends.

U.S. Census Bureau Health Insurance Coverage Statistics -- National data on uninsured population trends referenced in this report, including the rise from 42.8 million to 51.7 million uninsured between 2007 and 2010.

Kaiser Family Foundation / HRET Employer Health Benefits Survey -- The 2010 annual survey data on employer cost-shifting and benefit design changes cited in this study's analysis of rising cost barriers.

Health Affairs Journal -- Published the Martin et al. study on recession-driven slowdowns in national health spending growth referenced in this tracking report.

Centers for Medicare and Medicaid Services (CMS) Medicaid Overview -- Federal agency administering the FMAP increases, Medicaid DSH payments, and maintenance-of-effort requirements discussed in the safety net analysis.

CDC National Health Interview Survey (NHIS) -- One of the national surveys noted as producing broadly consistent estimates of unmet medical needs alongside the HSC Household Survey data.