Health Insurance and Health Care Use in the Aftermath of the Great Recession: Findings from the Michigan Recession and Recovery Study

Back to Policy Briefs

November 2015

Lucie Kalousová and Katrina Hauschildt
National Poverty Center, Gerald R. Ford School of Public Policy, University of Michigan

Download printable version pdf


The Great Recession, a major economic downturn that lasted from December 2007 to June 2009, meant instability and a decline in wealth for a large share of the American population. The recessionary tremors were felt more deeply in some communities than in others. Detroit, a city whose fate has been closely tied to that of the struggling auto industry, suffered more than most. Whereas the national unemployment rate peaked at about 10% in July 2009, 19% of workers in Wayne County, Michigan, were unemployed that month.

For individuals, employment shocks can have many possible adverse consequences, including a loss of employer-sponsored health coverage. Over the course of the recession, the percentage of Americans without any health insurance coverage rose from 16 to 20%.(1) An increase in the share of the population without health insurance could have important implications for population health. Insurance coverage is essential to mitigating the costs associated with adverse health events, but also enables access to preventive services such as flu vaccinations, cancer screenings, and blood pressure checks that ensure health in the long run.(2) This brief examines how health insurance coverage and health care use in Detroit area changed after the Great Recession. Fortunately, we find signs of recovery in health insurance coverage. A greater percentage of the population had health insurance in 2013 than in 2009, and this was true across the income spectrum. Nevertheless, large economic gaps in health care access persist, and people below the poverty line were significantly less likely to have a usual source of medical care in 2013, compared to those above the poverty line. As state, local, and federal agencies work to implement the Patient Protection and Affordable Care Act (PPACA), we encourage continued attention to measures of health care access.

Michigan Recession and Recovery Study (MRRS)

The MRRS is following a stratified random sample of English-speaking adults who lived in Southeastern Michigan (Macomb, Oakland, and Wayne counties) and were 19 to 64 years old at the first interview in late 2009/early 2010. The MRRS oversampled African Americans and includes mainly African American and non-Hispanic white respondents, reflecting the residential composition of the area. To date, respondents have been interviewed three times. During wave 1 in 2009-10, we interviewed 914 respondents, for a response rate of 83%. For wave 2 in 2011, we reinterviewed 847 of the original sample, for a response rate of 94% of survivors. In wave 3 in 2013, we re-interviewed 751 survivors for a 90% response rate.

The MRRS survey instrument is unique in its depth and breadth, covering many domains, including employment and the labor market, housing instability, material hardships, income, assets, financial problems, credit and debt, health and mental health, demographic characteristics, and the use of public programs and private charities. More information about the study, as well as related papers and policy briefs, can be found at

In this brief, we use data from all three waves of the MRRS to evaluate changes in health insurance coverage in Detroit’s three-county area and changes in health care use over time. We first consider the sociodemographic differences among residents who had private, public, or no health insurance at baseline. We then evaluate changes in the share of the population reporting any health insurance coverage in three different income-to-needs categories over the study period, and we also identify patterns of health care use in these three income groups at wave 3. Finally, we assess how likely people were to report having a usual source of medical care by income-to-needs groups and, for those who did, where they most often sought care. Survey weights are used in all analyses reported here to make our results representative of the population aged 19 to 64 in these metro Detroit counties.

Population Characteristics by Health Insurance Status in 2009/2010

The sociodemographic profiles of groups with private, public, or no health insurance were significantly different at baseline as shown in Table 1. (For more information on how all variables were measured, please see the appendix.)

Table 1: Population-Weighted Characteristics of the Michigan Recession and Recovery Study Respondent at Their First Interview (2009/2010); Stratified by Their Health Insurance Status

We find that people with private health insurance reported higher average household incomes compared to those with public insurance or none at all. Those with private health insurance were on average older and more likely to be married. Men were over-represented in the group with no health insurance, and so were Blacks, who comprised 25% of the MRRS sample overall, but represented nearly 34% of individuals who reported no insurance. Among those who had private insurance, 36% had a bachelor’s degree or more, while only about 8 and 7% of those with public health insurance or no health insurance respectively had achieved this level of education. The publicly insured were on the other hand living in low income households and typically unmarried. Women and African Americans were over-represented in this group.

Changes in Health Insurance Coverage Over Time

Figure 1: Percentage of the Population Reporting Any Type of Health Insurance at Each Survey Wave; by the Household Income to Needs Ratio

In Figure 1, we show the percentage of the population with any health insurance coverage at each survey wave, stratified by the household’s income to needs category (based on the federal poverty line at that time). Health insurance coverage was steady across waves for individuals with incomes of 300% or more of the poverty line, and coverage levels were very high, at about 95%. Those with incomes that fell below 100% of the poverty line were much less likely to report having any health insurance, but coverage rose slightly across waves, from 64% at wave 1 to 69% at wave 3. The group reporting incomes at 200-299% of the poverty line showed a steeper rise in coverage of 9 percentage points from wave 1 to wave 3, increasing from 75% to 84%. The coverage increase was the greatest among those in the 100-199% of the poverty line income group, rising by approximately 20 percentage points, from 55% at wave 1 to 75% at wave 3.

Health Care Use by Income to Needs

Table 2: Percentage of the Populaiton Reporting Different Types of Healthcare Use at Each Survey Wave

Patterns of health care use varied over time and by household income to needs ratio, as shown in Table 2. The share of the population that reported a physician visit increased between wave 1 and wave 3 in the two groups with the lowest incomes, but decreased among those at 200 to 299% of the poverty level, and stayed unchanged for the most advantaged group.

The percentage of people who had a regular dental checkup in past 12 months increased from wave 1 to wave 3 for all income to needs groups, thought the estimated increase was negligible in the group with incomes between 200 and 299% of the poverty line; it only rising from 55 to 56%. However, the group with income to needs of 100-199% reported the largest positive change, from approximately 48% to 59%. At all waves, about one in four respondents in households below the poverty line reported a recent overnight stay in the hospital, a figure considerably higher than for respondents in all other groups (e.g., 10% of respondents in the highest income category).

In waves 1 and 2, reports of forgoing a needed visit with a doctor or dentist in the last 12 months were most common among the group below the poverty line. However, at wave 3, the greatest percentage of people forgoing care was found in the group immediately above the poverty line. Nearly 49% of those below the poverty line reported forgoing needed care at wave 1, compared to 33% and 27% of individuals with income to needs at 100-199% and 200-299% of the poverty line, respectively. Respondents with income to needs at 300% or more of the poverty line were the least likely to report forgoing needed care at all waves (11%, 5%, and 7% at waves 1, 2, and 3, respectively). In spite of the increased health insurance coverage across waves, the prevalence of foregone medical care increased over time for those with incomes at 100 to 199% and 200 to 299% of the poverty line, from 33 to 47% and 27 to 33%, respectively.

Usual Source of Care in 2013

Figure 2: Predicted Probability of Having Any Usual Source of Care at the Third Wave of the Michigan Recession and Recovery Study (2013)

Note: 100 to 199% Above Poverty Line group had a statistically significantly lower predicted probability of having a usual source of care compared to the group at 300% or above the poverty line. Logistic regression model controls for age, race, gender, and self-rated health. (N= 748). Unadjusted percentages of respondents who have a usual source of care at wave 3 were: 72% for below poverty line, 67% for 100 to 199%, 74% for 200 to 299%, and 92% for 300%+.

An important measure of stable attachment to the health care system is having a usual source of medical care. In the third wave of data collection, in 2013, we asked respondents whether there was a place they usually go when they are sick or need advice about their health. Overall, 83% of the population said they had a place where they usually went for medical care. Respondents who reported having a usual source of care were also asked what type of facility they used. We assess the probability of having a usual source of care in a regression framework that helps account for other sociodemographic characteristics that might explain the association between income and having a usual source of care, including age, race, gender, health insurance status, and self-rated health. We found that the association was nonlinear; the greatest probability of having a usual source of care was evident among those with income at 300% of the poverty line or above. The group below the poverty line showed the next highest likelihood of reporting a usual source of care. The lowest probability was estimated for those at 100 to 199% of the poverty line. This finding, along with the results showing the greatest prevalence of foregone medical care in the 100 to 199% group, highlights the health care access vulnerability of the near poor.

Among those reporting a usual source of care, the respondents who were at or above 300% of the poverty line reported using a doctor’s office or HMO most often—at 90%. This contrasts starkly with the usual sources of care reported by the respondents in other income to needs groups. Only 54% of those whose income to needs fell below the poverty line, and 56% and 67% of those with income to needs at 100-199% and 200-299% of the poverty line typically visited a doctor’s office or an HMO. People with income to needs just above the poverty line used a clinic or health center as their usual source of care more often than those in all other groups, at 34%. The below the poverty line group reported that their usual source of care was a hospital outpatient department more often than those in other groups, at 8%.

Figure 3: Distribution of Usual Sources of Care at the Third Wave of the Michigan Recession and Recovery Study (2013), Conditional on Having a Usual Source of Care And Stratified by Household Income to Needs (N = 596)

Note: Respondents who reported the Emergency Room as their usual source of care were categorized as without a usual source of care.


The data collected by the Michigan Recession and Recovery Study between 2009 and 2013 show rising rates of health insurance coverage in the aftermath of the Great Recession in Southeastern Michigan. We surmise that there are at least two key reasons for this rise in coverage, one related to economic recovery and other to the recent health care reform. First, as some Detroiters regained employment, they also regained access to employer-sponsored health insurance or became able to afford health insurance on their own. Second, in anticipation of the Patient Protection and Affordable Care Act (PPACA) employer coverage mandate,(3) some employers may have already begun offering health insurance coverage to their previously uninsured workers. We are optimistic that as of this writing, several months after the official start date of employer coverage mandates and more than a year after the state expanded its Medicaid coverage to Michigan residents in households with income to needs up to 138% of the federal poverty line, the percentage of the local population with health insurance is even higher than what we can show with our data. In addition to the positive trend in health insurance coverage, our analysis suggested that there may have been a favorable decline in the number of adults foregoing medical care for cost reasons and a favorable rise in the fraction that had a dental checkup within the past twelve months.

The results demonstrating an increased percentage of residents just above the poverty line who reported foregoing medical care are less encouraging. This group also had the lowest probability of having a usual source of medical care, both before and after adjusting for their sociodemographic characteristics. Another worrisome result of our study revealed that a significant share of residents in these Southeastern Michigan counties, including many who live well above the poverty line, could not identify a stable place where they sought medical care. This is an important finding because past research has shown that people who report having no usual place of care are at a greater risk of not receiving appropriate preventative services(4) and might thus suffer from otherwise preventable health conditions in the future. Whether or not we will observe a rise in the share of Southeastern Michigan population with a regular place of medical care as a consequence of the health insurance coverage expansion remains an open question. We recommend continued close monitoring of this important indicator of access to care as the implementation of the health care reform progresses.


(1) C. White, J. Reschovsky, Great Recession Accelerated Long-Term Decline in Employment Coverage (2012), National Institute for Health care Reform, Research Brief No. 8. Full-text:

(2) S. McMorrow, G.M. Kenney, and D. Goin, Determinants of Receipt of Recommended Preventive Services: Implications for the Affordable Care Act (2014), American Journal of Public Health, 104(12), pp. 2392-2399.

(3) Starting with 2015, PPACA requires businesses with 100 or more full-time employees to offer affordable medical coverage to at least 70% of their full-time employees and their dependents below the age of 26.

(4) K.T. Xu, Usual source of care in preventive service use: A regular doctor versus a regular site (2002), Health Services Research, 37 (6), pp. 1509-1529.

Funding Sources

The Michigan Recession and Recovery Study was supported in part by grants from the Ford Foundation, the John D. and Catherine T. MacArthur Foundation, the Vice President for Research at the University of Michigan and the Office of the Assistant Secretary for Planning and Evaluation at the U.S. Department of Health and Human Services. For additional information, contact Sarah Burgard (

About the Authors

Lucie Kalousova is a Ph.D. student in the joint program in Sociology and Health Policy at the University of Michigan.

Katrina Hauschildt is a Ph.D. student in Sociology at the University of Michigan.