October 2013, VOLUME132 /ISSUE 4

Children’s Participation in Medicaid: A Matter of Time

  1. Andrew D. Racine, MD, PhD
  1. Albert Einstein College of Medicine and The Montefiore Medical Center
  • Medicaid
  • health insurance
  • pediatrics
  • child poverty

In their article, “A longitudinal view of child enrollment in Medicaid,” Simon and colleagues1 have done a considerable service to the pediatric community. By providing a more complete portrait of the degree to which children rely on Medicaid to secure health insurance coverage, they have helped explain how cross-sectional views of child participation in Medicaid significantly underestimate the reach of the program. Their point estimate that 32.8% of all children 0 to 13 years of age were enrolled in Medicaid in 2004 compares favorably with equivalent cross-sectional estimates based on the National Health Interview Survey and Medical Expenditure Panel Survey data for the same year, but this figure considerably underestimates the degree to which that same cohort of children will make use of Medicaid if the time horizon is extended to 5 years. Over that period, 41% or 2 in 5 children will, at 1 point or another, be enrolled in Medicaid. The longitudinal perspective provided by this report also allows readers to appreciate variation in how different age cohorts rely on Medicaid over time. For children between 0 and 3 years of age, the percentage of children who are enrolled in Medicaid during the entire 5-year period exceeds the 2004 cross-sectional estimate by 22% (47.3% compared with 38.8%). For older children, this discrepancy is even more pronounced where the 5-year estimate exceeds the 1-year cross-sectional estimate by close to 32%. Although the data do not permit us to distinguish how much of these observations are due to the age of the children, the particular characteristics of their cohort, or the time period during which the observations are made, the magnitude of these differences between 1-year and multiple-year estimates remains substantial. Moreover, as the authors correctly point out, the population of children using the Medicaid program is not homogeneous, and some within the program, notably children from households with lower parental income, fewer resources, and poorer health status, are more reliant on the program than their more advantaged counterparts.

These findings may be viewed from 2 different perspectives, which is to say they convey both good news and bad news. First, the good news: more children are covered by Medicaid than we might have thought. Why does this matter? It matters because health insurance is welfare-enhancing and because, as with education, the human capital investment embodied in providing children with adequate health services manifests robust returns.2 We know that health insurance is desirable in itself, because those who have the available resources are willing to pay money to secure it. This is no surprise. More than 60 years ago, Milton Friedman and L.J. Savage brought attention to the fact that so long as individuals face concave utility functions with respect to income (that is, that the value they place on 1 extra dollar of income diminishes the more money they have), they will be willing to pay someone to assume the risk of income loss on their behalf.3 In other words, they will buy insurance and pay a premium above the amount they might be expected to lose simply for the peace of mind of knowing they have limited their potential downside losses. That is how insurance markets produce net welfare gains for those who participate in them.4

Yet, markets cannot always be expected to work efficiently. The market failure of pricing medical care beyond the reach of a significant segment of the American public (initially the elderly and poor) has been remedied through government intervention, which is precisely the role of the Medicare and Medicaid programs. They increase social welfare by purchasing certainty for those who would like to have it but cannot afford it. In that regard, it is comforting to read in the present report that more children than we thought have been able to access this important resource.

Nor is certainty in some abstract sense the major source of welfare enhancement generated by health insurance coverage. In 2008, the state of Oregon, faced with limited resources, held a lottery to distribute access to its Medicaid program. Researchers who have looked at the resulting natural experiment concluded that those who were given the opportunity to participate in Medicaid had significantly fewer out-of-pocket expenditures and were referred to collection agencies significantly less often than those who did not get such coverage.5 Health insurance coverage is thus not merely a mechanism to allow individuals access to health care services, it also protects them against the possibility of significant income loss attendant on unexpected health care crises. Particularly for families living on the margin, this income protection can make the difference between being able to afford rent, food, or other basic necessities when illness strikes unexpectedly.

We also know that the increased participation in Medicaid means that more children than we might otherwise have suspected have used health care resources because of this coverage. Indeed, one of the main insights gleaned from the “Oregon Health Experiment” is that those who were given the option of Medicaid coverage increased their use of ambulatory and inpatient services. 6 What is more, the same social experiment provided evidence that coverage increased participants’ perceptions of their own health status, as those who received Medicaid coverage reported improved levels of health immediately after the coverage began.5 This factor is of particular relevance in view of Simon et al’s1 findings that children with poorer health status are more consistently covered by Medicaid over time than those with fewer medical problems.

Access to and use of health care services, particularly in childhood, represents a human capital investment that pays dividends in both future health and future productivity over an entire life span.7 In view of Medicaid’s positive impact on the health, economic security, and social welfare of poor and near-poor children and families, recognizing that its reach is greater than we imagined should be a source of considerable reassurance. The bad news is that Medicaid is a surrogate marker for poverty and it stands as a stark rebuke to the richest country in the world that, even before the current economic crisis, 2 in 5 of its children came from families living close enough to the poverty level to qualify for Medicaid. Therefore, even as the pediatric community acknowledges the pervasive need for the Medicaid program and defends its signature accomplishments, the findings of Simon et al’s report1 must focus a renewed commitment on a comprehensive sustained effort to eradicate childhood poverty in the United States.


    • Accepted August 5, 2013.
  • Address correspondence to Andrew D. Racine, MD, PhD, Montefiore Medical Center, OOP, 111 East 210th St, Bronx, NY 10467. E-mail: aracine{at}
  • FINANCIAL DISCLOSURE: The author has indicated he has no financial relationships relevant to this article to disclose.

  • FUNDING: No external funding.

  • POTENTIAL CONFLICT OF INTEREST: The author has indicated he has no potential conflicts of interest to disclose.

  • COMPANION PAPER: A companion to this article can be found on page 656, and online at