A full accounting of the excess burden of poor health in childhood must include any continuing loss of productivity over the life course. Including these costs results in a much higher estimate of the burden than focusing only on medical costs and other shorter-run costs to parents (such as lost work time). Policies designed to reduce this burden must go beyond increasing eligibility for health insurance, because disparities exist not only in access to health insurance but also in take-up of insurance, access to care, and the incidence of health conditions. We need to create a comprehensive safety net for young children that includes automatic eligibility for basic health coverage under Medicaid unless parents opt out by enrolling children in a private program; health and nutrition services for pregnant women and infants; quality preschool; and home visiting for infants and children at risk. Such a program is feasible and would be relatively inexpensive.
Differences between poor or minority children and other children in health, health care, and developmental outcomes impose an “excess burden” on poor and minority children.1 The high costs of these health disparities provide a reason for public policy to address them that goes beyond the humanitarian considerations involved. In the United States, policies for reducing health disparities usually center on extending eligibility for health insurance coverage. Although this is necessary, it is increasingly recognized that it is not sufficient. In addition to efforts to extend coverage, further efforts to reach eligible but unenrolled children are necessary. Moreover, an effective policy for addressing disparities must move beyond health insurance by improving the environments of poor and disadvantaged children. Extensions of existing programs, including early childhood intervention programs, the Supplemental Nutrition Program for Women, Infants, and Children (WIC), and nurse home-visiting programs, could help to accomplish this goal.
THE EXCESS BURDEN POSED BY DISPARITIES IN CHILD HEALTH
Table 1 lists disparities in the prevalence of 4 conditions: attention-deficit/hyperactivity disorder (ADHD), the most common childhood mental health condition; asthma, the most common chronic condition of childhood and the leading cause of pediatric hospitalizations and school absences; overweight/obesity, which affects 17% of US children; and injuries, the leading cause of death among children >1 year old. Important disparities exist in the incidence of all of these conditions.
In addition to costs of pain and suffering, the costs of the excess burden of disease posed by these disparities can be divided into 3 parts:
excess expenditure, defined as the reduction in expenditures that could be accomplished by reducing the burden of disease in the disadvantaged population to levels that prevail in the majority population;
long-term losses in economic productivity because of disease; and
costs that result from excess mortality.
Take the specific example of ADHD. Birnbaum et al9 have estimated, using administrative data from a single large company, the excess medical costs associated with ADHD. Their estimates summed the direct costs of treatment for ADHD, the additional medical costs associated with factors such as their higher injury rates,10 and the costs of lost work among adults. Their estimates imply that a child with ADHD costs $1954 more (in 2008 dollars) annually than a child without this problem.
Using this estimate, along with the differences in prevalence of ADHD between the poor and nonpoor from Table 1, we can calculate the first component of the savings that would be associated with eliminating disparities in incidence. In 2004, 13 million children aged 0 to 17 years were poor,11 which suggests that equating rates would reduce the number of children with ADHD by 338 000. If each additional child with ADHD costs $1954 per year, the medical and work-time cost savings attained by eliminating disparities would equal $660 million per year.
However, ADHD is likely to impose costs throughout the life course. Indeed, the effects of childhood ADHD on future productivity likely dwarf this figure. A diagnosis of ADHD is associated with a reduction in mathematics and reading test scores of approximately one third of an SD as well as with increases in the probability of repeating a grade.12 If these reductions lead to an increase in the probability of dropping out of high school, the effect on wages among affected children could easily swamp the entire direct cost of ADHD. Even compared with their own siblings, children with mental health problems in childhood have lower scores on standardized tests and a higher probability of using welfare.13 Mental health problems are 1 of the leading causes of days lost in the workplace, because these problems strike many people of working age.14 Hence, mental health problems with roots in childhood are likely to have a considerable impact on future productivity.
This lost productivity is likely to be in the billions of dollars. One longitudinal, 16-year follow-up study revealed that children with ADHD were much less likely to have graduated from college (14% vs 52%).15 Using the above estimate of 338 000 “excess” cases of ADHD among poor children thus implies a decrease of 128 440 college graduates in this cohort. In 2004, a college degree nearly doubled annual earnings relative to a high school diploma, from $27 915 to $51 206.16 So, loss in earnings for this group of children would be $3 billion. If the older cohorts (18- to 34-year-olds, 35- to 51-year-olds) suffered similar lost productivity, the annual cost could exceed $9 billion. Moreover, there may be additional costs of ADHD in terms of adult substance abuse, use of welfare, and criminal activity.
The third and final part of the excess burden, the costs associated with excess mortality that results from ADHD, are currently unknown. However, because children with ADHD are more likely to be injured and injury is the leading cause of death among children >1 year old, children with ADHD are probably at higher risk of death. A full valuation of the excess burden of ADHD would need to assign a value to the lives lost; the typical value used in such calculations is $6 to $7 million per life.17 Hence, if the higher probability of death suggests that an additional 100 lives would be lost, the lost value of life would exceed the $660 million cost of the additional medical care and lost parental work time for children with ADHD.
Although all such calculations are crude, a focus on medical costs alone clearly ignores the major component of the costs resulting from health disparities among children. In fact, the true social costs may far exceed those suggested by studies that have focused only on costs of medical treatment and parent's lost work.
POTENTIAL POLICY RESPONSES TO HEALTH DISPARITIES
Having established that child health disparities are costly, I next discuss several approaches to reducing them. The most commonly advocated approach is to reduce disparities in access to health care. Although this is important, it cannot eliminate disparities in health. What is needed are programs that will address disparities before they start. Several approaches are discussed below.
Reducing Disparities in Access to Health Care
Although lack of health insurance coverage remains a serious problem for many children, expansions of public health insurance under the Medicaid program and the State Children's Health Insurance Program (SCHIP) mean that the majority of poor and near-poor children are now eligible for public health insurance.
Table 2 suggests that if the cost of insuring currently uninsured children was similar to the cost of children on Medicaid, it would cost an additional $14.6 billion to extend health insurance coverage to the 9 million currently uninsured children.* This is a relatively small amount compared to, for example, federal disbursements for Medicare benefits, which totaled $380.4 billion in 2006, or for Medicaid benefits, which totaled $257.7 billion in 2004 (a relatively small share of Medicaid dollars are spent on families with children; the bulk is spent on the elderly and disabled).20,21
Although we need to protect the gains in insuring children that have been made to date, greater efforts must be made to ensure that eligible children receive coverage. Many eligible children do not sign up for public health insurance until they have an urgent medical problem. The children who have the poorest access to specialists are those with incomes between 125% and 200% of the federal poverty level, despite the fact that many of these children are eligible for the SCHIP.22
The large number of uninsured but eligible children suggests that a successful policy would make applying for Medicaid and maintaining Medicaid coverage much less onerous. Many states have implemented policies designed to streamline the Medicaid-application process (such as shortening application forms and allowing mail-in applications), with mixed results.23 An example on the positive side was a California program that paid community organizations to register families; enrollments increased, and preventable hospitalizations declined.24
Once enrolled, more children could be retained in Medicaid (and the SCHIP) by extending enrollment periods from 6 to 12 months or, in SCHIP programs where parents pay a premium, maintaining their enrollment unless the parents notify the program office of changes in eligibility.25
Considerable evidence shows that eligible people are more likely to “take up” their benefits in social programs with automatic enrollment. For example, when they turn 65, Americans are automatically enrolled in Medicare Part B unless they actively decline, with the result that enrollment levels are high. This example suggests that a more proactive approach to the problem of eligible, unenrolled children is warranted. We should make all children eligible for Medicaid services and charge premiums on a sliding scale, with the poorest children paying no fee. We should also allow families to opt out of Medicaid if they choose to purchase private insurance with at least equivalent coverage of basic services.26
Still, the available evidence suggests that further expansion of public health insurance is unlikely to ever entirely eliminate the relationship between socioeconomic status and health. The famous Black Report27 from Great Britain concluded that the relationship between socioeconomic status and health became more pronounced after the introduction of national health insurance, although it is possible that the differential would have widened even further in its absence. Similarly, in Canada, health and household income have a significant relationship even though Canadians have universal health insurance.28
Common sense indicates that if disparities exist in the incidence of health conditions, then better treatment of conditions once they have occurred cannot eliminate the excess burden of disease. We need to address the reasons for the higher incidence of health conditions in poor populations.
Eliminating Poverty Through Cash and “Near-Cash” Programs
Eliminating income poverty sounds like an ambitious goal, but Table 3 suggests that it would be surprisingly affordable. Ziliak has performed research quantifying the “poverty gap,” that is, the gap between the income a family has in the absence of social assistance and the US federal poverty line.29 In 2001, the total poverty gap was approximately $46.1 billion for the 5.3 million poor families with children. However, much of this gap is already being filled by various cash programs, including Supplemental Security Income (SSI), Temporary Assistance for Needy Families (TANF) (welfare), and the Earned Income Tax Credit (EITC), and by “near-cash” programs such as the Supplemental Nutrition Assistance Program (formerly the federal Food Stamp Program) and public housing (given the ambiguities surrounding the valuation of health insurance, its cost is not included in this gap valuation). These latter 2 programs are labeled as near-cash because providing for basic necessities (food and housing) that otherwise would have to be purchased has much the same effect as offering cash assistance. Ziliak has estimated that it would have taken a further $21.7 billion to close the gap, that is, to bring every family with children up to at least the poverty level. Compare this amount with the cost of Social Security for the elderly, which was $554 billion in 2006, or to the rising cost of the various recent financial “bailouts.”
The fact that lower incomes are associated with worse health does not, however, prove that giving money to poor families would dramatically improve children's health. Factors that are correlated with lower earnings, such as lack of parental education, may actually be more predictive of children's health status than income. It may seem logical to assume that giving money to parents would improve their children's health, but little evidence is available on this point one way or the other.
One large study of 5000 US children born between 1998 and 2000 revealed that measures of parenting skills and of the physical characteristics of the home were highly related to income.33 But, regardless of whether these measures were controlled for, the effect of income was small. The estimates implied that even cash subsidies that brought every family up to the poverty line would not eliminate disparities in child outcomes. Similar results have been found in the United Kingdom.34
Increasing evidence shows that a mother's own health and events in her early life matter more for her child's health than her current income. The life-course approach implies that child development occurs along a trajectory, and research about the genesis of health conditions increasingly supports this perspective. The “fetal origins hypothesis” associated with Barker35 posits that fetal conditions are related to adult risk of disease.36
Recent research highlights the effects of fetal conditions on future socioeconomic status. Compared with cohorts in utero just before or just after, children who were in utero during the influenza epidemic of 1918 (which affected one third of women of childbearing age) were 15% less likely to graduate from high school, and males suffered reductions in wages of 5% to 9%. Moreover, affected individuals were more likely to be poor, to be disabled, and to receive transfer payments as adults.37 Studies conducted by using large samples drawn from vital statistics records in Scotland, Norway, Canada, and the United States have shown a link between lower birth weight and lower educational attainment, even among siblings or twins.38–41 In the United States, analysis of data from the Panel Study of Income Dynamics revealed that low birth weight reduced adult annual earnings by 17.5%.42
Mothers' educational trajectories are also strongly influenced by factors early in life and are an important determinant of child health. For example, improvements in mothers' education have been linked to reductions in their children's mental health problems43 and in the incidence of low birth weight. Maternal education may improve child health through reductions in smoking, nonmarital childbearing, and parity and increases in the use of prenatal care.44 Hence, interventions that address disparities in educational attainment are likely to have intergenerational effects.
These examples suggest that the answer to the income paradox may be that short-term increases in the income of adult women have relatively little effect on their ability to parent their children but that the environment in which the mother is raised (and the fetal environment) matters greatly. Because we cannot change the past, we must look to more direct interventions to improve children's environments, reduce disparities in their health, and break the intergenerational cycle of poverty and poor health.
As discussed below, interventions early in the life cycle can be effective.* Heckman46 has provided a theoretical framework that explains why early interventions are often more effective than later ones: Children who do not receive sufficient investments early in life may suffer “capabilities” that are permanently lower than they otherwise would have been.
Early Childhood Intervention Programs
Early childhood intervention programs offer a promising avenue for reducing health disparities. Most such programs include significant health and nutrition components to address the needs of the “whole child.” For example, Head Start, the federal program that serves disadvantaged 3- to 5-year-old children, mandates that programs ensure that children receive the health assessments and services they need. Head Start has detailed performance standards for health services, and programs are regularly evaluated with respect to indicators such as the fraction of children who have received dental examinations, hearing and vision screenings, and immunizations. In recent years, the federal government has also created an experimental “Early Head Start” program to serve infants and toddlers.47 Evidence from the introduction of the Head Start program in the 1960s shows that it was associated with large reductions in mortality among children aged 5 to 9 years from causes that could have been affected by earlier participation in Head Start but not from other causes.48
Because today many different programs address specific health problems (including screening for lead poisoning, child nutrition, etc), it may seem duplicative for health to be a major focus of early intervention programs. However, parents who are struggling to put bread on the table may not be able to navigate health care bureaucracies. Quality early intervention programs offer “one-stop shopping” for the services a child needs. Staff members may also be both better at spotting problems and more knowledgeable about community resources than parents. However, the importance of the health services component of early intervention programs has yet to be assessed systematically.49
The available evidence suggests that children in Head Start are significantly more likely to have received preventive care than other similar children.50–52 For example, a large federal study found that children in Head Start were more likely to have health insurance and to have received dental care (on the basis of reports from the parents) than children in a control group, and they were in better overall health. These effects were larger for children of nonnative speakers, children with special needs, and children whose mothers were depressed at baseline, suggesting that early intervention programs can be particularly important for these vulnerable children.
The nutrition component of early intervention may also be important, given the alarming and rising levels of overweight among children. Children who moved from half-day to full-day Head Start programs as a result of changes in the availability of full-day programs were 4% less likely to be overweight (using a baseline of 33% of children being overweight).54
In summary, considerable scope seems to exist for reducing disparities in child health through the expansion of early intervention programs to all eligible children. These programs alter life-course trajectories by improving both health and educational outcomes.
One problem with offering health services through programs such as Head Start is that they do not reach all needy children; some eligible children do not enroll, and some needy children are not income-eligible. Home-visiting programs offer an alternative model that can prevent children from “falling through the cracks.” These programs form an important part of the public health service in all northern and western European countries and have been credited with reducing infant mortality, injuries, and postpartum depression. European programs employ professional visitors, usually nurses or social workers with public health training. They generally visit all newborns at least once, with subsequent visits targeted where they are most needed. Some programs offer prenatal visits.55
In the United States, the most successful program of this type is the Nurse Family Partnership (NFP) associated with the work of Olds et al.56,57 The NFP focuses on families in which the mother is young, poor, uneducated, and/or unmarried and involves nurse visits from the prenatal period up to 2 years after birth. NFP programs currently serve ∼20 000 families each year, and Colorado, Louisiana, Oklahoma, and Pennsylvania have launched statewide initiatives.
Randomized, controlled trials have shown many positive effects and have shown that the programs are cost-effective. The overall benefit/cost ratio has been estimated to be $2.88 saved for every dollar invested, with larger effects among mothers at high risk.58 As of age 2, children at 1 study site were much less likely to have been seen in a hospital emergency department for unintentional injuries or ingestion of poisonous substances. As of age 15, children of visited mothers were less likely to have been arrested or to have run away from home, had fewer sexual partners, and smoked and drank less. These children were also less likely to have been involved in verified incidents of child maltreatment. Effects have been largest for adolescent mothers, and large effects have been found on delaying the timing of second births among teenaged mothers, which is likely to have important positive effects on both mother and child lives.
These studies suggest that proactively attempting to locate children at risk and ensure that they receive necessary services would be a useful complement to other strategies for reducing disparities in child health. Home visiting promotes positive development and reduces the probability that poor children will suffer health shocks that could place them on permanently lower trajectories throughout their lives.
Targeting Mothers Through Programs Such as WIC
The literature on fetal origins suggests that one of the more effective ways to reduce health disparities in children may be to reduce disparities in maternal health during pregnancy. One way to do this is through programs such as WIC, which already serves much of this target population. WIC is a federal program that offers nutritious food supplements to pregnant, lactating, and postpartum mothers as well as to infants and children <5 years old. WIC programs facilitate access to medical services, and WIC offices are often run in medical clinics.
In any given month, half of the pregnant and postpartum women in the United States are eligible for assistance through WIC, and approximately two thirds of these women are served. In addition, the majority of all US infants are eligible for WIC and roughly half of them receive WIC benefits. Participation tends to drop off sharply after a child's first birthday, when WIC stops providing expensive infant formula.59
WIC already plays a large role in remediating health disparities. For example, WIC reduces the incidence of low birth weight and preterm birth, especially among black mothers,60 and the program has been credited with the dramatic decline in the incidence of anemia among young children that took place between 1975 (shortly after WIC was introduced) and 1985.61
In addition to providing nutritious food, the WIC program has several features that help it to address health disparities. First, it targets the critical prenatal and infant years. Second, WIC agencies have frequent contact with participants who typically come in at least quarterly to pick up their coupons and receive nutritional counseling. Third, WIC agencies are required to help participants obtain preventive health care by providing services on-site or through referrals. Finally, WIC is mandated to provide nutrition education. WIC already serves many children in utero and in their early years. Health disparities could be further reduced through the WIC program by taking steps to maintain children's enrollment until they can enter quality preschool programs.
NEXT STEPS FOR RESEARCH AND POLICY
The preceding discussion suggests a 2-pronged approach for policy. First, we need to make sure not only that all children are eligible for health insurance but also that they are covered. The most effective way to do this would be to extend basic Medicaid coverage to all children, with fees on a sliding scale (no fee for the poorest children), and an option for parents who purchased alternative coverage to opt out of Medicaid.
Second, we need to improve the environments facing young children from disadvantaged backgrounds. A first step toward accomplishing this goal would be to ensure that all eligible pregnant women receive WIC services and to improve these services. A second step would be to implement a home-visiting program that begins with a visit to all newborns. Children who need continuing follow-up could be tracked and receive priority referrals to early intervention programs. Quality early intervention programs should be expanded to cover all eligible children. These interventions would go a long way toward ensuring that all children enter the school years healthy and ready to learn.
The key question for researchers is how to accomplish these goals. For example, how would private insurance markets adjust if the default were that basic medical services for children would be paid for by public insurance? What is the best way to identify home-visited children who require additional follow-up? What is the most effective way to deliver WIC's educational activities? Can we obtain a more specific and, at the same time, comprehensive view of how health is affected by participation in early intervention programs (eg, does it reduce injuries, asthma or obesity?) Perhaps most important, could other interventions with a proven track record be brought “to scale”?. Recent initiatives to encourage responsible parenting by fathers, promote marriage, or provide integrated early intervention services to low-income children must all be carefully evaluated.
The costs of health disparities among children are large. A life-course perspective suggests incorporating future losses in health and productivity in these costs, and these losses dwarf the short-term costs of medical care and parents' lost work time. Expanding health insurance and eliminating income poverty are important social goals, but achieving them would not eliminate disparities in child health according to race and income. To eliminate health disparities, we need to take a more active role in improving the environments of young children so that disparities in the incidence of health conditions can be reduced.
The United States already has successful programs that are making an impact and that could do more. The total costs of providing these programs to all children are surprisingly modest, which suggests that the real problem has been lack of political will.
- Accepted July 20, 2009.
- Address correspondence to Janet Currie, PhD, Columbia University, National Bureau of Economic Research, Department of Economics, 420 W 118th St, New York, NY 10027. E-mail:
↵* An alternative estimate is available. Broaddus et al19 estimated that it would cost $9.5 billion per year to cover an estimated 6.5 million uninsured and eligible children. Scaling this number up to cover the million uninsured children yields an estimate of $13.2 billion per year. Neither this estimate nor the one in Table 2 accounts for “crowd-out” (ie, people switching from private to public health insurance).
The views presented in this article are those of the author, not the organizations with which she is affiliated.
Financial Disclosure: The author has indicated she has no financial relationships relevant to this article to disclose.
- ↵Currie J, Lin W. Chipping away at health: more on the relationship between income and child health. Health Aff (Millwood).2007;26 (2):331– 344
- Cuffe S, Charity M, McKeown R. ADHD symptoms in the National Health Interview Survey: prevalence, correlates, and use of services and medication. Poster presented at: 50th anniversary meeting of the American Academy of Child and Adolescent Psychiatry; October 20, 2003; Miami, FL
- Bloom B, Cohen RA. Summary health statistics for U.S. children: National Health Interview Survey, 2006. Vital Health Stat 10.2007;(234):1– 79
- Anderson PM, Butcher KF, Schanzenbach DW. Childhood disadvantage and obesity: is nurture trumping nature? NBER working paper No. 13479. Available at: www.nber.org/papers/w13479. Accessed September 1, 2009
- US Department of Health and Human Services, US Public Health Service. Mental health: a report of the Surgeon General. Available at: www.surgeongeneral.gov/library/mentalhealth/home.html. Accessed September 1, 2009
- ↵US Census Bureau. Table B-2. In: Income, Poverty, and Health Insurance Coverage in the United States: 2004. Washington, DC: US Census Bureau; 2005:52– 57. Report P60, No. 229
- ↵Currie J, Stabile M. Mental health in childhood and human capital. In: Gruber J, ed. An Economic Perspective on the Problems of Disadvantaged Youth. Chicago, IL: University of Chicago Press for NBER. 2009; In press
- ↵US Census Bureau. College degree nearly doubles annual earnings, Census Bureau reports [press release]. March 28, 2005. Available at: www.census.gov/Press-Release/www/releases/archives/education/004214.html Accessed September 1, 2009
- ↵Viscusi W, Aldy J. The Value of a Statistical Life: A Critical Review of Market Estimates Throughout the World. Boston, MA: Harvard University; 2002. Harvard Law School discussion paper No. 392
- ↵Kaiser Commission on Medicaid and the Uninsured. Monthly Medicaid enrollment for children. Available at: www.statehealthfacts.org/comparemaptable.jsp?cat=4&ind=612. Accessed July 14, 2008
- ↵Broaddus M, Greenstein R, Park E. Clearing up confusion on the costs of covering uninsured children eligible for Medicaid or SCHIP. Available at: www.cbpp.org/cms/?fa=view&id=1213. Accessed August 10, 2009
- ↵US Census Bureau. Table 137: Medicare disbursements by type of beneficiary—1990 to 2006. Available at: www.census.gov/compendia/statab/tables/08s0137.xls. Accessed September 1, 2009
- ↵US Census Bureau. Table 140: Medicaid—beneficiaries and payments: 2000 to 2004. Available at: www.census.gov/compendia/statab/tables/08s0140.xls. Accessed September 1, 2009
- ↵Kuhlthau K, Nyman RM, Ferris TG, Beal AC, Perrin JM. Correlates of use of specialty care. Pediatrics.2004;113 (3 pt 1). Available at: www.pediatrics.org/cgi/content/full/113/3/e249
- ↵Sommers B. Why millions of children eligible for Medicaid and SCHIP are uninsured: poor retention versus poor take-up. Health Aff (Millwood).2007;26 (5):w560– w567
- ↵Department of Health and Social Security. Inequalities in Health: Report of a Research Working Group [Black Report]. London, United Kingdom: Department of Health and Social Security; 1980
- ↵Ziliak J. Filling the poverty gap, then and now. In: Rubert P, ed. Frontiers in Family Economics. Vol1 . Amsterdam, Netherlands: Elsevier; 2009: In press
- ↵National Head Start Association. Head Start leaders have high hopes for new president and congress, but budget “double whammy” expected to force cuts of up to 14,000 child slots nationwide [press release]. Available at: www.supportheadstart.org/News/release2.cfm?releaseID=57. Accessed September 1, 2009
- ↵US Department of Agriculture, Food and Nutrition Service. WIC program participation and costs. Available at: www.fns.usda.gov/pd/wisummary.htm. Accessed July 13, 2008
- National Research Council. Estimating eligibility and participation for the WIC program: final report. Available at: http://nap.edu/openbook/030908962X/html/124.html. Accessed September 1, 2009
- ↵Berger L, Paxson CH, Waldfogel J. Income and child development: Center for Research on Child Wellbeing working paper No. 05-16-FF. Available at: http://crcw.princeton.edu/workingpapers/WP05-16-FF-Paxson.pdf. Accessed September 1, 2009
- ↵Burgess S, Propper C, Rigg J. The impact of low-income on child health: evidence from a birth cohort study: working paper No. 98. Available at: www.bristol.ac.uk/cmpo/publications/papers/2004/wp98.pdf. Accessed September 1, 2009
- ↵Barker DJP, ed. Mothers, Babies and Health in Later Life. 2nd ed. Edinburgh, Scotland: Churchill Livingstone; 1998
- ↵Gluckman P, Hanson M. The Fetal Matrix: Evolution, Development and Disease. Cambridge, England: Cambridge University Press; 2005
- ↵Lawlor DA, Clark H, Smith GD, Leon DA. Intrauterine growth and intelligence within sibling pairs: findings from the Aberdeen children of the 1950s cohort. Pediatrics.2006;117 (5). Available at: www.pediatrics.org/cgi/content/full/117/5/e894
- Black SE, Devereux PJ, Salvanes KG. The more the merrier? The effects of family size and birth order on children's education. Q J Econ.2005;120 (2):669– 700
- Oreopoulos P, Stabile M, Walld R, Roos L. Short, medium, and long term consequences of poor infant health: an analysis using siblings and twins: NBER working paper No. 11998. Available at: www.nber.org/papers/w11998.pdf. Accessed September 1, 2009
- ↵Johnson R, Schoeni R. The Influence of Early-Life Events on Human Capital, Health Status, and Labor Market Outcomes Over the Life Course. Berkeley, CA: Institute for Research on Labor and Employment. 2007. Population Studies Center research report 07–616
- ↵Carneiro P, Meghir C, Parey M. Maternal education, home environments and the development of children and adolescents: IZA working paper No. 3072. Available at: ftp://repec.iza.org/RePEc/Discussionpaper/dp3072.pdf. Accessed September 1, 2009
- ↵Currie J, Moretti E. Mother's education and the intergenerational transmission of human capital: evidence from college openings. Q J Econ.2003;118 (4):1495– 1532
- ↵Currie J. The Invisible Safety Net: Protecting America's Poor Children and Families. Princeton, NJ: Princeton University Press; 2006
- ↵Heckman JJ. The technology and neuroscience of capacity formation. Proc Natl Acac Sci U S A. 2007;104 (33):13250– 13255
- ↵Love JM, Kisker EE, Ross CM, et al. Making a Difference in the Lives of Infants and Toddlers and Their Families: The Impacts of Early Head Start. Washington, DC: Mathematica Policy Research; 2002
- ↵Ludwig J, Miller D. Does Head Start improve children's life chances? Evidence from a regression discontinuity design. Q J Econ.2007;122 (1):159– 208
- ↵Fosburg L, Goodrich N, Fox MK, et al. The effects of Head Start health services: report of the Head Start health evaluation. Available at: http://eric.ed.gov/ERICDocs/data/ericdocs2sql/content_storage_01/0000019b/80/2e/94/73.pdf. Accessed September 1, 2009
- US Department of Health and Human Services, Office of Planning, Research and Evaluation, Administration for Children and Families. Head Start Impact Study: First Year Findings. Washington, DC: Department of Health and Human Services; 2005
- ↵Frisvold D, Lumeng J. Expanding Exposure: Can Increasing the Daily Duration of Head Start Reduce Childhood Obesity? Ann Arbor, MI: University of Michigan Department of Health Management Policy; 2008
- ↵Kamerman S, Kahn A. Home health visiting in Europe. Future Child.1993;3 (3):39–52
- ↵Goodman A. The Story of David Olds and the Nurse Home Visiting Program. New Brunswick, NJ: Robert Wood Johnson Foundation; 2006
- ↵Karoly L. Investing in Our Children: What We Know and Don't Know About the Costs and Benefits of Early Childhood Interventions. Santa Monica, CA: Rand Corporation; 1998
- Copyright © 2009 by the American Academy of Pediatrics