Family Income Gradients in the Health and Health Care Access of US Children

Prior research has established consistent social class gradients in adult health that extend across disease categories (e.g., diabetes, cardiovascular disease, lung disease) and across indices of health status and well-being [20‐23]. This study reveals that even very early in life, substantial income gradients hold up consistently across a broad range of physical and developmental health measures. Health differences associated with family income were not confined to children in poor vs. non-poor families but rather extended across the income spectrum for most health indicators (although gradients were sometimes less steep at upper income levels). Despite their worse health status, children from lower income families also had health services marked by less continuity (e.g., no regular provider), more unmet needs, and fewer health visits outside of the emergency room. In multivariate models that control for a range of confounders plus health insurance status, striking income disparities in children’s health and health care access/utilization persist.

In the physical health domain, significant income gradients were noted across three general health indicators (global child health status, activity limitation, global condition of teeth) and a range of specific health conditions (overweight/obesity, asthma, vision/hearing problems, headaches, and ear infections). Diabetes and bone, joint, muscle problems did not manifest as gradients, although there were health differentials between the rich and the poor. Allergies (respiratory/hay fever, digestive/food, and skin) were the only physical health measures included in this study that did not show associations in the expected direction (and in fact showed gradients in the opposite direction, with those in the highest income category reporting worse health). This finding is consistent with a small number of studies in the US and other Western industrialized countries showing that hay fever and other allergies may be more common for higher SES individuals [24‐26]. One possible explanation for the reverse income-allergy gradient found in this study is that higher social class parents may be more likely to obtain a physician diagnosis of allergic disorder for their children due to better health care access or screening for these conditions. Less educated parents may also not recognize allergies, and classify similar symptom patterns as colds, variations of normal, and not notable as a medical condition. Alternatively, the “hygiene hypothesis” states that a lack of early childhood exposure to infectious agents and bacteria (which may be more common in higher SES families) increases susceptibility to allergic diseases by modulating immune system development [27].

In the developmental and behavioral health domain, gradients were noted for all health indicators except autism. Autism is known to have a strong genetic link [28, 29] and may be less susceptible to social environmental influences than other disorders like behavior/conduct problems and learning disabilities which showed steep family income gradients in this study. This could also relate to differential screening, as research indicates greater delays in diagnosis of autism for children from disadvantaged social groups and possibly even more misdiagnoses with other common co-morbid disorders like ADHD or adjustment problems [30‐32]. Although measurement issues and choice of health indicators complicate cross-domain comparisons, it is noteworthy that income gradients in children’s physical health appear of comparable magnitude to emotional, developmental, behavioral health in this comprehensive study. This is contrary to the findings of some studies with multi-dimensional measures of health that have found steeper social class gradients in the psychosocial domain during childhood [6], and highlights the important early influence of family income on general physical well-being and chronic/acute disorders, as well as emotional, developmental, behavioral health. Parent-rated global child health status showed a particularly steep income gradient.

The sub-optimal health and developmental status of children from lower income families would seem to suggest that these are the individuals most in need of attention by the health service system. On the contrary, this study uncovered substantially reduced access to health services for children from lower income families across the full range of health care access and utilization indicators available in NSCH, and these associations largely persisted in multivariate models with controls for confounders. The considerably lower rates of having a continuous health care provider for children in lower income families is disconcerting given the important role that such an individual could play as a long-term developmental advocate for the child [33]. Receiving primary care from the same physician has been associated with more frequent discussions about child health supervision topics and more preventive health visits [34‐36]. Sizable income disparities were also noted for unmet health needs (medical, dental, prescription) and use of routine and preventive health services and these disparities remained in models that included controls for health insurance status suggesting that although universal health coverage may be a necessary first step toward achieving equity in children’s access to health care, it may not be sufficient to close the large gaps in service between children in low- and high-income families. A multitude of financial and non-financial barriers beyond lack of health insurance coverage may inhibit access and use of health services by lower income families, including insufficient physician availability, health plan coverage limitations, high co-payments, transportation barriers, and time constraints [37, 38].

Addressing inequities in children’s health care coverage and access to quality health services seems an important component of any strategy to reduce income gradients in children’s health, but it is also likely that changes outside the health care system will be necessary [38‐42]. In this study, sizable income gradients in children’s health and development remained in multivariate models that controlled for health insurance status suggesting that other interventions that address the varied social determinants of children’s health (including social-emotional health) may be necessary. This might include macro-level social policy changes to address the upstream determinants of children’s health (e.g., income inequality, unemployment, inadequate time for parenting) as well as social, educational, and public health interventions designed to provide a wide variety of supportive services for children and families (e.g., parenting classes, family mental health services, comprehensive early intervention programs) [43]. The notion of a gradient suggests that not only do poor children fare worse than upper class children, but middle class children fare worse than upper class children too. Therefore policy strategies might very well combine interventions targeting lower socioeconomic groups, along with more broad-based universal programs needed to optimize the health and developmental trajectories of all children. The US literature has placed a great deal of emphasis on the problem of child poverty and the related policy debate often centers on whether to provide various forms of means-tested poor relief or encourage individual responsibility among the poor. However, by evaluating health differences in relationship to broader social gradients, the somewhat arbitrary distinctions imposed by a specific income level become less salient and policy considerations move from a focus on poverty to that of income inequality and the promotion of social inclusion through the provision of services to enhance health and well being for all children.

This study has several limitations. All measures of children’s health and access/utilization were reported by parents and subject to potential reporting bias. Some indicators of children’s health were indicated by parent report of a physician diagnosis, meaning that children with worse health care access may have had less opportunity for detection of the condition. The study is cross-sectional and therefore no causal inferences can be made. Children’s health conditions could, for example, influence family income levels instead of the reverse. In terms of measuring the shape of the income gradients in children’s health, it is important to note that the study was limited to a five-category income variable with an upper limit of 400% FPL or greater. This means that health indicators that showed consistent linear improvements with family income could still exhibit a trend of declining health improvement beyond 400% FPL or greater. The study does not address the mechanisms responsible for income differentials in children’s health. The strong associations between income and health suggest that income serves as a proxy for a range of different family and environmental factors relevant to children’s health.

Available cross-national research shows considerable variability in the magnitude of socioeconomic gradients in human health and development across industrialized nations [21, 44‐46]. Some evidence suggests that SES gradients may be less steep in countries that offer more egalitarian health and human development polices such as the provision of adequate family income, adequate time for parenting, and generous supportive services including universal health care coverage [44, 46, 47]. Life course theoretical models posit that the origins of adult social disparities in health can be traced back to the formative childhood years of development [44, 48, 49], and a substantial amount of research suggests that the health risks associated with early social disadvantage carry forward into adulthood and may be compounded across time [50‐53]. In this regard, it seems imperative that social and health policy attention be directed at efforts to prevent the development of the substantial and consistent income gradients in children’s health found in this study.