Pediatric Annals

Stress and Child Health: An Overview

W Thomas Boyce, MD

Abstract

The stresses of life have become a familiar and seemingly ubiquitous element in the day to day practice of pediatrics. Increasingly, pediatricians are called upon to manage not only the abundant stresses of their own lives, but those of their patients as well. Attempting to augment the pediatricians facility in the management of psychological stress, this article provides an overview of recent research on stress, social support, and child health. Three central questions are addressed. First, why should stress and social support be of any concern to pediatricians? Is this not an area of human experience best left exclusively to psychiatric and social science colleagues? Second, what is clearly known of the relationship between social experience and child health? What can be said with confidence of the link between psychological processes and the health and well-being of children? And third, what is not known or understood in the relationship of social experience to health? Particularly with regard to childhood, what are the questions that remain to be answered about the effects of stress and support on health and illness experience?

The primary and most important answer to the first of these questions is that children are heavily, perhaps even increasingly, exposed to psychosocial stress. More than one million children in this country, for example, are involved each year in a parental divorce. The Select Panel for the Promotion of Child Health has projected that, if current trends hold, by 1990 onethird of US children will have experienced a divorce and as many as half will have spent part of childhood with only one parent. l Even more unsettling is the fact that, as of 1982, more than two and a half million US children had lost at least one of their parents through death.2

In addition to the loss of parents through death or divorce, children are also exposed to many other forms of bereavement and psychological stress. Nearly 18% of US children move with their families to a new home each year, often disrupting important attachments to both people and places.2 A very conservative estimate is that at least 1% to 2% of children are physically or sexually abused in this country sometime during childhood, and three-quarters of a million new cases of child abuse are reported to child protection agencies each year. ' Finally, recent evidence indicates that over half of US teenagers believe there will be a nuclear war in their lifetime, and one-third report concerns about war as one of their top three worries. 3 While these observations are not meant to invoke nostalgia for some idyllic past, there can really be no doubt that the time in which we are living is a difficult and often stressful time to be a child.

Perhaps as a consequence of experiences such as these, it is also apparent that a growing proportion of the health concerns presented to pediatricians are social and psychological in nature. Table I shows the prevalence of psychosocial problems seen in pediatric primary care settings, as estimated in five separate reports published since 1979. 4'8 Each study was conducted in a relatively large patient population, and taken together, the studies cover a rather broad range of practice settings and childhood groups. While there is some variability in the prevalence rates found, it is clear that a substantial proportion of contemporary pediatric problems center on issues of behavioral, educational, and social concerns.

Table

Interestingly, in at least some of the studies concurrently assessing both stress and social support, there appears to be an interaction effect between these two categories of psychosocial variables. Figure 4 shows…

The stresses of life have become a familiar and seemingly ubiquitous element in the day to day practice of pediatrics. Increasingly, pediatricians are called upon to manage not only the abundant stresses of their own lives, but those of their patients as well. Attempting to augment the pediatricians facility in the management of psychological stress, this article provides an overview of recent research on stress, social support, and child health. Three central questions are addressed. First, why should stress and social support be of any concern to pediatricians? Is this not an area of human experience best left exclusively to psychiatric and social science colleagues? Second, what is clearly known of the relationship between social experience and child health? What can be said with confidence of the link between psychological processes and the health and well-being of children? And third, what is not known or understood in the relationship of social experience to health? Particularly with regard to childhood, what are the questions that remain to be answered about the effects of stress and support on health and illness experience?

The primary and most important answer to the first of these questions is that children are heavily, perhaps even increasingly, exposed to psychosocial stress. More than one million children in this country, for example, are involved each year in a parental divorce. The Select Panel for the Promotion of Child Health has projected that, if current trends hold, by 1990 onethird of US children will have experienced a divorce and as many as half will have spent part of childhood with only one parent. l Even more unsettling is the fact that, as of 1982, more than two and a half million US children had lost at least one of their parents through death.2

In addition to the loss of parents through death or divorce, children are also exposed to many other forms of bereavement and psychological stress. Nearly 18% of US children move with their families to a new home each year, often disrupting important attachments to both people and places.2 A very conservative estimate is that at least 1% to 2% of children are physically or sexually abused in this country sometime during childhood, and three-quarters of a million new cases of child abuse are reported to child protection agencies each year. ' Finally, recent evidence indicates that over half of US teenagers believe there will be a nuclear war in their lifetime, and one-third report concerns about war as one of their top three worries. 3 While these observations are not meant to invoke nostalgia for some idyllic past, there can really be no doubt that the time in which we are living is a difficult and often stressful time to be a child.

Perhaps as a consequence of experiences such as these, it is also apparent that a growing proportion of the health concerns presented to pediatricians are social and psychological in nature. Table I shows the prevalence of psychosocial problems seen in pediatric primary care settings, as estimated in five separate reports published since 1979. 4'8 Each study was conducted in a relatively large patient population, and taken together, the studies cover a rather broad range of practice settings and childhood groups. While there is some variability in the prevalence rates found, it is clear that a substantial proportion of contemporary pediatric problems center on issues of behavioral, educational, and social concerns.

Table

TABLE IPREVALENCE OF PSYCHOSOCIAL HEALTH ISSUES IN PEDIATRIC SETTINGS

TABLE I

PREVALENCE OF PSYCHOSOCIAL HEALTH ISSUES IN PEDIATRIC SETTINGS

There is also evidence that the proportion of children with significant psychosocial problems has been increasing steadily in recent years. Figure I shows data - taken from Roghmann's studies of psychiatric referrals among children in Monroe County, New York - that demonstrate a substantial, linear increase in the prevalence of problems requiring psychiatric intervention.9 Furthermore, the secular trend reflecting increasing need for such services, held for both white and non-white children over the past two and a half decades beginning in 1960. Finally, it has become apparent that children with psychosocial problems develop a disproportionate number of minor medical illnesses - such as respiratory disease and otitis media - and utilize health services at a frequency that exceeds that of their peers without psychosocial concerns. Figure 2, taken from the work of Dr. Barbara Starfield, shows the distributions of minor medical illnesses - categorized as low, average, or high in number - for two populations of children: those with psychosocial disorders and those with none. Clearly demonstrated is the tendency for children with psychosocial problems to fall disproportionately into the average or high categories in their rates of common medical illnesses, while children without psychosocial disorders display generally low rates of such illnesses.8 Children with psychological and social difficulties thus occupy an increasing share of the pediatrician's time - for both psychosocial and biomedical concerns.

The epidemiologic observation of a confluence between psychological and biological problems in the lives of certain children suggests the possibility of a causal relationship between these two categories of experience. It is now reasonably clear that a small, but reliable, association is found between experiences of stress and a variety of child health outcomes. Figure 3 shows, in part, the results of four representative studies in which a spécifie health outcome - the number of streptococcal illnesses, the number of injuries, the duration of respiratory illnesses, or overall morbidity - was examined as a function of the extent or intensity of stressful experience sustained by the study population of children.10"13 Each outcome was positively and significantly related to the level of stress experienced during the months immediately preceding the measurement of outcomes. In Meyer and Haggerty's study of streptococcal disease, for example, the mean number of illnesses increased progressively in children sustaining medium and high levels of stress, relative to those with low levels of stressful experience. I0 It is important to note that the relationship was generally small in terms of its clinical magnitude. For example, in the study of Boyce et al on respiratory disease, the mean duration of illnesses increased from 7 to about 10 days between children reporting low vs. high numbers of stressful events. 12 Nevertheless, there is a striking consistency in the general pattern of these results, suggesting that highly stressed children are at greater risk for changes in health.

Figure 1. Secular trends in the prevalence of childhood psychosocial problems (Adapted from Roughmann et al. 19829).

Figure 1. Secular trends in the prevalence of childhood psychosocial problems (Adapted from Roughmann et al. 19829).

Figure 2. Distributions of minor medical illnesses among children with and without psychosocial problems (Adapted from Starfield et al, 1984a).

Figure 2. Distributions of minor medical illnesses among children with and without psychosocial problems (Adapted from Starfield et al, 1984a).

Figure 3. The results of four studies in which specific health outcomes were examined as a function of the extent or intensity of stressful experience.

Figure 3. The results of four studies in which specific health outcomes were examined as a function of the extent or intensity of stressful experience.

In addition to the relationship between stress and child health, there is now substantial evidence of an association between the availability of social support and a child's risk of disease or disability. Table 2, for example, shows the results of a study relating the frequency of significant neonatal complications to the degree of maternal social support, in a population of single, adolescent mothers we have followed at the University of Arizona. H Social support was measured in an interview format, and scores were categorized into three groups of approximately equal size - reflecting low, medium, and high levels of social support. Among infants of young mothers with high social support, only 3.6% sustained complications. On the other hand, infants of mothers with medium or low social support experienced complications in 15.6% and 13.8% of cases, respectively. Relative to mothers with high support, those with only average or low support delivered infants with approximately four times the rate of neonatal complications.

Other work, such as that by Crockenberg shown in Table 3, has demonstrated the effects of social support on the security of maternal-infant attachment.15 In this study, mother- infant pairs with low support from family and friends demonstrated inadequate or insecure attachment in 55% of cases. By contrast, only 10. 7% of mothers and infants with high social support were found to have insecure attachments, indicating a five times greater risk among those with poor support. These studies and others provide tentative, but increasingly convincing, evidence that the character and frequency of a child's supportive social interactions are related in important ways to health and illness experience.

Table

TABLE 2RELATIVE RISK OF NEONATAL COMPUCATIONS BY SOCIAL SUPPORT SCORE (Boyce et al, 1984)

TABLE 2

RELATIVE RISK OF NEONATAL COMPUCATIONS BY SOCIAL SUPPORT SCORE (Boyce et al, 1984)

Table

TABLE 3RELATIVE RISK OF INSECURE MATERNAL-INFANT ATTACHMENT BY SOCIAL SUPPORT SCORE

TABLE 3

RELATIVE RISK OF INSECURE MATERNAL-INFANT ATTACHMENT BY SOCIAL SUPPORT SCORE

Figure 4. The interaction effects between low and high social support.

Figure 4. The interaction effects between low and high social support.

Interestingly, in at least some of the studies concurrently assessing both stress and social support, there appears to be an interaction effect between these two categories of psychosocial variables. Figure 4 shows the nature of this interaction in three study populations, the last of which was a childhood population.1,2,16,17 In each of these studies - which examined psychiatric symptoms, pregnancy complications, and the duration of respiratory illness - a major increase in illness outcomes was found only in the subpopulation with the combination of high stress and low social support. In study subjects with high social support scores, increasing levels of stressful experience resulted in no elevations in illness outcomes. Although interaction effects such as these have been inconsistent, it is reasonable to conclude that both stressful and supportive social experience act conjointly in some manner to affect children's overall susceptibility to disease.

A final consideration is the question of what is not known. It is tempting to answer that nearly everything is unknown in this particular area of investigation. Clearly, we have attained only the most coarse and rudimentary understanding of the relationship between psychological events and biological processes. A profusion of research questions await even preliminary investigation. Among those perhaps most interesting and worthy of study are the following:

1. What accounts for the variability in children's responses to stressful and supportive experience? Why do we reliably find such extreme differences in the resilience or vulnerability of individual children?

2. Why does a predisposition toward resilience or vulnerability persist over time? There is growing evidence, for example, that children who experience excessive morbidity may be systematically different in terms of their exposure to and/or response to psychological stress. What is the process by which this phenomenon of persistence begins?

3. What are the effects of child development on the relationship between stress and illness? The work of Lewis et al has suggested that the stresses of life, as perceived by children, may be quite different from those derived from an adult perspective.18 Other work, such as that by Birtchnell on parental death, indicates that the developmental timing of stressful events may be critical to their short- and long- term effects on children.19 How do developmental processes affect a child's response to stress and support?

4. What are thé biological pathways between stress and illness? A rapidly expanding area of research is the immunologie and neuroendocrine correlates of social experience. Human stress has now been associated, for example, with a variety of immunologic and reticuloendothelial changes, ranging from involution of the thymus and spleen to suppression of interferon production and impairment of lymphocyte cytotoxicity.20

5. Can social support become a strategy for intervention? In many cases, little can be done to avert the occurrence of stressful life events, in the lives of both children and adults. The implicit promise in research on the stress-modifying role of social support, however, has been the hope that strategies for fostering supportive interaction could be used in the effective prevention of disease.

A great deal remains to be learned. While it is increasingly apparent that stressful and supportive social experience figures prominently in the illnesses that children incur, the psychological and biological processes that underlie these associations remain largely unknown. And although social factors are now clearly among those which the pediatrician must address, the optimal approach to such factors in the clinical promotion of child health is far from clear. At the very least, pediatricians are being called by these epidemiologic observations to rethink the boundaries of their roles and to reevaluate the importance of supportive exchange in their interactions with children and families.

REFERENCES

1. Select Panel for the Promotion of Child Health: Better Health for Our Children: A National Strategy. Dept. of Health and Human Services, 1981.

2. U.S. Bureau of the Census: Statistical Abstract of the United States. U.S. Dept. of Commerce, 1984.

3. Goldenring JM, Doctor RM: Adolescents' concerns about the threat of nuclear war. Testimony before the U.S. House of Representatives Select Committee on Children. Youth and Families, 1983.

4. Goldberg ID, Regier DA, Mclnemy TK, et al: The role of pediatrician in the delivery of mental health services to children, Pediatrics 1979; 63:898-909.

5. Starfield B, Gross E, Wood M, et al: Psychosocial and psychosomatic diagnoses in primary care of children. Pediatrics 1979; 66: 1 59- 167.

6. Nader PR, Ray L, Brink SG: The new morbidity: Use of school and community health care resources for behavioral, educational, and social-family problems. Pediatrics 1984; 67:53-60.

7. Goldberg ID, Roghmann KJ, Mclnemy TK, et al: Mental health problems among children seen in pediatric practice: Prevalence and management. Pediatrics 1984; 73:278-293.

8. Starfield B, Kae H, Gabriel A, et al: Morbidity in childhood - A longitudinal view. N Engl; Med 1984; 310:824-829.

9. Roghmann KJ, Babigian HM, Goldberg ID, et al: The increasing number of children using psychiatric services: Analysis of a cumulative psychiatric case register. Pediatrics 1982; 70:790-801.

10. Meyer RJ, Haggerty RJ: Streptococcal infection in families: Factors affecting individual susceptibility. Pediatrics 1962; 29:539-549.

11. Padilla ER, Rohsenow DJ, Bergman AB: Predicting accidenr frequency in children, Pediatrics 1976; 58:223-226.

12. Boyce WT, Jensen EW, Cassel JC, et al: Influence of life events and family routines on childhood respiratory tract illness. Pediatrics 1977; 60:609-615.

13. Beautrais AL, Fergusson DM, Shannon FT: Life events and child morbidity: A prospective study. Pediatrics 1982; 70:935-940.

14. Boyce WT, Schaeter C, Uitti C: Permanence and change: Psychosocial factors in the outcomes of adolescent pregnancy. Presented at the Annual Meeting of the Society for Behavioral Pediatrics; San Francisco, April 1984.

15. Crockenberg SB: Infant irritability, mother responsiveness, and social support influences cm the security of infant-mother attachment. Child Dev 1981; 52.857-865.

16. Lin N, Simeone RS, Ensel WM: Social support, stressful life events, and illness: A model and an empirical test. J Health Soc Behav 1979; 20:108-1 19,

17. Nuckolls KB, Cassel J, Kaplan BH: Psychosocial assets, life crisis and the prognosis of pregnancy. Am J Epidemiol 1972; 95:431-441.

18. Lewis CE, Siegel JM, Lewis MA: Feeling bad: Exploring sources of distress among preadolescent children. Am J Public Health 1972; 74:117-122.

19. Birtchnell J: Early parent death and mental illness. Br J Psychiatry 1970; 116:281-288.

20. Palmblad J: Stress and immunologic competence: Studies in man, in Ader R (ed): Psychoneuroimmunology, New York. Academic Press, 1981.

TABLE I

PREVALENCE OF PSYCHOSOCIAL HEALTH ISSUES IN PEDIATRIC SETTINGS

TABLE 2

RELATIVE RISK OF NEONATAL COMPUCATIONS BY SOCIAL SUPPORT SCORE (Boyce et al, 1984)

TABLE 3

RELATIVE RISK OF INSECURE MATERNAL-INFANT ATTACHMENT BY SOCIAL SUPPORT SCORE

10.3928/0090-4481-19850801-05

Sign up to receive

Journal E-contents