- Email: [email protected]
Issues in Measuring Child Health
Perspective
Issues in Measuring Child Health Marie C. McCormick, MD, ScD Assessing child health is critical to a variety of child health, educational, and social programs, as well as to clinical research and practice. However, despite the apparent wealth of measures available in health, education, and legal and social welfare systems, little agreement exists as to what are the most important measures and for what domains of health. Development of improved measures may capitalize on advances in conceptualization of child health, including the dynamic nature of the interplay of child health and development over time. The need for enhanced measures can be illustrated by consideration of various aspects of pediatric care: well-child care, acute illness care, and the management of children with special health care needs. In particular,
the strong theoretical perspectives on aspects of development such as cognition and behavioral development need to be informed by the experience of various states of health and their developmental implications. If generalist academicians are to further their research and educational mission, they must engage, and urgently, in the development and measurement of child health.
I
cant side effects. For those conducting research relevant to educating practicing pediatricians, measures of child health that reflect the activities of the majority of pediatricians and their health promotion activities would be useful in informing educational programs and developing quality improvement strategies. In the program and policy arena, there is a need to be able to summarize across pediatric diagnoses at a more specific level for children with special health care needs so that meaningful estimates can be made of the need for pediatric subspeciality care, physicians, and others, both for chronic and acute conditions. For example, pediatric physical therapists are important for children with cerebral palsy and injuries. In addition, there needs to be a clear linkage between these more nuanced measures of child health and the indicators routinely used, so that appropriate monitoring and planning at the community level can be achieved. Addressing the issues in the measurement of child health would seem to be an important endeavor for generalist academic investigators as they interface with clinical medicine, population health, and the disciplines that contribute theoretical and empiric literature to child health. Although there appears to be a wealth of indicators of child health, many, if not most, are not well suited to many activities of general pediatricians. Newer conceptual models may provide better guidance in developing better measures for use, and potential applications of these models to well-child care, acute illness care, and the management of chronic illness in childhood can be illustrated. However, there are significant challenges to the development of more refined and meaningful measures of child health.
KEY WORDS: child health status; health services research; measurement Ambulatory Pediatrics 2008;8:77–84
f an investigator or a clinician wishes to measure general cognitive development in children, then a few well-known instruments come readily to mind—for example, the Bayley Scales of Infant Development1 for younger children and one of Wechsler instruments for older children.2 If resources do not permit such assessments, then a variety of questionnaires are available that provide reasonable information.3,4 Likewise, for behavior, the use of the appropriate Achenbach scale would likely be recommended.5 However, no such ready responses come to mind if the outcome is child health. Being able to characterize and measure child health is integral to a broad range of health and social activities and programs. We use measures of child health to monitor population trends in child well-being and to identify changes that might signal the need for new interventions. Clinically, measures of child health are important to characterize the severity of a health problem and to monitor the effect of treatment acutely and over time. If health is a goal to be aspired to, then a clear definition of that goal would seem to be imperative. Yet little consensus has emerged as to what constitutes child health. The lack of an integrated and useful definition of child health in a developmental context has implications for research in child health and for planning for interventions to improve health. For translational research involving specific conditions, moving beyond mortality averted and symptoms alleviated requires a broad and structured framework, or the work risks missing potentially signifiFrom the Department of Society, Human Development and Health, Harvard School of Public Health, Boston, Mass. Address correspondence to Marie C. McCormick, MD, ScD, Department of Society, Human Development and Health, Harvard School of Public Health, 677 Huntington Ave, Boston, Massachusetts 02115 (e-mail: [email protected]). Received for publication February 5, 2007; accepted November 21, 2007. AMBULATORY PEDIATRICS Copyright Ó 2008 by Ambulatory Pediatric Association
INDICATORS VERSUS MEASURES OF CHILD HEALTH Many of the routine statistics6–8 about child health tend to be a mixture of diagnoses or specific conditions, contextual characteristics (eg, proportion of children in poverty), risk
77
Volume 8, Number 2 March–April 2008
78
McCormick
(teenage pregnancies) or protective factors (wearing seat belts), and characteristics of the health care system (insurance coverage).9 They have often been selected for monitoring child health in the community or at a population level and may have little relevance to the health of an individual child. More often than not, they are cross-sectional in nature, reflecting the prevalence of these conditions or problems, and they do not convey a life-course approach. In addition, they often focus on the absence of health or negative states and provide little sense of the combinations of and interrelationships among the conditions or problems, even in individual children. Although such indicators provide important benchmarks for accountability for community health, they are of limited use to pediatric researchers and educators. Better approximations of the concept of health involve self-ratings or proxy ratings of overall health. Generally, these are questions that ask an individual to rate health along a 4 to 5–point scale from poor to excellent. Although such questions do correlate with other measures of current and future health states, there is insufficient variance in the response to make them useful, as only a tiny minority rates their health at the lower end of the scale.10 Moreover, the linkage between these types of questions and elements of health care delivery is not always straightforward. Finally, they provide little guidance to clinicians in measuring changes in health, as they tend to be stable over time for individuals and populations. CONCEPTUAL MODELS OF HEALTH One strategy to obtain a more useful operationalization of child health would be to rely on existing conceptual models of health and to identify measures that capture the elements in these models. Currently, there are 3 major conceptual models, each with attractive features for approaching the measurement of child health. World Health Organization Definition of Health The oldest of these conceptual models is the definition of health articulated by the World Health Organization (WHO): ‘‘state of complete physical, mental and social well-being not merely the absence of disease or infirmity.’’11 The advantages of this definition lie in its multidimensionality and its focus on positive well-being in several dimensions of health. Practically, this conceptualization of health has been incorporated into a family of measures of health for children, both general and disease specific.12 However, as an adult-focused model, this definition appears relatively static and fails to convey the dynamic nature of the interplay of child health and development over time. Moreover, the applicability of the dimensions of health to children is not straightforward. For example, cognitive development might have to be considered under mental health. Finally, because it focuses on optimal health, it provides little guidance about considerations of lesser states of health. International Classification of Functioning, Disability, and Health The latter point is directly addressed by the more recent addition to the WHO Family of International Classifica-
AMBULATORY PEDIATRICS
tions, the International Classification of Functioning, Disability, and Health.13 The underlying premise is that health and disability are not separate entities but points along a continuum of functioning that concludes with death. Moreover, experiencing even temporary degrees of disability is a near-universal experience, not the experience of a few; thus, the experience of disability becomes a normative one. The underlying thinking also emphasizes the role of the environment in worsening or ameliorating the functional impact of a given disability. The purpose for the development of this conceptual framework is to provide a common metric that can be used worldwide for assessments at the individual, institutional, and social level. Data-collection materials have been developed to assess major domains of impairments of body functions and structures, activity limitations and participation restriction, and environmental factors. Although the items are scaled, no overall scoring rules are provided. However, the intent is to measure improvement in terms of functioning regardless of actual diagnosis, including improvement in functioning as a result of the elimination of environmental factors. National Academy of Science Conceptualization In addition to a focus on function, the dynamism of the interplay of health and development over time has been incorporated into a multidimensional conceptualization by the Committee of Evaluation of Children’s Health of the National Academies of Sciences and the Institute of Medicine.14 This conceptualization begins with the following definition: Children’s health is the extent to which individual children or groups of children are able or enabled to (a) develop and realize their potential, (b) satisfy their needs, and (c) develop the capacities that allow them to interact successfully with their biological, physical, and social environments.14(p33) The report goes on to identify 3 specific domains of health. The first consists of health conditions, generally in the form of diagnoses, injuries, and common symptoms, acute or chronic, ‘‘that may be anatomical, physiological or psychosocial.’’14(p35) Functioning, the second domain, is considered to reflect ‘‘the direct and indirect effects of one or more health conditions and their treatments.[on] physical, psychosocial, cognitive and social functioning.’’14(p35) The final domain is health potential, which ‘‘includes both health assets that provide the capacity to respond to physical, psychological and social challenges and risk states that increase vulnerability to other aspects of poor health.’’14(p37) These domains are projected over multiple time frames and transitions to emphasize the potentially very long-term effects of child health on adult health. The first domain of this definition has clear parallels in the International Classification of Functioning, Disability, and Health (ICF),13 which postulates a progression from anatomic derangement through to symptoms. Both emphasize function as something separate but related. The National Academy of Science (NAS)14 framework is
AMBULATORY PEDIATRICS
much more dynamic and adds the concept of ‘‘potential,’’ an important concept in assessing children because the final ‘‘product’’ of health and development has yet to be realized. With regard to the issue of measurement, the NAS document critiques currently available measures as lacking consistency of definition and ascertainment across agencies, failing to consider clustering of conditions, underreporting some conditions, especially mental health problems, and lacking data on symptoms as opposed to diagnoses. The conclusion is that a consensus is needed on the critical components of child health and means of measuring them that captures the dynamic and transactional nature of child health across the life span.14(p101–107) These conceptual models, taken together, combine many of the desirable elements that should characterize measures of child health. These elements include multidimensionality, a life-course approach emphasizing the interaction among different dimensions of health and development, the minimization and amelioration of disability that results from health conditions, and the desirability of having a single metric ranging from perfect health to death. However, significant issues arise when it comes to operationalizing these constructs into measures useful at the individual and community level. Stein15 has argued that the assumptions of single construct of ‘‘health’’ and measurement of deviations from perfect health may not be possible. The best that can be achieved is clear specification of the underlying theoretical approach to measurement and an equally clear statement of the purpose for which the measure has been developed. Her concern is underscored by comparisons of the content of the more frequently used measures of child health status16,17 that reveal the lack of comparability among measures of health status, even for the same construct. In one such comparison17 of 7 generic child health status measures across 15 dimensions of health, agreement on the content was seen in only one area: physical functioning or ability to undertake the usual activities of daily living appropriate to a child. Some instruments include cognitive and behavioral health, as would the functional domain of the NAS conceptual framework. Some include symptoms; others do not. The lack of clear conceptualization is further indicated by the overlap of items in health and functional status measures with those of health-related quality of life.18 Finally, disease-specific measures also include generic measures of some of these domains, but may also differ in content.19 Stein15 further argues that even the best measures may be proxies for the construct of health, but despite the conceptual and methodologic challenges, the further development, even of these proxies, should be a high priority. IMPORTANCE TO GENERAL PEDIATRICS Why should such development of these proxies be a priority, and why should generalist academicians undertake this challenging endeavor? To illustrate the importance of developing a consensus on the relevant domains of child health and identifying appropriate measures, I consider the activities of pediatric practice in 3 main areas: well-
Issues in Measuring Child Health
79
child care, acute conditions, and chronic or longer-lasting conditions. Well-Child Care Well-child care has been defined to include health supervision, developmental surveillance, psychosocial assessment, coordination of care, immunizations, and additional screening.20 The general purpose of these activities is to assure adequate child physical and psychosocial development, prevention of disease, and early detection of problems. Our current measurement strategies for many of these activities rely on detecting deviations (generally 2 standard deviations) from some population mean. What might be different if the goal of well-child care were to achieve ‘‘well-being’’ or some measure of ‘‘optimal health’’? Let us take the example of monitoring physical growth— certainly a well-established routine in well-child care. One might argue that the current primary aim of measuring physical growth is to detect inadequate or excessive growth, and the major outcome of this activity is assessing children’s growth appropriate to age. Consideration of the literature on normal child cognitive and social development21 might suggest the need to take a more integrative approach. Studies suggest that growth that is at the extremes of normality or that is frankly abnormal may have effects on psychosocial outcomes and influence future disease states. The literature appears more consistent for obesity, which is associated with lower self-esteem and reduced economic mobility over and above the risks conferred for adult-onset chronic disease.22,23 The literature on the effects of short stature is more controversial, with some studies finding evidence of psychosocial disability in childhood24 or poorer educational attainment and employment among adults,25 but not all studies report these deficits.26,27 Likewise, nutritional problems that may lead to poor growth at either extreme may be associated with micronutrient deficiencies that may also contribute to cognitive and behavioral difficulties,28 and even in developed countries, the experience of hunger may lead to behavioral and emotional problems.29 Thus, it would appear that monitoring physical growth parameters might not be sufficient to capture all the issues surrounding adequate physical growth, and that other outcomes may be important. Reframing well-child care activities in terms of a general model of optimizing health might suggest needed outcomes or a different way of considering outcomes in other areas of care. For example, assessing the effect of recommendations for anticipatory guidance on physical activity30 might require measures of physical fitness, stamina, and the coordination and skills for participating in physical activities. Moreover, the arguments30 about the potential mental health benefits of participation in physical fitness activities underscore the importance of an integrative approach to assessing the effect of well-child care. Reconsideration of immunizations might argue for their role in a broader strategy for assuring optimal immune function, with breast-feeding as part of that strategy by potentially reducing the risk of allergy.
80
McCormick
In summary, the goals of well-child care would be seen as approaching this state of well-being as in the WHO definition and assuring potential, as in the NAS model. However, the measures of child health would have to be broadened to include not only the relevant psychosocial dimensions, but also important physiologic ones. Acute Illness A certain amount of experience with illness or injury is almost a normative experience; indeed, it may be necessary for the appropriate development of some physiologic systems31 and recognition of risky situations.32 Clearly, most parents also see this experience as unexceptional because, despite relatively frequent illness (which may occur monthly33), most parents consider their children to be ‘‘healthy.’’34 In this case, however, the developmental implications of the experience of brief illnesses is not clear in that the literature on child development provides little insight except to view illness as a familial stressor.35,36 As with well-child care, more integrative approaches might suggest potential outcomes for acute illness care besides symptomatic relief and prevention of more serious sequelae. For some families, the experience of frequent illness in the child may lead to parental distress with potential for developmental problems and increased use of medical care. Such families may be those with poor, depressed mothers37 or those who perceive that their child as more fragile after a diagnostic experience (the ‘‘labeling’’38–40 phenomenon) or after an illness perceived as severe (‘‘vulnerable child syndrome’’41,42). In addition, acute illness may have the potential to affect other aspects of development in unanticipated ways unless these effects are sought. For example, concerns about the effects of recurrent otitis media on the emergence of speech and language arose much earlier than concerns about effects on attention and behavior, but a consideration of the potential effect of dampened hearing on attention might have suggested the latter as well, although both concerns remain highly debatable and may reflect other characteristics of the environment.43–45 Alternatively, the experience of minor illness, which is likely to resolve successfully, may have a positive impact. Parmelee46 has suggested that the experience of acute health problems may foster the development of child resilience and management of discomfort. Theoretical models of the factors that may influence risk and resilience have been developed47 and could contribute to both the development of appropriate parental guidance and outcome measures, although most of this work has focused on chronic childhood conditions, not acute illness. However, if we are to operationalize Parmalee’s observation, then we need more nuanced measures of child resilience and ability to handle acute discomfort, of parental knowledge and confidence in managing acute illness, and of the delineation of those situations requiring more intensive intervention to assure that these outcomes will occur. Chronic Illness/Disability Recognition of the limits of the information from studies of single diseases and consideration of some of the overlap
AMBULATORY PEDIATRICS
among conditions has prompted a shift toward a noncategorical approach to chronic illness in childhood, such as children with special health care needs. The argument48–50 supporting such a shift contends that children with chronic illness and their families share common experiences, problems with accessing health care, needs for coordination of care, and other events that transcend the diagnosis. The noncategorical approach to the consideration of chronic illness in childhood has come to dominate the general pediatric literature, and indeed, population-based studies reveal distinctive patterns of sociodemographic characteristics, health care use, and insurance coverage by using a variety of definitions51–53 for the 15% to 20% of children with a physical, mental, or behavioral health problem.5,49,54–57 Similarities at the population level, however, mask substantial definitional and measurement issues for individual children. First, the stability of the prevalence across service sectors and over time may provide an illusory sense that these figures refer to the same children. Different classification algorithms identify different children as having special needs.58 There may also be imperfect correlation between those identified as having special health care needs and clinical diagnoses, with 20% of those identified by one algorithm lacking an associated diagnosis or condition.59 In addition, this classification, while identifying those with one or more problems, does not deal well with the clustering and overlap of various categories of disability.60 Different child service systems may identify different children as having special needs or characterize the same child in different ways. For example, the diagnosis used for classification for special education may not correspond completely with medical diagnoses. Thus, one important activity would be the development of a uniform characterization of children across different service sectors as to which children would fall into the category of children with special health care needs, perhaps by using a functional approach suggested in either the ICF or NAS models. That is not to say that measures of childhood functioning in the context of chronic illness do not exist, but this literature has been critiqued as having several flaws. A review has summarized these flaws as including a focus on the negative aspects of childhood chronic illness (using a deficit model); the lack of information on how the family responds and adapts to illness; the lack of longitudinal data; a focus primarily on maternal response to childhood illness, with less information about fathers and the children themselves; and the reliance on studies of small groups of children with a single disease.48 Thus, the current literature may not be sufficient to examine the assumption that the experiences and needs of these children and their families are sufficiently similar to justify such a broad grouping. This assumption may be useful in planning for and financing health and other services needs but may not be appropriate when considering the developmental impact of such ongoing conditions on both the child and the family. Moreover, current measures may not adequately capture family dynamics and positive adaptation to ongoing childhood illness.
AMBULATORY PEDIATRICS
Consideration of the heterogeneity of the conditions of children with special health care needs suggests that a more nuanced approach to conceptualization and measurement of outcomes is needed. Conditions vary considerably in the severity, course, and implications for the child and his or her family, and the developmental tasks that the family and child face. That different conditions are viewed differently by families, thereby potentially altering the family response to illness, is suggested by studies that indicate that illnesses characterized by invisibility, remitting-relenting courses, and uncertain prognosis cause more distress or have children viewed as less healthy than other conditions.60,61 Likewise, families may adjust better to conditions present at birth, even when producing more disability, than conditions that emerge in a previously healthy child, as illustrated by the fact that parents of children with spina bifida rate their children as healthier than those with children with juvenile rheumatoid arthritis, despite the fact that the latter have far fewer functional limitations.61 One might hypothesize that parents of children with conditions present at birth may recalibrate what it means for their child to be healthy, or have a different developmental task than parents with a previously healthy child who develops a chronic illness. One way this may occur is that parents may discriminate between the sequelae of a birth defect and other aspects of health. One mother of a child with spina bifida in this study61 reported that her child had a disability but was otherwise healthy. Moreover, in attempting to gauge the factors affecting parental reaction and adaptation to ongoing health problems in their children, we cannot assume that their understanding of the pattern of illness is equivalent to pathogenesis, even to the concept of ‘‘present at birth.’’ For example, in a study that involved a series of questions to ascertain child health problems from birth to the end of the first year of life,62 one parent reported that the child was not born with any problems, but was hospitalized 3 times in the first year (for congestive heart failure, cardiac catheterization, and corrective heart surgery) and was now followed in a program for children with Down syndrome. The importance of capturing other dimensions of ill health has been incorporated into the conceptualizations of chronic health conditions emerging from different theoretical perspectives,47,63,64 and 2 of these formulations63,64 incorporate consideration of relatively severe acute conditions. Among the dimensions considered important in these frameworks and suggested by others47,65 are the age at onset, duration, mobility-activity restrictions, visibility of signs and symptoms of the condition, course (eg, progressive, stable, unpredictable), prognosis, and effect on cognitive and emotional well-being. In a review of selected studies examining the effect of these dimensions on family and child adaptation, Thompson and Gustafson47 found that the impact of different aspects of ill health were sometimes, but not always, as the theory would predict. However, many of the studies reviewed involved relatively small samples and were restricted to single conditions, thus failing to capture the information that might be gleaned by cross-condition comparisons. In addition,
Issues in Measuring Child Health
81
more clinical and biologic observations are needed to inform the measurement. For example, child psychosocial adaptation to chronic illness appears worse in conditions affecting the brain.47 However, because much of the measurement of psychosocial adaptation concerns behavioral observations, it is not clear whether it is the child’s adaptation or that the observations reflect symptoms related to the primary neurological problem. Similar concerns have been raised about the extensive literature on the outcomes of very premature infants, which fail to consider the multiple health issues affecting these children that might affect their cognitive and social development other than direct neurological events.66 Taking a more integrated and developmentally focused approach toward chronic childhood conditions may yield insights into risk and protective factors for better or worse functional outcomes, as well as factors affecting treatment, as illustrated by a series of papers on attention-deficit/ hyperactivity disorder.67 In particular, this series provides an example of the application of the ICF to describe a condition68 and to guide the development of appropriate interventions.69 As indicated by these reports, a focus on function by using approaches like the ICF or the NAS model, with consideration of family adaptation, could provide needed information, both for the individual child and for larger groups of children. For the individual child, measures that could reflect improvement in function would offer a way of understanding the effects of individual referrals or therapeutic programs. If the measures had sufficient detail and could be compiled across children with different conditions, a better sense of the need for pediatric subspecialists and child-trained therapists would be possible. Further, there is an emerging theoretical framework70 to guide the examination of the effects of ill health on child development, although these models may need some reformulation to acknowledge that ill health may not just be another stressor but a more integral aspect of human development. However, understanding the full impact of ongoing conditions would require the dynamic and interactive approach suggested by the NAS model. CONCLUSION I began my career as a health services researcher interested in program evaluation, especially programs focused on improving pregnancy outcomes and early child health. That interest, in turn, led to research on the outcomes of very premature infants. Coming from the perspective of health services research, I was familiar with the approach that underlay instrumentation such as the Child Health Questionnaire17 and excited by what seemed to be a logical approach to thinking about measuring child health on the basis of a conceptual model (the WHO definition) and moving beyond diagnoses. However, in applying this approach to studies of the outcomes of very premature infants, especially those observed at school age, I realized that we were far from being able to characterize child health in a way that might capture the complexity of the situation of any one child, and from having an established
82
McCormick
armamentarium of measure to evaluate the effects of specific interventions and programs. We not only do not have a clear concept of health, but we also lack a conceptual framework linking biological and environmental factors in a way that would help us understand the effect of interventions, including the effects of many elements of general pediatric care. I am not alone in concluding that we lack of a clear definition or consensus on what is meant by child health.14(p101–107),17,70 Further, although most reviews agree that appropriate measures should be multidimensional,19,70 there is little concordance on what those dimensions should be from the perspective of the child, the family, or society at large. Moreover, because many of the child health measures reflect adaptation of adult instruments, measures for young children and infants tend to be sparse. The situation might not be so limited if clear theoretical perspectives linking health and development permitted capitalizing on the concepts and instruments available from studies of normal child development. As Eiser and Morse19 indicate, for consideration of ‘‘wellbeing,’’ for example, one could have a basic, generic measure of child health to which could be added measures of such constructs as self-esteem, self-competence, or emotional development, if relevant to the context. However, there is little theoretical guidance regarding when such constructs would be important child health outcomes. It would seem that achieving some consensus and either developing or assessing measures to characterize child health broadly would be a role for generalist academic pediatricians, both as researchers in child health broadly and as educators of generalist providers. Recommendations for the research training of academic generalists71 already include many of the research design and analytic methods needed for this endeavor, although greater depth in relevant theoretical areas might well be needed, particularly child development and measurement theory. Academic generalists also are in a unique position to bridge the worlds of social science and biology. Existing measures of child status have been criticized as too subjective for clinical and programmatic usefulness. Their usefulness would be greatly enhanced if these measures could be validated against other measures with physiologic or clinical significance. If, for example, a parent or child reports a limitation in activity due to health, like difficulty in running or walking certain distances, their response does not discriminate among difficulties in joint or muscle control or range of motion, decreased stamina, or developmental problems that result in coordination difficulties. A clearer tie to physical observations would assist clinicians in developing intervention strategies. Moreover, establishing such linkages would help in calibrating parental reports and contribute to the theoretical development of when parents report more or less difficulty than would be anticipated by the measurement of, for example, stamina. In this regard, a program of refining and validating measures of child health status might well be incorporated into the role of the new Clinical and Translational Science Awards72 units.
AMBULATORY PEDIATRICS
Because many academic generalists also bridge the worlds of pediatrics and public health, they could also work to develop clear linkages between what is measured in greater detail at the individual level and indicators of child health. Such linkages would facilitate program planning and interpretation in changes in indicators and variations in different groups. This argument is not meant to gloss over the considerable methodologic challenges involved in developing strong measures of child health, like the reliance on proxy reporters (caretakers) and cognitive development issues in eliciting child responses.18 However, the first step in this process is to begin to articulate a conceptual framework of the relevant domains of child health and aligning this framework with existing measures and their usefulness, and to identify measurement gaps. In addition, such an activity should also identify potential factors influencing health—for example, demographic ones, like maternal educational attainment, or health services ones, like type and quality of care—to generate hypotheses that would lead to the development of a more comprehensive construct of child health and its context. Even though it has its problems and raises concerns, psychologists have accomplished a lot with the IQ test. Maybe it is time to develop our own HQ (health quotient) test to improve children’s health, or at least to reduce the frustration of future health services researchers looking for useful measures of outcome. ACKNOWLEDGMENTS The author acknowledges the support of the Institute of Medicine as this work was done while a Scholar in Residence there. In addition, she thanks Elena Nightingale and Edward Schor for comments and suggestions on a draft. Supported in part by the Commonwealth Fund (grant 20040493).
REFERENCES 1. Bayley N. Bayley Scales of Infant Development. 3rd ed. San Antonio, Tex: Psychological Corporation; 2005. 2. Kaplan E, Fein D, Kramer J, Delis D, Morris R. Weschler Intelligence Scale for Children—Fourth Edition Integrated. San Antonio, Tex: Harcourt Assessment Inc; 2007. 3. Glascoe FP, Martin ED, Humphrey S. A comparative review of developmental screening tests. Pediatrics. 1990;86:547–554. 4. Glascoe FP, Sandler H. Value of parents’ estimates of children’s developmental ages. J Pediatr. 1995;127:831–835. 5. Achenbach TM, Edelbrock C. Manual for Child Behavior Checklist and Revised Child Behavior Profile. Burlington, Vt: Queen City Printers; 1983. 6. Child Trends DataBank. Available at: http://www.childtrendsdatabank. org. Accessed December 7, 2007. 7. http://www.fcd-us.org/issues/issues_show.htm?doc_id=447092. Accessed December 14, 2007. 8. National Institute of Child Health and Human Development, NIH, DHHS. America’s Children: Key National Indicators of Well-Being. Washington, DC: US Government Printing Office; 2007. 9. Peoples-Sheps MD, Guild PA, Farci AM, et al. Model indicators for maternal and child health: an overview of process, product and applications. Matern Child Health J. 1998;2:241–256. 10. Ried LD, Tueth MJ, Handberg E, Nyanteh H. Validating a self-report measure of global subjective well-being to predict adverse clinical outcomes. Qual Life Res. 2006;15:675–686. 11. Preamble to the Constitution of the World Health Organization as adopted by the International Health Conference, New York, NY,
AMBULATORY PEDIATRICS
12. 13.
14.
15. 16.
17.
18.
19. 20.
21.
22. 23. 24.
25.
26.
27.
28.
29.
30.
31.
32. 33.
34. 35. 36.
June 1946; signed July 22, 1946, by representatives of 61 states (Official Records of the World Health Organization, no. 2, p.100) and entered into force on 7 April 1946. QualityMetric: Health Outcomes Solutions. Available at: http://www. qualitymetric.com/products/. Accessed December 7, 2007. World Health Organization. International Classification of Functioning, Disability and Health (ICF). Available at: http://www.who.int/ classifications/icf/en/. Accessed December 7, 2007. Committee on Evaluation of Children’s Health. Children’s Health, the Nation’s Wealth: Assessing and Improving Child Health. Washington, DC: National Academy Press; 2004. Stein RE. Measurement of children’s health. Ambul Pediatr. 2004; 4(4 suppl):365–370. Olson LM, Lara M, Pat Frintner M. Measuring health status and quality of life for US children: relationship to race, ethnicity, and income status. Ambul Pediatr. 2004;4(4 suppl):377–386. Landgraf JM, Abetz L, Ware JE. Child Health Questionnaire (CHQ): A User’s Manual. Boston, Mass: Health Institute, New England Medical Center; 1996. 30–32. Drotar D. Validating measures of pediatric health status, functional status, and health-related quality of life: key methodological challenges and strategies. Ambul Pediatr. 2004;4(4 suppl):358–364. Eiser C, Morse R. Quality-of-life measures in chronic disease of childhood. Health Technol Assess. 2001;5:1–157. Kuo AA, Inkelas M, Lotstein DS, Samson KM, Schor EL, Halfon N. Rethinking well-child care in the United States: an international comparison. Pediatrics. 2006;118:1692–1702. Shonkoff JP, Philips DA, eds. From Neurons to Neighborhoods: The Science of Early Childhood Development. Washington, DC: National Academies Press; 2000. Must A, Strauss RS. Risks and consequences of childhood and adolescent obesity. Int J Obes Relat Metab Disord. 1999;23(suppl 2):S2–S11. French SA, Story M, Perry CL. Self-esteem and obesity in children and adolescents: A literature review. Obesity Res. 1995;3:479–490. Molinari E, Sartori A, Ceccarelli A, Marchi S. Psychological and emotional development, intellectural capabilities and body image in short normal children. J Endocrinol Invest. 2002;25:321–328. Zimet GD, Owens R, Dahms W, Cutler M, Litvene M, Cutler L. Psychosocial outcome of children evaluated for short stature. Arch Pediatr Adolesc Med. 1997;151:1017–1023. Downie AB, Mulligan J, Stratford RJ, Betts PR, Voss LD. Are short normal children at a disadvantage? The Wessex Growth Study. BMJ. 1997;314:97–100. Pilpel D, Lieberman E, Zadik Z, Carel CA. Effect of growth hormone treatment on quality of life of short stature children. Horm Res. 1995; 44:1–5. Lozoff B, De Andraca I, Castillo M, Smith JB, Walter T, Pino P. Behavioral and developmental effects of preventing iron-deficienccy anemia in healthy full-term infants. Pediatrics. 2003;112:846–854. Weinreb L, Wehler C, Perloff J, Scott R, Hosmer D, Sagor L, Gundersen C. Hunger: its impact on children’s health and mental health. Pediatrics. 2002;110:e41. Bright Futures at Georgetown University. Bright Futures in practice: physical activity. Available at: http://www.brightfutures.org/ physicalactivity/. Accessed December 7, 2007. Wills-Karp M, Santeliz J, Karp CL. The germless theory of allergic disease: revisiting the hygiene hypothesis. Nat Rev Immunol. 2001; 1:69–75. Robertson LS. Injury Epidemiology. Research and Control Strategies. 2nd ed. New York, NY: Oxford University Press; 1998. 117–123. Simpson L, Owens PL, Zodet MW, et al. Health care for children and youth in the United States: annual report on patterns of access, utilization, quality and expenditure by income. Ambul Pediatr. 2005;5:6–44. Hojer B, Sterky G, Widlund G. Acute illnesses in young children and family responses. Acta Paediatr Scand. 1987;76:624–630. Heymann SJ, Earle A, Egleston B. Parental availability for the care of sick children. Pediatrics. 1996;98:226–230. Heymann SJ, Toomey S, Furstenberg F. Working parents: what factors are involved in their ability to take time off from work when their children are sick? Arch Pediatr Adolesc Med. 1999;153:870–874.
Issues in Measuring Child Health
83
37. McCarthy P, Walls T, Cicchetti D, et al. Prediction of resource use during acute pediatric illnesses. Arch Pediatr Adolesc Med. 2003; 157:990–996. 38. Perrin EC, West PD, Culley BS. Is my child normal yet? Correlates of vulnerability. Pediatrics. 1989;83:355–363. 39. Bergman AB, Stamm SJ. The morbidity of cardiac non-disease in school children. N Engl J Med. 1967;276:1008–1013. 40. Fisher ES, Welch HG. Avoiding the unintended consequences of growth in medical care—how might more be worse? JAMA. 1999; 281:446–453. 41. Thomasgard M, Metz WP. The vulnerable child syndrome revisited. J Dev Behav Pediatr. 1995;16:47–53. 42. McCormick MC, Brooks-Gunn J, Workman-Daniels K, Peckham GJ. Maternal rating of child health at school age: does the vulnerable child syndrome persist? Pediatrics. 1995;16:47–53. 43. Roberts J, Hunter L, Gravel J, et al. Otitis media, hearing loss and language learning: controversies and current research. J Dev Behav Pediatr. 2004;25:110–122. 44. Bennett KE, Haggard MP. Behaviour and cognitive outcomes from middle ear disease. Arch Dis Child. 1999;80:28–35. 45. Hooper SR, Ashley TA, Roberts JE, Zeisel SA, Poe MD. The relationship of otitis media in early childhood to attention dimensions during the early elementary school years. J Dev Behav Pediatr. 2006;27: 281–289. 46. Parmelee AH. Illness and the development of social competence. J Dev Behav Pediatr. 1997;18:120–124. 47. Thompson RJ, Gustafson KE. Adaptation to Chronic Childhood Illness. Washington, DC: American Psychological Association; 1999. Chap 5. 48. Hayes VE. Families and children’s chronic conditions: knowledge development and methodological considerations. Sch Inq Nurs Pract. 1997;11:259–290. 49. Hobbs N, Perrin JM, Ireys HT. Chronically Ill Children and Their Families. San Francisco, Calif: Jossey-Bass Publishers; 1985. 50. Stein RE, Bauman LJ, Westbrook LE, Coupey SM, Ireys HT. Framework for identifying children who have chronic conditions: the case for a new definition. J Pediatr. 1993;122:342–347. 51. Newacheck PW, Taylor WR. Childhood chronic illness: prevalence, severity and impact. Am J Public Health. 1992;82:364–374. 52. Newacheck PW, Halfon N. Prevalence and impact of disabling chronic conditions on childhood. Am J Public Health. 1998;88: 610–617. 53. Newacheck PW, Strickland B, Shonkoff JP, et al. An epidemiologic profile of children with special health care needs. Pediatrics. 1998; 102:117–123. 54. Cadman D, Boyle M, Szatmari P, Offord DR. Chronic illness, disability and mental and social well-being. Pediatrics. 1987;79:805–813. 55. Roberts RE, Attkisson CC, Rosenblatt A. Prevalence of psychopathology among children and adolescents. Am J Psychiatry. 1998; 155:715–725. 56. Shaffer D, Fisher P, Dulcan MK, et al. The NIMH Diagnostic Interview Schedule for Children version 2.3 (DISC-2.3): description, acceptability, prevalence rates, and performance in the MECA Study. J Am Acad Adolesc Psychiatry. 1996;35:865–877. 57. Zill N, Schoenborn CA. Developmental, Learning and Emotional Problems. Health of Our Nation’s Children, 1988. Advance Data 190. Hyattsville, Md: National Center for Health Statistics; 1990. 58. Jessop DJ, Stein RE. Consistent but not the same. Effect of method on chronic condition rates. Arch Pediatr Adolesc Med. 1995;149: 1105–1110. 59. Liu C. Access, Quality, and Expenditures of Health Care for Children with Special Health Care Needs in the United States. Boston, Mass: Harvard School of Public Health; 2004:11. 60. Perrin JM, Kuhlthau K, McLaughlin TH, Ettner SL, Gortmaker SL. Chaning patterns of conditions among children receiving Supplemental Security Income disability benefits. Arch Pediatr Adolesc Med. 1999;153:80–84. 61. McCormick MC, Athreya B, Bernbaum JC, Charney EB. Preliminary observations on maternal rating of health of children: data from three subspecialty clinics. J Clin Epidemiol. 1988;41:323–329.
84
McCormick
62. McCormick MC, Wessel KM, Krischer JP, Welcher DW, Hardy JG. Preliminary analysis of developmental observations in a survey of morbidity in infants. Early Hum Dev. 1981;5:377–393. 63. Patterson J, Blum RW. Risk and resilience among children and youth with disabilities. Arch Pediatr Adolesc Med. 1996;150:692–698. 64. Perrin EC, Newacheck P, Pless IB. Issues involved in the definition and classification of chronic health conditions. Pediatrics. 1993;91: 787–793. 65. Roland JS. Families Illness and Disability. An Integrative Treatment Model. New York, NY: Basic Books; 1994:chap 2–3. 66. McCormick MC. The outcomes of very low birth weight infants: are we asking the right questions? Pediatrics. 1997;99:869–876. 67. Brown RT, Perrin JM. Measuring outcomes in attention-deficit/hyperactivity disorder. Ambul Pediatr. 2007;7(1 suppl):57–59.
AMBULATORY PEDIATRICS 68. Ustun TB. Using the International Classification of Functioning, Disease and Health in attention-deficit/hyperactivity disorder: separating the disease from its epiphenomena. Ambul Pediatr. 2007;7(1 suppl): 132–139. 69. Loe IM, Feldman HM. Academic and educational outcomes of children with ADHD. Ambul Pediatr. 2007;7(1 suppl):82–90. 70. Dreyer BP, Schonfeld DJ; for the Ambulatory Pediatric Association Research Committee. Content outline for a curriculum in research education for general academic fellowship programs. April 24, 2004. Available at: http://www.ambpeds.org/site/research/Draft Curriculum.htm. Accessed December 7, 2007. 71. http://ctsweb.org. Accessed December 19, 2007. 72. Szilagyi P, Schor EL. The health of children. Health Serv Res. 1998; 33(4 part 2):1001–1039.
Issues in Measuring Child Health Marie C. McCormick, MD, ScD Assessing child health is critical to a variety of child health, educational, and social programs, as well as to clinical research and practice. However, despite the apparent wealth of measures available in health, education, and legal and social welfare systems, little agreement exists as to what are the most important measures and for what domains of health. Development of improved measures may capitalize on advances in conceptualization of child health, including the dynamic nature of the interplay of child health and development over time. The need for enhanced measures can be illustrated by consideration of various aspects of pediatric care: well-child care, acute illness care, and the management of children with special health care needs. In particular,
the strong theoretical perspectives on aspects of development such as cognition and behavioral development need to be informed by the experience of various states of health and their developmental implications. If generalist academicians are to further their research and educational mission, they must engage, and urgently, in the development and measurement of child health.
I
cant side effects. For those conducting research relevant to educating practicing pediatricians, measures of child health that reflect the activities of the majority of pediatricians and their health promotion activities would be useful in informing educational programs and developing quality improvement strategies. In the program and policy arena, there is a need to be able to summarize across pediatric diagnoses at a more specific level for children with special health care needs so that meaningful estimates can be made of the need for pediatric subspeciality care, physicians, and others, both for chronic and acute conditions. For example, pediatric physical therapists are important for children with cerebral palsy and injuries. In addition, there needs to be a clear linkage between these more nuanced measures of child health and the indicators routinely used, so that appropriate monitoring and planning at the community level can be achieved. Addressing the issues in the measurement of child health would seem to be an important endeavor for generalist academic investigators as they interface with clinical medicine, population health, and the disciplines that contribute theoretical and empiric literature to child health. Although there appears to be a wealth of indicators of child health, many, if not most, are not well suited to many activities of general pediatricians. Newer conceptual models may provide better guidance in developing better measures for use, and potential applications of these models to well-child care, acute illness care, and the management of chronic illness in childhood can be illustrated. However, there are significant challenges to the development of more refined and meaningful measures of child health.
KEY WORDS: child health status; health services research; measurement Ambulatory Pediatrics 2008;8:77–84
f an investigator or a clinician wishes to measure general cognitive development in children, then a few well-known instruments come readily to mind—for example, the Bayley Scales of Infant Development1 for younger children and one of Wechsler instruments for older children.2 If resources do not permit such assessments, then a variety of questionnaires are available that provide reasonable information.3,4 Likewise, for behavior, the use of the appropriate Achenbach scale would likely be recommended.5 However, no such ready responses come to mind if the outcome is child health. Being able to characterize and measure child health is integral to a broad range of health and social activities and programs. We use measures of child health to monitor population trends in child well-being and to identify changes that might signal the need for new interventions. Clinically, measures of child health are important to characterize the severity of a health problem and to monitor the effect of treatment acutely and over time. If health is a goal to be aspired to, then a clear definition of that goal would seem to be imperative. Yet little consensus has emerged as to what constitutes child health. The lack of an integrated and useful definition of child health in a developmental context has implications for research in child health and for planning for interventions to improve health. For translational research involving specific conditions, moving beyond mortality averted and symptoms alleviated requires a broad and structured framework, or the work risks missing potentially signifiFrom the Department of Society, Human Development and Health, Harvard School of Public Health, Boston, Mass. Address correspondence to Marie C. McCormick, MD, ScD, Department of Society, Human Development and Health, Harvard School of Public Health, 677 Huntington Ave, Boston, Massachusetts 02115 (e-mail: [email protected]). Received for publication February 5, 2007; accepted November 21, 2007. AMBULATORY PEDIATRICS Copyright Ó 2008 by Ambulatory Pediatric Association
INDICATORS VERSUS MEASURES OF CHILD HEALTH Many of the routine statistics6–8 about child health tend to be a mixture of diagnoses or specific conditions, contextual characteristics (eg, proportion of children in poverty), risk
77
Volume 8, Number 2 March–April 2008
78
McCormick
(teenage pregnancies) or protective factors (wearing seat belts), and characteristics of the health care system (insurance coverage).9 They have often been selected for monitoring child health in the community or at a population level and may have little relevance to the health of an individual child. More often than not, they are cross-sectional in nature, reflecting the prevalence of these conditions or problems, and they do not convey a life-course approach. In addition, they often focus on the absence of health or negative states and provide little sense of the combinations of and interrelationships among the conditions or problems, even in individual children. Although such indicators provide important benchmarks for accountability for community health, they are of limited use to pediatric researchers and educators. Better approximations of the concept of health involve self-ratings or proxy ratings of overall health. Generally, these are questions that ask an individual to rate health along a 4 to 5–point scale from poor to excellent. Although such questions do correlate with other measures of current and future health states, there is insufficient variance in the response to make them useful, as only a tiny minority rates their health at the lower end of the scale.10 Moreover, the linkage between these types of questions and elements of health care delivery is not always straightforward. Finally, they provide little guidance to clinicians in measuring changes in health, as they tend to be stable over time for individuals and populations. CONCEPTUAL MODELS OF HEALTH One strategy to obtain a more useful operationalization of child health would be to rely on existing conceptual models of health and to identify measures that capture the elements in these models. Currently, there are 3 major conceptual models, each with attractive features for approaching the measurement of child health. World Health Organization Definition of Health The oldest of these conceptual models is the definition of health articulated by the World Health Organization (WHO): ‘‘state of complete physical, mental and social well-being not merely the absence of disease or infirmity.’’11 The advantages of this definition lie in its multidimensionality and its focus on positive well-being in several dimensions of health. Practically, this conceptualization of health has been incorporated into a family of measures of health for children, both general and disease specific.12 However, as an adult-focused model, this definition appears relatively static and fails to convey the dynamic nature of the interplay of child health and development over time. Moreover, the applicability of the dimensions of health to children is not straightforward. For example, cognitive development might have to be considered under mental health. Finally, because it focuses on optimal health, it provides little guidance about considerations of lesser states of health. International Classification of Functioning, Disability, and Health The latter point is directly addressed by the more recent addition to the WHO Family of International Classifica-
AMBULATORY PEDIATRICS
tions, the International Classification of Functioning, Disability, and Health.13 The underlying premise is that health and disability are not separate entities but points along a continuum of functioning that concludes with death. Moreover, experiencing even temporary degrees of disability is a near-universal experience, not the experience of a few; thus, the experience of disability becomes a normative one. The underlying thinking also emphasizes the role of the environment in worsening or ameliorating the functional impact of a given disability. The purpose for the development of this conceptual framework is to provide a common metric that can be used worldwide for assessments at the individual, institutional, and social level. Data-collection materials have been developed to assess major domains of impairments of body functions and structures, activity limitations and participation restriction, and environmental factors. Although the items are scaled, no overall scoring rules are provided. However, the intent is to measure improvement in terms of functioning regardless of actual diagnosis, including improvement in functioning as a result of the elimination of environmental factors. National Academy of Science Conceptualization In addition to a focus on function, the dynamism of the interplay of health and development over time has been incorporated into a multidimensional conceptualization by the Committee of Evaluation of Children’s Health of the National Academies of Sciences and the Institute of Medicine.14 This conceptualization begins with the following definition: Children’s health is the extent to which individual children or groups of children are able or enabled to (a) develop and realize their potential, (b) satisfy their needs, and (c) develop the capacities that allow them to interact successfully with their biological, physical, and social environments.14(p33) The report goes on to identify 3 specific domains of health. The first consists of health conditions, generally in the form of diagnoses, injuries, and common symptoms, acute or chronic, ‘‘that may be anatomical, physiological or psychosocial.’’14(p35) Functioning, the second domain, is considered to reflect ‘‘the direct and indirect effects of one or more health conditions and their treatments.[on] physical, psychosocial, cognitive and social functioning.’’14(p35) The final domain is health potential, which ‘‘includes both health assets that provide the capacity to respond to physical, psychological and social challenges and risk states that increase vulnerability to other aspects of poor health.’’14(p37) These domains are projected over multiple time frames and transitions to emphasize the potentially very long-term effects of child health on adult health. The first domain of this definition has clear parallels in the International Classification of Functioning, Disability, and Health (ICF),13 which postulates a progression from anatomic derangement through to symptoms. Both emphasize function as something separate but related. The National Academy of Science (NAS)14 framework is
AMBULATORY PEDIATRICS
much more dynamic and adds the concept of ‘‘potential,’’ an important concept in assessing children because the final ‘‘product’’ of health and development has yet to be realized. With regard to the issue of measurement, the NAS document critiques currently available measures as lacking consistency of definition and ascertainment across agencies, failing to consider clustering of conditions, underreporting some conditions, especially mental health problems, and lacking data on symptoms as opposed to diagnoses. The conclusion is that a consensus is needed on the critical components of child health and means of measuring them that captures the dynamic and transactional nature of child health across the life span.14(p101–107) These conceptual models, taken together, combine many of the desirable elements that should characterize measures of child health. These elements include multidimensionality, a life-course approach emphasizing the interaction among different dimensions of health and development, the minimization and amelioration of disability that results from health conditions, and the desirability of having a single metric ranging from perfect health to death. However, significant issues arise when it comes to operationalizing these constructs into measures useful at the individual and community level. Stein15 has argued that the assumptions of single construct of ‘‘health’’ and measurement of deviations from perfect health may not be possible. The best that can be achieved is clear specification of the underlying theoretical approach to measurement and an equally clear statement of the purpose for which the measure has been developed. Her concern is underscored by comparisons of the content of the more frequently used measures of child health status16,17 that reveal the lack of comparability among measures of health status, even for the same construct. In one such comparison17 of 7 generic child health status measures across 15 dimensions of health, agreement on the content was seen in only one area: physical functioning or ability to undertake the usual activities of daily living appropriate to a child. Some instruments include cognitive and behavioral health, as would the functional domain of the NAS conceptual framework. Some include symptoms; others do not. The lack of clear conceptualization is further indicated by the overlap of items in health and functional status measures with those of health-related quality of life.18 Finally, disease-specific measures also include generic measures of some of these domains, but may also differ in content.19 Stein15 further argues that even the best measures may be proxies for the construct of health, but despite the conceptual and methodologic challenges, the further development, even of these proxies, should be a high priority. IMPORTANCE TO GENERAL PEDIATRICS Why should such development of these proxies be a priority, and why should generalist academicians undertake this challenging endeavor? To illustrate the importance of developing a consensus on the relevant domains of child health and identifying appropriate measures, I consider the activities of pediatric practice in 3 main areas: well-
Issues in Measuring Child Health
79
child care, acute conditions, and chronic or longer-lasting conditions. Well-Child Care Well-child care has been defined to include health supervision, developmental surveillance, psychosocial assessment, coordination of care, immunizations, and additional screening.20 The general purpose of these activities is to assure adequate child physical and psychosocial development, prevention of disease, and early detection of problems. Our current measurement strategies for many of these activities rely on detecting deviations (generally 2 standard deviations) from some population mean. What might be different if the goal of well-child care were to achieve ‘‘well-being’’ or some measure of ‘‘optimal health’’? Let us take the example of monitoring physical growth— certainly a well-established routine in well-child care. One might argue that the current primary aim of measuring physical growth is to detect inadequate or excessive growth, and the major outcome of this activity is assessing children’s growth appropriate to age. Consideration of the literature on normal child cognitive and social development21 might suggest the need to take a more integrative approach. Studies suggest that growth that is at the extremes of normality or that is frankly abnormal may have effects on psychosocial outcomes and influence future disease states. The literature appears more consistent for obesity, which is associated with lower self-esteem and reduced economic mobility over and above the risks conferred for adult-onset chronic disease.22,23 The literature on the effects of short stature is more controversial, with some studies finding evidence of psychosocial disability in childhood24 or poorer educational attainment and employment among adults,25 but not all studies report these deficits.26,27 Likewise, nutritional problems that may lead to poor growth at either extreme may be associated with micronutrient deficiencies that may also contribute to cognitive and behavioral difficulties,28 and even in developed countries, the experience of hunger may lead to behavioral and emotional problems.29 Thus, it would appear that monitoring physical growth parameters might not be sufficient to capture all the issues surrounding adequate physical growth, and that other outcomes may be important. Reframing well-child care activities in terms of a general model of optimizing health might suggest needed outcomes or a different way of considering outcomes in other areas of care. For example, assessing the effect of recommendations for anticipatory guidance on physical activity30 might require measures of physical fitness, stamina, and the coordination and skills for participating in physical activities. Moreover, the arguments30 about the potential mental health benefits of participation in physical fitness activities underscore the importance of an integrative approach to assessing the effect of well-child care. Reconsideration of immunizations might argue for their role in a broader strategy for assuring optimal immune function, with breast-feeding as part of that strategy by potentially reducing the risk of allergy.
80
McCormick
In summary, the goals of well-child care would be seen as approaching this state of well-being as in the WHO definition and assuring potential, as in the NAS model. However, the measures of child health would have to be broadened to include not only the relevant psychosocial dimensions, but also important physiologic ones. Acute Illness A certain amount of experience with illness or injury is almost a normative experience; indeed, it may be necessary for the appropriate development of some physiologic systems31 and recognition of risky situations.32 Clearly, most parents also see this experience as unexceptional because, despite relatively frequent illness (which may occur monthly33), most parents consider their children to be ‘‘healthy.’’34 In this case, however, the developmental implications of the experience of brief illnesses is not clear in that the literature on child development provides little insight except to view illness as a familial stressor.35,36 As with well-child care, more integrative approaches might suggest potential outcomes for acute illness care besides symptomatic relief and prevention of more serious sequelae. For some families, the experience of frequent illness in the child may lead to parental distress with potential for developmental problems and increased use of medical care. Such families may be those with poor, depressed mothers37 or those who perceive that their child as more fragile after a diagnostic experience (the ‘‘labeling’’38–40 phenomenon) or after an illness perceived as severe (‘‘vulnerable child syndrome’’41,42). In addition, acute illness may have the potential to affect other aspects of development in unanticipated ways unless these effects are sought. For example, concerns about the effects of recurrent otitis media on the emergence of speech and language arose much earlier than concerns about effects on attention and behavior, but a consideration of the potential effect of dampened hearing on attention might have suggested the latter as well, although both concerns remain highly debatable and may reflect other characteristics of the environment.43–45 Alternatively, the experience of minor illness, which is likely to resolve successfully, may have a positive impact. Parmelee46 has suggested that the experience of acute health problems may foster the development of child resilience and management of discomfort. Theoretical models of the factors that may influence risk and resilience have been developed47 and could contribute to both the development of appropriate parental guidance and outcome measures, although most of this work has focused on chronic childhood conditions, not acute illness. However, if we are to operationalize Parmalee’s observation, then we need more nuanced measures of child resilience and ability to handle acute discomfort, of parental knowledge and confidence in managing acute illness, and of the delineation of those situations requiring more intensive intervention to assure that these outcomes will occur. Chronic Illness/Disability Recognition of the limits of the information from studies of single diseases and consideration of some of the overlap
AMBULATORY PEDIATRICS
among conditions has prompted a shift toward a noncategorical approach to chronic illness in childhood, such as children with special health care needs. The argument48–50 supporting such a shift contends that children with chronic illness and their families share common experiences, problems with accessing health care, needs for coordination of care, and other events that transcend the diagnosis. The noncategorical approach to the consideration of chronic illness in childhood has come to dominate the general pediatric literature, and indeed, population-based studies reveal distinctive patterns of sociodemographic characteristics, health care use, and insurance coverage by using a variety of definitions51–53 for the 15% to 20% of children with a physical, mental, or behavioral health problem.5,49,54–57 Similarities at the population level, however, mask substantial definitional and measurement issues for individual children. First, the stability of the prevalence across service sectors and over time may provide an illusory sense that these figures refer to the same children. Different classification algorithms identify different children as having special needs.58 There may also be imperfect correlation between those identified as having special health care needs and clinical diagnoses, with 20% of those identified by one algorithm lacking an associated diagnosis or condition.59 In addition, this classification, while identifying those with one or more problems, does not deal well with the clustering and overlap of various categories of disability.60 Different child service systems may identify different children as having special needs or characterize the same child in different ways. For example, the diagnosis used for classification for special education may not correspond completely with medical diagnoses. Thus, one important activity would be the development of a uniform characterization of children across different service sectors as to which children would fall into the category of children with special health care needs, perhaps by using a functional approach suggested in either the ICF or NAS models. That is not to say that measures of childhood functioning in the context of chronic illness do not exist, but this literature has been critiqued as having several flaws. A review has summarized these flaws as including a focus on the negative aspects of childhood chronic illness (using a deficit model); the lack of information on how the family responds and adapts to illness; the lack of longitudinal data; a focus primarily on maternal response to childhood illness, with less information about fathers and the children themselves; and the reliance on studies of small groups of children with a single disease.48 Thus, the current literature may not be sufficient to examine the assumption that the experiences and needs of these children and their families are sufficiently similar to justify such a broad grouping. This assumption may be useful in planning for and financing health and other services needs but may not be appropriate when considering the developmental impact of such ongoing conditions on both the child and the family. Moreover, current measures may not adequately capture family dynamics and positive adaptation to ongoing childhood illness.
AMBULATORY PEDIATRICS
Consideration of the heterogeneity of the conditions of children with special health care needs suggests that a more nuanced approach to conceptualization and measurement of outcomes is needed. Conditions vary considerably in the severity, course, and implications for the child and his or her family, and the developmental tasks that the family and child face. That different conditions are viewed differently by families, thereby potentially altering the family response to illness, is suggested by studies that indicate that illnesses characterized by invisibility, remitting-relenting courses, and uncertain prognosis cause more distress or have children viewed as less healthy than other conditions.60,61 Likewise, families may adjust better to conditions present at birth, even when producing more disability, than conditions that emerge in a previously healthy child, as illustrated by the fact that parents of children with spina bifida rate their children as healthier than those with children with juvenile rheumatoid arthritis, despite the fact that the latter have far fewer functional limitations.61 One might hypothesize that parents of children with conditions present at birth may recalibrate what it means for their child to be healthy, or have a different developmental task than parents with a previously healthy child who develops a chronic illness. One way this may occur is that parents may discriminate between the sequelae of a birth defect and other aspects of health. One mother of a child with spina bifida in this study61 reported that her child had a disability but was otherwise healthy. Moreover, in attempting to gauge the factors affecting parental reaction and adaptation to ongoing health problems in their children, we cannot assume that their understanding of the pattern of illness is equivalent to pathogenesis, even to the concept of ‘‘present at birth.’’ For example, in a study that involved a series of questions to ascertain child health problems from birth to the end of the first year of life,62 one parent reported that the child was not born with any problems, but was hospitalized 3 times in the first year (for congestive heart failure, cardiac catheterization, and corrective heart surgery) and was now followed in a program for children with Down syndrome. The importance of capturing other dimensions of ill health has been incorporated into the conceptualizations of chronic health conditions emerging from different theoretical perspectives,47,63,64 and 2 of these formulations63,64 incorporate consideration of relatively severe acute conditions. Among the dimensions considered important in these frameworks and suggested by others47,65 are the age at onset, duration, mobility-activity restrictions, visibility of signs and symptoms of the condition, course (eg, progressive, stable, unpredictable), prognosis, and effect on cognitive and emotional well-being. In a review of selected studies examining the effect of these dimensions on family and child adaptation, Thompson and Gustafson47 found that the impact of different aspects of ill health were sometimes, but not always, as the theory would predict. However, many of the studies reviewed involved relatively small samples and were restricted to single conditions, thus failing to capture the information that might be gleaned by cross-condition comparisons. In addition,
Issues in Measuring Child Health
81
more clinical and biologic observations are needed to inform the measurement. For example, child psychosocial adaptation to chronic illness appears worse in conditions affecting the brain.47 However, because much of the measurement of psychosocial adaptation concerns behavioral observations, it is not clear whether it is the child’s adaptation or that the observations reflect symptoms related to the primary neurological problem. Similar concerns have been raised about the extensive literature on the outcomes of very premature infants, which fail to consider the multiple health issues affecting these children that might affect their cognitive and social development other than direct neurological events.66 Taking a more integrated and developmentally focused approach toward chronic childhood conditions may yield insights into risk and protective factors for better or worse functional outcomes, as well as factors affecting treatment, as illustrated by a series of papers on attention-deficit/ hyperactivity disorder.67 In particular, this series provides an example of the application of the ICF to describe a condition68 and to guide the development of appropriate interventions.69 As indicated by these reports, a focus on function by using approaches like the ICF or the NAS model, with consideration of family adaptation, could provide needed information, both for the individual child and for larger groups of children. For the individual child, measures that could reflect improvement in function would offer a way of understanding the effects of individual referrals or therapeutic programs. If the measures had sufficient detail and could be compiled across children with different conditions, a better sense of the need for pediatric subspecialists and child-trained therapists would be possible. Further, there is an emerging theoretical framework70 to guide the examination of the effects of ill health on child development, although these models may need some reformulation to acknowledge that ill health may not just be another stressor but a more integral aspect of human development. However, understanding the full impact of ongoing conditions would require the dynamic and interactive approach suggested by the NAS model. CONCLUSION I began my career as a health services researcher interested in program evaluation, especially programs focused on improving pregnancy outcomes and early child health. That interest, in turn, led to research on the outcomes of very premature infants. Coming from the perspective of health services research, I was familiar with the approach that underlay instrumentation such as the Child Health Questionnaire17 and excited by what seemed to be a logical approach to thinking about measuring child health on the basis of a conceptual model (the WHO definition) and moving beyond diagnoses. However, in applying this approach to studies of the outcomes of very premature infants, especially those observed at school age, I realized that we were far from being able to characterize child health in a way that might capture the complexity of the situation of any one child, and from having an established
82
McCormick
armamentarium of measure to evaluate the effects of specific interventions and programs. We not only do not have a clear concept of health, but we also lack a conceptual framework linking biological and environmental factors in a way that would help us understand the effect of interventions, including the effects of many elements of general pediatric care. I am not alone in concluding that we lack of a clear definition or consensus on what is meant by child health.14(p101–107),17,70 Further, although most reviews agree that appropriate measures should be multidimensional,19,70 there is little concordance on what those dimensions should be from the perspective of the child, the family, or society at large. Moreover, because many of the child health measures reflect adaptation of adult instruments, measures for young children and infants tend to be sparse. The situation might not be so limited if clear theoretical perspectives linking health and development permitted capitalizing on the concepts and instruments available from studies of normal child development. As Eiser and Morse19 indicate, for consideration of ‘‘wellbeing,’’ for example, one could have a basic, generic measure of child health to which could be added measures of such constructs as self-esteem, self-competence, or emotional development, if relevant to the context. However, there is little theoretical guidance regarding when such constructs would be important child health outcomes. It would seem that achieving some consensus and either developing or assessing measures to characterize child health broadly would be a role for generalist academic pediatricians, both as researchers in child health broadly and as educators of generalist providers. Recommendations for the research training of academic generalists71 already include many of the research design and analytic methods needed for this endeavor, although greater depth in relevant theoretical areas might well be needed, particularly child development and measurement theory. Academic generalists also are in a unique position to bridge the worlds of social science and biology. Existing measures of child status have been criticized as too subjective for clinical and programmatic usefulness. Their usefulness would be greatly enhanced if these measures could be validated against other measures with physiologic or clinical significance. If, for example, a parent or child reports a limitation in activity due to health, like difficulty in running or walking certain distances, their response does not discriminate among difficulties in joint or muscle control or range of motion, decreased stamina, or developmental problems that result in coordination difficulties. A clearer tie to physical observations would assist clinicians in developing intervention strategies. Moreover, establishing such linkages would help in calibrating parental reports and contribute to the theoretical development of when parents report more or less difficulty than would be anticipated by the measurement of, for example, stamina. In this regard, a program of refining and validating measures of child health status might well be incorporated into the role of the new Clinical and Translational Science Awards72 units.
AMBULATORY PEDIATRICS
Because many academic generalists also bridge the worlds of pediatrics and public health, they could also work to develop clear linkages between what is measured in greater detail at the individual level and indicators of child health. Such linkages would facilitate program planning and interpretation in changes in indicators and variations in different groups. This argument is not meant to gloss over the considerable methodologic challenges involved in developing strong measures of child health, like the reliance on proxy reporters (caretakers) and cognitive development issues in eliciting child responses.18 However, the first step in this process is to begin to articulate a conceptual framework of the relevant domains of child health and aligning this framework with existing measures and their usefulness, and to identify measurement gaps. In addition, such an activity should also identify potential factors influencing health—for example, demographic ones, like maternal educational attainment, or health services ones, like type and quality of care—to generate hypotheses that would lead to the development of a more comprehensive construct of child health and its context. Even though it has its problems and raises concerns, psychologists have accomplished a lot with the IQ test. Maybe it is time to develop our own HQ (health quotient) test to improve children’s health, or at least to reduce the frustration of future health services researchers looking for useful measures of outcome. ACKNOWLEDGMENTS The author acknowledges the support of the Institute of Medicine as this work was done while a Scholar in Residence there. In addition, she thanks Elena Nightingale and Edward Schor for comments and suggestions on a draft. Supported in part by the Commonwealth Fund (grant 20040493).
REFERENCES 1. Bayley N. Bayley Scales of Infant Development. 3rd ed. San Antonio, Tex: Psychological Corporation; 2005. 2. Kaplan E, Fein D, Kramer J, Delis D, Morris R. Weschler Intelligence Scale for Children—Fourth Edition Integrated. San Antonio, Tex: Harcourt Assessment Inc; 2007. 3. Glascoe FP, Martin ED, Humphrey S. A comparative review of developmental screening tests. Pediatrics. 1990;86:547–554. 4. Glascoe FP, Sandler H. Value of parents’ estimates of children’s developmental ages. J Pediatr. 1995;127:831–835. 5. Achenbach TM, Edelbrock C. Manual for Child Behavior Checklist and Revised Child Behavior Profile. Burlington, Vt: Queen City Printers; 1983. 6. Child Trends DataBank. Available at: http://www.childtrendsdatabank. org. Accessed December 7, 2007. 7. http://www.fcd-us.org/issues/issues_show.htm?doc_id=447092. Accessed December 14, 2007. 8. National Institute of Child Health and Human Development, NIH, DHHS. America’s Children: Key National Indicators of Well-Being. Washington, DC: US Government Printing Office; 2007. 9. Peoples-Sheps MD, Guild PA, Farci AM, et al. Model indicators for maternal and child health: an overview of process, product and applications. Matern Child Health J. 1998;2:241–256. 10. Ried LD, Tueth MJ, Handberg E, Nyanteh H. Validating a self-report measure of global subjective well-being to predict adverse clinical outcomes. Qual Life Res. 2006;15:675–686. 11. Preamble to the Constitution of the World Health Organization as adopted by the International Health Conference, New York, NY,
AMBULATORY PEDIATRICS
12. 13.
14.
15. 16.
17.
18.
19. 20.
21.
22. 23. 24.
25.
26.
27.
28.
29.
30.
31.
32. 33.
34. 35. 36.
June 1946; signed July 22, 1946, by representatives of 61 states (Official Records of the World Health Organization, no. 2, p.100) and entered into force on 7 April 1946. QualityMetric: Health Outcomes Solutions. Available at: http://www. qualitymetric.com/products/. Accessed December 7, 2007. World Health Organization. International Classification of Functioning, Disability and Health (ICF). Available at: http://www.who.int/ classifications/icf/en/. Accessed December 7, 2007. Committee on Evaluation of Children’s Health. Children’s Health, the Nation’s Wealth: Assessing and Improving Child Health. Washington, DC: National Academy Press; 2004. Stein RE. Measurement of children’s health. Ambul Pediatr. 2004; 4(4 suppl):365–370. Olson LM, Lara M, Pat Frintner M. Measuring health status and quality of life for US children: relationship to race, ethnicity, and income status. Ambul Pediatr. 2004;4(4 suppl):377–386. Landgraf JM, Abetz L, Ware JE. Child Health Questionnaire (CHQ): A User’s Manual. Boston, Mass: Health Institute, New England Medical Center; 1996. 30–32. Drotar D. Validating measures of pediatric health status, functional status, and health-related quality of life: key methodological challenges and strategies. Ambul Pediatr. 2004;4(4 suppl):358–364. Eiser C, Morse R. Quality-of-life measures in chronic disease of childhood. Health Technol Assess. 2001;5:1–157. Kuo AA, Inkelas M, Lotstein DS, Samson KM, Schor EL, Halfon N. Rethinking well-child care in the United States: an international comparison. Pediatrics. 2006;118:1692–1702. Shonkoff JP, Philips DA, eds. From Neurons to Neighborhoods: The Science of Early Childhood Development. Washington, DC: National Academies Press; 2000. Must A, Strauss RS. Risks and consequences of childhood and adolescent obesity. Int J Obes Relat Metab Disord. 1999;23(suppl 2):S2–S11. French SA, Story M, Perry CL. Self-esteem and obesity in children and adolescents: A literature review. Obesity Res. 1995;3:479–490. Molinari E, Sartori A, Ceccarelli A, Marchi S. Psychological and emotional development, intellectural capabilities and body image in short normal children. J Endocrinol Invest. 2002;25:321–328. Zimet GD, Owens R, Dahms W, Cutler M, Litvene M, Cutler L. Psychosocial outcome of children evaluated for short stature. Arch Pediatr Adolesc Med. 1997;151:1017–1023. Downie AB, Mulligan J, Stratford RJ, Betts PR, Voss LD. Are short normal children at a disadvantage? The Wessex Growth Study. BMJ. 1997;314:97–100. Pilpel D, Lieberman E, Zadik Z, Carel CA. Effect of growth hormone treatment on quality of life of short stature children. Horm Res. 1995; 44:1–5. Lozoff B, De Andraca I, Castillo M, Smith JB, Walter T, Pino P. Behavioral and developmental effects of preventing iron-deficienccy anemia in healthy full-term infants. Pediatrics. 2003;112:846–854. Weinreb L, Wehler C, Perloff J, Scott R, Hosmer D, Sagor L, Gundersen C. Hunger: its impact on children’s health and mental health. Pediatrics. 2002;110:e41. Bright Futures at Georgetown University. Bright Futures in practice: physical activity. Available at: http://www.brightfutures.org/ physicalactivity/. Accessed December 7, 2007. Wills-Karp M, Santeliz J, Karp CL. The germless theory of allergic disease: revisiting the hygiene hypothesis. Nat Rev Immunol. 2001; 1:69–75. Robertson LS. Injury Epidemiology. Research and Control Strategies. 2nd ed. New York, NY: Oxford University Press; 1998. 117–123. Simpson L, Owens PL, Zodet MW, et al. Health care for children and youth in the United States: annual report on patterns of access, utilization, quality and expenditure by income. Ambul Pediatr. 2005;5:6–44. Hojer B, Sterky G, Widlund G. Acute illnesses in young children and family responses. Acta Paediatr Scand. 1987;76:624–630. Heymann SJ, Earle A, Egleston B. Parental availability for the care of sick children. Pediatrics. 1996;98:226–230. Heymann SJ, Toomey S, Furstenberg F. Working parents: what factors are involved in their ability to take time off from work when their children are sick? Arch Pediatr Adolesc Med. 1999;153:870–874.
Issues in Measuring Child Health
83
37. McCarthy P, Walls T, Cicchetti D, et al. Prediction of resource use during acute pediatric illnesses. Arch Pediatr Adolesc Med. 2003; 157:990–996. 38. Perrin EC, West PD, Culley BS. Is my child normal yet? Correlates of vulnerability. Pediatrics. 1989;83:355–363. 39. Bergman AB, Stamm SJ. The morbidity of cardiac non-disease in school children. N Engl J Med. 1967;276:1008–1013. 40. Fisher ES, Welch HG. Avoiding the unintended consequences of growth in medical care—how might more be worse? JAMA. 1999; 281:446–453. 41. Thomasgard M, Metz WP. The vulnerable child syndrome revisited. J Dev Behav Pediatr. 1995;16:47–53. 42. McCormick MC, Brooks-Gunn J, Workman-Daniels K, Peckham GJ. Maternal rating of child health at school age: does the vulnerable child syndrome persist? Pediatrics. 1995;16:47–53. 43. Roberts J, Hunter L, Gravel J, et al. Otitis media, hearing loss and language learning: controversies and current research. J Dev Behav Pediatr. 2004;25:110–122. 44. Bennett KE, Haggard MP. Behaviour and cognitive outcomes from middle ear disease. Arch Dis Child. 1999;80:28–35. 45. Hooper SR, Ashley TA, Roberts JE, Zeisel SA, Poe MD. The relationship of otitis media in early childhood to attention dimensions during the early elementary school years. J Dev Behav Pediatr. 2006;27: 281–289. 46. Parmelee AH. Illness and the development of social competence. J Dev Behav Pediatr. 1997;18:120–124. 47. Thompson RJ, Gustafson KE. Adaptation to Chronic Childhood Illness. Washington, DC: American Psychological Association; 1999. Chap 5. 48. Hayes VE. Families and children’s chronic conditions: knowledge development and methodological considerations. Sch Inq Nurs Pract. 1997;11:259–290. 49. Hobbs N, Perrin JM, Ireys HT. Chronically Ill Children and Their Families. San Francisco, Calif: Jossey-Bass Publishers; 1985. 50. Stein RE, Bauman LJ, Westbrook LE, Coupey SM, Ireys HT. Framework for identifying children who have chronic conditions: the case for a new definition. J Pediatr. 1993;122:342–347. 51. Newacheck PW, Taylor WR. Childhood chronic illness: prevalence, severity and impact. Am J Public Health. 1992;82:364–374. 52. Newacheck PW, Halfon N. Prevalence and impact of disabling chronic conditions on childhood. Am J Public Health. 1998;88: 610–617. 53. Newacheck PW, Strickland B, Shonkoff JP, et al. An epidemiologic profile of children with special health care needs. Pediatrics. 1998; 102:117–123. 54. Cadman D, Boyle M, Szatmari P, Offord DR. Chronic illness, disability and mental and social well-being. Pediatrics. 1987;79:805–813. 55. Roberts RE, Attkisson CC, Rosenblatt A. Prevalence of psychopathology among children and adolescents. Am J Psychiatry. 1998; 155:715–725. 56. Shaffer D, Fisher P, Dulcan MK, et al. The NIMH Diagnostic Interview Schedule for Children version 2.3 (DISC-2.3): description, acceptability, prevalence rates, and performance in the MECA Study. J Am Acad Adolesc Psychiatry. 1996;35:865–877. 57. Zill N, Schoenborn CA. Developmental, Learning and Emotional Problems. Health of Our Nation’s Children, 1988. Advance Data 190. Hyattsville, Md: National Center for Health Statistics; 1990. 58. Jessop DJ, Stein RE. Consistent but not the same. Effect of method on chronic condition rates. Arch Pediatr Adolesc Med. 1995;149: 1105–1110. 59. Liu C. Access, Quality, and Expenditures of Health Care for Children with Special Health Care Needs in the United States. Boston, Mass: Harvard School of Public Health; 2004:11. 60. Perrin JM, Kuhlthau K, McLaughlin TH, Ettner SL, Gortmaker SL. Chaning patterns of conditions among children receiving Supplemental Security Income disability benefits. Arch Pediatr Adolesc Med. 1999;153:80–84. 61. McCormick MC, Athreya B, Bernbaum JC, Charney EB. Preliminary observations on maternal rating of health of children: data from three subspecialty clinics. J Clin Epidemiol. 1988;41:323–329.
84
McCormick
62. McCormick MC, Wessel KM, Krischer JP, Welcher DW, Hardy JG. Preliminary analysis of developmental observations in a survey of morbidity in infants. Early Hum Dev. 1981;5:377–393. 63. Patterson J, Blum RW. Risk and resilience among children and youth with disabilities. Arch Pediatr Adolesc Med. 1996;150:692–698. 64. Perrin EC, Newacheck P, Pless IB. Issues involved in the definition and classification of chronic health conditions. Pediatrics. 1993;91: 787–793. 65. Roland JS. Families Illness and Disability. An Integrative Treatment Model. New York, NY: Basic Books; 1994:chap 2–3. 66. McCormick MC. The outcomes of very low birth weight infants: are we asking the right questions? Pediatrics. 1997;99:869–876. 67. Brown RT, Perrin JM. Measuring outcomes in attention-deficit/hyperactivity disorder. Ambul Pediatr. 2007;7(1 suppl):57–59.
AMBULATORY PEDIATRICS 68. Ustun TB. Using the International Classification of Functioning, Disease and Health in attention-deficit/hyperactivity disorder: separating the disease from its epiphenomena. Ambul Pediatr. 2007;7(1 suppl): 132–139. 69. Loe IM, Feldman HM. Academic and educational outcomes of children with ADHD. Ambul Pediatr. 2007;7(1 suppl):82–90. 70. Dreyer BP, Schonfeld DJ; for the Ambulatory Pediatric Association Research Committee. Content outline for a curriculum in research education for general academic fellowship programs. April 24, 2004. Available at: http://www.ambpeds.org/site/research/Draft Curriculum.htm. Accessed December 7, 2007. 71. http://ctsweb.org. Accessed December 19, 2007. 72. Szilagyi P, Schor EL. The health of children. Health Serv Res. 1998; 33(4 part 2):1001–1039.