THE UNIQUE NATURE OF DEVELOPMENTAL-BEHAVIORAL PEDIATRIC RESEARCH

The scope of research in the field of developmental-behavioral pediatrics (DBP) is as diverse and rich as the clinical field itself. A wide range of research methods and analytical techniques accounts for both its depth and its complexity. The same characteristics of research in the field that render the potential for its findings to be of such practical significance and relevance often pose critical challenges to ensuring its scientific validity.

The research and the associated research teams are often multidisciplinary, permitting an application of various methodological approaches. The field of DBP permits integration of complementary theoretical perspectives and methods, such as the blending or juxtaposition of quantitative methods characteristic of medical science with qualitative approaches more typical of social science research. Research training in the field is therefore more eclectic and broader than in subspecialties that rely almost exclusively on basic science techniques. The field does not have one well-circumscribed set of research methods that can be mastered in a relatively short time. For quality research in DBP, multidisciplinary teams must consist of individuals who can each contribute their own perspective and skills, and each team member must be adequately informed of the basic principles inherent in the research approaches of the other disciplines.

DBP research often aims to study the full spectrum of child development and behavior: from normal variations to concerns or problems to clinical disorders. One of the driving forces for establishing the American Psychiatric Association’s Diagnostic and Statistical Manual of Mental Disorders was to standardize the diagnostic criteria for disorders to foster consistency in research in mental illness.¹ Research that incorporates the continuum of developmental and behavioral difficulties must establish reliable and valid outcome measures for subthreshold or problem conditions or criteria for identifying where on the bell-shape curve of behavior or development is the appropriate cutoff for defining a concern or a problem. Although achieving reliability in delineating the diagnostic criteria for a mental illness may be challenging, it is often even more elusive for a behavioral problem or personality trait. One common approach is to inquire whether the characteristic of interest (e.g., attention) is believed to occur significantly more often in one person than in typical peers of the same age or developmental level and to require an association with some perceived impairment (e.g., attention-deficit/hyperactivity disorder [ADHD]). This approach often introduces a reliance on subjective, self-reported measures of perceived impairment or relative deviation from perceived norms that can compromise validity and produce a reporting bias.

Research in DBP often addresses more abstract issues, such as community support or adjustment to illness. Because much of the research addresses such common topics, the researcher may assume that the methodology is therefore “simple.” But, in fact, operationalizing these variables and developing and validating relevant measures are difficult. Much of the research in DBP involves measuring constructs for which validated measures do not already exist and for which objective, concrete biological outcome measures are not feasible.

Because DBP often assumes an ecological perspective, researchers are more apt to look critically at sociocultural influences on child development and behavior. Such factors are difficult to measure, even harder to report accurately, and far more difficult to interpret or explain. The use of race and ethnicity as explanatory variables illustrate the complexity of this issue.² Researchers who understand the complexity of social and cultural influences appreciate the futility of controlling for all relevant influences within an ecological model.

Despite these challenges, the complexity of research design issues in DBP fosters its richness. The multiple perspectives and theories and the diversity of available methodological approaches enable the construction of rich, multidimensional theoretical models. Researchers must necessarily explore not only outcome measures but also mediators and moderators (see Chapter 2). The complexity is increased by the factor of time and the challenges inherent in measuring one construct in the context of a child’s developmental trajectory. For example, in studies of the influences of early childhood experiences on later language outcomes, investigators need to consider not only the multiple environmental, familial, cultural, and community factors that may influence language development but also the reality that developmental processes are not static in the individual child. Parsing out how much of change in language development is attributable to the normative process of child development ot to inherent deficits in the child, social, environmental, family or community factors, or the unanticipated effect of uncontrolled historical events (such as changes in preschool policy or educational interventions) can be daunting.

CROSS-CUTTING METHODOLOGICAL AND THEORETICAL ISSUES

The nature of DBP research introduces a range of cross-cutting methodological and theoretical concerns that must be addressed to ensure the validity of the findings. This section highlights select examples that illustrate the complexity of the issues that are involved.