Methodological Considerations

Before embarking on this review of the pediatric literature, it may be especially instructive to identify some of the limitations and weaknesses inherent in much of the research conducted to date. Two broad problem areas are especially prevalent across studies: the failure to accurately and comprehensively ascertain the nature, severity, and course of the disease and its impact on both the child and the family, and the absence of large-scale studies from multiple research centers that consistently use a core battery of child-appropriate, valid, and reliable neurocognitive assessment measures.

When studying children with any type of chronic disorder, it is critically important to be able to identify and document, for each child, the nature and extent of their disease process from both a biomedical and a psychosocial perspective. For the diabetic child, we ought to - but hardly ever - have medical, metabolic, and psychosocial data from diagnosis onward. Did the child experience ketoacidosis and/or cerebral edema around the time of diagnosis or anytime thereafter? Since diagnosis, how often and for what duration did the child experience excessively low - and excessively high - glucose values, and how were these episodes of hypoglycemia and hyperglycemia operationalized? When did the child begin to show evidence of microvascular complications and other comorbid conditions like blood pressure elevations, and how did these progress over time? How did the child cope psychologically with the diagnosis of diabetes and with diabetes-related events, like the occurrence of a hypoglycemic seizure or coma? Each of these biomedical and psychosocial variables may influence the phenomenology, measurement, and/or etiology of neurocognitive outcomes, but they are rarely captured by researchers or incorporated into analyses.

It is also critically important for researchers to be able to delineate, in a meaningful and reliable fashion, the neurocognitive characteristics of each child and any acute state - depression, anxiety, low blood glucose values -that might influence cognitive performance at the time of that assessment. Ordinarily, this is accomplished by administering a battery of psychome-trically sound neuropsychological tests, assessing mood state, and measuring blood glucose periodically during the assessment session. Unfortunately, there are no universally agreed upon standards for selecting such tests, particularly when assessing children (or adults!) with diabetes, despite pleas for the establishment of a "core" battery (5), as has happened in research with other neurocognitive disorders like dementia (6, 7). Lack of consistency in neuropsychological assessment across research groups challenges our ability to rationally aggregate results from many smaller cross-sectional studies.

This problem has been compounded further by researchers who have either failed to assess neuropsychological function in a comprehensive fashion or have used tests with questionable, or unknown, psychometric properties for a pediatric population. As an example, a number of seminal studies either ignored the assessment of learning and memory skills completely (8), used tests that had been developed for adults but not children (9), or relied on mnestic tests from the research laboratory with questionable or unproven psychometric properties (10, 11). This would, at least in part, explain why there exists no consensus about the impact of diabetes on a cognitive process which is a key determinant to classroom success: effective learning and memory abilities (11-14).

Longitudinal studies in which subjects are evaluated repeatedly over an extended period of time present an additional set of problems (15,16). Not only may practice or familiarity effects from repeated exposure to the same materials reduce the sensitivity of the test to subtle declines in cognition (17), but the use of different cognitive tests (which can have very different psychometric properties) at different ages to account for qualitative and quantitative variations in the normal course of cognitive development may further complicate the researcher's ability to detect meaningful changes over time (18).

Those methodological issues notwithstanding, extant research on diabetic children's brain function has identified a number of themes that emerge in this review. All other things being equal, children diagnosed with type 1 diabetes early in life - within the first 5-7 years of age - have the greatest risk of manifesting neurocognitive dysfunction, the magnitude of which is greater than that seen in children with a later onset of diabetes. The development of brain dysfunction seems to occur within a relatively brief period of time, often appearing within the first 2-3 years following diagnosis. It is not limited to performance on neuropsychological tests, but is manifested on a wide range of electrophysiological measures as marked neural slowing. somewhat surprisingly, the magnitude of these effects does not seem to worsen appreciably with increasing duration of diabetes - at least through early adulthood.

Delicious Diabetic Recipes

Delicious Diabetic Recipes

This brilliant guide will teach you how to cook all those delicious recipes for people who have diabetes.

Get My Free Ebook

Post a comment