NHGRI logo

NHGRI Special Populations Research Program Collaboration with University of Antioquia

Genetic Studies of Attention Deficit Hyperactivity Disorder in a Colombian Population Isolate

Attention Deficit Hyperactivity Disorder (ADHD) is the most common behavioral disorder of childhood, affecting approximately 3 to 5 percent of children between the ages of 7 and 18. The estimated male/female ratio ranges from 4:1 to as low as 2.5:1. In the Diagnostic and Statistical Manual of Mental Disorders, the American Psychiatric Association defines three categories of ADHD: predominantly inattention type; predominantly impulsive type; and a combined type with symptoms of both. A number of instruments, administered to parents, children and teachers, are used to assist in making the diagnosis.

Family studies suggest that there is a genetic component to ADHD. When the rate of illness in relatives of ADHD probands is compared to the rates of ADHD in controls, it is clear that first- and second-degree relatives of ADHD probands are at increased risk for the disorder. Twin studies, which yield concordance estimates of 80 percent for monozygotic twins vs. 32 percent in same sex dizygotic twins, indicate that this familial clustering is due to shared genetic risk factors rather than to shared environment. The genetic risk, however, does not appear to follow a simple Mendelian pattern. The recurrence risk in full siblings, for example, is approximately 17 to 25 percent. These data are consistent with the view that ADHD is a complex genetic disorder in which two or more genes are involved as well as predisposing environmental risk factors.

Linkage analysis is a method for identifying the location of a locus of interest by determining whether it is in close proximity to known markers. It is done by looking for cosegregation of ADHD alleles at an ADHD locus with a close by marker allele at a known locus. Sometimes, genes on the same chromosome can cosegregate when they are far apart from each other because crossovers occur. Recombinant types of gametes result from crossovers. There is an excess of parental type gametes when linkage does occur and an equal amount of each of the four possible gametes when there is no linkage. In reality, there are many confounding factors, so we need to look at the lod score, which is the probability of what is observed if there is linkage.

A fruitful approach to uncovering the genetic component of complex disorders is to use population isolates in which the disease of interest appears to be segregating. Antioquia, Colombia, South America is an isolated region located between the Central and Western branches of the Andes with a population of about 5 million. The inhabitants, who refer to themselves as "Paisas," are endogamic. Up until the 1950s, the families had large sibships with sizes ranging between five to 20 children (average sibship size: 8) and occasionally even more. Based on anthropologic/genetic studies, the population appears to be genetically homogeneous. Started from less than 30 founding families (Spanish, Basque and Jewish), it has been in a geographically and culturally isolated setting for about 20 generations (over 450 years).

A segregation analysis in 53 nuclear families using POINTER, rejected models of cohort effect (non-inheritance), multifactorial inheritance, recessive major gene, non-major gene component and non-transmission of major gene. By contrast, dominant and codominant major gene models plus a non-multifactorial component could not be rejected. The best fitting model gave an estimate of major susceptibility allele frequency of 0.03 in the general population of Antioquia, with lifetime penetrance of 30 percent.

Simulation studies of the 27 most informative appearing pedigrees suggest that we will have excellent power even in the presence of genetic heterogeneity. Simulation studies of the 27 most informative appearing pedigrees plus 13 less informative appearing pedigrees indicate that even with genetic heterogeneity there is power to detect linkage. In the 27 most informative appearing families, the power simulations suggest that we will have excellent power to detect linkage when only 50 percent of the families are linked to the same region. Power is moderate when only 25 percent of families are linked to the same region. However, the addition of more families increases the power such that expected power is good even when only 25 percent of families are linked.

Families from a large Colombian isolate in which ADHD is segregating are being used for a genome wide screen to search for candidate regions and for identification of disease genes. The goal of this study is to elucidate the biochemical pathways involved in ADHD pathogenesis. Data from this study will also be useful for the second phase of our study of ADHD in North American affected sib pairs. Thus, the objective of this study is to obtain phenotypic information and DNA from up to 100 Paisa families (30 to 35 per year) in whom ADHD is present in multiple relatives based on locally-validated interviews, rating scales, neuropsychological evaluations, and who meet exclusion/inclusion criteria.

Last updated: November 15, 2012