The convenience sample as a source of data in the study of Down syndrome

Epidemiologic and health care research on Down syndrome are frequently based on cases known to a specific hospital, clinic or social service agency. To determine if there are any systematic differences between such readily available cases and their birth cohort that might impair their use for research, we compared a total population of children with Down syndrome live-born in Montreal in 1975 and 1976 (n = 88) with two convenience samples drawn from this population: cases known to a hospital genetic counselling centre and cases participating in an early intervention programme. Complete information on maternal age and parity, birth weight and sex of child, presence or absence of congenital heart disease and residential placement was obtained for eighty-three members of the cohort. Analysis revealed that the cases known to the genetic counselling centre were more likely to have younger mothers caring for them at home than the remaining cases in the Down syndrome population; children participating in an early intervention programme were also more likely to be residing in their natural homes than those cases in the remaining comparison population. The majority of potential cases from the original population were not included in either case group. The findings suggest that "convenience" samples differ significantly from their parent population and that caution is required when applying data generated from selected groups to the remaining population.