Factors influencing family participation in a longitudinal study: comparison of pediatric and healthy samples.

Compared participation levels in a longitudinal study of parent-child relationships from infancy to 4 years in families of children with cystic fibrosis, congenital heart disease, and with no chronic illness. Demographic (parent's age, education) and child, parent, and family variables (medical status, family environment) were investigated for their predictive value of families' participation. 34% of families (71/209) were lost to the study at a later date. Families of children with cystic fibrosis were the least likely to be lost. Parents' age and/or education predicted participation in all groups. Families in both pediatric samples participated less when parental well-being was less optimal, and the level of mother-infant attachment organization was lower. Unlike demographic factors, family factors have differential impact on participation in families in pediatric and nonpediatric samples.