Generational differences in the age at diagnosis with Ibd: genetic anticipation, bias, or temporal effects

Previous cross-sectional research has demonstrated generational differences in age at diagnosis (AAD) in inflammatory bowel disease (IBD). This observation has at times been ascribed to genetic anticipation, but could also be due to biases related to case ascertainment or follow-up or to temporal changes in IBD epidemiology. We aimed to explore this issue using a population-based database.

In 1995 we used the comprehensive administrative databases in the province of Manitoba, Canada to establish a population-based IBD Research Registry that includes clinical and demographic information for persons. We contacted those subjects within our Research Registry who reported having any family members with IBD and their family members for verification of diagnosis and AAD. Differences in AAD between familial pairs were calculated. In addition, to assess whether duration of follow-up accounted for generational differences in AAD, we computed the mean AAD for subjects with and without family histories of IBD based on age at the time of interview (i.e., < 45 and > or = 45 yr of age).

Of the 2445 persons with IBD in the Research Registry, 548 reported positive family histories, and 315 of these (58%) were reached by telephone. There were 169 Crohn's disease and 146 ulcerative colitis subjects with positive family histories. The mean AADs for the parents, aunts/uncles, and grandparents were significantly greater than the mean AADs for the children, nieces/nephews, and grandchildren, respectively. There was a doubling of the mean AAD when comparing the grandparent/grandchild cases with the parent/child or aunt/uncle-niece/nephew cases. No statistically significant difference in anticipation was observed, whether or not the older generation was male or female or had Crohn's disease or ulcerative colitis. The AAD was substantially greater for those interviewed at > or = 45 yr of age for subjects with and without family histories. However, there was no substantial difference in mean AAD between familial and nonfamilial subjects.

The present study has demonstrated that there is a tendency for children to be younger than their parents at the time of diagnosis of familial IBD, and that this difference in mean AAD is almost doubled for grandparent/grandchild pairs. However, we conclude that these differences are most likely due to a bias based on length of follow-up or recent multigenerational temporal changes in the risk of IBD, or both.