Objective. Rheumatoid arthritis (RA) is the most common form of inflammator
y arthritis. Although there is a large body of evidence suggesting that RA
is immune mediated, the etiology remains unresolved. Twin studies have show
n disease concordance rates of similar to 15% in monozygotic twins and 4% i
n dizygotic twins, while the estimated risk ratio for siblings of RA patien
ts ranges from 5 to 8. Our goal was to use genealogic data from Iceland to
further investigate the genetic component of RA.
Methods. Data were obtained from a population-based, computerized genealogy
database that was developed to examine multigenerational relationships amo
ng individuals in the relatively homogeneous population of Iceland. Using a
n algorithm, the minimum founder test, we calculated the least number of fo
unders required to account for a list of RA patients, and compared it with
1,000 sets of same-sized matched control groups. In addition, we estimated
the kinship coefficient and risk ratios for relatives of the RA patients.
Results. Several familial clustering tests demonstrated that the RA patient
s were more related to each other than were the average control set of Icel
anders. A significantly fewer number of founders was necessary to account f
or our patient list than for the random sets of matched controls (P < 0.001
), and the average pairwise identity-by-descent sharing was greater among t
he patients than among the control sets (P < 0.001). In addition, there was
an increased risk of RA in first- and second-degree relatives of the patie
nts; e.g., for siblings, the risk ratio was 4.38 (95% confidence interval 3
.26-5.67), and for uncles/aunts, the risk ratio was 1.95 (95% confidence in
terval 1.52-2.43).
Conclusion. The familial component of RA is shown to extend beyond the nucl
ear family, thus providing stronger evidence for a significant genetic comp
onent to RA.