Autoimmune disease in families with multiple sclerosis: a population-based study

Evidence of an association between multiple sclerosis (MS) and other autoimmune diseases would substantiate the hypothesis that MS is an autoimmune disease, and implicate a common mechanism. We aimed to investigate and compare the rate of autoimmune disease in MS patients, in their first-degree relatives, and in their unrelated spouses.

We used data from a national, multicentre, population-based sample to investigate the rate of autoimmune disease in 5031 MS patients, 30 259 of their first-degree relatives, and 2707 spousal controls. We asked patients and controls whether they had any of ten autoimmune diseases: Crohn's disease, ulcerative colitis, rheumatoid arthritis, type 1 diabetes, psoriasis, pernicious anaemia, systemic lupus erythematosus, autoimmune thyroid disease, vitiligo, and myasthenia gravis. MS probands were also asked whether their first-degree relatives had Crohn's disease, ulcerative colitis, rheumatoid arthritis, or type 1 diabetes.

After correction for age and sex, we did not identify any increased risk of autoimmune diseases in MS patients compared with their spousal controls (odds ratio [OR]=1.07, 95% CI 0.86-1.23, chi(2)=0.47, p=0.49), or in the first-degree relatives of MS probands compared with controls (OR=0.89, 0.63-1.17, chi(2)=1.11, p=0.29). However, the reported frequency of autoimmune diseases did differ according to the sex of the interviewee (female vs male patients chi(2)=92.2, p<0.0001; female vs male spousal controls chi(2)=87.1, p<0.0001). MS patients had slightly higher rates of thyroid disease and pernicious anaemia than did controls, which is consistent with MHC associations for these diseases, but this effect disappeared when results were adjusted for sex. For eight other diseases the rates were similar in MS patients and controls. Families with multiple cases of MS were no more likely to report autoimmune diseases than families with single MS cases.

When data were adjusted for sex, no excess of common autoimmune diseases could be identified in MS patients or their families, including multicase pedigrees. Our results suggest that women are more aware of family medical histories than men, which emphasises the potential for ascertainment bias in unstratified data for a sex-limited disease. Family histories should thus be taken from male patients in the presence of a spouse.