Parental Age, Family Size, and Risk of Multiple Sclerosis

Montgomery, Scott M.*†; Lambe, Mats‡; Olsson, Tomas§; Ekbom, Anders*‡

doi: 10.1097/01.ede.0000142138.46167.69
Original Article

Background: Family structure, such as having siblings, provides proxy measures for a variety of characteristics relevant to disease risk. The etiology of multiple sclerosis (MS) is not well defined and analysis of family structure may provide etiologic clues. We conducted a case–control study to examine possible associations.

Methods: Using the Swedish Inpatient Register, we identified 4443 patients with a diagnosis of MS. From the general Swedish population, using birth and death registers, we selected 24,194 controls with similar characteristics for year, county of birth, and survival until at least age at diagnosis of the matched cases. The Multi-Generation Register linked data on siblings and parents. The Census provided father's social class based on occupation.

Results: Having 3 or more younger siblings, compared with none, produced an adjusted odds ratio (OR) for MS (with 95% confidence interval) of 0.80 (0.70–0.92) (adjusting for number of siblings, twins, maternal and paternal age, parental MS, sex, father's social class, county and year of birth). With 3 or more older siblings, the adjusted OR was 0.83 (0.72–0.96). Different-sex twin pairs compared with singletons had an OR of 0.59 (0.37–0.95) for MS. The risk of MS increased steadily with father's age but not mother's age, up to 2.00 (1.35–2.96) for 51- to 55-year-old fathers (compared with 21- to 25-year-old fathers).

Conclusions: Parents who have offspring with MS may have subtly impaired fertility. The unexpected association with paternal age may be the result of an increased risk of accumulating germ cell mutations among older men.

Author Information

From the *Clinical Epidemiology Unit, Department of Medicine, Karolinska Hospital, Karolinska Institutet, Stockholm, Sweden; the †Clinical Research Centre, Örebro University Hospital, Örebro, Sweden; the ‡Department of Medical Epidemiology and Biostatistics, Karolinska Institutet, Stockholm, Sweden; and the §Neuroimmunology Unit, Department of Medicine, Karolinska Hospital, Karolinska Institutet, Stockholm, Sweden.

Submitted 2 April 2003; final version accepted 12 July 2004.

Correspondence: Scott M. Montgomery, Enheten för klinisk epidemiologi, Karolinska sjukhuset M9:01, SE-171 76, Stockholm, Sweden. E-mail:

© 2004 Lippincott Williams & Wilkins, Inc.