Environmental influences on multiple sclerosis and other autoimmune disorders

The risk of autoimmune disorders such as multiple sclerosis (MS) may be influenced by environmental factors early in life. This epidemiological thesis provides new knowledge on the role of early life environmental factors in MS in particular, for which the specific aetiology is unknown. Two main potential environmental determinants of MS, ultraviolet radiation (UVR) and infections, are explored through an analysis of their timing of action in the life course and a consideration of their possible protective effects. Other organ-specific autoimmune disorders whose aetiology is also unknown, including type 1 diabetes and rheumatoid arthritis (RA), are also assessed for links with UVR for comparison with MS. Two existing national Australian datasets provided the outcome data for the analyses. The 1995 Australian National Health Survey (NHS) provided summary prevalence estimates for ecological (population-level) analysis of four immune disorders other than MS. The 1981 Australian MS Survey, used for the largest part of the thesis, provided individual-level MS-case data for 1981 throughout Australia, and was further modified to construct a longitudinal MS-rates study dataset for analysis of timing of birth. Ecological analysis of the 1995 NHS data showed that geographic latitude was positively associated, and regional ambient UVR inversely associated, with the prevalence of type 1 diabetes in Australia. Ambient UVR exposure may thus be a protective factor against such disease at the population level. The association supports previous ecological findings for MS in Australia and adds to the evidence that UVR exposure might be a modifiable determinant for autoimmune disease generally. Longitudinal analysis of the reconstructed 1981 Australian MS Survey dataset showed that increased MS risk of around 30% was evident in Australians born in November to December (southern hemisphere early summer) compared with those born in May to June (early winter). This MS-risk pattern, indicating environmental factor(s) acting around the time of birth, mirrored (seasonally) that seen for MS in the northern hemisphere, suggesting globally similar perinatal environmental determinants modifying the risk of MS onset. Most importantly, this Australian pattern was found to be fully accounted for by individual, regional (state) and seasonal ambient UVR levels specific to the prenatal period seven to eight months before birth. Low ambient (maternal) UVR exposure in the first trimester of pregnancy thus appears to be associated with a higher risk of MS in the offspring. Birth-order analysis of cases in the 1981 MS Survey further showed that early birth order was independently associated with MS risk, MS cases being more likely to be one of the older siblings in their sibships. Consistent with the hygiene hypothesis, this result suggests that a lack of microbial exposure in early childhood may increase MS risk later in life. Population health implications of these findings are discussed. In particular, safe sun exposure and/or vitamin D supplementation during early pregnancy may help prevent subsequent onset of high-morbidity, long-duration and presently incurable autoimmune disorders such as MS in the offspring