Cohort analysis of Parkinson's syndrome: Evidence for a single etiology related to subclinical infection about 1920

Abstract

Cohort analysis was undertaken of all cases of Parkinson's syndrome seen at the Massachusetts General Hospital between 1875 and 1961. There were 1383 recorded cases which were suitable for analysis. Only 24 cases were located in the 43 years prior to 1918, while 1359 cases were found in the ensuing 42 years. For all cases, the mean year of birth was 1894.3 and the range ( mean ± 2 × s.d. ) was 1866.7 to 1921.9. Between the group with onset in 1920–1924 and the group with onset in 1955–1959, a period of 35 years, the mean age of the population with Parkinson's syndrome has increased 27 years. When the cases were examined by year of diagnosis, a similar trend was found. Patients with Parkinson's syndrome appear therefore to be part of a cohort of the population aging together. The curve of increasing age was found to correspond well with the mean ages of a population which ranged between ages 5 and 59 in 1920. When a series of cases with known encephalitis lethargica between 1918 and 1920 were examined, it was discovered that they had a mean age of 33.1 ( mean ± 2 × s.d. was 3.7–62.5 years). The data indicate that a population very similar to that in which von Economo's disease occurred during the epidemic of 1918 to 1921 in the United States has aged at a rate which is remarkably close to that of the increase in age for successive new cases of Parkinson's syndrome as demonstrated in this study. Only 11.2 per cent of our cases gave a history of encephalitis, however. These data would indicate that most cases of Parkinson's syndrome have a single etiology possibly related to subclinical infection in 1918 to 1920 resulting in the onset of Parkinson's syndrome up to 40 or more years later. A corollary of the hypothesis is that a period of 40 or more years may elapse between an initial, presumably infections insult and the development of a secondary manifestation such as Parkinson's syndrome. If the hypothesis is correct, then the cohort with Parkinson's syndrome will gradually disappear and the number of cases diminish. An attempt has been made to project and predict the occurrence of Parkinson's syndrome in Massachusetts. It appears that a precipitous drop will occur prior to 1980 in the number of cases.