Estimation of Survival with Parkinson's disease in the Canadian Longitudinal Study on Aging: An analysis using only current disease durations observed at baseline

Previous studies of survival with Parkinsons disease (PD) have relied primarily on survival data from incident PD cohort studies with follow-up, prevalent PD cohort studies with follow-up, or a combination of these two study types. Each imposes logistical and resource constraints because of the length of follow-up required. Here, by using only the current disease durations for prevalent PD cases at study baseline we propose a strategy that does not require follow-up when estimating survival with PD. We apply our methods to data collected as part of the Canadian Longitudinal Study on Aging (CLSA). Using the reported disease durations for 110 CLSA participants classified as having prevalent Parkinsons Disease at baseline, the estimated median survival was 5.72 years or 5.94 years, depending on how we accounted for possible uncertainty in the recalled diagnosis dates. Our results suggest a probability of at least 0.12 of surviving more than 15 years with PD. Our estimates are lower than those from other studies, which have entailed lengthy follow-up. We use only baseline data from a study that it is projected to last for twenty years. Under certain constraints our methods can be applied to other diseases and where "early" results are desired.