Statistics from Altmetric.com
The publication of the 10 year mortality data from the Sydney multicentre study of Parkinson's disease serves as a timely reminder that patients with this condition still die at a rate in excess of their peers despite advances in therapeutics and surgery.1
This fact has been lost on many of our colleagues working in this area, both on the clinical and the research fronts. On many occasions at local and national meetings, I have been forced to remind people that levodopa has not normalised mortality rates in this condition.
Figure A indicates, as Hely et al 1 point out, the fall in standardised mortality rates (SMRs) in the early years of levodopa use but a return to mean SMRs of between 1.5 and 2.0 over the past decade. All of the studies over the past 10 years show a statistically significant difference as the 95% confidence intervals (95% CIs) do not embrace 1. In some, the upper 95% CI overlaps the original Hoehn and Yahr study in the prelevodopa era. These results mirror national mortality statistics for England and Wales (figure B). The fall in death rates in the late 1970s and early 1980s has now returned to a steady rise, thought to be due to the aging of the population.2
These data must act as a spur to attempts to develop neuroprotective or restorative therapies which substantially reduce mortality from Parkinson's disease. Large pragmatic studies in the future which examine novel treatments or approaches in early Parkinson's disease must consider not only quality of life and health economics issues, but also mortality in the hope of establishing reduced death rates.