In Reply We agree with the concern of Freedman and Pfeiffer that in addition to genetic background, various levels of disease ascertainment in different health care systems might be an important factor for opposite associations between cancer and Parkinson disease (PD) reported in our study and previous ones. The nationwide health care system in Taiwan makes a thorough investigation on disease association possible, provided the analyses are comprehensive.
To make our findings even more convincing, we performed additional analyses with adjustment on the number of physician’s visits 1 year prior to index date. Patterns of medical surveillance among the 2 cohorts were thus adjusted. The overall association between cancer and PD remained the same (HR, 1.37 [95% CI, 1.30-1.44]; P < .001), although the association turned out to be nonsignificant in gallbladder (HR, 1.48 [95% CI, 0.97-2.24]; P = .07), pancreatic (HR, 1.26 [95% CI, 0.91-1.74]; P = .17), and bladder (HR, 1.26 [95% CI, 0.98-1.61]; P = .07) cancers. The reason could be attributed to relatively small case numbers of these 3 cancers under the multivariate Cox regression model.
Lin P, Lin C, Yang P. Factors in Association Between Parkinson Disease and Risk of Cancer in Taiwan—Reply. JAMA Oncol. 2016;2(1):145-146. doi:10.1001/jamaoncol.2015.4148