Answers are grounded in this paper and cite their sources.
Background/Objectives: Parkinson’s disease (PD) progression is highly heterogeneous, and prognostic biomarkers guiding personalized management remain elusive. Complementary therapies are widely used by people with PD. Their real-world associations with long-term symptom progression remain poorly characterized, and conclusions drawn from observational data depend on the analytic approach used. We examined baseline complementary therapy use in relation to symptom severity and its rate of change using four complementary analytic approaches. Methods: Data came from the Modifiable Variables in Parkinsonism prospective cohort study (2013-present, n = 2656 at baseline). Participants reported consistent use of 51 complementary therapies over the prior six months, and the outcome was the Patient-Reported Outcomes in PD (PRO-PD) score. Four approaches were applied: (1) separate regression models for each therapy, (2) a multivariate model including the 24 therapies with ≥100 users, (3) unadjusted smoothed trajectories, and (4) linear mixed-effects models with a Time × therapy interaction. Adjusted models accounted for age, gender, income, and years since diagnosis. Results: Vitamin D (63.1%), vitamin B12 (40.6%), and fish oil (39.5%) were most frequently used. In the full 24-therapy baseline model, low-dose lithium, fish oil, and coenzyme Q10 were associated with lower baseline symptom severity, whereas oral glutathione was borderline (p = 0.067) and reached p = 0.045 after backward elimination. After false-discovery rate control, baseline use of oral glutathione and coconut oil was associated with slower accumulation of PRO-PD points, and NAC and probiotics with faster accumulation; four further therapies reached unadjusted p < 0.05 only. Associations frequently changed across analytic approaches, and several therapies reversed direction between cross-sectional and prognostic analyses. Conclusions: Associations between complementary therapy use and PD symptoms depended strongly on the analytic approach, and cross-sectional associations were a poor guide to later symptom change. Long-term pragmatic follow-up identified over-the-counter therapies associated with slower (oral glutathione, coconut oil) and faster (NAC, probiotics) symptom accumulation. These findings do not establish causation; rather, they illustrate the value of systematically examining longitudinal patient-reported data in identifying biologically plausible patterns that emerge amid the complexity of real-world care and can guide subsequent explanatory research.
No takes yet. Share an insight, caveat, or question.
Mischley et al. (2026) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: