Differences in Parkinson’s Disease populations: teaching hospitals versus other settings and implications for clinical trials

Background — Parkinson disease (PD) disease-modifying therapy (DMT) trials generally recruit individuals from teaching hospitals. Whether these participants represent the broader PD population is unclear.

Objective — The objective was to compare individuals with PD seen by neurologists in teaching hospitals early in their disease—a proxy for DMT trial-eligible cohorts—with individuals seen in other settings.

Methods — This retrospective cohort study using population-based data from Ontario (Canada) included individuals with PD from 1995 to 2017. Individuals with ≥1 PD visit with a teaching hospital neurologist within 3 years served as a proxy for DMT trial-eligible participants. Comparators were individuals with PD seen in other settings. We compared age, sex, income, rurality, marginalization, and comorbidities. We measured time to milestones, including drug escalation, surgical/infusion therapies, home care, dementia, long-term care admission, and death.

Results — We identified 19,948 individuals with PD, of whom 4386 (22.0%) were seen by a teaching hospital neurologist and 15,562 (78.0%) in other settings. Compared to other settings, individuals with teaching hospital neurology visits were younger, belonged to socioeconomically advantaged neighborhoods, and had fewer comorbidities. They had more drug escalation (unadjusted hazard ratio = 1.30; 95% confidence interval [CI] = (1.21, 1.38), surgical or infusion therapies (2.35 [2.09, 2.64]), and home care (1.06 [1.02, 1.10]). They had less dementia (0.813 [0.77, 0.86]), long-term care admission (0.62 [0.58, 0.67]), and death (0.68 [0.64, 0.72]).

Conclusions — Individuals with PD seen early by teaching hospital neurologists exhibited differences from the PD population. DMT trials in PD may exclude individuals with faster PD progression and from marginalized groups, impacting generalizability. Our study highlights health equity issues.