Neurologist-associated reduction in PD-related hospitalizations and health care expenditures

Objective: To investigate the impact of neurologist care on Parkinson disease (PD)-related hospitalizations. Recent data indicate that neurologist treatment in PD may be associated with improved survival, yet is underutilized. Factors contributing to this improved survival remain unknown, but may be due in part to optimal disease treatment or avoidance of disease-related complications. Methods: This was a retrospective cohort study of Medicare beneficiaries diagnosed with PD in 2002 and still living in 2006. Hospitalization for PD-related (neurodegenerative disease, psychosis, depression, urinary tract infection, and traumatic injury) and general medical (hypertension, diabetes, congestive heart failure, angina, and gastrointestinal obstruction) illnesses was compared by PD treating physician specialty using Cox proportional hazard models, adjusting for confounders. Secondary analyses included PD-related rehospitalization and cost stratified by frequency of neurologist care. Results: We identified 24,929 eligible incident PD cases; 13,489 had neurologist care. There were 9,112 PD-related hospitalizations, and these occurred and recurred less often among neurologist-treated patients. Neurologist PD care was associated with lower adjusted odds of both initial and repeat hospitalization for psychosis (hazard ratio [HR] 0.71, 95% confidence interval [CI] 0.59-0.86), urinary tract infection (HR 0.74, 0.63-0.87), and traumatic injury (HR 0.56, 0.40-0.78). PD-related outcomes improved with frequency of neurologist care in a stepwise manner. Odds of general illness hospitalization or hospitalization did not differ by neurologist involvement. Conclusions: Regular neurologist care in PD is specifically associated with lower risk of hospitalization and rehospitalization for several PD-related illnesses. This may reflect an improved ability of neurologists to prevent, recognize, or treat PD complications.