An algorithm (decision tree) for the management of Parkinson’s disease (2001):

Parkinson’s disease (PD) is named in honor of James Parkinson, whose classic monograph, “An Essay on the Shaking Palsy,” written in 1817, has provided an enduring description of the clinical features of this disorder.1 PD is an age-related neurodegenerative disorder with an average age at onset of 60 years. An estimated 1 million persons in the United States suffer from PD,2 and there are approximately 60,000 new cases each year. United States Census Bureau projections indicate that there will be a substantial increase in the number of at-risk individuals 60 years of age and older, and therefore the prevalence of PD is likely to increase in the coming decades.

The introduction of levodopa in the late 1960s represented a major therapeutic advance in the management of PD,3 providing clinical benefit to virtually all patients and reduced mortality. However, it soon became apparent that long-term treatment with levodopa is complicated by the development of adverse events that include motor fluctuations, dyskinesias, and neuropsychiatric complications.4-6⇓⇓ In addition, with disease progression, patients develop features that do not respond well to levodopa therapy, such as freezing episodes, autonomic dysfunction, falling, and dementia. As a consequence, despite levodopa treatment, most PD patients eventually suffer disabilities that cannot be satisfactorily controlled with existing medical therapies. Therefore, there has been an intensive effort to develop new treatments that reverse disabilities in patients with advanced disease, that provide enhanced clinical benefits with a reduced risk for adverse events, and that slow the rate of disease progression. This has led to an explosion of new laboratory and clinical information and to a variety of new treatment strategies for the management of PD. Physicians who treat PD patients must now assimilate a considerable body of data to optimally manage patients with this complex disorder. …