Are demented patients with Parkinson's disease accurately reflected in prevalence surveys? A survival analysis

Two of the primary statistics which might be gathered by an epidemiologist studying a particular disease are prevalence and incidence. Prevalence is the number of cases existing in the population, and is usually expressed in terms of cases per 100,000 people. Incidence is the number of new cases which appear each year, and is usually expressed and cases per 100,000 people per year. While the two different statistics both give some idea about the occurrence of the disease, the inappropriate statistic may also given a misleading idea. This is the case when trying to determine how many patients with Parkinson's disease (PD) are also affected with dementia. Several problems face the researcher trying to tabulate the relation between parkinsonism and dementia. Simply counting the number of patients with both Parkinson's disease and dementia is quite difficult in itself, due to the problems inherent in diagnosing dementia. However, a subtler problem also exists. In a previous study, it was found that 10.9 percent of patients with Parkinson's disease also had dementia. That is, the prevalence of dementia among the PD patients was 10.9 percent. The subsequent identification of patients missed by the first study raised the prevalence to 15.9 percent. However, this number does not take into account any possible difference in the death rate of the two groups. For example, even if dementia affected 90 percent of the patients, if the demented patients died quickly, then a population survey would find primarily non-demented patients. Investigation found that demented PD patients do indeed have a higher death rate than do PD patients without dementia. In a review of cases after almost five years, 17 deaths occurred among 65 demented PD patients, in contrast with 28 deaths among 168 nondemented patients. These deaths decrease the prevalence of dementia among PD patients. The prevalence of dementia among PD patients is roughly 4.5 times greater than in a comparable control population. However, the rate at which cases of dementia develop, that is, the incidence, among Parkinson's patients is 6.65 times greater than a comparable control group. To appreciate the clinical association between Parkinson's disease and dementia, incidence is a more appropriate statistic than prevalence. (Consumer Summary produced by Reliance Medical Information, Inc.)