Projected number of people with Parkinson disease in the most populous nations, 2005 through 2030

The article by Dorsey et al is an important update of a lingering question about the worldwide incidence of Parkinson's disease (PD). [1] While they report an increasing prevalence, there are questions about their study design. They state the study examines the incidence and prevalence in the ten most populous countries of the world, yet the data from four of the ten are manipulated.

In the absence of data, the authors should have stated that fact and then proceed to use the data from neighboring countries. It would have been more effective if the authors used data from the ten most populous countries where data is available and then point out that specific countries should have been included but there were no data.

Neighboring countries can have markedly different populations, environment and geography so data extrapolated from one country to another may not be accurate. Estonia is ethnically, environmentally and geographically different from Russia as is Singapore from Indonesia. It appears that there is data from India, Pakistan and Bangladesh when it is actually one set of data for all three.

For example, people in India, Pakistan and Bangladesh have a curiously flat prevalence from age 60 to 80+ when actually it is only India which reflects this incidence. The way the data is presented it initially appears that the prevalence remains flat in this vast area of the world. Russia and Indonesia have a very high prevalence in the 80+ group, when it is Estonia and Singapore that show this trend.

Presenting the data more clearly would have enhanced the article. Their conclusions are perhaps valid but their extrapolation of data from one country to another may not be.

References

1. Dorsey ER, Constantinescu R, Thompson JP, et al. Projected number of people with Parkinson disease in the most populous nations, 2005 through 2030. Neurology 2007; 68: 384-386.

Disclosure: The author is an investigator for Schwarz Pharma and Kyowa Pharma.

We thank Dr. Calabrese for his interest in our work. Research is often limited by the availability of data. In our study, we sought to overcome the lack of country-specific data by two means to generate estimates of the number of people with Parkinson disease.

In the first, we used age-specific prevalence by country where available and where not available, we used estimates from neighboring countries as Dr. Calabrese points out. Our use of data from neighboring countries is clearly highlighted in the Methods, in the Table, and explicitly discussed as a limitation in the Discussion. Other investigators when confronting scant epidemiological data to generate regional estimates for the prevalence of dementia, for example, have used methods that rely on the opinion of experts to generate estimates (e.g., a Delphi consensus approach). [2] Such estimates, even with their limitations, are needed (and are vastly preferable to no estimate) as they help guide health, social, and economic policy.

Our second method, which was used to generate an aggregate estimate for the 15 countries examined, used the median prevalence estimate from 14 door-to-door studies conducted from around the world to make projections. Both methods returned similar estimates for the number of people with Parkinson disease in 2030 in the most populous nations (8.7 million for the first method and 9.3 million for the second) and highlight the growing global burden of Parkinson disease.

References

2. Ferri CP, Prince M, Brayne C et al. Global prevalence of dementia: a Delphi consensus study. Lancet 2005;366:2112-2117.

Disclosure: Dr. Dorsey is a strategic advisor to Avid Radiopharmaceuticals, Inc., and has received monetary compensation in excess of $10,000 and stock options. Avid Radiopharmaceuticals had no role in the analysis, preparation, or submission of this article.