Before everyone gets overly excited about carbidopa and risk of death, let's think this through rationally. Whether carbidopa contributes to an increase in PD death rate is beyond the scope of my knowledge. The science discussed in the review may all make sense. However, let's keep in mind who the lead author is and remember, we can always do plenty of fancy things with statistics to make unrelated outcomes appear to be something factual. Although a case is made for a relationship between the start of carbidopa use and the beginning of an increase in PD death rate, it may also be unrelated to a causal effect and related to some other common variable.

For example, we know two facts that have occurred over the same time period:
1} Average life expectance has increased significantly from 1970 to 2010, and we know that PD, more than anything, is a disease of aging. The rate of occurrence per thousand increases dramatically every year we age, as does the rate of death.

2) There is a much better understanding today of PD, and subsequently many more people getting diagnosed that may have been either undiagnosed back then or misdiagnosed.

3) The interaction of the two items above. Older people have more problems, get more illnesses, and die more often. Many times the death is related to several variables. In the '70s, I never remember someone dying from PD (what actually is that anyway). In fact, I probably never even heard of PD. The cause of a death would probably be listed as the problem that PD instigated. For instance, an 85 year old with PD falls, breaks a hip, gets pneumonia from being bed ridden, and eventually dies from congestive heart failure. What does the doctor put down on the death certificate as the cause?

Finally, percentage increases (as Hinz reports) are usually dramatically huge when starting from a low point, yet not always practically significant. Case in point some numbers from a CDC article that I have (sorry no link so you'll have to trust me). This is a table of deaths per 100,000 for PD over the 10 year time period 2002 - 2011.


Parkinson's disease death rates by age groups:

2011 * * * * * * 0.1 1.3 12.8 76.0 168.1 7.0
2010 * * * * * * 0.2 1.3 11.8 74.8 165.9 6.8
2009 * * * * * * 0.2 1.3 11.2 70.8 157.0 6.5
2008 * * * * * * 0.2 1.2 12.3 71.2 157.4 6.6
2007 * * * * * * 0.1 1.2 11.7 71.5 157.0 6.5
2006 * * * * * * 0.2 1.2 12.0 69.5 157.6 6.5
2005 * * * * * * 0.2 1.4 12.8 71.1 156.0 6.6
2004 * * * * * * 0.2 1.2 11.9 67.4 145.1 6.2
2003 * * * * * * 0.2 1.3 12.6 67.6 145.8 6.3
2002 * * * * * * 0.1 1.2 12.1 63.8 142.2 6.0

Sorry, but it's hard to post a table here. The first column after year is Ages 45-54, then 55-64, 65-74, and 85-94. The last column is the age adjusted average. As can be clearly seen in the first three age groups there is hardly any change in average death rate attributed to PD. The increase does become more apparent in the 75-84 and 85-94 year old groups. And, clearly the trend on age adjusted average is increasing. However, although the increase overall from 6/100,000 to 7/100,000 may be statistically significant, given the size of the population, is it really practical? Is it just possible that people who die with PD have multiple complications at the end and it's just easier for the assigning doctor to list PD as the cause?