Reader Comments

Post a new comment on this article

Fluoxetine, Suicide and the Economy

Posted by plosmedicine on 30 Mar 2009 at 23:59 GMT

Author: Carlos A Camargo
Position: MD/ Emeritus Professor of Clinical Medicine
Institution: Stanford University School of Medicine
Additional Authors: Daniel A Bloch PhD, Professor of Health Research and Policy, Stanford University
Submitted Date: September 01, 2006
Published Date: September 6, 2006
This comment was originally posted as a “Reader Response” on the publication date indicated above. All Reader Responses are now available as comments.

Milane et al report a strong negative correlation between the number of prescriptions for fluoxetine in the United States, and the age-adjusted suicide rates for the years 1988 to 2002. On the basis of this association, and an arbitrary baseline period (1960-1987) they calculate the number of lives saved for men and women by the use of fluoxetine.

This approach is fraught with dangers. A simple statistical association should not be confused with causality! For example, the authors did not take into account the possible influences upon suicide rates of the well-known economic expansion in the United States during the 1990s. For more than a century, scholars have mentioned the intricate and complex association between rates of suicide and economic parameters (e.g., poverty, unemployment, Gross National Product per capita, etc). For example, after the great crash of the stock market in the late 1920s and the ensuing Great Depression, the rate of suicide in the USA increased dramatically for several years, and then decreased as the economy improved.

On this matter, the literature is quite clear and references abundant.

Since Milane et al failed to take into account any of these confounding
factors, we decided to look at the correlations of several economic
parameters and the rates of suicide for the same years: 1988-2002. The unemployment rate and the percentage of the population eligible for the Food Stamp Program (a reasonable indicator of poverty rates) were both highly correlated with the suicide rates. The Dow Jones industrial average for each of those years, when compared with the US suicide rate, gives an even higher (negative) correlation: r=-0.98, p below 0.0001.

We also calculated the correlation between fluoxetine prescriptions and the Dow Jones average. Not surprisingly, there is a very strong positive correlation: r=0.89, p below 0.0001. And most certainly, the increasing sales of fluoxetine did not cause the marked rise of the Dow Jones during those years! If one looks at the rates of crimes against property in the US for the same period they also show very high correlations with fluoxetine prescriptions. Most scholars would relate the decrease in crime rates to the improvement of the economy during those years.

Finally, we decided to explore the relationship of suicide rates with both fluoxetine prescriptions and Dow Jones averages as potentially confounding factors in a single multivariate model. Results were clear: The statistically significant association between increasing fluoxetine sales and decreasing suicide rates disappeared (r=-0.18, p=0.54), whereas the negative correlation between the Dow Jones and the suicide rates remained high at -0.88 (p below 0.0001).

In conclusion, we believe that there is little likelihood that the increasing sales of fluoxetine from 1988 to 2002 were the cause of the modest decrease in the suicide rate during those years. It appears more likely that factors such as those connected with the sustained economic recovery of the 1990s were responsible.

(A full version of this letter, with the original data, details of the statistical analysis, tables and figures, references, etc will be submitted soon to PLoS Medicine.)

Competing interests declared: None