A very good point by Jyllian Kemsley

William Banholzer and a group of other high-level chemical corporation executives wrote a letter in this week's C&EN, where they really took academic chemical safety to task* ** -- and used an interesting metric to do it:

The facts are unequivocal. Occupational Safety & Health Administration statistics demonstrate that researchers are 11 times more likely to get hurt in an academic lab than in an industrial lab. There have been serious accidents in academic labs in recent years—including fatalities—that could have been prevented with the proper use of protective equipment and safer laboratory procedures.

 Most chemistry and chemical engineering graduate students will find employment in industry. As new hires come on board, many companies spend weeks on remedial safety training before new hires are allowed to work in their labs. This clearly shows that the current state of graduate safety education is lacking and that there is a clear need to address it. If the report is supposed to focus on “preparing graduate students, about the future,” how can this not be a relevant topic?

Jyllian Kemsley unequivocally rejects the use of the metric yesterday (but agrees with them on the general need for academic chemical safety reform):

The “11 times more likely” statistic is inaccurately framed. I followed up on it with the letter authors and Lori Seiler, Dow’s associate director for environmental health and safety in research and development. The numbers actually compare the overall injury and illness rate for academic institutions (including those that might occur, for example, in grounds keeping or a dining hall as well as in laboratories) to Dow’s overall rate. Seiler adds that the injury and illness rate for Dow’s research laboratories is consistent with the company’s overall rate, when calculated per employee. 

That said, it seems like it would be wise for the academic community to take this letter to heart.  Banholzer, Calabrese, and Confalone are not writing in a vacuum—they see the skills that chemistry graduates lack, and those skills are necessary whether those graduates are going on to work in industry, academia, or elsewhere.

This is not the first time that Dr. Banholzer has used this metric; here's a tweet where he said it in February. (Interestingly, Jyllian Kemsley raised the same concerns. Even more interestingly, the statistic seems to have grown from a 7:1 academia/Dow ratio to 11:1.) I believe that this metric was used in Dr. Banholzer's presentation to the President's Council of Advisers on Science and Technology.

While I am also in agreement that academic chemical safety could learn from industry, I think that Dr. Banholzer is missing the regulatory aspect of the different approaches between corporations and universities. Corporations are employers, and thus have legal responsibilities to provide a safe workplace and state and federal regulatory agencies to help make that happen. There is no similar requirement or regulatory agency for graduate students and universities. (Note that Sheri Sangji was a research technician and an employee of UCLA and Professor Harran, thus involving Cal/OSHA -- if she had been a graduate student, I'll bet the outcome would have been very different.)

I also really doubt this comment in the letter:

As new hires come on board, many companies spend weeks on remedial safety training before new hires are allowed to work in their labs. This clearly shows that the current state of graduate safety education is lacking and that there is a clear need to address it.

I doubt that it is literally "weeks", and if so, I suspect that it's 6 days of bureaucratic nonsense about the structure of EH&S management and accident reporting, and 4 days of actual useful training. I'm more than willing to believe that industry does a better job of safety training, but I just don't think this is a very probative statement.

Finally, I also think that it is unfair for large corporations (Dow, DuPont and Corning) to compare the safety rates of experienced employees (what is the median age of a Dow bench scientist? 42? 45? The median age of an industrially employed ACS member in 2010 was 48) versus the safety rates of the relatively young graduate students and postdocs that populate academia. They're very different populations with very different risk assessment capabilities.

*Background: Some people, including Richard Zare, were quoted by Celia Henry Arnaud in her comprehensive article that the safety section in the report seemed a little out of place.  

** While we're at it, it is frustrating to me that this report makes vague references to recent incidents in academic chemical safety, yet refuses to talk about specific cases (e.g. Yale/Dufault, TTU/Brown, UCLA/Sangji.)