Year of Publication



Martin School of Public Policy and Administration

Executive Summary

Recent news stories about laboratory accidents in which students were severely injured and killed have brought much-needed attention to lab safety. Creating a positive safety climate on college and university campuses is very important in reducing accidents. At the University of Kentucky, all laboratories are inspected once every year, and the results are recorded in a lab inspection database. These data include the department, building, lab classification, type of violation, room number, and the name of the PI (Principal Investigator). I want to analyze inspections to find out whether department and the lab classification are significant in looking at violations. Knowing this will help safety officials provide better, more specific training to those who work in these areas, and will provide those in authority with better tools to reach people at the most appropriate level. Violations can be looked at as potential accidents and potential fines. Acting on the findings of these inspections is crucial in preventing accidents from occurring on campus.

In looking at the idea of safety climate, I considered which available variables would be best to reach this concept. I thought about using building, but it is collinear with department. Departments to some extent share space in a building. Multiple departments may be housed in a building, or one department may be split between a couple of buildings. If I used this and department, my results would be difficult to interpret Departments are under the same leadership, and lab inspection reports are distributed to department chairs as well as the PI, and other safety officials. Lab classification is important because it defines the storage and use of chemicals in that lab facility. There are four lab classifications that range from broad use and storage of chemicals to no use or storage of hazardous chemicals. There are no data on specific chemicals used, or the type of experiments that are conducted. Looking at the classification was also the best way to look at potential risk with the available data. The question I hope to answer is: Do department and classification of labs affect the likelihood of violations? I looked at the average violations per inspection by department. I also looked at the average violations per inspection by classification. Fixed effects and random affects regressions were run with inspections as the unit of analysis, looking at classification and department.

According to the regression, the null hypothesis that lab classification is unrelated to the number of violations can be rejected. The P-values are less than 0.05, which makes lab classifications statistically significant. The results indicate that the laboratories that are equipped to handle the most hazardous chemicals are more likely to have violations, whereas the laboratories that are more restricted in the use of chemicals have fewer violations. This may simply occur because there is greater risk in a laboratory where there is broad use of chemicals, as opposed to those where chemicals are simply to be stored. The coefficient increases steadily along with the classification of the lab. This reinforces the finding that labs that are equipped to handle more chemicals are more likely to have violations. Department is also an important indicator of violations. Even when the classification is accounted for, violations per department are statistically and managerially significant and vary by more than 0.5 violations above and below a mean 0.678 violations per lab. This finding would allow further investigation and targeted training to departments that need it. I would also like to know more about the type of violations by department, to learn more about trends or possible causes.

The Occupational Health and Safety Department (OHS) at UK has no authority to force labs to address any violations. Department chairs, any safety officials they designate, and most importantly, the Vice President of Research has enforcement authority. There must be procedures in place at the departmental level to ensure a commitment to safety that is perceived by employees and practiced in their daily work. Safety must be easily accessible to all employees. Proper equipment, information, and a climate where safety is the main priority are crucial. I recommend that the Vice President of Research receive a quarterly report of inspections by department and lab classification from OHS, since he has the authority to enforce lab inspection findings, and it is in his interest to make sure no accidents occur, and no fines are assessed to the University by the Occupational Safety and Health Administration (OSHA), the government agency that sets the standard for occupational exposure to hazardous chemicals in laboratories. Department chairs and PIs already receive the inspections, but having additional oversight may improve the departments who rank lower in their inspections. Because reporting at the departmental level has significance, a simplified report can be created rather than one that lists the results of every inspection on campus.