This chapter examines the interdependencies of the actors involved and the contextual features that make the academic research laboratories unique. Important among these features are the influence of personnel within the academic hierarchy, pressures for scientific productivity, feedback and communication channels, and the influences of external sources (e.g., funders, journals, and competitors). The chapter identifies well-recognized systems, lab processes, and practices that can improve safety performance in academic research labs. This coverage recognizes the complex and dynamic nature of the environment in which academic administrators, researchers, and students must work.
While large and small institutions have different resources to implement an effective culture of safety, it is also true that all institutions must meet certain safety requirements to operate and conduct scientific research. Positive safety performance is more difficult for some institutions to achieve given their resources, but none are absolved of the responsibility to provide a safe environment for their employees and students. Moreover, positive safety results can be an effective tool for recruiting and sustainability. Many of the same organizational processes, pressures, and practices apply to most academic organizations independent of their size.
Finally, while examples of practices from national laboratories are included in this discussion, we recognize that there are similarities and differences between these environments and the academic landscape. Nevertheless, organizations are encouraged to take advantage of lessons
learned as good practices to be considered. Learning organizations take advantage of these successes and find ways to implement versions for their own purposes.
Although there is still some debate on just how true it is that certain national laboratories have models for safety that go far beyond what is observed in a university setting, with some information in hand, it does seem true that academics can learn from these models and start to initiate their own. For example, a recent visit to a national lab (National Renewable Energy Laboratory) showed an impressively high level of safety precautions. From extremely safe and easy-to-use engineering controls in laser labs, to very high levels of documentation of chemicals and materials, this lab was a model for what many research labs should seek. Many of the procedures and precautions used in this lab can actually be found on the Web.1 It is the desire of the Department of Energy to carry out precautions in a way that illuminates any possible weakness in their system regarding handing of chemicals or radiation exposure. The key safety personnel for each chemistry department can easily access this information and start to initiate their own departmental safety protocol.
A strong, positive safety culture instills thinking and behavior that assigns a high priority to safety. Such a culture encourages all concerned to have a questioning attitude about anything related to safety, to adopt a prudent approach to all aspects of their jobs, and to welcome open communications among different levels in the organization about safety issues. Chemistry laboratories are affected by hierarchies in the university, in the wider professional arena, in funding agencies, and in research organizational contexts. There are well-defined hierarchies within these entities that influence their ability to realize a vibrant safety culture. Several important factors influence this process on a variety of levels.
Chemistry departments house academic teaching and research functions and there are subunits within the department that can operate with a fair degree of autonomy. Department chairs, principal investigators, lab
managers, and graduate students head these units. Experience and anecdotal evidence support the description of the research units as academic “fiefdoms” where principal investigators have significant authority over their own research and operate autonomously as long as they do not intrude into other “fief” territories.2 “Each fiefdom has an intellectual or administrative territory over which he or she reigns.”3
There are hierarchies within these independent silos that can impede developing a culture of safety. First, the department head has administrative responsibility for safety in the department. The managerial responsibility of department chairs may conflict with their role as the principal investigator. Second, principal investigators may regard safety practices, such as inspections by outsiders, as a barrier to their research projects and violation of their academic freedom. Third, the individuals within the unit (lab managers, graduate students, and staff) are dependent, financially and educationally, upon a principal investigator’s grant or research project. Taken together, these factors make it difficult to communicate safety concerns, raise awareness, or suggest changes.
At the majority of U.S. institutions that conduct chemistry research, the faculty are expected to develop independent research programs and generate, from external sponsors, much, if not most, of the financial support necessary to support the equipment, supplies, and personnel, often including support for graduate students, required for research. As noted elsewhere,4,5 these expectations and traditions of academic advancement create substantial pressure. Funding and publications are often given priority in decisions about advancement, salary, space, and other reputational issues. These pressures, combined with minimal if any training in personnel or laboratory management during the doctoral and postdoctoral periods or “on the job” in most universities, create challenges for the academic safety culture.
Within the hierarchy, graduate students’ goals are aligned with these productivity goals because as one student succinctly captured it, “time is thesis.” The more the researchers produce, the faster they can graduate.
2 U.S. Chemical Safety and Hazard Investigation Board. Texas Tech University Laboratory Explosion: Case Study. Case No. 2010-05-I-TX. Washington, DC, October 19, 2011.
3 Vangelisti, A. L., J. A. Daly, and G. W. Friedrich, eds. Teaching Communication: Theory, Research, and Methods. Routledge, New York, 2013.
4 U.S. Chemical Safety and Hazard Investigation Board. Texas Tech University Laboratory Explosion: Case Study. Case No. 2010-05-I-TX. Washington, DC, October 19, 2011.
5 American Chemical Society. Advancing Graduate Education in the Chemical Sciences: Full Report of an ACS Presidential Commission. American Chemical Society, Washington, DC, 2012.
There is a pressure to publish, but there is also the pressure to come up with results that the leader (professor, principal investigator) is seeking. This leads to quantitative workload stress, derived from the need to keep working to retain one’s job and avoid getting “scooped” by a colleague or competitor. It also produces demands in terms of qualitative workload stress—that is, the need to keep working until you find the results you targeted in your research project. Finally, the power differences between the principal investigator and graduate students can inhibit the reporting of hazards, incidents, shortcuts, or near misses. This is relevant because of educational hurdles as well as keeping the funding for the research unit.
The pressures to produce results are further fueled by the fact that financial support for graduate students relies heavily on individual research grants. This reliance on grants to support students creates a potential conflict between a culture of safety and productive grant-supported research.6 Decoupling graduate students’ dependence on grants for financial support may provide a useful way to enhance the development of positive safety culture in research groups.
There is evidence that the social context that these productivity pressures create can cause injuries. External loads, organizational factors, and social contexts were hypothesized to have a relationship to repetitive strain injuries. Since then, there has been evidence that emotional and psychological demands can have effects on biomechanical functioning.7 Injuries further erode the culture of safety within the unit.
Because of the composition of academic laboratories, it is important to make special mention of evidence that young people differ from more experienced researchers in their perceptions about risks that affect their behavior. A National Academies study examined how youth are different and are affected by the way that work is organized and managed, with possible generalization to postsecondary students.8 While university students are not children or adolescents, there is certainly a range of maturity and development within the university community and some of these trends may be applicable.
7 Marras, W. S., K. G. Davis, C. A. Heaney, A. B. Maronitis, and W. G. Allread. The influence of psychosocial stress, gender, and personality on mechanical loading of the lumbar spine. Spine 2000; 25(23): 3045-3054.
8 National Research Council. Protecting Youth at Work: Health, Safety, and Development of Working Children and Adolescents in the United States. National Academy Press, Washington, DC, 1998.
Young workers are often engaged in work with high turnover, little on-the-job training and limited discretion, uncertain hours, low pay, and few benefits. Jobs with these qualities tend to be more dangerous, and are often found in small businesses, much like a laboratory setting. Studies point to a negative relationship between an organization’s size and risk of injury or death. Like small businesses, university labs may have high turnover, leaving more inexperienced workers in charge of potentially dangerous tasks. University labs are also more exposed to market pressures, which may lead them to ignore safety procedures by cutting corners. A National Institute of Occupational Safety and Health survey found that smaller organizations (fewer than 100 workers) provided less training, conducted fewer inspections, and used fewer professionals in their safety programs.
How young people recognize and assess risks and how they decide on which courses of action are important to all aspects of university life. As children develop into adults they begin to generate options, look at situations from different perspectives, anticipate consequences, and evaluate the credibility of sources. By mid-adolescence, young people can make decisions similar to those of adults.
There are data to indicate that injured teens may have taken on tasks to prove that they are responsible and independent. They performed these tasks despite knowing that they were dangerous or violated laws but acted in fear of losing their jobs. There may be analogies between behavior in these situations and in university laboratories.
Another report, Improving the Health, Safety, and Well-Being of Young Adults: Workshop Summary,9 highlights the differences between younger workers and adults and the interventions that seem to be effective in improving health and safety:
- Young adults tend to have the lowest awareness of risk and the least access to health care and insurance.10
- Brief interventions, including skills-based interventions, motivational interviewing, and personalized normative feedback are
9 National Research Council. Improving the Health, Safety, and Well-Being of Young Adults: Workshop Summary. The National Academies Press, Washington, DC, 2013.
10 Id., p. 45.
effective methods for reducing risky behavior, such as drinking among college students.11
- Peer-to-peer interventions can achieve buy-in, trust, and rapport in creating effective change.12
- Rewarding those young people for good positive behavior, rather than punishing bad negative behavior, may achieve getting young people involved in reducing undesirable actions. For example, if someone is in trouble for drinking or drug use, the person offering the help should not get in trouble for reporting the problem. Young people can provide resources for their friends by becoming involved. They should not be punished for this reporting.13
Communication about lab safety is couched in the language of compliance. There is a stronger emphasis on compliance than on safety. Understandably, administrators are keenly aware of managing perceptions about organizational safety and its impact on the institution. This leads to the enactment of policies and procedures designed to mitigate these risks. This is often done as a top-down approach to creating change. At the same time, technical support staffs (including environmental health and safety [EHS] and chemical safety personnel) are familiar with mandated standards that must be met to comply with regulations. Professional staffs have a sense of urgency because they understand the technical aspects of the requirements and regulations and because of their genuine interest in mitigating risks to people. The actions they produce are often grounded in regulatory directives, or prohibitions to autonomously functioning individuals and research units.
Most of the measures reviewed from chemistry laboratories are lagging indicators of safety performance. That is, they record what has already occurred, tend to have a negative tone, and seem to be affixing blame. To change behavior and the culture, organizations should be monitoring leading indicators—measures that can prevent incidents and mitigate risks. Lagging indicators are more typical of a compliance-based, reactive approach. Typical lagging indicators would include parameters such as the number of accidents, incident rates, deaths, body part affected, time
11 Id., p. 87.
12 Id., p. 88.
of injury, reasons why the injury occurred, profile of the injured worker, direct or worker compensation costs, and number of lost workdays.
Leading indicators could include, but are not limited to, near misses; lessons-learned databases; research group meetings focused on safety; job safety analyses completed and trends therein; surprise inspections and their results; case studies highlighting good practices; results of suggestion programs and changes made; training opportunities, requirements, and resources; awards for positive actions; behavioral observations completed; principal investigator coaching; intra-lab coaching and information sharing; and safety perceptions about how people throughout the organization view safety. These kinds of data highlight the importance of changing behavior and allow information to flow upward in a hierarchy. Moreover, if the leading indicators were to be tied to decisions such as promotion, salary increases, and resource allocation, they could influence peoples’ behaviors in meaningful ways.
A systems approach is needed to manage any changes and to avoid serious injuries. A thorough analysis of risk in complex systems considers more than the technological and engineering solutions. It requires addressing the psychological, social organizational, and political processes that contribute to incidents.14 One implication is to understand the leading indicators to change individual and organizational behavior. Human factors and ergonomics principles and systems safety have been used to change many complex systems using leading indicators.15
Although the context differs from industrial examples, the same principles can be applied to academic chemistry research laboratories. A proactive systems approach is needed to influence individual and organizational behavior. Forward-looking methodologies and metrics can avoid the unintentional blindness caused by a compliance-based approach.
The top-down approach is often met with resistance, in part, because the policies and procedures may not seem to make sense, or have any real validity, or may be perceived as being at odds with research productivity. This is especially true when requirements are promulgated by those without any experience in a specific research area or when a policy or procedure is expected to cover a wide range of applications. Further, if the demand for action is perceived as a response to litigation or as a defensive
14 Bea, R., I. Mitroff, D. Farber, Howard Foster, and K. H. Roberts. A new approach to risk: The implications of E3. Risk Management 2009; 11(1): 30-43.
15 National Research Council. Macondo Well Deepwater Horizon Blowout: Lessons for Improving Offshore Drilling Safety. The National Academies Press, Washington, DC, 2011.
action, the approach may be seen as geared to match compliance demands rather than as an active attempt to improve safety culture. The negative reaction is both predictable and understandable.
A less than enthusiastic response can be expected when professional staff assert that the reason for doing something is because “it’s the law.” Moreover, when managers and responsible individuals are threatened by regulations, the modus operandi is to practice avoidance behavior rather than proactively seek positive outcomes. Finally, when policies and procedures establish minimum standards, these become the target (“satisficing”).16 Instead, a true culture of safety should involve optimizing conditions through desired behaviors.
A prime example of this was found with the University of California’s response to its settlement with Cal/OSHA. It developed laboratory safety policies for Laboratory Safety Training, Personal Protective Equipment, and Minors in Laboratories and Shops. After its initial draft and much negative reaction from researchers, the policy had to be reworked to make reasonable accommodation for practical implementation by laboratories.
More than a set of standard operating procedures and policies, a culture of safety extends beyond departments to all members of the organizational community. This will require a campus-wide approach to changing the safety culture. Partnering with other labs, departments, and colleges can have a much higher synergistic effect than a single laboratory making changes in isolation.
The motivation for changing practices should be to improve the working conditions in laboratories to enhance the quality of research, protect its people, and create sustainable results. If done correctly, compliance will follow. Focusing on a compliance strategy alone has a less likely chance of developing a positive safety culture.
Leaders at all levels in the organization must demonstrate that safety is a value and must convey their expectations to their followers. Who are these leaders? Similar to other industries and organizations, “the ultimate responsibility for creating a safe environment and for encouraging a culture of safety rests with the leadership of the organization and its operating units.”17
The investigation of the 2010 incident at Texas Tech revealed “safety
16 March, J. G., and H. A. Simon. Organizations. Blackwell, Cambridge, MA, 1958.
17 National Research Council. Prudent Practices in the Laboratory: Handling and Management of Chemical Hazards, Updated Version. The National Academies Press, Washington, DC, 2011.
policies either did not exist or were not enforced. No single person or entity within the university was accountable for ensuring that the CHP was up-to-date, enforced, and applicable to the laboratories it was meant to regulate.”18
Often, researchers who manage projects are unaware that they are the persons responsible for safety in their organization. Clear lines of authority and responsibilities that come with positions should be articulated clearly to everyone.
Analysis of tragic events in complex systems19 have shown that failures
can be traced back to management processes that did not provide adequate controls over the uncertainty of human decision making, … Management processes failed to adequately identify and mitigate risks created by operational decisions prior to the blow out, communicate critical information, train key engineering personnel, and ensure measures taken to save time and reduce costs did not adversely affect overall risk.20
The lesson learned here is that leadership needs to be exerted at all of these levels to create a culture of safety.
The U.S. Chemical Safety and Hazard Investigation Board (CSB) investigation points out that there is no single point of failure in serious incidents. The event is the result of a complex interaction among diverse actors across levels of the organization. While most accidents focus on the human error, or mistakes made by the person directly involved, deficiencies can be found throughout the organization that contributed through inaction, poorly defined roles or expectations, training, enforcement, and/or monitoring. Therefore, strong leadership should be taken at all levels of the academic institution. Moreover, leadership should address not only the technical and engineering aspects of safety, but also the psychological, social, organizational, and political processes involved in causing injury events.
18 U.S. Chemical Safety and Hazard Investigation Board. Texas Tech University Laboratory Explosion: Case Study. Case No. 2010-05-I-TX. Washington, DC, October 19, 2011: 14.
19 National Research Council. Macondo Well Deepwater Horizon Blowout: Lessons for Improving Offshore Drilling Safety. The National Academies Press, Washington, DC, 2011.
20 Id., p. 76.
The daily routine of most faculty members is filled with many responsibilities. These responsibilities range from educational activity and academic research to administering, planning, and executing new initiatives as well as departmental service (which includes teaching and committees). Generally, these responsibilities constitute the bulk of the evaluation of the annual success of each faculty. These are the core parameters for which promotion and tenure as well as merit-based salary standards are set. Thus, great attention is placed on these areas of activity in each department each year. This is already a substantial set of responsibilities, which indeed keeps faculty members who run research laboratories very busy. In a research leadership position, laboratory safety is also a major responsibility. However, the level of importance that is placed on laboratory safety in various chemical laboratories in reference to the overall evaluation of a faculty member’s performance is not as certain. This leads to the question of how much a faculty member’s safety practices should be weighed in considering advancement within a department.
The question of impact (reward) of a faculty member’s safety practices is as much a matter of research and scientific discipline as it is a matter of culture. The first important issue to remember is that the need for laboratory safety is not only good for the health of the students and researchers involved but also in educating and providing a positive example to younger scientists that laboratory research can be done safely and, at the same time, efficiently. The practice of laboratory safety is ultimately left up to the individual, and in most cases the importance of doing research safely is learned from others in the same lab. A faculty member’s leadership skills are truly tested in both illustrating the importance of lab safety and enforcing its practice at all times. There exists a temptation to sacrifice this responsibility, at times, out of a perceived need to conduct particular experiments when time and/or resources are limited. The faculty member’s leadership and exemplary discipline in carrying out proper safety precautions is needed most in these situations. When safety precautions are neglected in the lab, it is the responsibility of the faculty member to use measures necessary to eliminate this behavior so as not to harm others in the lab. This is indeed a matter of scientific or research discipline. However, because the previous research experiences of each lab member may vary, so too will their level of discipline in safety behavior. Thus, there is a cultural aspect to the demonstration of a faculty member’s attitude toward safety. While it is certain that each faculty member may have experienced varying cultural attitudes toward safety, it is now clear there
is little room for this diversity in allowing bad safety habits to exist and ultimately to harm those who are present in the laboratory. Department practices that place real importance on safety during annual or advancement evaluations of each faculty member could have a large impact on changing the culture of safety in academic laboratories.
There are several reasons that would justify a department using laboratory safety as one measure of a faculty member’s advancement. Perhaps the most important of these is that this could be a good preemptive strategy for preventing accidents or injuries. If this is a generally accepted practice and each faculty member is aware that his or her annual evaluation is partly dependent upon their safety practices and safety management, this may provide more uniform safety behavior (culture), which is safety culture in the department before an accident actually happens. Unannounced safety checks may also provide a good measure of each faculty member’s performance in this regard. This would also be a good way for the department to evaluate the progress of each of its faculty members over time in providing a safe research environment. As mentioned above, while the discipline of performing research is dependent on the leadership of the faculty member and ultimately the individual doing the research, the possibility of changing the cultural attitude toward safety is also the responsibility of the department and research/university community. An additional reason is that including safety in annual and advancement evaluations allows faculty members to document and report the substantial work required to develop and sustain a strong, positive safety culture in their laboratories. It encourages faculty to measure and report leading indicators for their groups, as metrics of adaptation to rapidly changing research programs. Using more direct and formal methods of evaluating a faculty member’s discipline, leadership, and, ultimately, cultural attitude toward doing laboratory research in a safe manner could make a difference in reducing the number of incidents each year.
Publication is a major component of academic life. As mentioned previously, it is also a major factor in promotion and tenure decisions. It is the driving force behind the hard work and effort of aspiring graduate students. Because this high level of ambition and enthusiasm may at times cause some scientists and engineers to make hasty decisions about safety, the publication process may also be used to define, describe, and defend the important safety precautions and practices necessary to carry out research. Some journals encourage the inclusion of safety information when particularly hazardous materials are used in experiments documented in an article. Since many experiments involve potentially
hazardous procedures, making safety information a regular component of most or all experimental papers would provide a strong incentive to the development of more widespread safety culture.
The manner in which this could be enforced by particular journals (in chemistry) is relatively straightforward. In each publication of a full article (or even in communications), there is a section for experimental details. This section should be expanded to include strategies for hazard identification and risk mitigation. The purpose of this expansion is not only to inform future researchers about the hazards of carrying out a reported procedure, but also to allow the young scientists writing the papers to recognize that this is a professional requirement. Much of this can be formulated into procedures that many of the lab members can utilize in their own papers. In instances in which unanticipated hazards or risks are discovered during data acquisition or analysis, safety information must be included in results and discussion sections and in the abstract and any publicity about the work.
Hazard analysis involves the identification, assessment, and mitigation of hazards and their associated risks. It is a process to assess risks and ensure that those risks are mitigated or eliminated before initiating any laboratory work. These are critical skills for an individual to know and apply. One should assume that no activity is guaranteed to be absolutely risk-free, especially when some hazards may not have been identified, assessed, or properly mitigated. In addition, one cannot assume that hazards remain unchanged even on routine jobs or with any task requiring job hazard analysis.
For the hazard analysis to be successful, all individuals involved are required to participate and be able to recognize and identify hazards. Hazard recognition and identification can only be obtained through training and continuous feedback (e.g., during walk-throughs, observations, and peer-to-peer feedback). This learning process must be extended to all individuals involved in research: undergraduate and graduate students, postdocs, faculty/teachers, principal investigators, laboratory managers, coordinators, etc. To build a long-term, well-informed/educated culture of safety, this process should start at the undergraduate levels and be incorporated into academic research at all levels, including thesis and dissertation proposals, laboratory notebooks, presentations, and publications.
The designs associated with safe, efficient laboratories have evolved over time. In synthetic chemistry laboratories, two factors that have changed significantly are the ratio of hood to open bench space, and the relative locations of space in which active experimentation is going on and space in which writing, computations, and other desk work are being carried out. Academic laboratories built before 1950 had significant bench space but little associated fume hood space. It was common practice to carry out chemical reactions (even ones involving highly toxic chemicals) on laboratory benches where the researcher was not protected by a fume hood. Gradually, the ratio of hood to bench space increased as new buildings were constructed, but the common standard of one hood per researcher was not institutionalized in many laboratories until the past 10 or 20 years. There are still many laboratories in which the available hood space per researcher is limited, resulting in experimental procedures involving hazardous chemicals and gases are being carried out on bench-tops or on vacuum lines situated outside of fume hoods.
Physical and biological laboratories raise potentially problematical issues. In years past, most physical chemistry groups performed relatively few syntheses. However, with increasing interest in novel and functional materials, such groups have been carrying out more synthetic work. The amount of fume hood space in typical physical chemistry laboratories, as well as the perception of the risks involved in carrying out synthetic procedures, is often too limited. Biological laboratories face a similar risk. Groups that work with highly toxic organisms or certain radioactive materials have special laboratories designed to protect workers from those hazards, and there appear to be good protocols and campus oversight for those activities.21 However, it is clear that in many routine situations, many researchers in the biological sciences feel their experiments are free of chemical hazards, perhaps because they are performed mostly in aqueous media. This leads both principal investigators and researchers to believe that common chemistry laboratory safety practices, such as wearing safety glasses, lab coats, and protective footwear and gloves, are unnecessary—even in cases in which biological materials are being modified with potentially hazardous chemical reagents.
Many physical chemistry laboratories have an additional possible concern that may need to be addressed. This involves the use of lasers. This equipment raises the important issue of eye damage from accidental exposure of co-workers’ eyes to laser irradiation. This can be prevented
by the rigorous installation of interlocks and, better yet, the installation of devices that allow the positioning of the elements used in laser experiments (e.g., mirrors, detectors, and spectrometers) by remote control, which minimizes the accidental exposure of the experimenter to laser beams (Box 4-1).
Laser Safety Anecdote
The use of Class 3 and 4 lasers in academic and other research institutions has become commonplace. There have been a number of serious accidents involving exposure to laser beams. These may result from a lack of training, experience, or safety culture for those involved, or possibly point to the need for critical, yet costly, engineering controls. These incidents have involved both new and experienced scientists and engineers. For example, one recent incident involved a graduate student and a visiting scientist with more than 15 years of laser experience. Both researchers were working with a Class 4 multiple laser system at full power when the scientist was struck in the right eye by specular reflection, resulting in a retinal burn and a loss of acuity in the eye. Neither researcher was wearing laser eye protection while repositioning a mirror element that investigators believe caused the beam to reflect off a stainless steel mounting post. Laser eyewear was not worn so that the researchers would see a small amount of visible light from the laser while aligning the mirror. This was a clear violation of standard operating procedures that specified the use of laser eye protection.1
A number of safety precautions were overlooked in this incident. For example, the potential for eye exposure while repositioning optical elements was not even considered during the work planning process. Even if it had been, would (or could) the incident have been prevented with the use of specific engineering controls? At a minimum, it is clear that lasers should be equipped with a protective housing, a clearly identified aperture, and a clearly marked switch to deactivate the laser or reduce its output to less than maximum permissible exposure. However, in this example, as is the case in a large number of the accidents involving laser exposure, the laser light came from a specular reflection, not directly from the laser beam. The use of engineering controls2 is thus necessary to protect all individuals in a laser room, even those who are not actually performing the experiment on the laser table.
1 United States Department of Energy. Special Operations Report: Laser Safety. U.S. Department of Energy, Washington, DC, February 2005. Available at http://jrm.phys.ksu.edu/Safety/DOE_Laser_Safety_Report-Mar-05.pdf.
2 University of Waterloo Safety Office. Engineering Controls. Available at http://www.safetyoffice.uwaterloo.ca/hse/laser/documents/control__engineering.html. Accessed July 8. 2014.
Engineering controls, with complete elimination of a hazard, are at the top of the hierarchy for safe experimental design. A number of research institutions have used engineering controls to remove a hazard or place a barrier between the worker and the hazard. Well-designed engineering controls can be highly effective in protecting workers and will typically be independent of worker interactions to provide this high level of protection. The initial cost of engineering controls can be higher than the cost of administrative controls or personal protective equipment. This is especially true in dealing with engineering controls for electronic or laser equipment. However, over the longer term, operating costs are frequently lower, and in some instances can provide cost savings in other areas of the process.
As noted earlier, many universities offer research workers the option of obtaining free prescription safety glasses. However, in some places the cost is charged to the principal investigator’s research grants. Similarly, chemical waste disposal in many institutions is covered by university funds, but for others, the cost is also recharged to research grants, just as it is for safety glasses. Recharging safety glasses and hazardous waste disposal costs to grants incentivizes researchers to take shortcuts that could result in injury or damage to the environment.
In a strong, positive safety culture, researchers are encouraged to care about working safely and are rewarded, rather than sanctioned, for this philosophy. One of the most recalcitrant problems in many chemistry laboratories is the attitude, unfortunately often reinforced by principal investigators, that safety practices are time-wasting inhibitions to research productivity. Efforts must be found to convince such people that working safely enhances, rather than inhibits, research productivity. Certainly, an accident is one of the most serious inhibitors of research productivity. Thus, one would think that principal investigators would have a strong incentive, for that reason as well as many others, to foster a positive safety culture in their laboratories.
Strong, positive safety cultures will develop when researchers care about and promote working safely, and when institutions have an obligation to monitor working conditions to ensure that they are safe and that the procedures being used are safe. The classical approach involved is enforcement, that is, strong sanctions for people who do not work safely.
Although this may be necessary in some cases, and is one of the factors that maintains safety in industrial research laboratories where people can be fired for safety violations, we believe that using only the “stick” rather than the “carrot” is not the most rational way to ensure a strong safety culture. We believe that encouragement and rewards for good safety practices are both more effective and result in a more collegial and safe university laboratory environment.
In this context, the following characteristics should be sought and encouraged in laboratory environments to ensure that laboratories have strong safety cultures:
- Laboratory safety culture is strongly influenced by the extent to which research workers are consulted about safety rules and procedures. Rules handed down from the administration in the absence of such consultation tend to be designed in a one-size-fits-all manner, which may apply reasonably well to one type of research laboratory, but not very well to others. This not only creates inefficiencies, but also produces hard feelings on the part of research workers, which can erode any hope of developing a culture that encourages researchers to care about working safely.
- There are facets of a rational award structure that can be improved in most universities. One important target should be the group meetings that almost all research groups hold on a weekly or other regular basis. If incentives could be found to devote some period of time every week to safety issues at these local meetings, it would go a long way toward the establishment of a positive safety culture in specific laboratories.
- Funding agencies may choose to include these factors in grant evaluations, for example, as part of the “broader impacts” sections that are now being required in National Science Foundation grant proposals.
- Even if a reward structure for working safely can be developed, administrators have an obligation to make sure that proper safety procedures are being followed in their institutions. Most chemistry departments (or at least universities) have one or more safety officers who are responsible for monitoring laboratory environments and working out ways to deal with problems that arise. The existence of such positions is important, but note that in many institutions the safety officer is overwhelmed by the large number of laboratories that he or she is responsible for, and especially by the diversity of activities (e.g., synthetic work, laser experiments, biological studies) subject to monitoring. To alleviate this burden, and also to improve administrator–research
worker interactions, students and faculty should be involved in both monitoring and establishing safety procedures, perhaps by the appointment of one or more faculty members as “safety advocates” rather than safety officers, and by membership on a departmental safety committee.
- It is essential that some kind of laboratory inspection schedule—without prior announcement—be established. If these are handled in a collegial way, the inspections can have a positive effect on the development of laboratory safety culture. The inspections, as well as other interactions with departmental safety committees and/or advocates, could also play a role in encouraging intra-lab coaching/collaboration and teaching researchers how to politely approach their peers about potential safety hazards that should be corrected. In this way, a positive safety learning environment in the laboratory can be created.
Currently most departments require formal safety training for incoming graduate students. This typically involves communicating information on the proper use of protective gear, such as lab coats, safety glasses, proper foot and head protection, and fume hoods, along with scenarios of accidents that have occurred when such precautions were not taken. It also provides information about what to do when an accident occurs, which requires knowledge of emergency phone numbers, location of safety showers, etc., and includes hands-on training in the use of fire extinguishers. Some of this training includes advice about what to do in case of a fire or the occurrence of a natural disaster. It is important to include instructions about procedures to follow in the event of chemical spills and explosions.
For institutions that are still not providing such training, it should be made part of the curriculum. In addition, it should be a requirement not only for students, but for postdocs and other researchers as well. Training for non-student researchers hired directly by the principal investigator that may not arrive on a specific schedule, as graduate students usually do, and may not pass through an institutional safety training program is a concern. Particularly problematical are research workers who enter university laboratories with their own funding, which often means that it may not be possible to use a payroll roster to screen them for safety training. However, in nearly all cases that we are aware of, research workers are given keys or electronic card access to the buildings and laboratories in which they work. The key-issuing office may be used as a checkpoint for determining whether incoming laboratory workers have received
appropriate safety training, irrespective of whether they claim to have had such training in another institution. They should not be given their building access until someone has signed off on this training.
While this report is written to help improve the safety culture of academic laboratories in the United States, it is realistic to recognize that there will be a minority of principal investigators and research groups that will be resistant to the development of a positive safety culture in their laboratories. In such situations, the ultimate responsibility for ensuring safe working conditions rests with the department chair and the university administration. Although the prospect of shutting down a principal investigator’s laboratory is an unfortunate action, it cannot be taken completely off the table as a last resort way of making sure that research workers in a university are protected.
There are many advantages in promoting a safety culture and environment in a chemistry department. Ultimately, the results of providing a culture of safe and reliable scientific practices can be leveraged to enhance the overall success of a chemistry department and possibly increase its competitiveness. While there are some numerical metrics that might be used to characterize the success of a particular department, one measure might be the quality and competitiveness of the department to attract the very best talent in chemistry. For example, if a particular department is noted for establishing a good and safe culture in doing scientific research, then this may attract highly competitive graduate students through their recruiting efforts. Most chemistry departments have extensive events to recruit excellent students each year to their programs. Some of these events are extremely costly to these departments. The department could optimize their investments in this process by including information in the message to future students that ONLY a safe and welcoming culture in doing scientific research would be allowed in their department. This message, and the data to support the claim, would be a very powerful point to make to those intending to do research as they undertake careers in the sciences and would offer respect and assurance to the students that the department cares about their safety. Often, the issues of laboratory safety are overlooked during recruiting weekends and events, and providing this information to prospective students (and postdocs) would make a
strong point that their safety is important to that particular department. This approach may also be an advantage for future grant proposals by the department, both external (federal) and internal (university-wide).
While it is clear that many departments gain more resources by virtue of their accomplishments in publishing papers, acquiring research grants, and ultimately in national rankings, there should be a level of appreciation and reward for practicing safe methods in doing research as well. It is the responsibility of the entire university community to promote safe environments for research. A measure of its impact would be to have safety as a measure of success of a department or college. Because resources are heavily contended throughout the university, the administration or leadership could set a standard that it expects its faculty to uphold in providing a safe environment for the many students that do research. As mentioned previously, the faculty salary program could have safety as a measure of success. Also, the resources used for start-up funds and other renovations could be allotted in part based on a department’s or unit’s safety practices.
In regard to creating a culture that is conducive to safety, there needs to be a nonthreatening atmosphere. This requires the principal investigator to be able to make observations and, subsequently, suggestions in a proactive manner. If the methods mentioned above regarding near-miss reporting are to make an impact, then it should not matter who reported the incident or who is the primary person involved. Instead, the focus should be on addressing the threat of the danger and eliminating it as quickly as possible. If a culture that is created for doing research safely is to be successful, this step is critical. It makes all those involved know that everyone is responsible and that no one should harm themselves in the important research they are doing.
Competitive academic programs of teaching and research require investments not only in buildings, equipment, and infrastructure, but also in excellent personnel (senior leaders, faculty, staff). These personnel need both scientific expertise and skills of leadership and management to establish and sustain a strong, positive safety culture. Provosts, deans, and chairs need to work with faculty who lead academic research laboratories
University of Minnesota Safety Program
One approach to changing academic laboratory safety culture is illustrated by the collaboration among the Department of Chemistry and the Department of Chemical Engineering & Materials Science at the University of Minnesota (Twin Cities, MN) and the Dow Chemical Company (Midland, MI).a Faculty, department chairs, graduate students, and postdoctoral associates from the departments partnered with EHS professionals to develop awareness and practices to foster safety. This “bottom-up” approach was developed and implemented by groups of volunteer laboratory safety officers (LSOs)—graduate students and postdoctoral associates—from each of the 59 research laboratories housed in the two departments. Organized as a “joint safety team” and charged with developing the program, the LSOs developed a safety approach focused on day-to-day attitudes, values, and practices.
Initial activities of the LSOs included surveys of safety attitudes and practices among faculty, staff, and students; tours of a wide variety of other laboratories in their home institution, as well as a visit to the Dow facilities in Midland. Each of these activities created opportunities for the LSOs to align their perceptions and expectations about safety practices with actual laboratory practices that were both inside and outside of their own areas of specialization, and in a research facility outside an academic department.
Supported by seed funding from the heads of the two departments, the LSOs developed a set of recommendations (CARE) focused on four areas: Compliance (roles and expectations), Awareness (signage, regular discussion of safety, e-mail updates), Resources (equipment, infrastructure, waste management), and Education (with a particular focus on lab-specific topics). Initial activities, again formulated by the LSOs, targeted areas such as peer tours of laboratories, personal protective equipment, a public website (http://www.jst.umn.edu/), and a lab cleanup week. The LSOs also instituted a practice of beginning group meetings and all departmental seminars with “safety moments” and, as an example, created an illustrative slide (Figure 4-1) that contains a safety topic relevant to the group or seminar topic, educational content, and one or more key citations. These “safety moments” are a striking example of strategies that make safety topics normal parts of academic culture and direct attention to the practice and science of safety.
The UMN program is exciting, but it remains to be seen how it will affect the overall safety of the department over time. More data are needed about whether
FIGURE 4-1 An example slide that may be developed for discussion around a safety topic.
SOURCE: Reprinted with permission from McGarry, K.A, et. al. Student Involvement in Improving the Culture of Safety in Academic Laboratories. J. Chem. Educ. 2013, 90(11): 1414-1417. Copyright 2013. American Chemical Society.
it is sustainable and scalable within its home institution, whether it will produce long-term changes in its home institution, or which of its features will be adaptable or critical to other departments and institutions. McGarry et al.b suggest several features that may be important to the program’s success and sustainability, but emphasize thatthe program’s characteristic “bottom-up” approach may be particularly important as it builds on the drive and future focus of the next generation of academic scientists.
aUniversity of Minnesota, Department of Chemistry. “Dow + U = lab safety.” Available at http://discover.umn.edu/news/vision-leadership/u-and-dow-chemical-team-lab-safety/. Accessed March 12, 2014. bMcGarry, K.A, et. al. Student Involvement in Improving the Culture of Safety in Academic Laboratories. J. Chem. Educ. 2013, 90(11): 1414-1417.
to identify the variety of leadership challenges they face and provide explicit tools and professional development opportunities that address these challenges. Useful tools and professional development include
- Resources for hazard analysis,22,23 which might include support for faculty to attend workshops on hazard analysis offered by groups such as the ACS Division of Chemical Health and Safety and expectations that they do so as part of their faculty role.
- Introductions to guidance and processes available from institutional human resources and mental health services, with examples of how to approach difficult personnel issues and of when and how individuals can be referred to the services.
- Development and mentorship programs focused on leadership skills. Resources such as Making the Right Moves,24At the Helm,25 and the American Association for the Advancement of Science Careers site26 contain practical information and guidance from experienced researchers; resources such as Training Scientists to Make the Right Moves27 provides guidance for institutions, and institutional programs can be tailored to specific challenges faced by faculty in a local environment.
- Institutional support for development and dissemination of lab-specific safety information, for expectations that faculty and trainees will regularly include EHS professionals in research planning, and for involvement of students and postdoctoral students in safety programs (indeed, the Minnesota program described in Box 4-2 suggests that institutions may need to empower and support trainees as leaders of departmental programs).
22 National Research Council. Prudent Practices in the Laboratory: Handling and Management of Chemical Hazards, Updated Version. The National Academies Press, Washington, DC, 2011.
23 American Chemical Society Committee on Chemical Safety. Identifying and Evaluating Hazards in Research Laboratories: Guidelines Developed by the Hazards Identification and Evaluation Task Force. American Chemical Society, Washington, DC, 2013.
24 Burroughs Wellcome Fund and Howard Hughes Medical Institute. Making the Right Moves: A Practical Guide to Scientifıc Management for Postdocs and New Faculty, 2nd Ed. Available at http://www.hhmi.org/sites/default/files/Educational%20Materials/Lab%20Management/Making%20the%20Right%20Moves/moves2.pdf.
25 Barker, K. At the Helm: Leading Your Laboratory, 2nd Ed. Cold Spring Harbor Laboratory Press, Long Island, NY, 2010.
26 Bea, R., I. Mitroff, D. Farber, Howard Foster, and K. H. Roberts. A new approach to risk: The implications of E3. Risk Management 2009; 11(1): 30-43.
27 Burroughs Wellcome Fund and Howard Hughes Medical Institute. Training Scientists to Make the Right Moves: A Practical Guide to Developing Programs in Scientifıc Management. 2006. Available at http://www.hhmi.org/sites/default/files/Educational%20Materials/Lab%20Management/Training%20Scientists/training-scientists-fulltext.pdf.
- Integration of safety work into promotion and recognition programs at all levels of the institution, so that the work required to advance academic laboratory safety becomes a “normal” part of performance expectations and of academic discourse (group meetings, seminars, dissertations, publications).
This page intentionally left blank.