BEHAVIOR-BASED SAFETY

Does BBS Work?

O
As behavioral safety nly a few years ago, many WHAT QUALITIES DEFINE BBS TODAY?
safety and health profession- As any complex effort to prevent
continues to gain visibility als knew little about behav- threats to people’s wellness, BBS contin-
ior-based safety (BBS)—a ues to evolve. In many settings, behav-
safety professionals must systematic approach to pro- ioral safety is now an important facet of
moting behavior supportive the total injury prevention package—
assess the efficacy of injury prevention. Today, integrated with areas such as ergonomics,
of the methods. From a the media and many professional organi- engineering, training, and occupational
zations are giving increased attention to health and safety management.
conservative perspective, the topic. Safety professionals want to To illustrate this integrated aspect, an
learn more about the way this strategy investigation of a behavior-based strate-
the question, “Does behavior- operates and what it accomplishes—in gy to prevent back injuries included
based safety work to decrease other words, does it really work? pinpointing of specific skills; training
That may not be the best question to maintenance workers in the ergonomics
incidents in every case?” pose, however. This article briefly de- of moving materials safely; bio-feedback
scribes behavioral safety systems and on muscle tension; and verbal feedback
can be difficult to answer. their fundamental elements. It discusses and rewards for progress in practicing
several (of myriad) factors that potential- those skills safely (Rosado). In another
However, sufficient data ly influence the operation of these meth- illustrative case, McCann and Sulzer-
are available to demonstrate ods in a given situation. Facts about its Azaroff selected a set of behavioral tar-
impact on incidence rates are also re- gets for computer-terminal operators
that the approach can viewed to help readers judge the effec- seeking to reduce their risk of injury;
tiveness of BBS and the appropriateness these targets were based on recommen-
accomplish that end. of posing the question in this fashion. dations of an occupational health physi-
This article also examines FUNDAMENTAL ELEMENTS
cian (277+).
Safety management was a focus area
a related question: “How can Although specific behavioral safety of a program implemented in a telecom-
systems may vary in form and complexi- munications plant (Sulzer-Azaroff, Loaf-
safety professionals make ty, at their most basic level they share sev- mann, et al 99+). Managerial pinpoints
eral common elements: included providing weekly safety per-
behavior-based safety 1) Identify (or target) behaviors that formance updates to the safety director;
work better?” Depending on impact safety. participating in the awards program; pro-
2) Define these behaviors precisely viding reinforcement to supervisors of
the purpose of the enough to measure them reliably. departments that met or exceeded goals;
3) Develop and implement mechan- helping to develop pinpoints; and en-
intervention, several isms for measuring those behaviors in couraging safety suggestions. Safety-
suggestions are offered. order to determine their current status related behaviors include not only those
and set reasonable goals. of workers, but also those of supervisors,
4) Provide feedback. managers and others within the system
5) Reinforce progress. whose support is crucial.
AMERICAN SOCIETY OF SAFETY ENGINEERS JULY 2000 19
 WHAT DOES “DOES BBS WORK” MEAN? and delivering feedback and reinforce- Again, researchers may define these
Beyond its basic elements, BBS can vary ment may be applied differently from one components differently. Goals may be set
from time to time and place to place. program to another, as may supplemen- on a participative or assigned basis (Fellner
Depending on an organization’s needs, tal elements. These variations can affect and Sulzer-Azaroff 3+). Many do not spec-
resources and objectives, each system will the outcome. (It should be noted that ify how goals were set (Cohen and Jensen
have uniquely customized features. So, unless each individual element has been 125+). Some install highly structured sys-
when potential users ask, “Does BBS carefully isolated from the rest, the extent tems to support problem solving (Krause),
work?” it is difficult to provide a definitive of any one facet’s influence generally while others provide less-formal support
answer. Several variables account for the remains ambiguous. In practice, these (Laitinen, et al 35+).
diversity of results one might encounter. variations tend to be combined and
viewed as elements of the total interven- Variations in the Role of System Coordinator
Target Behaviors tion “package.”) The role of the individual who designs
Data on efficacy are influenced by the Feedback strategies are one such and coordinates the system may influ-
behaviors targeted for change. An inter- example. Behavior-analysis textbooks ence results as well. To the applied behav-
vention’s ultimate value can only be as offer diverse definitions for at least 12 dif- ioral researcher, the “does it work”
good as the precise behaviors selected. ferent types of feedback (Sulzer-Azaroff question usually means, “Can a function-
These behaviors must closely relate to and Mayer; J. Cooper, et al). The process al (cause-and-effect) relation be demon-
improving well-being—in other words, of implementing feedback strategies can strated between a particular set of BBS
when practiced consistently, they will also vary. For example, Balcazar, et al variables and those that make up safety
reduce risks and improve the physical identified six characteristics of feedback performance?” Relying primarily on
well-being of employees and manage- that varied widely between studies (65+). measured changes in rates of safe behav-
ment. Attempting to change what people McAfee and Winn commented on these ior and perhaps incidents, this researcher
say, think or feel is not the same as chang- fluctuations when, in a set of studies they approaches the question by tightly con-
ing what they do. In addition, the de- reviewed, they found feedback assuming trolling those variables thought to affect
scription of a safe practice must be many different forms (7+). the targeted outcome measures.
objective, correct and complete (i.e., One can also cite variations in how In fact, however, many of the varia-
valid). No matter how much individuals agents define, supplement and imple- tions in intervention strategies have more
improve the way they perform a task, if ment these components. Consider this to do with achieving experimental con-
key components of that task are irrele- brief list of potential variations in the trol (e.g., establishing inter-rater reliabili-
vant, omitted or wrong, materials han- feedback-delivery process. ty, using the researcher to observe and
dlers will remain in danger. •Communicate general approval or provide feedback) than anything else. If
disapproval in the form of a gesture, a spo- the effect of the well-controlled interven-
Purpose of Intervention ken “good,” “okay” or “safe” for one or tion can be shown by using valid, reliable
When setting behavioral targets, one more safe performance targets, or share outcome measures, the researcher con-
must consider the objective of the inter- more-specific information in words or cludes, “That intervention had a demon-
vention. The purpose of an endeavor can visually (e.g., graph) to describe what strable effect on safety performance” (in
influence the outcome as well as the man- about the performance was safe or unsafe. other words, BBS worked in that case).
ner in which findings are interpreted and •Comment on positive or negative One example of this is a study
reported. The form of evaluation can actions or offer constructive suggestions designed to explore the impact of a self-
depend on whether the intervention is for future behavior, or some combination. generated and/or an external feedback
part of a controlled laboratory investiga- •Offer praise, recognition and/or package on changes in the body postures
tion, a systematic field study or a local rewards to individuals or groups. among a small group of volunteer com-
assessment. •Provide one-on-one feedback, or ask puter terminal operators (McCann and
Scientific research attempts unam- peers, supervisors, subordinates, safety Sulzer-Azaroff 277+). Following the suc-
biguously to sort out competing explana- personnel, managers or external sources cessful demonstration and follow-up
tions for results of given interventions. to deliver it. assessments, these researchers concluded
For example, an investigator may study •Offer feedback publicly or privately that their particular BBS package worked
the differential influence of massed ver- to individuals, groups or both. for this new class of behaviors. Insuffi-
sus spaced practice as a method of pro- •Provide feedback verbally, in written cient resources prevented a wide-scale
moting rapid improvement in the way form (e.g., graph or digitally), via tangi- implementation of the methods to test the
healthcare workers lift and transfer ble reward or some combination. finding’s validity and generality.
patients (Alavosius and Sulzer-Azaroff •Deliver feedback according to a fixed Researchers also want to be as certain
151+). Program evaluators, on the other or variable schedule (often, seldom or as possible that their purported “treat-
hand, may ask whether specific varieties somewhere in-between). ments” actually occur as planned. Conse-
of BBS are effective tools for improving •Combine feedback with goal-setting quently, they or trained assistants deliver
safe practices or reducing injury rates (interim, final or both). intervention components instead of inter-
within their organizations (Alavosius in •Precede feedback by a baseline phase nal organizational personnel (Alavosius
press; Krause, et al 1+). that includes all elements except feedback. and Sulzer-Azaroff 151+; Lingard and
Of course, both researchers and practi- Many researchers and practitioners Rowlinson 243+; Ludwig and Geller 253+;
tioners are concerned that results be as also build-in components beyond identi- Reber, et al 51+; Saarela 177+). This causes
clear as possible. However, the scientist fication, observation and feedback. They one to question whether the organization
takes far greater care to ensure that com- may incorporate goal-setting (Austin, et could obtain similar results by applying
peting explanations can be rejected, while al 49+; M. Cooper, et al 219+; Fellner and those methods independently.
the practitioner is usually more concerned Sulzer-Azaroff 7+; Reber and Wallin To the safety professional or business
with effectiveness and cost-efficiency. 544+; Reber, et al 51+) or problem-solving manager, however, the “does it work”
(Fellner and Sulzer-Azaroff 7+; Killimett question can mean something entirely dif-
Variations in Application 209+; Krause; Laitinen, et al 35+; Walters ferent. Typically, publishing their results is
Basic components, such as identifying 34+). Often, training is a key aspect as not a high priority. Nor do they have time
and defining safe practices, observing, well (Sulzer-Azaroff, Fox, et al). to wait for scientific confirmation of the
20 PROFESSIONAL SAFETY
 In many settings, behavioral safety is now an important facet
of the total injury prevention package—integrated with areas such as
ergonomics, training, and occupational health and safety management.
results. Immediate decisions must be the effects of performing less-expedi- its reliability. It is also possible that some
made to address human and cost issues. tiously but more safely. Others may seek inconclusive or negative incidence/injury
Therefore, managers must make the change in safety climate or culture. data simply were not submitted or accept-
best judgment based on the information Individuals with power or control of ed for publication. [It should be noted that
available. Their questions often encom- reinforcers will focus on those outcomes in most fields involving intervention (e.g.,
pass a wider domain of practical con- of concern to them. This fact, alone, will clinical health, social service, psychothera-
cerns: Will it work in this organization, in influence results because what gets mon- py), there is potential bias because poor,
this work culture? Will all levels of the itored gets reinforced and what gets rein- negative or non-results may not be submit-
workforce buy into the program’s con- forced is repeated more frequently ted for publication—or when they are,
cepts? Are the resources available to (Komaki 270+). The opposite is also they are often rejected.]
make this work over the long term? If true—those elements monitored less
employees buy-in and resources are carefully or ignored tend to diminish. ARTICLE ANALYSIS
available, should the firm expect both Of the 83 data-based evaluations of
immediate and continuous improve- Variations in Support Structures behavioral safety programs, the authors
ment—not only in safe performance, but The organizational support structures identified 33 that reported data on
also in fewer incidents and injuries? in place also influence long-term results. changes in incidence rates. Table 1 pro-
A business-oriented intervention typi- Many experts agree that the following fac- vides details of these studies.
cally calls on the organization’s work- tors can affect the durability of any BBS Two researchers then independently
force to deliver system components, effort: clear, visible, ongoing senior execu- categorized changes in incidence rates in
frequently with each level playing a key tive commitment; institutionalized mech- each study. In all but five cases, entries
role. In these cases, individuals who anisms such as actively participating matched, producing an agreement score of
champion the program provide the nec- safety/steering committees; data collec- 85.7 percent [(agreements)/(agreements +
essary training and guidance (Alavosius tion (e.g., minutes of meetings; observing disagreements) x 100]. Three disagree-
in press; Krause, et al 1+; Krause). That and recording numbers of observations ments occurred because one reviewer con-
person or group is then challenged to conducted, feedback charts up-to-date); verted the reported incidence frequencies
maintain the system’s integrity—to verify feedback routines supporting system to percentage of changes in incidence rates,
that it is transpiring as planned. maintenance; encouragement of involve- while the other reviewer did not. In addi-
ment by all personnel (Sulzer-Azaroff and tion, in three cases, one reviewer over-
Variations in Priorities Lischeid 31+). looked a portion of the reported data.
While the universal goal is to reduce Therefore, to portray the results as accu-
incident and severity rates, other specific WHAT ABOUT RESULTS? rately as possible, the disagreements were
outcomes can be affected, depending on What would many consider the bot- discussed and ultimately reclassified.
the role of the person/group within the tom-line measure when attempting to
organization. answer the “does it work” question? In ANALYSIS OF RESULTS
For example, executives, stockholders most cases, incidence rates. Therefore, the Participants & Settings
and directors focus on the bottom line: authors surveyed literature on BBS to dis- The number of participants ranged
How much does it cost and what is the cover what its impact has been on “acci- from five (Haynes, et al 407+; Larson, et al
return? BBS data are beginning to show dents” or incidence and injury rates. 571+) to 39,664 that one organization
short- and long-term savings due to The Network on Behavioral Safety of tracked (Krause, et al 1+). The number of
reduced injuries and their associated costs the Assn. for Behavior Analysis main- sites evaluated ranged from single loca-
(insurance rates, direct medical, equip- tains an exhaustive list of reports on this tions to 73 (Krause, et al 1+). Settings
ment replacement and repair, down-time topic. Eighty-three of those listed de- included construction sites, grocery
and lost-time costs). scribe case demonstrations or experi- distributors, electrical and gas utilities,
Of course, figures can diverge, de- ments that contained hard numerical manufacturing plants, mines, police
pending on methods used to calculate data. “Promote safe performance” was departments, railroads, shipyards and
costs. For example, Veltri offered a direct- the main focus of most of these studies. transit systems. In addition, although
cost impact model designed to demon- Many reports omitted correlated data most studies were conducted in the U.S.,
strate how reducing accidents can lower on accident/injury rates, the measure of some involved sites in Chile, Cuba,
the degree of operating leverage (67+). concern in this article. Several factors may Finland, Hong Kong, Spain and the U.K.
Sulzer-Azaroff, Loafmann, et al used a have driven these omissions: 1) the investi-
less-formal, more-conservative basis to gation or implementation was conducted Injury Outcomes
estimate that in three small departments over too short a time period to produce Of the 33 articles reviewed, 32 report-
within a large manufacturing plant, meaningful differences; 2) the number of ed reductions in injuries. In many cases,
$55,000 was saved due to reduced injury participants was too small to generate however, the reporting format differed.
rates over a six-month period (99+). meaningful injury data (McCann and Some listed numbers of lost-workdays;
Production and quality managers may Sulzer-Azaroff 277+); 3) the base rate of others, numbers of accidents. In addition,
be concerned with production rates and injuries was too low to be meaningful some accident rates were calculated on
quality, the amount of time employees (Sulzer-Azaroff 11+); and 4) data was kept the basis of hours worked or miles dri-
would be away from work for training, confidential for business purposes or ven. (In the future, researchers should be
observing and providing feedback, and because researchers had misgivings about encouraged to report rates—not just raw
JULY 2000 21
 TABLE 1 BBS Studies Reporting Impact on Injury Rates
NUMBER OF
STUDY AUTHOR(S) SETTING REDUCTION IN ACCIDENT/INCIDENT RATES
PARTICIPANTS
M. Alavosius (in press) 5 – 500 50 small companies Lost-workdays per 100 workers: 184 pre-intervention; 111
during; 84 post-intervention (six months) and 58 (12 months).
M.D. Cooper, et al 540 Construction industry From 6.33 prior to 3.88 at end; from 3.3 to 0.56 on checklisted
(1994) items.
D.J. Fellner and 158 Paper mill Significant difference between pre- and during-feedback—from
B. Sulzer-Azaroff (1984) 6.9 percent to 4.9 percent.
F. Fiedler (1987) 500 Mine Baseline = 226 percent; follow-up two percent over industry
average.
D.K. Fox, et al (1987) 1,754 Coal mine Range: 15 to 32 percent.
D. Harshbarger and a) 100 a) Bedding Lost-time accidents a) 95 percent; b) 87 percent.
T. Rose (1991) b) 350 – 400 b) Footwear
R.S. Haynes, et al (1982) 100 Urban transit 24.9 percent.
B.S. Karan and R.E. Not reported Vehicular and industrial 2.2 percent and 4.0 percent.
Kopelman (1987)
J.L. Komaki, et al (1978) 38 Food manufacturing plant Injuries fell to “. . . less than 10 lost-time accidents per million
hours worked, a relatively low number” (pg. 441).
J.L. Komaki, et al (1980) 55 Vehicle maintenance Decline from 3.0 lost-time injury rate per month preceding to 0.4
during and 1.8 following intervention.
T.R. Krause, et al (1999) 51 to 3,000 per 73 facilities participating Year 1: 26 percent.
site (39,664 for up to five years Year 2: 42 percent.
across 73 Year 3: 50 percent.
sites) Year 4: 60 percent.
Year 5: 69 percent.
H. Laitinen, et al (1998) 300 Engineering workshop 46 percent reduction in absenteeism.

B. Loafmann (1998) Not reported Utility company Treatment group about 78 percent; control group had a
50-percent increase.
L. Lopez-Mena and 914 Forestry (2) and 62.8 percent; maintained for three years.
J.V. Antidrian (1990) cement factory (1)
L. Lopez-Mena and 41 Electrical distribution 84.9 percent in one setting; 60.8 percent in a second setting.
R. Baynes (1988) system
L. Lopez-Mena, et al 191 Electrical energy 34.3 percent.
(1988) distribution system
M. Mattila and M. 100 Building construction Accident rate per 100 workers at control site higher during (166)
Hyödynmaa (1988) and after (55) than experimental site—94 and 47, respectively.
T. McSween (1995) Not reported Gas pipeline company 35 percent lost-time accidents.

T. McSween (1995) Not reported Chemical company From four to zero the next 18 months.
(union-coordinated)
R. Montero (1996) Not reported Industry (general) “Rate dropped almost to zero.”
M. Näsänen and 32 Shipyard 50- to 75-percent reduction in accidents.
J. Saari (1987)
D. Petersen (1984) Not reported Railroad “Experimental groups had fewer injuries than control [groups].”
R.A. Reber, et al (1984) 105 Farm machinery 53.83 percent.
manufacturing
R.A. Reber, et al (1990) 44 Manufacturing 50 percent.
K.L. Saarela (1989) 2,800 Shipyard Modest, non-significant reduction in accident frequency. This
intervention involved a poster campaign, not a full behavioral
program, and feedback to supervisors.
K.L. Saarela (1990) 24 Shipyard 20 percent during; about 40 percent after.
K.L. Saarela, et al (1989) 58 Shipyards No significant differences (poster campaign; general subject
feedback).
K.L. Saarela (in press) >900 Shipyard 60+ percent.
J. Saari and M. Näsänen 24 Shipyard 25-percent reduction in accidents; 30-percent reduction in
(1989) injuries.
R. Schwartz (1989) 110 Grocery distribution 39.4 percent.
workers
M. Smith, et al (1978) 44 Shipyard Average decrease in eye injuries of 7.48 per 100 workers; control
group average reduction of 1.16.
B. Sulzer-Azaroff, et al 140 Paper mill From 19 recordable incidents during baseline to two after
(1986) feedback given for three behaviors.
B. Sulzer-Azaroff, et al 225 Telecommunications parts Number of OSHA recordables dropped by 17 from prior to
(1990) manufacturing plant during intervention; lost-time from 14 to 1.

22 PROFESSIONAL SAFETY
numbers—as well as the basis for the rate promise for attaining improvement in be- growing interest in this technology is justi-
calculation.) Differences in experimental haviors of concern within a given context; fied by the gains many companies have
designs also affected data reports. Those •data which demonstrate that the reportedly achieved. Most essential to
using group comparisons reported infor- experiment was conducted as described. those gains are the training, organizational,
mation on statistical significance, while However, when the purpose is to managerial, follow-through and other fac-
those using repeated measure or inten- implement a process in the most-effec- tors discussed. To ensure that the system
sive designs (e.g., multiple baselines) tive, efficient way within a given organ- works most effectively, these factors, along
relied, according to convention, primarily ization, one must look beyond the basics with the findings generated from ongoing
on graphic presentations. To simplify the of the technology itself to organizational and future research, should be considered
descriptive analysis for this article, when- issues that impact how the technology is when designing any BBS system. 䡲
ever feasible, average changes were cal- implemented. Are mechanisms in place
culated from pre- to during and/or to sustain longevity? Is the method com- REFERENCES
post-intervention. patible with the site implementing it? Alavosius, M. “Behavioral Approaches to
To illustrate the difficulties encoun- Does a reasonable balance exist between Organizational Safety.” In The Handbook of
tered in drawing unambiguous conclu- objectives and resources? Applied Behavior Analysis, J. Austin and J.E.
sions about the efficacy of a given BBS Others share a similar perspective. Carr, eds. Reno, NV: Context Press (in press).
intervention, consider this example of After evaluating the outcomes of 41 Alavosius, M. and B. Sulzer-Azaroff.
Saarela’s (in press) thwarted efforts to efficacy studies of safety training, John- “Acquisition and Maintenance of Health-
maintain tight experimental controls dur- ston, et al concluded that effective training care Routines as a Function of Feedback
ing an intervention at a shipyard. programs shared four common character- Density.” Journal of Applied Behavior Analy-
sis. 23(1990): 151-162.
Following a two-hour training seminar, istics: 1) needs assessment; 2) program
Austin, J., et al. “Using Feedback and
safety teams from each production depart- development; 3) goal setting; and 4) knowl-
Reinforcement to Improve the Performance
ment identified departmental safety prob- edge of results through feedback (147+). and Safety of a Roofing Crew.” Journal of
lems, then set goals, solved problems, In a related analysis, Hidley identified Organizational Behavior Management. 16(1996):
implemented improvements, monitored seven factors critical to the success of BBS 49-75.
results and provided feedback. According implementations: Balcazar, F., et al. “A Critical, Objective
to the original research plan, only three 1) Use a process blueprint character- Review of Performance Feedback.” Journal
departments were to be involved, with ized by structure and rigorous implemen- of Organizational Behavior Management.
others joining later; however, when those tation, while allowing flexibility to adapt 7(1985/6): 65-89.
not included saw the results, they de- technologies to site-specific needs. Cohen, H.H. and R.C. Jensen. “Measuring
manded immediate participation. 2) Create buy-in through communica- the Effectiveness of an Industrial Lift Truck
The success of those tactics was under- tion. Highest levels of leadership, manage- Safety Training Program.” Journal of Safety
scored when results were analyzed. By ment and labor must be committed to Research. 15(1984): 125-135.
the time the 900 workers in all 10 depart- improvement—in both actions and words. Cooper, J.O., et al. Applied Behavior
ments were actively participating in the 3) Demonstrate leadership and active Analysis. Columbus, OH: Merrill, 1987.
process, the shipyard’s accident rate was support for the change process. Cooper, M.D., et al. “Reducing Accidents
more than halved. Subsequently, the pro- 4) Ensure implementation-team com- Using Goal Setting and Feedback: A Field
gram was continued under the coordina- petence through training and follow-up. Study.” Journal of Occupational and Organiza-
tion of the shipyard’s safety personnel. 5) Provide action-oriented skills train- tional Psychology. 67(1994): 219-240.
Beyond experimental-design issues, ing in a structured, safe environment in Fellner, D.J. and B. Sulzer-Azaroff.
“Increasing Industrial Safety Practices and
other cautions must be considered. For which trainees can practice new skills.
Conditions Through Posted Feedback.”
example, how much room remained for 6) Use feedback data based on safe
Journal of Safety Research. 15(1984): 7-21.
improvement? Sites with high injury behavior under the control of employees Fellner, D.J. and B. Sulzer-Azaroff.
rates have a larger potential for improve- to remove barriers. “Occupational Safety: Assessing the Impact
ment than those with low rates. One must 7) Ensure that adequate technical assis- of Adding Assigned or Participative Goal-
also consider problems in the reliability tance (internal or external) is available Setting.” Journal of Organizational Behavior
of reporting of incidence rates. from individuals who make practical, fea- Management. 7(1986): 3-24.
sible and well-considered recommenda- Fiedler, F. “Structured Management
THE REAL CHALLENGE: tions for process success (30+). Training in Underground Mining: Five
HOW TO MAKE BBS WORK BETTER A similar list describing a quality Years Later.” Proceedings of Bureau of Mines
Despite the limitations noted, the data behavioral safety system was detailed in a Technology Transfer Seminar. Pittsburgh:
gathered support the conclusion that BBS survey of behavioral safety specialists Bureau of Mines, 1987. 149-153.
systems appear to have helped reduce (Sulzer-Azaroff and Lischeid 31+). That list Fox, D.K., et al. “The Long-Term Effects
injuries on many occasions. In light of included: well-defined, correct, relevant of a Token Economy on Safety Performance
that finding, the next question is, “Given target behaviors; worker participation; in Open-Pit Mining.” Journal of Applied
these particular purposes and circum- managerial involvement; observational Behavior Analysis. 20(1987): 215-224.
stances, how can this organization make data collection; training/education; posi- Harshbarger, D. and T. Rose. “New
BBS work as well as possible?” tive reinforcement; interventions based on Possibilities in Safety Performance and the
The answer depends on the main pur- sound behavioral technology; behavioral Control of Workers’ Compensation Costs.”
pose of the intervention. If it is research Journal of Occupational Rehabilitation. 1(1991):
feedback; and organizational systems 133-143.
seeking to demonstrate a clear causal structured to support the effort. Haynes, R.S., et al. “Reducing Accident
relation between a given class of behav- Rates with Organizational Behavior Modifi-
iors and a particular change method, then CONCLUSION cation.” Academy of Management Journal.
finely honed, tightly controlled experi- The studies evaluated suggest that 25(1982): 407-416.
ments must be designed. These include: despite some reservations about how accu- Hidley, J.H. “Critical Success Factors for
•valid, reliable measures of perfor- rately the published literature reflects real- Behavior-Based Safety: Avoiding Common
mance and side effects, under baseline, ity, in many cases, incidence rates have Pitfalls and Achieving Real Gains.” Profes-
treatment and afterward; been reported to decline following imple- sional Safety. July 1998: 30-34.
•change strategies that hold the most mentation of BBS system. Presumably the Johnston, J.J., et al. “The Efficacy of
JULY 2000 23
Training for Occupational Injury Control.” moting Job Safety on Building: An Smith, M., et al. “Behavioral Modifica-
Occupational Medicine: State of the Art Reviews. Experiment on the Behavior Analysis tion Applied to Occupational Safety.”
9(1994): 147-158. Approach.” Journal of Occupational Acci- Journal of Safety Research. 10(1978): 87-88.
Karan, B.S. and R.E. Kopelman. “The dents. 9(1988): 255-267. Sulzer-Azaroff, B. “Behavioral Ecology
Effects of Objective Feedback on Vehicular McAfee, R.B. and A.R. Winn. “The Use and Accident Prevention.” Journal of Organi-
and Industrial Accidents: A Field Experi- of Incentives/Feedback to Enhance Work zational Behavior Management. 2(1978): 11-44.
ment Using Outcome Feedback.” Journal Place Safety: A Critique of the Literature.” Sulzer-Azaroff, B. Who Killed My Daddy?
of Organizational Behavior Management. Journal of Safety Research. 20(1989): 7-19. A Behavioral Safety Fable. Concord, MA: Cam-
8(1987): 45-56. McCann, K.B. and B. Sulzer-Azaroff. bridge Center for Behavioral Studies, 1998.
Killimett, P.T. “Completing the Loop: “Cumulative Trauma Disorders: Behavioral Sulzer-Azaroff, B., C. Fox, et al. “Feed-
Action Planning for Continuous Safety Injury Prevention at Work.” Journal of back and Safety: Involving Workers.”
Improvement.” Tappi Journal. 75(1992): 209- Applied Behavioral Science. 32(1996): 277-291. Final report to National Institute of Occupa-
211. McSween, T. The Behavior-Based Safety tional Safety and Health. Grant #1 RO1-
Komaki, J.L. “Toward Effective Super- Process. New York: Van Nostrand Reinhold, OHO1928-01, 1986.
vision: An Operant Analysis and Compari- 1995. Sulzer-Azaroff, B. and W.E. Lischeid.
son of Managers at Work.” Journal of Applied Montero, R. “How Much ‘Participation’ “Assessing the Quality of Behavioral Safety
Psychology. 71(1986): 270-279. Does a Behavior Modification Program Need Initiatives.” Professional Safety. April 1999:
Komaki, J.L., et al. “Effect of Training for Improving Safety?” Presentation at the 31-36.
and Feedback: Component Analysis of a International Conference on Occupational Sulzer-Azaroff, B., B. Loafmann, et al.
Behavioral Safety Program.” Journal of Health, Estocolmo, Suecia, Sept. 1996. “Improving Occupational Safety in a Large
Applied Psychology. 65(1980): 261-270. Näsänan, M. and J. Saari. “Effects of Industrial Plant: A Systematic Replication.”
Krause, T.R. The Behavior-Based Safety Positive Feedback on Housekeeping and Journal of Organizational Behavior Manage-
Process: Managing Involvement for an Injury- Accidents at a Shipyard.” Journal of Occupa- ment. 11(1990): 99-120.
Free Culture. 2nd ed. New York: Van tional Accidents. 8(1987): 237-250. Sulzer-Azaroff, B. and G.R. Mayer. Be-
Nostrand Reinhold, 1997. O’Toole, M. “The Relationship Between havior Analysis for Lasting Change. Fort
Krause, T.R., et al. “Long-Term Evalua- Employees Perceptions of Safety and Worth: Harcourt, Brace, Jovanovich, 1991.
tion of a Behavior-Based Method for OSHA Lost-Time Injury Rates.” Funded Sulzer-Azaroff, B. and K.E. McCann.
Improving Safety Performance: A Meta- project in progress (2000). “Promoting Health and Safety on the Job.”
Analysis of 73 Interrupted Time-Series Petersen, D. “An Experiment in Positive In Performance Management: Behavior Analy-
Replications.” Safety Science. 32(1999): 1-18. Reinforcement.” Professional Safety. May sis in Organizations, C.M. Johnson, W. Red-
Laitinen, H., et al. “Improving Physical 1984: 30-35. mon and T. Mawhinney, eds. New York:
and Psycho-Social Working Conditions Reber, R.A., et al. “Improving Safety Per- Springer Press (in press).
Through a Participatory Ergonomic Proc- formance with Goal Setting and Feedback.” Veltri, A. “An Accident Cost Impact
ess.” International Journal of Industrial Ergo- Human Performance. 3(1990): 51-61. Model: The Direct Cost Component.”
nomics. 21(1998): 35-45. Reber, R.A. and J.A. Wallin. “The Effects Journal of Safety Research. 21(1990): 67-73.
Larson, L.D., et al. “Reduction of Police of Training, Goal Setting and Knowledge of Walters, H.A. “Identifying & Removing
Vehicle Accidents Through Mechanically Results on Safe Behavior: A Component Barriers to Safe Behaviors.” Professional
Aided Supervision.” Journal of Applied Analysis.” Academy of Management Journal. Safety. Jan. 1998: 34-36.
Behavior Analysis. 13(1980): 571-581. 27(1984): 544-560.
Lingard, H. and S. Rowlinson. “Behav- Reber, R.A., et al. “Reducing Industrial
ior-Based Safety Management in Hong Accidents: A Behavioral Experiment.” Beth Sulzer-Azaroff, Ph.D., is a professor emeri-
Kong’s Construction Industry.” Journal of Industrial Relations. 23(1984): 119-125. tus at the University of Massachusetts and an
adjunct professor at Florida International Univer-
Safety Research. 28(1997): 243-256. Rosado, W.T. “The Application of Be-
sity, Miami. In addition, as a performance manage-
Loafmann, B. “Behavior-Based Safety: havioral Principles to Promote Safe Practices ment consultant, she develops and writes about
Power & Pitfalls.” Professional Safety. Aug. and Reduce Accidents and Injuries.” Unpub- behavioral systems and how they promote perform-
1998: 20-23. lished master’s thesis. Miami: Florida Inter- ance on the job. Her Ph.D. was earned at the
Lopez-Mena, L. and J.V. Antidrian. national University, 1999. University of Minnesota.
“Applicaciones Del Refuerzo Positivo A La Saarela, K.L. “A Poster Campaign for
Reduccion De Accidents En El Trabajo. Improving Safety on Shipyard Scaffolds.” John Austin, Ph.D., is an assistant professor in the
[Applications of Positive Reinforcement to Journal of Safety Research. 20(1989): 177-185. Dept. of Psychology at Western Michigan Univer-
Reduction of Work Accidents.]” Revista Saarela, K.L. “An Intervention Program sity in Kalamazoo. At the university, he teaches
Latinoamericana de Psicologia. 22(1990): 357- Utilizing Small Groups: A Comparative organizational and behavioral psychology, and
371. Study.” Journal of Safety Research. 21(1990): behavior-based safety courses. Austin is also co-edi-
Lopez-Mena, L. and R. Baynes. “Preven- 149-156. tor of the Journal of Organizational Behavior
Saarela, K.L. “An Intervention Study for Management and serves on the editorial board of
cion de Riesgos en el Trabojo: Effectos de la
Improving Safety at a Shipyard.” In People at the Journal of Applied Behavior Analysis. Austin
Retroalimentacion y La Participacion. [Pre-
Work, H. Laitinen and J. Saari, eds. Finland: holds a B.A. from the University of Notre Dame and
vention of Work Accidents: Effects of Feed- an M.S. and Ph.D. from Florida State University.
back and Participation.]” Avances en Psicologia Finnish Dept. of Occupational Health, Div. of
Clinica Latinamericana. 6(1988): 53-65. Occupational Safety (in press).
Lopez-Mena, L., et al. “Beneficios Econó- Saarela, K.L., et al. “The Effects of an
micos Pbtenidos con Un Programa Conduct- Information Safety Campaign in the Ship-
ual en Seguridad Del Trabajo. [Economic building Industry.” Journal of Occupational READER FEEDBACK
Benefits Obtained with a Behavioral Work Accidents. 10(1989): 255-266.
Safety Program.]” Psicologia del Trabajo y de Saari, J. And M. Näsänen. “The Effect of Did you find this article interesting
Las Organizaciones. 4(1988): 74-86. Positive Feedback on Industrial House- and useful? Circle the corresponding
Ludwig, T.D. and E.S. Geller. “Assigned keeping and Accidents: A Long-Term Study number on the reader service card.
Versus Participative Goal-Setting and Re- at a Shipyard.” International Journal of Indus-
sponse Generalization: Managing Injury trial Ergonomics. 4(1989): 201-211. YES 28
Control Among Professional Pizza Deliver- Schwartz, R. “Cognition and Learning in SOMEWHAT 29
ers.” Journal of Applied Psychology. 82(1997): Industrial Accident Injury Prevention: An
253-261. Occupational Therapy Perspective.” Health
NO 30
Mattila, M. and M. Hyöynmaa. “Pro- Promotion Prevention Programs. 6(1989): 67-85.
24 PROFESSIONAL SAFETY