Atlanta, GA: U.S. Department of Health and Human Services, Public Health Service, Centers for Disease Control and Prevention, National Institute for Occupational Safety and Health, R01-CCR-413067, 2000 Jun; :1-293
The goal of this study was to develop and test the effectiveness of a safety training intervention for employees, supervisors, and owners of small construction firms (= 10 employees) in Kentucky. The project involved a partnership between the Kentucky Injury Prevention and Research Center (KIPRC) and Kentucky Employers' Mutual Insurance (KEMI). The project was in two phases over a three year interval. Phase 1 (Year 1) consisted of intervention development and authentication activities. Phase 2 (Years 2 and 3) consisted of intervention dissemination and evaluation. The intervention consisted of a series of six reality-based latent-images, narrative simulation exercises targeted to the prevention of falls or back injuries. They were designed to emphasize the economic impact of injuries and benefits of individual and organizational injury prevention strategies. The simulations, developed in Year 1, were based upon findings from a series of focus groups of small construction company owners and employees held in each of eight regions in Kentucky. In each year of Phase 2, three of the simulations were administered. The study design used to test the intervention in Phase 2 was quasi-experimental. In each year of Phase 2, intervention participants took pre-test (PRE) and immediate post-test (IPT) measures concurrently with the simulations. Delayed post-test (DPT) and concurrent retrospective pre-test (RPT) measures were administered to intervention participants three to four months after the intervention. In Year 2, the intervention group was subdivided into participants who attended a group administration meeting and participants who completed simulations at home. Control group subjects took concurrent PRE and RPT measures. Hypotheses tested were: (1) workers who participate in a training program consisting of six simulations (3 for fall prevention and 3 for back injury prevention) experience fewer injuries and submit fewer worker compensation claims than those workers who do not participate; and (2) worker compensation claims filed by Kentucky Employers Mutual Insurance (KEMI) companies that participate in a simulation training program differ significantly from claims submitted by non-participating KEMI companies and companies not insured through KEMI. Additional research aims included: (3) examining the differences in safety climate, safety attitude, and self-efficacy among the home-administration and control groups; and (4) exploring reasons for non-participation in a safety training program. The intervention group consisted of owner-operators, supervisory personnel, and employees. In Year 2, the treatment was stratified into home or group administration, but participation in group sessions was poor (n = 38). Therefore, in Year 3, group administration ceased, while home administration was maintained (n = 260 for Years 2 and 3). In both years, a no-treatment control group (n = 95 for Years 2 and 3) consisted entirely of owner-operators or supervisory personnel. Participant and control owners/supervisors were highly experienced and employed full time. The mean years of construction experience among non-supervisory employees was at least 9 years in both years of Phase 2. Control subjects were somewhat more experienced in construction and put in longer hours than owners/supervisors in the intervention group, but both groups were highly experienced in construction (means = 18 years) and worked full-time. There were no differences between intervention participants and controls on a pre-test measure of safety climate in either year. In both Year 2 and Year 3, simulation performance scores were relatively high (overall means = 79% of maximum), and simulation evaluations were generally favorable (overall means = 82% of maximum). In Year 2, there was no significant difference between PRE and DPT scores on the safety climate measure. However, scores on the RPT (a mean of 2 items) were significantly lower than the scores on the corresponding PRE and DPT items. This suggested a possible response-shift (i.e., the intervention may have influenced subjective calibration judgments with respect to the measure being used). In Year 3, using a revised and more reliable safety climate measure (10 items; alpha = .89) there was no significant difference between PRE and RPT or PRE and DPT measures in the intervention group. In terms of claims experience, between the inception of KEMI in 1995 and May 1999, 61 claims of any type occurred in 19 of 147 companies (13%) who participated in the intervention in Year 2 or Year 3. Of those 61 claims, 7 (from 3 companies) were for back injuries and 12 (from 8 companies) were claims resulting from falls. In the control group, over the same interval, 82 claims of any type occurred among 58 of 389 companies (15%). Of those 82 claims, 17 were for back injuries (16 companies) and 25 were fall-related (22 companies). There were not enough cases for meaningful inferential analysis of a treatment effect on claims. There were no significant differences in the number or cost of claims between intervention participants and controls over this claims history period. It also turned out that there was no feasible way to link Kentucky Department for Workers Claims insurer data to individual claims. Therefore, comparisons between KEMI and non-KEMI claims could not be made. The most frequently reported reasons for not participating in Years 2 and 3 were: "Time involved too much for me"; and "Company already has good safety record". The most frequently selected reasons for participating were "Will get insurance discount" (80%, n = 123) and "Safety is a priority for our company" (68%, n = 104). In the small construction outfits whose owners agreed to participate or were willing to serve as controls, safety climate is a valued and stable characteristic. Despite not showing an impact on safety climate or claims, simulation evaluations suggested that participants found them realistic and worthwhile. Simulation exercises may provide a reinforcement of good safety practices rather than new knowledge or impetus for behavioral change in highly experienced workers. Claims experience was far less than expected. Smaller construction outfits may actually be safer overall than larger construction companies; despite being relatively less regulated.
University of Kentucky, Kentucky Injury Prevention and Research Center (KIPRC), Lexington, KY 40504
University of Kentucky, Kentucky Injury Prevention and Research Center (KIPRC), Lexington, Kentucky