study (from CSST annual reports), for this review’s purposes, is a surrogate for OHSMS implementation and may be limited in terms of validity and accuracy.
4.4.5 Summary of results in studies of mandatory OHSMSs The five studies involving mandatory OHSMSs were conducted in Norway, Quebec, and Ontario. The Lewchuk et al. (1996) and Dufour et al. (1998) studies were based on data before and after the Quebec and Ontario regulations were enacted in 1979. The health and safety regulations that were introduced in Quebec and Ontario were progressive for their time, but in terms of contemporary approaches to integrated health and safety management their requirements were limited. The 1992 Norwegian Internal Control regulation, on the other hand, incorporated systematic management systems which were becoming more common as best practice models in business at that time. Therefore, the two studies of the Quebec and Ontario legislations were based on regulatory efforts that were more limited in the application of OHSMS concepts; whereas, the Norwegian-based studies focused on the IC regulation which was, and is still, highly focused on OHSMSs.
Three of the five studies were conducted on the Norwegian IC regulation and they provided evidence for positive effects across a variety of outcomes. Four years after the IC regulation, 45 per cent of surveyed firms had fully implemented IC (Nytro et al., 1998). At the same point in time, Torp et al. (2000) found significant relationships between measures of IC implementation (four years post IC regulation) in their cross-sectional study and intermediate variables that included satisfaction with HES activities, satisfaction with the physical working environment, HES related management support, and workers’ participation in HES activities. Final OHS outcomes were reported in two of the studies. Saksvik and Nytro (1996) reported a significant decline in absenteeism in firms with higher levels of IC implementation, one year after implementation. A decline in accident rates with IC implementation was observed but was not statistically significant. Torp et al. (2000) found a significant negative relationship between IC implementation and musculoskeletal symptoms in their study of garage workers, indicating that firms with higher implementation had fewer workers reporting symptoms. A statistically insignificant negative relationship of IC implementation and sickness absence was also reported.
These three studies of the Norwegian IC regulation, which all showed positive effects across a range of outcomes, were limited in their study design. In two, subjects were selected by randomized quota sampling (Saksvik and Nytro, 1996; Nytro et al., 1998) which is vulnerable to selection bias, but of which the authors did not acknowledge. Selection bias also could not be ruled out in Torp et al. (2000), as their sample included a large number of managers who were scheduled to receive OHS training.
Effectiveness of Occupational Health & Safety Management Systems: A Systematic Review