4.4.1 Studies of implementation Two of the five studies provided evidence on the implementation of the Norwegian Internal Control (IC) regulation (Nytro et al., 1998; Saksvik and Nytro, 1996). The first of these studies was a cross-sectional investigation by Saksvik and Nytro (1996) conducted in 1993, one year after the IC regulation came into effect (January 1992). The investigation had multiple objectives including the determination of the prevalence of implementation of IC systems, identification of IC induced organizational changes, and discovering whether absenteeism and accident rates were related to implementation. The survey, conducted amongst a randomized quota sample of 2092 public and private enterprises in Norway, was implemented using computer assisted telephone interviewing. The respondents, as described by the authors, “…for the most part were manager representatives” of each of the firms. The health and safety data that were obtained were derived primarily from archival records but the authors indicated some were based on recall or estimates provided by the respondent.
The 45 questions in the survey had been pre-tested in previous research involving 500 enterprises that had implemented IC prior to 1992. Of these questions, four provided information pertaining to IC implementation (this study also provided evidence for intermediate and final OHS outcomes). Study findings were reported based on a sample of 915.12 Findings (percent firms) for reported changes due to the implementation of IC were as follows: more clear lines of responsibility (58 per cent), better documentation (58 per cent), more/better risk assessment (46 per cent), and new strategic plans (42 per cent). No statistical tests of significance or precision were conducted. An additional finding that did not fit into any of the review’s evidence categories was that 53 per cent of respondents reported integration of IC with Total Quality Management as a result of IC implementation.
Findings for implementation outcomes from the Saksvik and Nytro (1996) study should be interpreted with caution as these were not the investigators’ primary research questions. Additionally, the overall study design raises serious concerns about internal and external validity. Randomized quota
12 It was not clear to the reviewers how this number of respondents was obtained as 66 per cent of respondents (n=1381) reported not having begun to implement the IC regulation (excluding these respondents from the data would leave 711 subjects). The authors contacted Dr. Saksvik and he suggested that the “extra” respondents to these questions were likely those who had developed systems similar to the IC system (e.g., TQM) yet answered “no” to actually have begun implementing IC.
Effectiveness of Occupational Health & Safety Management Systems: A Systematic Review