The aim of the investigation by Lewchuk et al. (1996) was to examine whether the legislation in Ontario, Bill 70 had any impact on workplace level OHS performance. The authors stated that with Bill 70 “…a number of the rules of the game were changed at that time and caused changes in the workplace – whether or not these were immediately translated into JHSCs (which became mandatory for most workplace after Bill 70).” Their research question was whether or not there was a reduction in work-related injury and illness frequency after the act came into effect.
To address this question, a multiple, pooled cross-sectional time series study with a comparison group was performed. The sample consisted of 637 workplaces (436 in the manufacturing sector and 201 in retail) who were participants in a previous survey in 1991, along with 140 additional workplaces who were non-participants in the 1991 survey (selection method not specified). To be eligible for the previous survey, manufacturing workplaces had to have 50 employees and retail workplaces 20 employees in 1988. For these workplaces, annual data from 1976 to 1989 for loss time claims records, short and long run compensation costs, and employment level by rate group were obtained from the Workers’ Compensation Board. Regression analyses were conducted for each sector with the indicator variable, Bill 70, assigned a value of 0 for the years up to 1979 and a value of 1 for 1980 and after. The authors felt that the retail sector could be thought of as a control group since critical sections of the legislation (including formation of JHSCs) did not apply.
The results of the regression analyses, in which the dependent variable was injury/illness frequency and included employment, time, union status, and industry as control variables, indicated that Bill 70 had a significant effect in the direction of lower frequencies for manufacturing whereas it was not significant for the retail sector. The authors indicated that the size of the coefficient for Bill 70 was such that the reduction in lost-time accident frequencies was in the order of 18 per cent for manufacturing workplaces. The time variable was significant and positive in both regression analyses, indicating growing accident/illness rates in both sectors; the authors mentioned that they conducted other analyses using the time variable, including fitting time trend as a quadratic in the models, but that the final linear model was appropriate.
The important finding from Lewchuk et al. (1996) is that there was a significant reduction of work-related injury and illness rates in manufacturing after Bill 70 compared to before the regulation whereas no significant reduction was observed in the retail sector. They also found that the effects were progressively larger for years 1980, 1981, and 1982 in manufacturing, indicating that the impact of the legislation grew over the study period (this trend was not observed in the retail group). The study nonetheless has a number of limitations. The authors suggested that a number of other variables may have had an impact on OHS performance and
Institute for Work & Health