Lost time claim frequency data

This case provides relatively little support for Worksafe s claims. There was a dramatic drop in the total number of injuries and total injury costs, but this was largely a consequence of the big cut in the workforce. Most of the safety data presented in the study do not take account of this reduction in employee numbers and so are essentially uninterpretable. The one exception is the lost-time injury frequency rate which does show a decline. The new technology played a part in this decline since it is inherently safer than the old. Presumably, attention to OHS also played a part. Finally, it must be stressed that the very substantial increase in productivity which the mill has experienced has nothing to do with improved safety and everything to do with the state-of-the-art technology installed. Any suggestion that the productivity gains at the plant were due to a focus on health and safety would be entirely fallacious. 

Some of the data which apparently support the productivity claim relate to NSW coal mines over the decade of the 1980s and into the 1990s. Figure 10.1, which presents the data relied on by Mathews (1993, p. 48), shows that during the period 1982-1989 the lost-time injury frequency rate, i.e. the number of lost-time injuries per million hours worked, went steadily down while productivity went steadily upwards. 

The data in Table 8.2 is in concert, generally, with the trending previously shown for the years 1999 through 2003 in Table 8.1. Reductions in less serious injuries are down substantially more than that for more serious injuries. In the August 2011 NCCI Research Brief, NCCI also says that workers compensation claim frequency for lost-time claims has increased 3% in 2010. This represents the first increase since 1997 and only the third time that frequency has increased in the last 20 years. This upward claim frequency trend for lost time cases is relative to the increase in fatalities for 2010 as shown in Table 8.3. 

Shannon et al. (1996) is an example of a study limited in both of these ways. It doesn t have any variables on employees involvement with the company s financial returns, and it doesn t present the results of the analysis in a multivariate ftamewoik. The researchers examine a matched sample of questionnaires sent to firms in Ontario with data on those firms lost-time frequency rates (equivalent to the analysis of claim frequency given below). Survey questionnaires include responses from both workers and management. The paper discusses only univariate statistical analyses that compare various workplace practices to whether a firm has a low, medium, or high claim frequency. A multivariate regression analysis is mentioned but not reported on in the paper. 

---