Workplace Fatality Figures:

CPM 8 and Beyond

The latest Comparative Performance Monitoring Report (CPM 8) raises several issues for industry, because this report is one of the main measures by which industry performance is assessed.  It is of concern that the data is continually revised (upwards) and that its authors proclaim such a high level of uncertainty as to its accuracy. This is data by which industry is judged, and forms a large part of regulator justification for their policies and regulatory action.

As part of the review which preceded the publication of the revised CPM 8, a Statement of Purpose was developed to “address the needs of key users”, which included providing “measurable information to support policy making and program development” and “identification of factors contributing to improved OHS and workers compensation performance”.

The “high level” information provided is, at best, summary data that does not disclose the factors underlying the trends which are so briefly described. Without an understanding of these factors, the CPM reports are limited in achieving their purpose of identifying factors that contribute to improved OHS and workers compensation performance.

Consequently, the CPM reports in this format should not be published without explanatory detail, or at the very least, an accompanying Compendium of Workers Compensation Statistics, which provide a fuller picture of what underlies the data and presents the information in a different light. This is important for inter-industry data and for understanding the nature of injuries and the mechanism/agent of injury, neither of which are reported in the CPM reports.

The shortcomings of the CPM reports are particularly evident in the cursory presentation of fatalities data. We are told that:

• in 2004-05 there were 214 accepted compensation claims for a work related fatality from disease and injury/musculoskeletal disorder.
• along with other data, “this number is expected to rise” due to acceptance of more claims. 

CPM 8 makes large upward revisions of previous data (see table below) but no explanatory detail is provided, other than late acceptance of claims and “auditing of data extraction methodologies undertaken by many jurisdictions to ensure greater comparability” (page 8).  For 2003-04 this “auditing” has resulted in an increase of 67 workplace deaths nationally, with Victorian fatality numbers increasing by over 35% in one year and Queensland and Tasmania by over 50%.  

Then in this “substantially reduced in size” publication, considerable space is devoted to explaining the limitations and understatement of the National Data Set (NDS), and pointing out that the data is subject to considerable revision. Still more space is devoted to the alternative data source – the Notified Fatalities dataset.

Yet from the table and analysis below it would appear that the Notified Fatalities Report is not generating a very different picture on work fatalities to the NDS (particularly from a trend perspective). Whilst more likely to capture data from “high” fatality industries, it reflects a lesser number of fatalities than NDS (144 work related traumatic fatalities in 2003-04) even allowing for its exclusion of traffic accidents. In this regard, it should be noted that the basis for distinguishing between vehicle, traffic and “fatality from other causes” (such as hit by moving object and crushing) is not clear (see Appendix).

A similar number of fatalities and large proportion of road and commuting deaths was also reported in the 2006 ASCC report “Estimating The Number Of Work-Related Traumatic Injury Fatalities In Australia 2003-04.”  Of the 332 estimated deaths, 163 were road crash or commuting accidents.

What would have been of more use in understanding trends and setting priorities would have been an explanation of just what constituted the “162 fatalities from injury and musculoskeletal disorder”.  It is clear from all current and historical data sources that vehicle/traffic accidents constitute a very high proportion of workplace deaths in Australia.  The Compendium of Workers Compensation Statistics, the Notified Fatalities Report and the ASCC 2006 report estimates all demonstrate this fact.

Yet instead of explaining the high component of vehicle/traffic fatalities in this data and the other two major contributors to workplace deaths, Being Hit by Moving Objects and Chemicals and other Substances (mesothelioma), CPM 8 emphasises the limitations and understatement of the NDS. No reference is made to the Compendium.

Given the current regulatory focus on the transport industry and in particular on regulation for fatigue management, it is important for industry to have accurate and reliable data about the causes of fatality involving vehicles and certainty as to the basis for classification as to “vehicle”, “traffic” “commuting and the re-categorisation of these fatalities.

The number of fatalities overall does not constitute massive data. Analysis of the circumstances of each death could be published and the decisions made in categorising this data made more transparent. This task is already undertaken at State and national levels. If regulators are serious about using evidence-based approaches to formulating policy and legislation, the evidence should be made known in detail. Obviously, this would be of assistance to duty holders in undertaking risk assessment.

The inadequacies in reporting on fatalities in Australian workplaces is further demonstrated by:

• the prominence given in CPM 8 to Australia’s performance in international comparison.
• the WRMC aspirational goal of having the lowest rate of traumatic fatalities by 2009.

Acknowledging the important effect that traffic/vehicle accidents have on our workplace fatality rates (particularly in comparison with countries such as the UK who explicitly exclude such accidents from work fatality data) would give readers a more meaningful understanding. It would also help to explain to readers that the fatalities incidence rate can be volatile because the fatality numbers are low. Statements such as “consistent improvement needs to be achieved” provide little assistance. We have had consistent improvement, at least on previous, unadjusted data. But this may not be reflected in lower volatility in the incidence rate where numbers of deaths remain relatively small in terms of the size of the working population.

Whilst it may be desirable to provide a summary publication of key performance indicators, the usefulness of such indicators is limited where the data is open to interpretation and no explanation of the data is provided. The substantial upward revision of claims numbers is also a matter for considerable concern and is particularly pronounced in CPM 8. Such large upward revision cannot be explained away by pointing to late acceptance of claims and different historical data due to changes in data development.

 

Comparison of Total Fatalities

• Original preliminary data from CPM 6 and CPM 7
• CPM 8 revised data in bold brackets

 

	2002-03			2003-04			2004-05
	Workplace	Road	Total	Workplace	Road	Total	Workplace	Road	Total
CPM Compensated claims for work related fatalities; journey claims not included but includes “deaths from traffic incidents which account for around 30% of all compensated claims” (CPM 8 page 10)			257 (283)			189 (256)			214
Notified Fatalities Report Worker only, by-stander in square brackets, said not to include “traffic accidents” but includes “vehicle accidents” (CPM 8 report page 10).      See Appendix.				113 [12]	13 [6]	126 [18]	102 [7]	25 [5]	127 [12]
Compendium of Workers Compensation Statistics Compensated work related fatalities, journey claims data provided but not included in Compendium total	211	53	211 (283)	189	60	189 (256)	n/a	n/a	214

 

 

Appendix

Inconsistency in reporting Traffic, Vehicle fatality and their re-categorisation as death by other causes

 

(1)  The Notified Fatalities Statistical Report July 2003 to June 2004

The Notified Fatalities Statistical Report July 2003 to June 2004 states that:

• “many work related transport fatalities are not notified to OHS jurisdictions and are therefore not reflected in these figures”.

• “Several jurisdictions do not cover work related deaths due to traffic accidents in their notification systems although they are covered in their workers compensation systems”.

• “At present, most jurisdictions do not collect reports of work related deaths due to traffic or other transport related accidents through their notification systems”.

So the report excludes traffic accidents but includes vehicle accidents. Whilst it is understood that States classify and report differently, the basis for the distinction between “traffic and “vehicle” in what is reported appears to be unclear. Further, the report subsequently re-categorises what would appear to be vehicle (traffic?) accidents.

This report states that there were 144 notified work related fatalities consisting of 126 workers and 18 by-standers. We are told that vehicles were involved in 37 fatalities of which:

• 19 were due to vehicle accidents.
• 12 were pedestrians hit by a vehicle.
• 6 involved persons hit or crushed by their unattended vehicle.

 

In “causes of fatalities” the 12 pedestrians hit by a vehicle and the 6 persons crushed or hit by a vehicle were classified not as vehicle accidents but as “Being hit by a moving object”.

We are then told that 58 notified fatalities were caused by “Mobile plant and transport”. Of these:

• Trucks, semi trailers and lorries were responsible for 14 fatalities.
• Cars, station wagons, vans and utilities were responsible for 7 fatalities.

 

It is unclear which of the above vehicle accidents are also included in this category or have been re-categorised elsewhere because the total of fatalities by cause (page 6) exceeds the 144 notified fatalities:

 

Being hit by moving objects	29
Trapped between stationary/moving objects/trapped by moving machinery	24
Vehicle accidents	19
Falls from height	22
Electrocution	12
Mobile plant and transport	58
Machinery and mainly fixed plant	24
Outdoor environment	21
Total	209

 

(2)  The Notified Fatalities Statistical Report July 2004 to June 2005

This report tells us that there were 139 notified work related fatalities, comprising 127 workers and 12 bystanders. Of these, there were 30 notified work related fatalities involving vehicle accidents. The report cross-classifies the mechanism of fatality with the breakdown agency to see “what type of equipment or circumstances are particularly associated with causes of fatality”.

However, this does not provide greater clarity. For example, 60 deaths are classified as mobile plant and transport fatalities, with the mechanism of injury for 30 deaths being “vehicle”  and 27 deaths being “crushing” or “being hit by moving objects”. It would appear that some vehicle accidents have been categorised as “crushing”, as the summary of notification narrative states that 6 fatalities involved being crushed between a vehicle and 4 involved being crushed under a vehicle.

 

(3)  The CPM and Compendium of Workers Compensation Statistics

CPM data is derived from the National Data Set (SDS) whose data is more fully reported in the Compendium of Workers Compensation Statistics. CPM 8 states:

“These data are known to understate the true number of fatalities from work related causes, particularly deaths from diseases, deaths in the agricultural sector and potentially deaths from road traffic accidents for which compensation may be alternatively sought through the Compulsory Third Party insurance scheme”.

The most recent Compendium 2003-04 (provisional) reports there were 189 work related fatality claims, with an additional 60 compensated fatality journey claims. The most common mechanism for compensated fatalities, accounting for 67 fatalities (35% of claims), was Vehicle accidents.

The most common breakdown agency for compensated fatality claims in
2003–04 was Mobile plant and transport. These 80 fatalities represented 42% of compensated fatality claims. Of these claims, 66 involved Road transport, mostly comprising Trucks, semi-trailers, lorries (32 fatalities) and Cars, station wagons, vans, utilities (27 fatalities). Some vehicle/traffic accidents would also be categorised as “hit by moving object” which accounts for a further 31 fatalities.