J.E. Gaskin and R.J. Parker
J.E. Gaskin and R.J. Parker are, respectively, Director and researcher at the Logging Industry Research Organisation, Rotorua, New Zealand.
Use of an Accident Reporting Scheme (ARS) to identify aspects of the logging operation that need investigation to reduce accident frequency and severity. Objective risks for various logging tasks, injury sites and felling and trimming situations are compared with loggers' subjective assessments of risk.
Protective leg-wear saved this logger from a serious chainsaw injury
Internationally, the accident rate in logging is high. Education and equipment can be used to reduce this rate but, in the face of limited resources (including in New Zealand), efforts must be directed to where they are most needed and can do most good. Detailed accident information is the key to identifying target areas and guiding the efforts of researchers, trainers and manufacturers to reduce accident frequency and severity.
In New Zealand, the Logging Accident Reporting Scheme (ARS) is used by the Logging Industry Research Organisation (LIRO) to drive its research and development programme for investigating safer work methods and protective equipment. The first section of this article describes the ARS, providing examples of data derived from its use and illustrating how the scheme has been used to determine research and development interventions. The second section discusses the use of the scheme to measure the accuracy of loggers' perceptions of risk.
In 1980, Swedfor Consultancy of Sweden was contracted to develop "safer felling and delimbing techniques for New Zealand conditions". The Logging Industry Research Association (LIRA) renamed Logging Industry Research Organisation in 1991 was asked to provide the consultants with specific information on felling and delimbing accidents. At the time, only limited records were found to be available, and these were typically restricted to the larger forest companies. No industry-wide collection of accident records had been undertaken apart from incomplete data managed by the Department of Labour. There was clearly a need for a comprehensive accident data base containing accurate, up-to-date information on the number and type of accidents occurring in logging. A 15-month pilot, voluntary ARS was tested in the Bay of Plenty region. Based on positive reactions to the test, in 1983 the ARS was introduced on a national basis for all of New Zealand's logging workforce (Prebble, 1984).
TABLE 1. Number of lost-time accidents by activity and injury, 1985-1991
|
Operation |
Type of injury |
Percentage of total |
||||||||||
|
Laceration |
Fracture |
Bruising |
Multiple |
Other |
Overall |
|||||||
|
Felling |
180 |
(14) |
63 |
(39) |
118 |
(8) |
30 |
(37) |
61 |
452 |
(18) |
27.3 |
|
Trimming |
357 |
(12) |
14 |
(34) |
41 |
(6) |
6 |
(7) |
36 |
454 |
(12) |
27.5 |
|
Landing work |
120 |
(14) |
53 |
(43) |
88 |
(6) |
13 |
(23) |
85 |
359 |
(15) |
21.7 |
|
Breaking out |
39 |
(5) |
34 |
(29) |
63 |
(6) |
14 |
(29) |
64 |
214 |
(12) |
13.0 |
|
Other |
52 |
|
25 |
|
27 |
|
9 |
|
61 |
174 |
|
10.5 |
|
Total |
748 |
(12) |
189 |
(37) |
337 |
(7) |
27 |
(29) |
307 |
1653 |
(14) |
- |
|
Percentage |
45.3 |
|
11.4 |
|
20.4 |
|
4.4 |
|
18.6 |
- |
|
100.0 |
Note: Figures in parentheses are the average number of days lost per accident.
Data collection and analysis
Under the ARS, logging companies are required to send summaries of accidents to LIRO at regular intervals. Individual contractors who are not coveted by a company ARS send accident reporting forms directly to LIRO. All accident records are anonymous. The records are added to a computerized data base and quarterly summaries and annual reports are prepared and disseminated.
The data received on the accident report forms enable LIRO to analyse specific elements related to logging accidents and also to paint a broad picture of trends regarding the overall situation in New Zealand, including the number, type and severity of accidents.
According to the 1991 analysis of lost time accidents (Parker, 1992), felling and trimming accidents are the most common (accounting for more than 50 percent of the total), followed by landing work and breaking out. These results confirm long - term trends (see Table 1). A recent increase in hauler operations (from 10 percent in 1987 to 21 percent in 1991) is considered to have resulted from the increased number of new, inexperienced crews starting up in recent years.
TABLE. 2. Location of laceration injuries, 1985-1991
|
Location of laceration |
Number |
Percentage of total |
|
Head |
72 |
9.6 |
|
Eye |
22 |
2.9 |
|
Arm |
82 |
11.0 |
|
Hand |
216 |
28.9 |
|
Upper torso |
18 |
2.4 |
|
Lower torso |
6 |
0.8 |
|
Upper leg |
57 |
7.6 |
|
Lower leg |
98 |
13.1 |
|
Foot |
167 |
22.3 |
|
Other |
10 |
1.3 |
|
Total |
748 |
100.0 |
A total of 218 lost-time accidents (accidents causing a worker to lose at least one full day's work) were reported in 1991, with the average number of days lost per accident being 15. About 42 percent of the accidents resulted in less than five days' absence from work but there were seven fatal accidents.
The feet, lower legs and hands continued to be the most frequently injured parts of the body and accounted for 41 percent of all lost time injuries in 1991. However, there was a marked reduction in hand injuries in 1991 (16 percent) compared with the overall trend (see Table 2). The ARS data also indicate that lacerations are the most common type of injury, followed by fractures, bruises and other miscellaneous injuries (see Table 1).
LIRO also analyses logging accidents by day of the week and time; similar to results from other countries (ILO, 1991), the first full hour of work after the weekend break in the case of New Zealand, Monday morning between 9 a.m. and 10 a.m. - is the most dangerous period.
Use of the ARS to direct research and development
The data from the ARS help to establish which aspects of the logging operation need investigation to increase safety, guide the development of new techniques and monitor the success of new equipment and methods of work.
For example, in 1985 chainsaw protective leg-wear was made compulsory in the New Zealand logging industry. While the garments are not an absolute barrier to the chainsaw, they have reduced the severity of chainsaw lacerations. In fact, the proportion of injuries to the leg decreased from that year. However, the proportion of leg injuries is again increasing with time (see Table 3). This increase could be the result of a deterioration of protective gear with age; cuts, nicks and exposure to solvents such as chainsaw fuel and oil over time may be reducing its effectiveness. LIRO has initiated a study to measure the effective life of protective leg-wear.
Further analysis of chainsaw-related injuries indicates that 33 percent of all chainsaw injuries to the lower leg are inflicted to the unprotected back of the leg and to the left foot behind the protective steel toecap (see Table 4). This information is being used by protective leg-wear and footwear manufacturers to direct future product developments.
The ARS data revealed that, in 1991, two fatal and eight lost-time accidents occurred because the loggers were "not seen" by machine operators or other loggers. These lost-time accidents each resulted in a mean of 31.5 days off work. LIRO has embarked on a project to develop highly visible and practical clothing for loggers to increase their conspicuousness on the landing and in the foliage beneath trees.
Trimming under a hang-up - a dangerous practice
TABLE 3. Chainsaw injuries to the lower leg as a proportion of all chainsaw injuries
|
Accident |
1985 |
1986 |
1987 |
1988 |
1989 |
1990 |
|
Total lost-time chainsaw accidents |
114 |
95 |
76 |
113 |
75 |
53 |
|
Chainsaw cuts to the lower leg |
24 |
11 |
8 |
15 |
11 |
8 |
|
Percentage of total chainsaw injuries |
21.1 |
11.6 |
10.5 |
13.3 |
14.7 |
15.0 |
Future developments
In addition to the Logging ARS, a Forestry ARS has been in operation for one year (Gaskin, 1991). Forestry includes nursery work, establishment, precommercial thinning, pruning and accidents associated with forest maintenance, but it excludes logging. Data collection, analysis and dissemination will proceed in a similar fashion to the Logging ARS. LIRO is currently investigating the feasibility of establishing a Wood Processing ARS to summarize sawmill accidents.
TABLE 4. Chainsaw cuts to legs and feet, 1985-1991
|
Part of body |
Chainsaw injury |
|
|
To back or side |
Total |
|
|
Upper leg |
9 |
46 |
|
Lower leg |
28 |
84 |
|
Foot |
- |
163 |
As discussed in the previous section, the ARS data indicate that most logging injuries occur during the felling and trimming phases (Gaskin, 1990). However, a question that arises is "how do loggers perceive the risks inherent in the various aspects of their occupation?"
The following section details an ongoing three-part effort, based on responses to questionnaires, to compare loggers' subjective assessments of risk with the actual objective measure of risk as revealed by the ARS.
Part I. Assessment of personal risk of injury by logging task and part of the body. The summaries of accident frequency and severity reported by the ARS detail the results of an accident but not how or why loggers got into the situation which resulted in injury. In addition, data alone are not enough to determine how loggers perceive the relative risk of injury in their occupation and if their (subjective) perception of risk corresponds with the actual (objective) risk derived from the ARS. Therefore, Part I of the investigation assessed whether loggers were aware of which parts of a logging operation were most hazardous and which parts of the body were most at risk (Tapp, Gaskin and Wallace, 1990).
TABLE 5. Surveyed loggers' subjective ranking and ARS objective ranking of the likelihood of injury performing various logging tasks
|
Situation |
Survey1 |
ARS2 |
Rank |
|
|
Survey |
ARS |
|||
|
Felling (general) |
3.2 ± 0.1 |
26.9 ± 4.2 |
1 |
1 |
|
Trimming (general) |
2.9 ± 0.1 |
28.3 ± 3.3 |
3 |
1 |
|
Landing work |
2.4 ± 0.1 |
26.4 ± 4.4 |
3 |
1 |
|
Breaking out |
2.3 ± 0.1 |
13.4 ± 1.5 |
3 |
2 |
1Mean score (± 95 percent CI).
2Mean percentage (± 95 percent CI) of lost - time accidents per year, 1985-1991.
The analysis of 225 questionnaires completed by a representative sample of forest workers indicated that loggers considered felling the most hazardous of logging tasks, followed by trimming, landing work and breaking out (freeing felled trees from understorey ground cover or humus layer), (see Table 5). In terms of injury risk to specific body parts, the loggers considered the feet, hands and lower leg to be the most at risk (see Table 6). Injury to the torso was considered the least likely.
Comparison of these results with objective data from the ARS provides valuable insights. In contrast with the workers perspectives, lost-time accident data from the ARS indicate that felling, trimming and landing work have accounted for roughly equal percentages of injuries since 1985. Perhaps the perceived risk of felling is greater because the largest proportion (55 percent) of fatalities occur in the felling phase (Gaskin, 1988) and fatalities are well publicized in the industry. Moreover, the dramatic sight and noise of a large tree falling to the ground may increase the perceived risk of felling operations. The lower subjective rankings given to trimming work, despite the fact that it accounts for nearly the same proportion of injuries as felling, indicate the need for more awareness-building among workers in this regard. Of particular concern is the workers' perception of landing work and breaking out as being equally dangerous, despite the fact that landing accounted for nearly twice as many lost-time accidents.
The loggers surveyed demonstrated a good knowledge of which body parts were the most likely to be injured (feet, hands and lower legs). The head, eye, face and lower torso were rated lower on the scale of injury risk and ARS data supports this assessment.
However, there are a number of inconsistencies between subjective and objective results that bear examination. For example, the upper leg is injured only half as frequently as the lower leg, yet loggers rated the likelihood of injury to the two as being nearly equal. The chainsaw power head is often rested on the thigh for much of the working day and this close proximity may have increased the perceived risk of injury to the upper leg. It is also noteworthy that the subjective rating of injury to the eye was higher than that indicated by the ARS data. Wood chips and dirt often enter the eyes and cause pain and irritation, which may be the reason for the higher subjective assessment of the likelihood of eye injuries.
Of greater concern is the logger's significant underestimation of the risk of injury to the upper torso. Although worker's rated the upper torso as least likely to be injured, actual data indicate more injuries to the upper torso than to the upper leg, the arm, the head or the eye. Here again, more attention to awareness building is necessary.
TABLE 6. Surveyed loggers' subjective ranking and ARS objective ranking of the likelihood of injury to specific body parts while logging
|
Body part |
Survey1 |
ARS2 |
Rank |
|
|
Survey |
ARS |
|||
|
Foot |
3.5 ± 0.1 |
17.3 ± 3.7 |
1 |
1 |
|
Hand |
3.4 ± 0.1 |
16.5 ± 3.8 |
1 |
1 |
|
Lower leg |
3.3 ± 0.1 |
17.7 ± 2.7 |
2 |
1 |
|
Upper leg |
3.0 ± 0.1 |
7.9 ± 2.5 |
3 |
2 |
|
Arm |
2.9 ± 0.1 |
8.1 ± 1.1 |
3 |
2 |
|
Head |
2.8 ± 0.1 |
7.5 ± 1.6 |
4 |
2 |
|
Eye |
2.8 ± 0.1 |
3.3 ± 1.0 |
4 |
3 |
|
Upper torso |
2.3 ± 0.1 |
10.4 ± 1.8 |
5 |
2 |
|
Lower torso |
2.3 ± 0.1 |
6.8 ± 2.7 |
5 |
2 |
1Mean score (± 95 percent CI),
2Mean percentage (± 95 percent CI) of lost-time accidents per year, 1985-1991.
TABLE 7. Felling situations ranked in order of perceived risk
|
Situation |
Perceived risk rating | |
|
Felling the supporting tree in a hang-up |
0.31 |
a |
|
Felling within two tree lengths of another person |
0.22 |
b |
|
Leaving a hang-up and having a meal break or going home |
0.17 |
b |
|
Overcutting the back cut |
0.09 |
c |
|
Driving a tree at right angles across a hang-up |
0.09 |
c |
|
Felling against the wind or lean |
0.05 |
d |
|
Felling without clearing around the butt |
0.04 |
d |
|
Trying to bring down a tree by rocking it |
0.02 |
c |
|
Felling into standing trees |
0.02 |
c |
Note. Risk ratings followed by a different letter are significantly different, p<0.05.
TABLE 8. Trimming situations ranked in order of perceived risk
|
Situation |
Perceived risk rating | |
|
Trimming in front of a scarfed and back-cut tree |
0.40 |
a |
|
Trimming under a hang-up |
0.40 |
a |
|
Using a saw behind the legs |
0.06 |
b |
|
Standing on a stem under tension while cutting head of stem |
0.05 |
bc |
|
Trimming a tree which is well off the ground |
0.04 |
c |
|
Using the saw above shoulder height |
0.04 |
c |
|
Trimming on the downhill side on steep country |
0.01 |
d |
|
Overreaching to trim |
0.006 |
c |
|
Walking along the top of the stem to trim |
0.002 |
f |
Note: Risk ratings followed by a different letter are significantly different, p<0.05.
Part II. Loggers' ranking of felling and trimming hazards. This part of the study assessed the accuracy with which a group of 137 loggers and 46 logging contractors judged felling and trimming hazards by paired ranking (Parker, 1991). Loggers were shown a booklet containing combinations of pairs of diagrams with captions describing each situation. They were asked which diagram of the pair portrayed the more hazardous situation. The perceived relative risk of each situation was determined by measuring the proportion of respondents who rated one situation more hazardous than another. Analysis was by logistic regression.
Loggers ranked "felling the supporting tree in a hang-up" more hazardous than any other felling situation (see Table 7). "Trimming in front of a scarfed and back - cut tree" (trimming in front of a tree which has had the scarf removed and has been cut but still remains standing) and "trimming under a hang-up" were ranked equally as the most hazardous trimming situations (see Table 8).
In regard to felling, comparison of risk ratings with 20 years of records of fatal New Zealand logging accidents (Gaskin, 1988) indicated that loggers are correct in identifying hang-ups as a serious hazard; hang-ups were the greatest cause of death (causing 16 Of 49 fatalities) over the period 1968-1987. According to ARS data, although hang-ups accounted for only 7.6 percent of lost-time felling accidents, they accounted for 31 percent of fatalities. Of concern is the very low rating given to "felling into standing trees", despite the fact that broken branches and cones falling from trees accounted for 20 percent of lost-time accidents in the period 1985-1991.
The loggers' ranking of hazardous trimming situations is consistent with fatality results (Gaskin, 1988) which show that most fatalities can be attributed to these causes. When a fatality occurs, it is well publicized and loggers become aware of the cause. The risk of injury while "walking along the top of the stem to trim" was rated lowest but resulted in a high number of lost-time accidents (10 percent of the total) from 1985 to 1991. Falling from a log while trimming has not caused any fatalities in New Zealand and so this type of accident is not well publicized.
LIRO is alerting the logging workforce to these hazards through increased publicity and is promoting the use of spiked boots among falters.
Part III. On-site assessment of the hazards confronted by fallers and how hey deal with them. This part of the study is currently in progress.
This article has demonstrated the utility of accident reporting, in general, and the ARS, in particular. The value of the ARS is evident in monitoring trends in accident occurrence as well as in identifying those parts of the logging operation where more training or safety equipment could reduce the frequency and severity of injury. In addition, when used in conjunction with questionnaires, the ARS has enabled the loggers' subjective assessment of risk at work to be compared with an objective measure of accident frequency to gain a greater understanding of the causes of logging accidents.
Dunn, J.G. 1972. Subjective and objective risk distribution: a comparison and its implications for accident prevention. Occup. Psychol., 46: 183-187.
Gaskin, J.E. 1988. Analysis of fatal logging accidents, 1968-1987. LIRA report, Vol. 13, No. 20. Rotorua, New Zealand, LIRA.
Gaskin, J.E. 1990. Analysis of lost-time accidents - 1989 (Accident Reporting Scheme statistics). LIRA report, Vol. 15, No. 5. Rotorua, New Zealand, LIRA.
Gaskin, J.E. 1991. Lost-time accidents in forestry 1990 (Accident Reporting Scheme statistics). LIRA report, Vol. 16, No. 10. Rotorua, New Zealand, LIRA.
ILO. 1991. Proc. Occupational Safety and Health in Forestry. Forestry and Wood Industries Committee, Second Session. Geneva, ILO.
Ostberg, O. 1980. Risk perception and work behaviour in forestry: implications for accident prevention policy. Accident Anal. Prev., 12: 189-200.
Parker, R.J. 1991. Loggers' ranking of felling and trimming hazards. LIRA report, Vol. 16, No. 4. Rotorua, New Zealand, LIRA.
Parker, R.J. 1992. Analysis of lost-time accidents 1991 (Accident Reporting Scheme Statistics). LIRO report, Vol. 17, No. 8. Rotorua, New Zealand, LIRO.
Prebble, R.L. 1984. A review of logging accidents in New Zealand. In Proc. LIRA Seminar Human Resources in Logging. Rotorua, New Zealand, LIRA.
Tapp, L., Gaskin, J.E. & Wallace, K 1990. Loggers, assessments of risks in their work. LIRA report, Vol. 15, No. 1. Rotorua, New Zealand, LIRA.
