Abstract
The objective of this study was to identify and quantify the hazards present during arboricultural operations. The Occupational Safety and Health Administration Fatality and Catastrophe incident database and other Bureau of Labor Statistic sources were analyzed for arboricultural operation incidents within the 17-year period from 2001 through 2017. There were 865 fatal and 441 nonfatal incidents reviewed from this period. The leading four fatal incidents, from the largest to the smallest number of fatalities, were climber falls, workers struck by a falling tree, workers making indirect contact with an electric current, and workers struck by a falling branch. Climber falls were also the leading incident for severe nonfatal injuries, followed by ground workers struck by a falling branch, workers struck by a chain saw, and falls by aerial device operators. The American National Standards Institute Z133 American National Standard for Arboricultural Operations—Safety Requirements establishes safety requirements and recommendations for arboricultural operations in the United States. It addresses common hazard sources and has guidelines to avoid, eliminate, or reduce them. Safety training programs should emphasize the most common hazard sources for fatal and nonfatal incidents and follow the ANSI Z.
INTRODUCTION
Occupational injuries were once accepted as an inherent element of hazardous work (Dedobbeleer and Beland 1991). Today a safety culture prevails with a focus on understanding why injuries occur, not merely accepting them (Griffin and Neal 2000). One reflection of this change is the replacement of the word “accident” with “incident” for unplanned and undesired events. Accidents are unplanned and uncontrolled events in which an action results in personal injury (Heinrich et al. 1980). The word “uncontrolled” implies the action is random and unavoidable (Salomone and Pons 2007). Incidents, while also unplanned, are avoidable and preventable. The U.S. Department of Labor, Occupational Safety & Health Administration (OSHA) recommends using the word “incident” for work-related events in which an injury occurred (OSHA 2015a).
Since the 1930s, compilations of arboricultural operation incidents have been published from personal observations, government statistics, and industry surveys. Regardless of the period or survey methodology, the most common incidents were contact with electric current, falls, and being struck by an object (Kiplinger 1938; Kummerling 1948; McGarry 1962; Wiatrowski 2005; Castillo and Menendez 2009; Ball and Vosberg 2010). While these authors separated incidents in broad event categories (e.g., contact with an object, falls), they did not identify specific hazards.
A hazard is any source—equipment, machinery, or activity—of potential harm to people (Oleske et al. 1989; OSHA 2012). Identifying specific hazard sources is essential to managing safety (Northwood et al. 2012). A beginning step to avoid, eliminate, or reduce hazards is identifying sources most often associated with injury, the frequency of these incidents, if the injuries are fatal or nonfatal, and the traumatic injuries common to the nonfatal incidents.
It is equally important to know what industry standards are already in place to avoid, eliminate, or reduce hazard sources. The ANSI Z133 American National Standard for Arboricultural Operations—Safety Requirements provides the safety criteria for arboricultural operations in the United States. Arboricultural industry representatives developed this consensus standard following the requirements of the American National Standards Institute. This publication, hereafter referred to as the ANSI Z, is reviewed and updated about every five years with the last completed in 2017.
The objective of this study was a review of arboricultural operation incidents, their specific hazard sources, relative frequency, and injuries. This review also included ANSI Z standards to avoid, eliminate, or reduce hazards associated with these incidents.
METHODS
The data collection and analysis were like the Grant and Hinze (2014) review of the OSHA Integrated Management Information System (IMIS) fatalities and catastrophe incident database for construction worker fatalities. Arboricultural operation incidents were extracted from this same all-industry database inclusive of 2001–2017. Each incident in the database has an investigation summary describing the events leading to the incident, hazard sources, and injuries.
The U.S. Department of Labor, Bureau of Labor Statistics (BLS) codes arborist occupational injuries into the Standard Industrial Classification (SIC) code, 0783, Ornamental Shrub and Tree Services. This code is within the North American Industry Classification System (NAICS) 56173, Landscape Services, which combines Ornamental Shrub and Tree Services with Lawn and Garden Services (SIC 0782) and Landscape Counseling and Planning (SIC 0781). Arboricultural operation incidents were identified in all three 078 SICs.
Arboricultural operation incidents reviewed in this study were not limited to NAICS 56173. There were also incidents found in NAICS 238910 (Site Preparation Contractors), and OSHA combines this NAICS code with Landscape Services for their reports of tree trimming and clearing incidents (OSHA 2015b). Some arboricultural operation incidents were also identified within NAICS 113310 (SIC 2411), Logging, and NAICS 2211 (SIC 4911), Electrical Services.
Since 2011, the BLS Injuries, Illness, and Fatality (IIF) program has used the Standard Occupational Classification (SOC) to code occupations. Data for this review was also obtained for SOC 37-3013, Tree Trimmers and Pruners, and SOC 37-3011, Landscapers and Grounds Workers.
The incidents analyzed in this study were cases for which inspections were conducted. The primary source was the OSHA IMIS database, which includes investigation summaries of occupational incidents. The National Institute for Occupational Safety and Health (NIOSH) Fatality Assessment and Control Evaluation (FACE) investigation reports were also reviewed along with state-based FACE reports. Incident reports and investigation summaries from these sources were compared to eliminate duplicates.
The OSHA IMIS database has detailed investigation summaries prepared by a compliance officer that includes a narrative of the incident along with injuries and other information. However, OSHA posts fewer inspections than the actual number of incidents they investigate. It has been estimated that only about one-third of the work-related fatalities are posted from the IMIS database (American Public Health Association 2015). In addition, OSHA state plan agencies that conduct investigations do not always contribute reports to the database.
The FACE program has detailed reports on the circumstances surrounding the incident. It is focused on research to determine trends and patterns and is not charged with enforcing compliance with any safety standards. The FACE reports, however, do discuss prevention strategies. NIOSH targets specific industries and activities, so their reports do not include all fatal incidents. States also voluntarily notify NIOSH of specific traumatic injury incidents. The FACE program focus is on fatal incidents and does not cover incidents that result in nonfatal injuries.
The two database sources, OSHA IMIS and the NIOSH FACE, do contain incidents with companies employing fewer than 11 workers, though most of the data comes from incidents in larger companies. The data includes only incidents that occur in private companies. There are also more fatalities investigated than ones involving nonfatal injuries, so direct comparisons cannot be made between the fatal and nonfatal incident data. While all these factors limit the applicability of the data for research purposes, it can still be useful for exploring trends and patterns.
The incidents compiled in this review were placed into BLS Occupational Injury/Illness event or exposure categories (BLS 2012; OSHA 2012). The event or exposure category, usually referred to as the event, is the manner in which the injury occurred. The arboricultural operation incidents were coded into one of six event or exposure categories:
Contact with Objects and Equipment (e.g., caught in a chipper, struck by a falling branch, struck by a falling tree);
Exposure to Harmful Substances or Environments (e.g., contact with electric current, hyperthermia);
Falls, Slips and Trips (e.g., fall from an aerial device, fall from a tree);
Fire and Explosions (e.g., gasoline vapor explosion);
Transportation Incidents (e.g., struck by passing traffic, struck by a train); and
Violence and Other Injuries by Persons or Animals (e.g., bee sting, drug overdose, intentional self-harm).
Within each event category, the data was placed into hazard sources based upon the narrative in the investigation summaries and reports. The traumatic injuries for nonfatal incidents were also noted. Only severe nonfatal injuries were included in this analysis. A severe nonfatal injury is defined as at least one-day in-patient hospitalization, work-related amputation, or loss of an eye (OSHA 2014).
The arboricultural operation incidents were analyzed according to these variables:
Event category (e.g., Contact with Objects and Equipment, Transportation Incidents);
Hazard source (e.g., electric current, falling tree);
Severity of injury (fatal or nonfatal); and
Traumatic injuries in nonfatal incidents (e.g., fracture, laceration).
Descriptive statistics were generated for these variables to obtain information on the hazard sources of arboricultural activities (e.g., brush chipping, manual tree felling) and the consequences of an incident (e.g., death, fractures). Comparative analysis was based on proportions, not rates, as there is no data on the number of workers performing specific activities. Unpaired t-tests were used to determine if there was a statistically significant difference between fatal and nonfatal fall heights for different sources and activities.
Since the incident reports for nonfatal are not included in all databases, the number of nonfatal incidents reviewed in this study are fewer than the number of fatal incidents. This may give the false impression that there are fewer nonfatal incidents than fatal incidents during arboricultural operations. The fatal and nonfatal incidents for specific hazard sources are identified as percentages of the respective number of incidents reviewed to provide a better comparison of the data among sources.
The number of incidents, as well as its percentage of fatal or nonfatal incidents, is identified for each hazard source (e.g., struck by falling branch). But the number of incidents related to specific activities (e.g., aerial device operator, climber, or groundworker struck by a falling branch) are given as fractions. This is done to highlight that these are approximations since the exact number of incidents is unknown.
The time period 2001 to 2017 spanned three ANSI Z editions, 2000, 2006, and 2012. The most recent ANSI Z edition was published in November 2017. This most recent edition, ANSI Z-2017, is referenced in the discussion. However, if a specific standard was absent or worded differently in any of the three previous editions, it was noted in the discussion.
RESULTS AND DISCUSSION
There were 865 fatal incidents reviewed in this study, about 51 per year, lower than the average number of fatal incidents per year identified in other studies. The BLS identified about 67 tree trimmer fatalities (SOC 37-3013) per year between 2003 and 2016 inclusive (BLS 2018). Other studies that reviewed BLS data during the past decade or two have found between 56 and 80 tree worker fatalities per year (Wiatrowski 2005; Buckley et al. 2008; Castillo and Menendez 2009). Our review has fewer incidents, as it includes only incidents in which there was a complete investigation summary to determine sources and activities.
There were 441 nonfatal incidents reviewed in this study. There are few published surveys of nonfatal incidents among any landscape services occupations. One study of trimming and pruning injuries among individuals including homeowners estimated more than 650,000 hospital emergency department visits between 1990 and 2007, with slightly more than 2% requiring hospitalization (Watson et al. 2012).
The incidents are discussed by event or exposure category beginning with the category with the most reviewed incidents, Contact with Objects and Equipment, to the least number of incidents, Fire and Explosions.
Contact with Objects and Equipment
There were 761 fatal contact incidents across all United States industries in 2016—almost 15% of the total occupational fatalities (BLS 2018). A falling object was the source for about 70% of contact incidents (BLS 2014). Across all industries, falling trees have been one of the most common sources for fatal contact incidents (Personick 1998).
In our study, there were 359 fatal (41.5%) and 189 nonfatal (42.8%) arborist incidents reviewed within this event category (Table 1). This is within the percentage range of the Contact fatal incidents obtained from the other surveys (Table 2). There were two subcategories within the event category Contact with Objects and Equipment that included arboricultural operation incidents. The subcategory Struck by Object or Equipment, in which the motion producing the contact is the source of the injury (e.g., falling tree or branch), accounted for most of the arboricultural operation fatal and nonfatal incidents. The second subcategory, Caught in or Compressed by Equipment or Objects, in which a worker’s body or body part was compressed or crushed in operating equipment or materials (e.g., pulled in a chipper), comprised the remainder.
Struck by Object or Equipment
There were 286 fatal (33.1%) and 146 nonfatal (33.1%) incidents in Struck by Object or Equipment subcategory. These included struck by falling, flying, rolling, or swinging objects.
Struck by a Falling Tree or Its Debris
Chain saw operators and other ground workers struck by a falling tree or its debris during manual tree felling operations accounted for 145 fatal (16.8%) and 24 nonfatal (5.5%) incidents reviewed in this study (Table 1). The incidents were almost equally divided between the chain saw operator and other ground workers.
There were 55 fatal (6.4%) and 5 nonfatal (1.1%) manual felling incidents to chain saw operators struck by the tree they were felling. Almost all these incidents occurred near the base of the tree. Chain saw operators remained standing near the tree or continued to cut as it fell and were struck when the tree kicked back off the stump or fell in an unintended direction. Chain saw operators must be moving along the retreat path once the felling cuts have been completed or the tree begins to fall (ANSI Z-2017 8.6.5.3).
There were 44 fatal (5.1%) and 6 nonfatal (1.4%) manual tree felling incidents where ground workers were struck by the falling tree. About half of these incidents involved ground workers struck by the falling tree while walking or dragging brush through the strike zone. A command and response system, where the chain saw operator calls “stand clear” or a similar command and waits for a response before making the back-cut, is mandatory (ANSI Z-2017 8.6.15). The incident summaries describing these incidents did not mention whether a command and response system was used, but its use may have alerted the ground worker that the back-cut was about to be made and the tree would be falling.
Another quarter of these incidents were to ground workers standing nearby awaiting further instructions. During manual felling operations, non-involved ground workers are to remain a distance from the tree at least two times its height (ANSI Z-2017 8.6.4) and this requirement had not changed from the previous editions. Either ground workers ignored these requirements or were unable to accurately judge the tree’s fall distance.
Slightly less than a quarter of these incidents occurred to involved workers, ground workers tending taglines, and those standing within the strike distance of the tree. Ground workers tending taglines must be positioned at a distance at least 1.5 times the tree’s height (ANSI Z-2017 8.6.4). This standard also appeared in the ANSI Z-2012 8.5.10, but only as an advisory recommendation and not as a mandatory requirement. The previous two editions, 2000 and 2006, set this distance as at least 1 times the tree’s height. Either workers were unable to judge tree heights and the appropriate distance or disregarded these requirements. Setting the distance at 1.5 times the tree’s height may reduce future incidents since workers have a greater margin of error in determining heights.
Getting struck by debris, branches, or treetops that snapped off the tree as it impacted the ground was the cause for 13 fatal (1.5%) and 3 nonfatal (0.7%) incidents. These were almost equally divided between chain saw operators and other ground workers. Adhering to the previously mentioned ANSI Z-2017 8.6.4 may reduce the number of these incidents. However, the incident summaries do not indicate the distance the debris flew before striking the worker.
Most of the remaining manual felling incidents, 22 fatalities (2.5%) and 8 nonfatal injuries (1.8%), affected chain saw operators and other ground workers struck by an adjacent tree that broke when hit by the tree being felled. These trees were in the strike zone of the tree being felled and the chain saw operator either misjudged the distance to the adjacent tree or did not control the direction of the fall.
There were also nine fatalities (1.0%) and two nonfatal (0.5%) injuries to chain saw operators struck by a tree that rolled as they were bucking it. There were also two incidents of ground workers fatally struck by a tree near the job site that collapsed without warning.
Struck by a Falling Branch
A ground worker, climber, or aerial device operator struck by a falling cut branch was the source for 104 fatal (12.0%) and 56 nonfatal (12.7%) incidents (Table 1). The most common incidents affected ground workers walking through or working in a drop zone, sometimes during chipping operations, who were struck by a free-falling branch, one not attached to a rigging line. This source was described in 54 fatal (6.3%) and 36 nonfatal (8.3%) branch struck-by incidents. Establishing communication with the arborist aloft can reduce the risk of a ground worker being struck by a falling branch. The ANSI Z-2017 8.3.2 and 8.5.11 require a command and response communication system be established and used during pruning and rigging operations.
Several of the branch struck-by incident summaries noted that “headache” was shouted by the chain saw operator before making the cut. Whether the operator waited for a response is not mentioned. Many of the fatal incident investigation summaries mentioned the deceased ground worker was not wearing a helmet and suffered a skull fracture. All workers must wear head protection (ANSI Z-2017 3.3.4).
There were 21 fatal (2.4%) and 12 nonfatal (2.7%) incidents to aerial workers struck by a branch they cut, with most involving climbers. The branches were often positioned above the climbers and the cut branch fell, striking them on the head or chest. These incidents resulted in skull or shoulder fractures and lacerations. A few similar incidents occurred to aerial device operators. There were also a couple of fatal and nonfatal incidents of workers standing on ladders and being struck by a branch they cut.
There were 29 fatal (3.4%) and 8 nonfatal (1.8%) struck-by incidents involving a controlled detached branch, one that was attached to a rigging line. Most of these incidents were to ground workers struck by a falling branch due to a failure in the rigging system, either the branch anchoring the block failed or the block was not properly secured. ANSI Z-2017 8.5.2 requires that rigging points be assessed. About a fourth of these incidents occurred to aerial device operators or climbers that were struck when they misjudged the swing of a cut branch on a line. ANSI Z-2000 9.4.2 recommended workers position themselves above or to the side of limbs being lowered. This language was changed to a requirement rather than a recommendation in ANSI 2006 8.4.15 and the subsequent edition. The 2006, 2012, and 2017 editions added language that the climber should also have a retreat/escape plan prepared. Incidents of climbers or aerial device operators struck by a rigged branch that swung into them occurred during each ANSI Z edition period.
Struck by a Line or Projectiles from a Chipper
Most chipper operation incidents were in the event subcategory Caught in or Compressed by Equipment or Objects, but some were placed within Struck by Object or Equipment. There were 13 fatal (1.5%) and 15 nonfatal (3.4%) incidents involving chipper operators struck by branches, lines (throwlines, climbing lines, rigging lines, or winch lines), or projectiles (hood cover or broken blades)(Table 1). About half of the fatal incidents and one-third of nonfatal incidents were to operators struck in the face or chest by a projectile, usually the hood, when they opened the access cover before the blades stopped rotating. ANSI Z-2017 8.7.11 prohibits opening hood covers while any part of the chipper is turning or moving.
The remaining chipper operation struck-by incidents were mostly to operators struck by lines, including winch lines, that were inadvertently pulled into the chipper with the brush. Several incidents involved a rope or winch line striking the chipper operator, causing their head to hit the side of the chipper. Workers are required to keep all lines clear of the chipping area (ANSI Z-2017 8.7.5).
There were a few chipper operation incidents where an operator was struck in the head by a branch that whipped to the side when fed into a disk chipper. These were nonfatal incidents that resulted in face lacerations. ANSI Z-2017 8.7.9 requires operators to hand-feed brush and logs from the side of the feed table center line and to step away from the table when the brush is taken into the rotor or rollers.
Struck by a Chain Saw
There were 13 fatal (1.5%) and 41 nonfatal (9.3%) incidents where a chain saw was identified as the hazard source. Chain saw injuries probably occurred in more arboricultural operation incidents but were ones that did not involve hospitalizations. Most individuals treated for chain saw injuries in hospital emergency departments are released the same day (Hammig and Jones 2015).
Climbers were involved in about two-thirds of the fatal and half the nonfatal chain saw incidents, with only a few incidents occurring to aerial device operators. Since the injuries in most of these incidents were lacerations to the left upper extremities, hands, or arms, the chain saws were most likely being operated with only the right hand. Several investigation summaries noted the worker was holding the branch with the left hand and the chain saw with the right.
ANSI Z-2000 7.2.7 mandated that hands encircle both handles with an exception if the employer demonstrated a greater hazard for two-hand operation in a particular situation. The exception did not apply to chain saws under 6.8 kg (15 lb) used in the tree. ANSI 2006 6.3.7 mandated the use of two hands on the saw with no exceptions. ANSI 2012 6.3.6 was like the 2000 standard, but specified that the left hand and thumb shall be wrapped around the forward handle and the right hand shall grip the rear handle except when the employer could demonstrate a greater hazard is posed by operating a saw in this manner in a specific situation. ANSI Z-2017 6.3.5 requires two-handed use at all times but with a subpart, 6.3.5.1, to the standard that allows for the switching of hands (right hand on forward handle, left hand on rear handle) if the employer demonstrates a greater hazard is posed by the more common configuration of the left hand on the forward handle and the right hand on the rear handle. Incidents of climbers and aerial device operators being struck by the chain saw were consistently reported during each ANSI Z edition.
Slightly less than a third of the fatal chain saw incidents and nearly half of the nonfatal were to ground workers. These occurred while chain saw operators were bucking and limbing fallen trees. Linear or rotational kickback of the chain was mentioned in some of the investigation summaries. The legs were the most common body part that experienced lacerations. Whether the chain saw operator was wearing cutresistant leg protection is not known for many of these incidents, but some investigation summaries mentioned the operator was not wearing any personal protective equipment (PPE). The ANSI Z-2017 3.3.8 requires that cut-resistant leg protection be worn when operating a chain saw on the ground.
Other Struck-by Incidents
The remaining fatal and nonfatal incidents in the Struck by Object or Equipment subcategory came from numerous sources. These involved workers struck by logs rolling off trucks, workers struck by unoccupied trucks rolling on the work site, workers struck by tree parts supported by a crane, and a worker lacerated by an electric hedge trimmer among others.
Caught in or Compressed by Equipment or Objects
There were 73 fatal (8.4%) and 43 nonfatal (9.8%) incidents that were the result of being crushed. These occurred mostly during chipping operations.
Pulled in a Chipper
Most chipper incidents were coded in this event subcategory and were the source for 34 fatal (3.9%) and 23 nonfatal (5.2%) arboricultural operation incidents reviewed in this study (Table 1). Almost all these incidents involved body parts entering the feed rollers, a pattern also noted by Zhu and Gelberg (2018) in their review of chipper operation fatalities. The worker entered the hopper hands-first in about half the fatal incidents reviewed in this study, feet-first in about one-third and the remainder were unknown as no one witnessed the incident. Most of the nonfatal incidents involved the loss of arms or hands from operators reaching in while pushing small brush into the hopper. ANSI Z-2017 8.7.8 and chipper manufacturer operating instructions prohibit placing hands and other body parts beyond the plane of the infeed hopper. Small branches shall be fed into chippers either mixed in with longer branches or pushed with a stick or tool designed for such purpose (ANSI Z-2017 8.7.14).
A few of the chipper pull-in incidents involved lines (throwlines, climbing lines, rigging lines, or winch lines) that wrapped around a worker when the line became entangled in brush and was fed into the rollers. As the line was pulled into the chipper, it wrapped around either the arm or neck of the chipper operator. These incidents resulted in amputations, fractures, or asphyxiation from being strangled by the line. The ANSI Z-2017 8.7.16 prohibits the winch line from passing beyond the plane of the infeed hopper and stowing this line before the brush is engaged by the feed wheels. ANSI Z-2017 8.7.5 requires that all other lines, for example, throwlines, climbing lines, and rigging lines, be clear of the chipping area during chipper operations.
Caught in Collapsing Palm Skirts
Palm skirts present a unique risk to climbers and were a source for 25 fatal (2.9%) and 3 nonfatal (0.7%) incidents reviewed by this study (Table 1). These incidents may be underreported, as Southern California alone may have as many as three or more a year (Jergler 2015). Most of these incidents reported by fire and rescue departments occurred to self-employed landscape gardeners.
As climbers cut the dead or detached fronds forming a skirt above them, the fronds may suddenly dislodge and collapse upon the worker. The weight of these detached interlacing fronds pushes down on the climber’s head, forcing the chin to the chest, which restricts breathing and results in restrictive (compression) asphyxiation. ANSI Z-2017 8.3.8 requires that skirts be removed from the top down and, when possible, the worker be positioned above the skirts. The three previous editions of the ANSI Z specified that skirts with three years or more of growth be removed from above.
Caught in a Stump Cutter Grinder Wheel
There were two fatal (0.2%) and nine nonfatal (2.1%) stump grinding incidents reviewed in this study (Table 1). Almost all involved the operator or assistant clearing debris away from the spinning cutting wheel. Most of these incidents happened while the operator was working alone, but a few were to assistants pushing the debris away from the moving cutting wheel. Lacerations or amputations of the foot or lower leg were the traumatic injury associated with most of the nonfatal incidents, with a few involving hand or finger amputations. The two fatal incidents were the result of severe hemorrhaging with the loss of the lower leg and foot when the operators were caught by the wheel. Stump cutters must have an enclosure or guard to protect the operator (ANSI Z-2017 5.5.2). ANSI Z-2017 5.5.8 requires that operators using remote controls maintain a safe working distance determined by the equipment or manufacturer. This is a new ANSI Z guideline.
Other Caught In
The remaining fatal and nonfatal incidents in the Caught in or Compressed by Equipment or Objects subcategory were associated with many other sources. These include fingers lacerated or severed by a wood splitter, workers crushed between the arms and cage of a wheeled or tracked material handler, and workers crushed between a truck dump box and frame, among other incidents.
Falls, Slips, and Trips
Incidents involving falls to the same level, slips, and trips are now incorporated into the Falls event category and the title expanded to reflect these new sources (Northwood et al. 2012). The majority of fatalities in this event category across all industries are falls to a lower level. Eight hundred forty-nine US workers suffered fatal falls during 2016, which was about 16% of the total fatal occupational injuries (BLS 2018). Falls from trees may account for almost 3% of all fatal falls in the United States (Braddee et al. 2000). Two hundred ninety-three fatal (33.9%) and two hundred six nonfatal (46.7%) incidents involving falls were reviewed (Table 1).
Falls to Lower Level
The subcategory Falls to Lower Level accounted for almost all the fatal and nonfatal arboricultural operation fall incidents. Falls to lower levels is where the motion of the person and impact were generated by gravity. Falls to Lower Level includes falls from a collapsing structure or equipment such as trees or nonmoving vehicle platforms. Falls to lower levels, typically from trees or aerial devices, comprised about one-third of all arborist fatalities in other studies (Table 2).
Climber Falls
Most of the fall incidents involved climbers. They were identified in 173 fatal (20.0%) and 136 nonfatal (30.1%) incidents. There were 56 fatal and 69 nonfatal falls from trees where the height was mentioned in the investigation summary (Table 3). The mean height of a fatal fall was 13.6 m and ranged from 4.6 m to 30.5 m. The mean height of a nonfatal fall was 6.4 m and ranged from 1.5 m to 22.8 m. There was a significant difference in height between fatal and nonfatal climber falls (P < 0.0001).
Thirty-one fatalities (3.6%) and twenty-seven nonfatal injuries (6.2%) affected climbers disconnecting from their climbing system. These frequently occurred while transferring from ascending to working in the tree or repositioning during the climb. Climbers must stay secured through another line, for example, a climbing line or work-positioning lanyard, while repositioning their climbing line (ANSI Z-2017 8.1.6). Several of these falls occurred when climbers were transferring from or to a ladder in the tree. Climbers shall be tied-in before working from or leaving the ladder to enter the tree (ANSI Z-2017 8.1.7). A few of these falls occurred after climbers were hoisted into the tree by a crane and then detached from the crane before being secured to the tree.
The failure of the climbing line anchor point resulted in 27 fatal (3.1%) and 29 nonfatal (6.6%) incidents. About two-thirds of these incidents involved the climber pruning off the branch or treetop that was the anchor. These most often occurred when the climber became preoccupied or distracted and forgot their climbing line anchor was on the part of the tree being cut. Fatigue is sometimes mentioned as a contributing factor in these investigation summaries. There were also a few incidents where a climber on gaffs/spurs had their work-positioning lanyards pull over the top of the spar.
The remaining anchor failures occurred as climbers were ascending a tree. There was a requirement that climbing lines be placed around the trunk using a branch union as a stop (ANSI Z 1994, 9.1.3). This was removed in the next edition (ANSI Z-2000). If the climbing line is isolated on the anchor branch, a failure means the climber will fall to the ground rather than a lower branch serving as a stop. The criteria for selecting and evaluating an anchor was not addressed in the ANSI Z-2000, 2006, or 2012. ANSI Z-2017 8.1.11 requires that the anchor point be inspected from the ground and, if suitable, subject to a load of about twice the weight of the climber. Some investigation summaries mentioned the crew tested the anchor by having two workers pull on the line before the climber began to ascend. In these failures, the climber often ascended 6 m or more before the branch broke. The two-worker load test may not be adequate, as it does not account for the repeated cyclic load as a climber ascends (Kane 2018a) or the amplitude of load on the branch by the climber during an ascent (Kane 2018b).
A climber falling when their climbing lines or work-positioning lanyards were severed by a saw or another line running over them resulted in 26 fatal (3.0%) and 25 nonfatal (5.7%) incidents. Climbers shall have two means of being secure, a climbing line and at least one other line, for example, a second climbing line or a work-positioning lanyard, and use both when deemed necessary (ANSI 2017 8.1.4). Climbers are also required to use two means of being secured when operating a chain saw in the tree (ANSI Z-2017 6.3.6).
There were 14 fatal (1.6%) and 10 nonfatal (2.3%) falls while climbing without using a climbing system. The ANSI Z-2017 8.1.6 requires climbers be secured while ascending a tree and to remain so until returning to the ground, and similar language was in the 2006 and 2012 editions. ANSI Z-2000 9.1.2 (2) allowed an unsecured ascent if the branches prevented the use of a climbing system. The ANSI Z 1994 9.1 exempted ascending from the requirement for the climber to be tied in. Ascending without being secured was a common practice in earlier decades (Thompson 1956).
There were two fatal (0.2%) and nine nonfatal (2.0%) incidents during which a climber came off the rappel on a figure eight or the rope terminal pulled through the climbing hitch. The ANSI Z-2017 8.1.13 requires a stopper knot be placed in the end of the climbing line when the climber is working at heights more than one-half the length of the line. Most of these falls occurred from less than 5 m.
Climbing gaffs/spurs kicking out were a source for one fatal (0.1%) and nine nonfatal (2.0%) fall incidents. The climbers fell when the gaffs/spurs kicked out and they slid to the ground with the lanyard still around the tree. These incidents occurred near the ground and generally resulted in nonfatal injuries, typically lower leg fractures or sprains.
Climbers also fell with a failing tree, and this source was identified in 41 fatal (4.7%) and 14 nonfatal (3.2%) incidents. Trees stressed by load changes while rigging large limbs or treetops typically failed either just beneath the climber or near the root collar. Arborists shall visually inspect the tree, including the root collar, before climbing or working in a tree (ANSI Z-2017 3.4.8). They are also required to assess the tree’s structural integrity before rigging (ANSI Z-2017 8.5.1). Either these assessments were not performed or the crew misjudged the tree’s structural integrity. Some investigation summaries mentioned that the crew thought the tree was sound or could withstand the rigging forces.
There were an additional 31 fatal (3.6%) and 13 nonfatal (2.9%) climber falls where the friction hitch loosened, the climbing line was caught by passing traffic and pulled the climber from the tree, and the climbing line wrapped around the climber’s neck, among other sources.
Aerial Device Operator Falls
Aerial device operators were identified in 93 fatal (10.7%) and 32 nonfatal (7.3%) fall incidents. The most common falls from aerial devices occurred when the operator fell out of the bucket (49 fatal, 5.8%; 9 nonfatal, 2.0%). Most of the fatal falls and about half the nonfatal falls from buckets occurred when a strike by a falling branch or tree jarred the boom and the operator was ejected. About a fourth of the fatal and nearly half of the remaining nonfatal incidents were to operators overreaching and falling out. The operators were either not wearing or using fall protection in all but a few of these fatal incidents. Operators are required to use an approved system of personal fall protection while aloft (ANSI Z-2017 5.2.7). The failure to use fall protection is a common factor in aerial device incidents across all industries (Pan et al. 2007).
There were 14 fatal and 5 nonfatal falls from aerial devices where the height was mentioned in the investigation summary (Table 3). The mean height of a fatal fall was 11.0 m and ranged from 3.0 m to 18.2 m. The mean height of a nonfatal fall was 7.3 m and ranged from 4.6 m to 10.7 m. There was no significant difference in height between fatal and nonfatal aerial device operator falls from buckets (P = 0.1065).
Falls from collapsing aerial devices due to mechanical or hydraulic failures resulted in 34 fatal (3.9%) and 16 nonfatal (3.6%) incidents. Aerial devices shall be inspected daily and an operational check completed before use (ANSI Z-2017 5.1.2). The investigation summaries did not give details on the boom failure, other than it was either mechanical or hydraulic. The investigation summaries also did not mention whether there were records of boom inspections. However, operational checks may have alerted the operator to problems before the failure. There is also the possibility that the aerial devices were not being properly maintained or had been overloaded during previous use, but this could not be determined from the investigation summaries.
There were 12 fatal and 5 nonfatal falls with a failing aerial device where the height was mentioned in the investigation summary (Table 3). The mean height of a fatal fall was 11.3 m and ranged from 6.1 m to 18.2 m. The mean height of a nonfatal fall was 8.8 m and ranged from 7.6 m to 10.7 m. There was no significant difference in height between fatal and nonfatal falls with a failing aerial device (P = 0.1965).
About a fourth of the fatal aerial device failures occurred while the boom was being used as a hoist and the joint load of the operator and cut branch either exceeded the rated capacity of the device, or the device was not designed for this use. Not all aerial devices are approved for use as cranes or hoists. ANSI Z-2017 5.2.3 prohibits the use of aerial devices as cranes or hoists unless they are specifically designed by the manufacturer for this task.
Aerial devices incidents also involved tip-overs, and this source resulted in 10 fatalities (1.1%) and 7 nonfatal (1.6%) incidents. Outriggers were not used or were set on unstable ground in these incidents. ANSI Z-2017 5.2.5 requires the use of pads. This is a change from the previous edition, ANSI Z-2012 5.2.6, where pads under the outrigger feet were only needed to provide firm footing. Aerial device operators may have misjudged the need for pads to improve the outrigger stability in the past, and this new requirement may decrease these incidents.
Ladder Falls
Ladders were a source for 4 fatal (0.5%) and 22 nonfatal (5.0%) incidents. Falls from ladders often occurred when the worker had the ladder struck by a falling branch. Ladder falls also resulted when a branch bend beneath the ladder sprung out as the load shifted, knocking the ladder off the tree.
There were three fatal and eight nonfatal falls from ladders where the height was mentioned in the investigation summary (Table 3). The mean height of a fatal fall was 8.6 m and ranged from 6.1 m to 12.1 m. The mean height of a nonfatal fall was 3.2 m and ranged from 1.8 m to 4.6 m. There was a significant difference in height between fatal and nonfatal falls from ladders (P = 0.0015).
Other Falls
There were 23 fatal (2.7%) and 11 nonfatal (2.5%) incidents involving falls from crane attachments, roofs, flatbeds, or aerial device cab protectors; falling out of the bucket while the aerial device vehicle was in motion; or falling while climbers practiced aerial rescue, among others.
Falls to Same Level
There were five nonfatal (1.1%) incidents of a worker tripping on the jobsite, either stepping in a divot made by the impact of a falling branch or tripping over debris. These resulted in sprains and, in one instance, a fractured ankle, where the workers were hospitalized overnight for observation. There are probably more tripping and slipping incidents, but these generally do not result in injuries requiring hospitalization.
Exposure to Harmful Substances or Environments
Exposure to Harmful Substances or Environments includes exposure to electricity, noise, inhaled substances, and environmental conditions (Northwood et al. 2012). There were 518 fatal incidents occurring across all industries within this category during 2016 (BLS 2018). This was about 10% of all fatal industrial incidents, with nearly half occurring from contact with an electric current. One hundred forty-four fatalities (16.6%) and twenty nonfatal (4.5%) arboricultural operation incidents were reviewed in this event category (Table 1).
Exposure to Electric Current
Arborists had about 1.5% of all work-related electrical injuries in a Texas 20-year study (Arnoldo et al. 2004). The Exposure to Electric Current subcategory had the highest number of injuries reviewed by this study, 135 fatal (15.6%) and 18 nonfatal (4.1%) incidents. Electrocution, a fatal electrical shock, is one of the most common mortality events during arboricultural operations, representing about 16% of the fatalities, though much fewer of the nonfatal ones (Table 2). Electrical contact incidents across all industries are disproportionately fatal (Cawley and Homce 2003).
Forty-six of the one hundred thirty-five electrocution investigation summaries identified the voltage (Table 4). Most of the electrocutions were on primary conductors with voltages between 5.1 kV to 15 kV. Five were from indirect contact with secondary conductors (120 volts). One electrocution involved indirect contact with a transmission line (138 kV) when a chain saw operator had the tree fall into the conductor.
Observing the minimum approach distance (MAD) from conductors reduces exposure to this hazard. The MAD is based on the nominal voltage and qualification of the worker. Elevation also is a factor for qualified line-clearance arborist MAD. The investigation summaries did not provide enough detail to determine the decedent’s qualifications for working near electrical conductors. However, 24 electrocutions occurred during line clearing operations where employers are required to certify that each employee is qualified to work within the proximity to electrical hazards (ANSI Z-2012 4.3.2).
There has been a change in ANSI Z-2017 from previous ANSI Z editions where there was only one category of worker qualified to work where an electrical hazard exists, the qualified line-clearance arborist (ANSI Z-2012 4.2.3). ANSI Z-2017 4.2 introduced a new category, the incidental line-clearance arborist. The difference between the qualified line-clearance arborist and the incidental line-clearance arborist is that the line-clearance arborist is performing the work for the utility as their employee or as an employee of the line-clearance contractor, whereas the incidental line-clearance arborist is working for the tree’s owner and the line is incidental to their work. The training requirements for these two categories are similar, though their MADs and work procedures differ.
Indirect Contact with an Energized Electrical Conductor
Indirect contact is made by touching a conductive object in contact with an energized electrical conductor. This was the most common source of contact and resulted in 106 fatal (12.3%) and 13 nonfatal (2.9%) incidents. These involved climbers (58 fatal, 6.7%; 9 nonfatal, 2.0%), ground workers (30 fatal, 3.5%; 0 nonfatal, 0.0%), and aerial device operators (18 fatal, 2.1%; 4 nonfatal, 0.9%).
The two most commonly described conductive objects in climbing incident investigation summaries were a pruner/saw and cut branches. The saw, usually a metal pole pruner, though occasionally a chain saw held by a climber, touched a primary conductor in 31 fatal (3.6%) and 2 nonfatal (0.5%) incidents. Indirect contact via a cut branch or frond falling onto the climber while also touching an energized conductor resulted in 23 fatal (2.7%) and 4 nonfatal (0.9%) incidents.
These two sources, pruner/saws and cut branches, were also common in indirect incidents involving aerial device operators. There were 11 fatal (1.3%) and 3 nonfatal (0.7%) incidents where the indirect contact was through metal pole pruners or chain saws. There were also seven fatal (0.8%) and one nonfatal (0.2%) incidents where the indirect contact was through a cut branch. The dielectric properties of these aerial devices were not mentioned in the investigation summaries.
Indirect contact through climbing lines or work-positioning lanyards in contact with a primary or secondary conductor resulted in four fatal (0.5%) and three nonfatal (0.7%) incidents. The ANSI Z-2017 4.1.11 states ropes that are wet, contaminated, or not insulated for the voltage involved shall not be used within MAD. Synthetic lines were once considered insulated and recommended for breaking contact between a worker and an electrical conductor (Abbott 1994). Since this review found climbing lines, work-positioning lanyards, and even taglines providing the indirect path for a fatal electric current, this ANSI Z standard may need to be reworded. We are not aware of any manufacturer of ropes commonly used in arboriculture that will attest to the dielectric properties of their ropes under field conditions.
The most common indirect contact electrocutions involving ground workers were touching metal pole pruners (9 fatal, 1.0%), aerial devices (7 fatal, 0.8%), fallen trees (5 fatal, 0.6%), ladders (4 fatal, 0.5%), taglines (3 fatal, 0.4%), and a crane (2 fatal, 0.5%) that was in contact with the conductor. The sources of the remaining indirect contacts to ground workers were unknown.
Direct Contact with an Energized Electrical Conductor
Direct contact is made when any part of the human body touches an energized conductor, and contact resulted in 29 fatal (3.4%) and 5 nonfatal (1.1%) incidents. Direct contact incidents most often involved aerial device operators (23 fatal, 2.7%; 2 nonfatal, 0.5%). The direct contacts were mostly from a falling branch or tree forcing the boom and operator into a primary conductor. In a few instances, the bucket controls became entangled in telephone lines or branches, causing the boom and bucket to move into a primary conductor.
Direct contact incidents also involved climbers and ground workers. Fatal direct contact to three (0.4%) ground workers was through primary conductors pulled off the poles when struck by a falling tree or branch. Direct contact to climbers was from an anchor failure causing the worker to swing into the primary conductor. This was the source for three (0.3%) fatal and three (0.7%) nonfatal incidents.
Exposure to Temperature Extremes
There were eight fatal (0.9%) heat stroke incidents in the Exposure to Temperature Extremes subcategory. These were ground workers raking, dragging brush, or chipping. Some investigation summaries mentioned this work was being performed during heat advisories when the air temperature was above 35 °C.
These heat stroke incidents may have been avoided if the crews were aware of the signs and symptoms of hyperthermia and gave first aid while calling 911 for emergency medical services. There was also a single incident of heat exhaustion requiring hospitalization, along with an incident of frostbite. The frostbite occurred to a climber stuck in a tree and suffering hypothermia while awaiting help to descend. While the ANSI Z-2017 3.2.3 requires employers to give instructions for preventive measures and first aid for poisonous plants and pests, environmental emergencies are not addressed.
Oxygen Deficiency
One fatality occurred in the subcategory Oxygen Deficiency. A chain saw operator drowned after wading into a pond to buck and limb a fallen tree. The fallen tree rolled off the stump and the branches trapped him under the water.
Transportation Incidents
Transportation Incidents accounted for about 40% of all fatal occupational incidents in the United States during 2016, with about half involving roadway collisions of motor vehicles (BLS 2018). Transportation incidents cover driving, either highway or non-highway, and injuries to drivers or occupants of a vehicle during working hours. Incidents while driving on public roads are within U.S. Department of Transportation jurisdiction, which preempt OSHA, so many driving incidents are not in the OSHA Fatality and Catastrophe Investigation database. This event category would likely have a higher percentage of arborist incidents if it included all transportation-related incidents.
Fifty-three fatalities (6.1%) and twenty-one nonfatal (4.8%) injuries were reviewed. The percentage of fatal incidents is like that reported in other studies (Table 2). Transportation incidents include flaggers and other workers struck by passing traffic or struck by vehicles driving on the work site (Northwood et al. 2012). However, if the vehicle is on-site, unoccupied, and rolls striking a worker, the incident is coded within the Contact with Objects and Equipment event subcategory.
Pedestrian Vehicular Incidents
The Pedestrian Vehicular Incidents subcategory includes workers struck by passing traffic. This was the source for 26 fatal (3.0%) and 11 nonfatal (2.5%) injuries. These involved flaggers directing traffic, workers setting out traffic cones, and workers struck by vehicles that drove into the work zone. Employers shall train workers in the usage and placement of traffic cones for temporary traffic control zones (ANSI Z-2017 3.5.2). Unfortunately, inattentive drivers may still drive through these control zones, and it is imperative that arborists working along roads during arboricultural operations keep a high situational awareness and not solely rely on cones and traffic devices to prevent vehicles from entering the work zone.
Seventeen fatal (2.0%) and six nonfatal (1.4%) incidents involved ground workers struck by company vehicles on the job site. About a third of these incidents occurred when a company truck backed over a worker who was fueling or working on a chain saw. Drivers must make a visual inspection before backing up and have a designated observer or the vehicle equipped with a reverse signal alarm audible above the ambient noise level (ANSI Z-2017 5.1.13). The remaining incidents to workers on the job site were to workers struck by wheeled or tracked material handlers.
Roadway Incidents Involving Motorized Land Vehicles
There were six fatal (0.7%) and three nonfatal (0.7%) incidents while driving on public roadways. One incident involved the driver falling asleep at the wheel and striking an oncoming vehicle. The driver, who was wearing the restraint system, survived, but the two passengers suffered fatal injuries when they were ejected from the truck. ANSI Z-2017 5.1.10 requires the use of the safety restraint system for drivers and passengers when the vehicle is in motion. The other incidents involved drivers striking vehicles, but no further details are known.
Rail Vehicle Incidents
There were two fatal (0.2%) incidents in the Rail Vehicle subcategory, which includes collisions with vehicles and pedestrians. One was a worker struck by a train as he walked on the tracks of a railroad bridge and another to a company truck driver who attempted to drive across the tracks before the train entered the intersection.
Non-roadway Incidents Involving Motorized Land Vehicles
There were two fatal (0.2%) and one nonfatal (0.2%) incidents of workers crushed by the wheeled or tracked material handlers they were operating when the vehicles rolled on steep slopes.
Violence and Other Injuries Caused by Persons or Animals
There were 866 fatal incidents in 2016 across all industries in the event category Violence and Other Injuries Caused by Persons or Animals, with about half involving homicide (BLS 2018). Employers must report all workplace homicides and assaults. However, these are not within OSHA jurisdiction, so they may be missing from the OSHA Fatality and Catastrophe Investigation database. There were eight fatalities (0.9%) and two nonfatal injuries (0.5%) in this event category (Table 1).
Intentional Injuries by Person
The Intentional Injuries by Person subcategory includes self-inflicted injuries, and there was one workplace suicide. The worker died of carbon monoxide poisoning by running the engine of a company truck in a garage bay after sealing the doors.
Unintentional Injury by Person
This event category also has a subcategory, Unintentional Injury by Person, and this subcategory includes drug overdoses. There were two deaths from cocaine toxicity while the workers were on the job site. There were several investigation summaries that mentioned the injured worker tested positive for cannabinoids or methamphetamines, but it was unclear if these drugs played a role in the incident.
Animal- and Insect-Related Incidents
Fatal insect stings were once placed within the event category Exposure to Harmful Substances or Environments, but these incidents were moved to this event category in 2010. There is about one death per year in the landscaping services industry involving insects (Pegula and Kato 2014). There were five fatal (0.6%) incidents in the Animal- and Insect-Related Incidents subcategory reviewed in this study. These workers died of anaphylactic shock from a bee or wasp sting. One worker was hospitalized after being bitten by a rattlesnake, and another had a finger amputated due to toxins from a spider bite.
Fire and Explosions
There were 88 fatal occupations incidents across all industries in this category during 2016 (BLS 2018). There were no fatalities reviewed in this event category, but there were three nonfatal (0.7%) injuries (Table 1). These incidents identified the ignition of gas vapors by a cigarette during chain saw fueling as the source. Two may have been avoided by following the requirement of no smoking while refueling (ANSI Z-2017 3.6.5). However, the ignition of gasoline vapors by a cigarette has not been proven in experiments (Marcus and Geiman 2013).
Occupational Illnesses
There were eight fatal (0.9%) occupational illnesses reported in the investigation summaries (Table 1). Six were heart attacks (acute myocardial infarctions), three of which occurred during lifting or moving brush while chipping, two while climbing, and another when a worker came upon the fragmented body of another worker who had been partially fed through a chipper. There were two deaths to an unknown illness.
CONCLUSIONS
There are limitations to the use of the incident databases, as some have minimal descriptions for the incident. They also do not include descriptions for all known incidents. There are also an unknown number of incidents that are not reported in these databases. These limitations are further compounded by the challenge of determining who is an arborist and what is an arboricultural operation. Regardless of these difficulties, the data is useful for looking at trends and patterns.
Arboricultural operation incidents differ from the all-industry percentages by event categories. Tree worker incidents, as a percentage, are more likely to be in the contact with an object or equipment, falls, and exposure to harmful substances or environment categories and less likely in the transportation and violence categories. Struck by falling trees and branches, pulled-in chippers, falls from/with aerial devices and trees, and electrocuted by indirect contact with an electric current are the dominant hazards that occupy the daily life of many arborists.
While fatal incidents in the arboriculture profession often draw attention, injuries requiring hospitalization should not be overlooked, as they can have lifelong consequences. Injured workers may not be able to return to work as production arborists due to the loss of mobility.
The sources for these nonfatal injuries differ from fatalities. While contact with electric current is one of the most common sources of fatal injuries, it is associated with relatively fewer hospitalizations. The most common hospitalization was a fall by a climber, and the traumatic injuries were often fractures and lacerations. Arborist safety training must not only focus on the leading sources for fatalities but also those associated with hospitalizations, as the two are not the same.
Arborists also need to be familiar with the American National Standard for Arboricultural Operations—Safety Requirements. Most incidents in this review may have been avoided had the ANSI Z been followed. Arboricultural operations job briefings and training should incorporate the ANSI Z and its appropriate sections to the task at hand. Unfortunately, ANSI Z compliance varies greatly within the arboriculture industry (Julius et al. 2014). While the ANSI Z needs to be periodically updated as our technology and techniques change, it appears that the greatest challenge is lack of compliance with the standard.
While the ANSI Z-2017 covers many of the hazard sources that resulted in incidents reviewed in this study, there are two that need to be addressed. The ANSI Z does not provide adequate guidelines for setting the branch anchor point or gauging its strength. The ANSI Z also does not address heat stroke. A paragraph noting the importance of judging weather conditions for the day’s activities during the job briefing may alert workers to this risk during heat advisories and to take proper precautions.
Footnotes
Conflicts of Interest:
The authors reported no conflicts of interest.
- © 2020, International Society of Arboriculture. All rights reserved.