 |
 |
 |
 |
 |
 |
 |
 |
|
|
Prepared For presentation at the 1997 Summer Meeting of the NATIONAL
INSTITUTE FOR FARM SAFETY, INC.
Indianapolis, Indiana
June 22-26, 1997
| JAMES M MEYERS University of California, Berkeley |
IRA JANOWITZ University of California, San Francisco |
| DIANA G TEJEDA University of California, Davis |
JOHN A MILES University of California, Davis |
| VICTOR DURAJ University of California, Davis |
JULIA FAUCETT University of California, San Francisco |
| JOHN KABASHIMA University of California, Irvine |
Abstract:
The University of California is conducting a three year intervention trial in cooperation with commercial nurseries to use engineering controls to reduce the incidence of musculoskeletal disorders among workers performing selected highly repetitive lifting and carrying tasks. Previously the UC team identified eighty-five such injuries involving 1246 lost work days among the cooperating pool of 1290 workers for the years 1993 and 1994. The team and cooperators then selected specific high risk jobs for intervention. A set of hand tools for carrying and moving plant containers were developed along with workstation redesign in one job. These engineering controls are now being subjected to a twelve month cooperative intervention trial, after which their effect on incidence of musculoskeletal disorders and related symptoms associated with these jobs will be evaluated.This paper reports on the impact the types of handles for carrying and moving plant containers and the one work station redesign have on specific ergonomic risk factors identified in these tasks and associated with the targeted disorders. Preliminary ergonomic data provided initial guidance to engineering control design parameters. Ergonomic data were collected using trunk inclination observation and protractor measurement; using the Lumbar Motion Monitor, a computer-based instrument which gives multiple dynamic measurements of back and upper torso movement; using a Chatillon dynamometer with a project-designed adapter to measure maximum voluntary grip capacity; and use of the 1993 NIOSH Lifting Equation to assess effects on repetitive lifting and lowering. All measures indicate large magnitude reductions in targeted risk factor exposures including, extreme stooped postures, repetitive finger pinch grip, and repetitive lifting and lowering of containers.
This project is supported by NIOSH Cooperative Agreement U05/CCU911435-01.
INTRODUCTION
Review of reported injury data for a ten year period for California agriculture by AgSafe (AgSafe, 1990) which suggested that sprain/strain type injuries and overexertion as a cause of injury were both of very high incidence across all crops and commodities. In response a multi-disciplinary team of UC researchers determined to apply ergonomics methods to the analysis and prevention of these injuries. Plant nurseries were selected as an initial intervention site because they offer a stable workforce, are a significant and growing sector of the California agricultural economy, and share some characteristics with both field agriculture and manufacturing industries. A review of data on injuries (AgSafe, 1990) showed that nurseries share with agriculture generally, high rates of sprain and strain injuries (48.9% of reported injuries) and over-exertion as a cause of injury (30.2% of reported injuries.
Three nursery companies are formally cooperating in the project. All three specialize in container grown outdoor bedding and ornamental plants, primarily for delivery to retail nurseries. These cooperators are all large operations by industry standards, and between them account for some 1290 employees at the involved worksites. This industry is almost completely non-union in California, and there is no active union representation at any of the cooperator sites. The majority of workers in these operations are Spanish-speaking, from Mexico or elsewhere in Central America. They earn an average of about $5.00 per hour. Nursery work is considered a comparatively good job by most California agricultural workers because it is year-round, relatively well-paid, and at these sites includes health benefits. All of these cooperators have active and well-planned injury and illness prevention programs on-site. Provision of worker's compensation insurance benefits is required in California.
HIGH RISK JOB TASKS
In order to identify high risk job tasks, cooperators injury and first aid records were reviewed, all jobs were described and screened for ergonomic risk factors using a checksheet method, and workers and supervisors were asked to identify difficult job tasks. Together with cooperators the project team selected job tasks identified as involving high risk for musculoskeletal injury for intervention. Among high priority job tasks were:
It should be obvious that in a fully cooperative project, all interventions must have the full approval of both workers and employers if they are to be given fair and full trial. Once priority job tasks for intervention were agreed upon, design constraints for intervention development were agreed upon. Among design constraints considered were the following:
At first, the handles utilized a hand grip set at a 90 degree angle from the stem. This type of handle is still used for ground-to-ground movement of containers and is included in the intervention trial. However, over time it was realized that the 90 degree handles forced workers moving containers from trailer-to-ground or ground-to-trailer to employ very awkward upward reaches with the forearms and shoulders to reach the trailers. Again, a variety of approaches were explored. The most successful and the approach selected for intervention trial employs a smooth curved handle which affords either a low grip when containers are elevated or a high grip when they are on the ground. This handle allows for much more natural forearm and shoulder positions when lifting containers to or from trailers. When lifting from a trailer, the handle is gripped low and then the weight of the container slides the handle through the hand until it reaches a stop at the handle end. Thus, containers can be placed on the ground without stooping.
Other primary handle designs included in the intervention trial are multi-container handles used for ground-to-ground spacing and for loading trailers at the canning line (where plants are inserted into containers). For ground-to-ground spacing handles are designed which pick up every other container so that they can be set down in the new spacing arrangement with out further manual handling. Multi-container handles are also used in combination with a work-station redesign scheme for loading trailers at one nursery. Where workers used to stand at ground level and extend containers with long physical reaches, they now stand on the trailers themselves (augmented by catwalks between the trailer and the canning line) and use the handles to move containers from the line to the trailer.
ERGONOMICS DATA COLLECTION METHODS
There is always some difficulty in applying research instrumentation in a real working environment. At minimum, neither workers nor their supervisors want work inhibited, and doing so would distort results. While a variety of explanatory ergonomics data on the interventions under trial are being collected, the focus is on those ergonomic improvements which the interventions are targeted on. Those are:
Extremes of observed trunk inclination were measured by using a protractor on still photographs of workers at extreme stooping moments both with and without using handles. The same workers were used in both measurement conditions keep comparisons consistent with individual differences. With the protractor in place a line was drawn from the hips through the shoulder. Observed trunk inclination is a combination of trunk flexion and hip flexion and is limited to static measurement. Dynamic postural data are being collected via the Lumbar Motion Monitor, but have not been fully analyzed at this time.
The Lumbar Motion Monitor (Marras, Sudhakar, Lavender. 1989). also uses the dynamic data it collects to calculate the formula for the revised NIOSH Lifting Equation. This calculation is used to assess tool effects on load lifting capacity. The product of the NIOSH equation is the "recommended weight limit" or RWL. The RWL is defined as the load weight that most healthy workers could sustain for a period of up to 8 hours without increased risk of lower back pain. The RWL is calculated via the formula:
Four workers were evaluated using the NIOSH Lifting Equation (Waters, et al, 1993), two using 5 gallon tools and two using 1 gallon tools. Where a worker's performance was evaluated more than once, results were averaged. Workers lifted containers from a trailer and placed them on the ground. This is the most common and most difficult field container movement. Vertical and horizontal distances varied somewhat due to worker anthropometry, but the trailers are approximately 3 feet high. There was no asymmetry involved. Lift frequency was standardized for all workers at 4.6 per minute for both workers using the 5 gallon tools. However, the workers using the 1 gallon tools did not have extensive experience with them and as a result their frequencies with the tools were less than without (3:2.3 and 3.5:2.8). Duration was set for an 8 hour shift. Object coupling was designated as poor without the tool and good with the tool.
Finally, there was need to compare the ergonomic effects of the handles on the hands in shifting from a finger pinch grip to a full-hand power grip on the handle. In the pre-intervention condition, workers routinely lift three 1 gallon containers or one 5 gallon container in each hand using a finger pinch grip. The intervention calls for use of one of three handles which eliminate the pinch grip and substitute a hand or power grip.
Because of the uniqueness of the gripping task, no standardized instrument was available for this measurement. Three adapters were developed to assist in the comparison of carrying forces for pinch grips versus hand grips. All utilized a Chatillon force measurement dynamometer to measure force exerted. The dynamometer was connected to a plywood platform on which the workers stood while lifting plant containers.
Two adapters utilized rods through the sidewall drainage holes of one 5 gallon and three I gallon containers. A ring was attached to the rod(s) which allowed for connection to the Chatillon dynamometer. Workers could then grasp three I gallon or one 5 gallon plant container with a finger pinch grip as is done normally and then pull with full force against the resistance of the dynamometer to give a measurement of exerted force. The ring is the attachment point and when experiencing a load positions itself directly under the pinch grip. As increasing loads were applied to the attachment point little, if any, rotation was induced and meaningful force measurements were obtained.
The third adapter utilized the same Chatillon dynamometer and platform, but linked to a 5 gallon handle. This handle was designed with the capability to accept all three hand grips included in the trial. The handle base has a multi-holed bar which provides for varying attachment points for connection to the dynamometer. This is to accommodate the changing center of gravity induced by the different hand grips. The attachment point is adjusted by repositioning an eye-bolt and wing nut into the appropriate hole.
Nine workers were tested lifting one 5 gallon container with and without tools using this system. Because two workers did not customarily handle the smaller containers, only seven workers data was included for lifting three I gallon containers. Results were measured in pounds of force by the Chatillon dynamometer. These data were then converted into a maximum voluntary grip ratio (MVG) by dividing the object weight by the maximum upward pull force. These ratios were the used to calculate comparative percentages in MVG for the different conditions. The determination of overall actual-to-maximum lift force ratios for the pre- and post-intervention task requirements are expressed as percentage of grip capability.
RESULTS
Results from focused measurements of ergonomic effects of interventions subjected to trial are very positive. The tool and work station designs are significantly reducing the ergonomics risk factors targeted.
Postural data collected via protractor show important differences in observed trunk inclination between conditions of tool use and non-use. Two workers demonstrating very different lifting styles were observed and documented. Worker A employed an extreme stooped posture to place containers on the ground. Worker B employed a very upright posture, counterbalanced by one leg, when placing containers on the ground. These subjects postures were examples of opposite extreme endpoints in observed lifting styles, with most workers approaches falling in between.
Worker A showed the most improvement, a reduction of 38 degrees in observed trunk inclination when using the tools to place containers on the ground. Worker B showed less improvement, a reduction of 17 degrees in observed trunk inclination when using tools.
| Subject | Without Tool (degrees) | With tool (degrees) | Reduction (degrees) |
|---|---|---|---|
| Worker A | 84 | 46 | 38 |
| Worker B | 62 | 45 | 17 |
Measurement of maximum voluntary grip capacity (MVG) with and without
tools also showed important differences. Workers realized a 52% reduction
in required MVG when using the handle to lift one 5 gallon container (wt.
with adapter- 13.54#). Said another way, workers had to exert more than
twice the MVG effort when not using the tool. Workers realized a 78%
reduction in required MVG when using the handle to lift three I gallon
containers (wt. with adapter-9.4#). Workers had to exert more than 4.5
times the MVG when not using the tool to make the same lift. As expected,
the tools enabled the workers to exert a maximum voluntary grip force
measured at some 8 times that of the unaided finger pinch on three I
gallon containers and some 3.5 times that of the most common unaided
finger pinch on one 5 gallon container.
| Average Maximum Voluntary Grip Force Exerted (lbs.) | ||
|---|---|---|
| without tool | with tool | |
| (3) 1 gallon containers | 14.96 | 120.48 |
| (1) 5 gallon containers | 34 (palm down) 44 (palm up) |
120.48 |
| Maximum Voluntary Grip (MVG) Ratio | ||
| without tool | with tool | |
| (3) 1 gallon containers | 0.414 | 0.091 |
| (1) 5 gallon container | 0.247 | 0.118 |
| Percent of MVG Capacity Required Using Tool (tool compared to non-tool) |
||
| with tool | ||
| (3) 1 gallon containers | 21.98% | |
| (1) 5 gallon container | 48% | |
| Percent Reduction of MVG Capacity Required Using
Tool (tool compared to non-tool) |
||
| with tool | ||
| (3) 1 gallon containers | 78.02% | |
| (1) 5 gallon container | 52.39% | |
Four workers were evaluated using the NIOSH Lifting Equation, two using 5
gallon tools and two using 1 gallon tools. Where practical, worker
performance was evaluated twice to increase confidence. Results
consistently yielded an increased Recommended Weight Limit when the tools
were employed.
| Container | Worker | RWL | No Tool | With Tool | % RWL Increase |
|---|---|---|---|---|---|
| 5 gallon | 1 | Origin Destination |
8.89 8.42 |
12.21 12.94 |
37.3% 53.71% |
| 2 | Origin Destination |
7.74 7.01 |
12.60 11.20 |
62.65% 59.63% |
|
| 1 gallon | 3 | Origin Destination |
9.49212 9.5931 |
22.5365 15.8523 |
137.4% 65.25% |
| 4 | Origin Destination |
12.8486 8.34978 |
19.8166 13.4696 |
54.23% 61.3% |
DISCUSSION
Field measurement of ergonomics risk factors is fraught with difficulty. At minimum, neither workers nor employers want work disrupted. In agricultural settings, job tasks often change quickly and sometimes unpredictably. Finally, in a complex work setting there are any number of complicating or intervening elements or occurrences with potential for affecting data collection. In most cases, we have found standardized instruments such as the Lumbar Motion Monitor (Marras, Sudhakar, Lavender. 1989). to be acceptable and effective in the nursery workplace. It does require removing a worker from regular work for a period but does allow for performance of the targeted job task in the field setting. Data used to calculate the NIOSH Lifting Guideline (Waters, et al, 1993) as reported here, were collected via the Lumbar Motion Monitor.
Perhaps most problematic with the NIOSH Lifting Guideline data is the fact that the workers using the 1 gallon tools did not have extensive experience with them, which resulted in reduced frequency of repetition with tools than without. Also, because of data collection difficulties, no angle was added for task asymmetry. In fact, in the field these tasks do sometimes involve asymmetry. Subsequent data collection with the Lumbar Motion Monitor in 1997 should correct these deficiencies. However, review of the input data does show that horizontal distances were reduced by the tools, which is an important factor in the calculation of the RWL. So, deficiencies aside, it is clear that the tools are having a large effect on a task performance variable critical to RWL calculation.
Calculation of the maximum voluntary grip as performed here, is subject to some concern. First, since it is not a standardized measure, there are no accepted base data for comparison. Second, despite care in design of the coupling of the objects lifted to the dynamometer there is still potential for error, especially with reference to possible container rotation and resulting torque. Finally, it must be noted that this remains a measurement of psychophysical grip exertion, not direct measurement of finger compression force. That is, workers were asked to exert their maximum lift force under two conditions (i.e., with and without handles) for comparison. This approach is potentially subject to each worker's perception of stress and strength. Though it may have been more appropriate to use finger compression forces, especially with the pinch grips, the substantially different nature of biomechanical forces of the pinch grip and hand, or power, grip would have required a much more complex and detailed approach. Design and development of such instrumentation was beyond the scope and resources of this project. Subsequent investigations may provide opportunity to develop the more complex instrumentation required for such direct measurement of finger forces. The psychophysical grip methodology employed does provide for practical determination of overall actual-to-maximum lift force ratios for the pre- and post-intervention task requirements, expressed as percentage of grip capability. Additionally, the approach of calculating a percentage of each worker's own voluntary capability helps control for individual variations.
Use of a protractor to calculate observed postures is an accepted field measurement method. As employed here it is subject to some concerns. First, is acceptance that the photographs used did catch the intended moments of extreme trunk posture in task performance. Second, this is a static measurement which combines two different factors; flexion of the hip and of the lumbar spine, expressed together as trunk inclination. Third there is necessarily some error in calculation of the exact degree of inclination. Practitioners generally assume that all other concerns aside a three to five degree error is possible with this method. However, this is a practical and cost effective approach to field measurement and with the use of photographs is documentable and replicable. Finally, the large differences observed here give confidence that even allowing for potential error of three to five degrees, large and important changes in trunk inclination are taking place. Here, again, subsequent planned data collection using the full capabilities of the Lumbar Motion Monitor will add preciseness of detail and include consideration of dynamic motion.
It must be reiterated that despite concern for potential sources of error in methods employed, the large differences in results in each area for task performance with and without tools inspires confidence that a measurable effect is occurring as described.
CONCLUSIONS
This project has three primary goals. First to demonstrate that ergonomics methods and approaches have practical application to agricultural work. Second, demonstrate that ergonomics risk factors identified can be targeted and either eliminated or significantly reduced through engineering intervention acceptable to workers and employers. And, finally, to demonstrate that reduction of exposure to selected ergonomics risk factors will result in reduced incidence of musculoskeletal disorders and symptoms.
The first goal was readily achieved in the practical and useful description of extant ergonomics risk factors associated with nursery jobs and the productive use of this information to design tools to correct the deficiencies observed. The second goal is addressed by this paper. The third goal will be addressed in the project's final reports in which final data on MSD incidence and symptoms are analyzed and reported.
While the data reported here are largely preliminary and will be supplemented and augmented by further data collection and analysis still underway, they do show persuasive evidence of the impact of the designed intervention tools and strategies on the tar geted risk factors.
Repetitive or sustained stooping is a serious risk factor for chronic back injury which is present in much agricultural field work. Stooped posture or trunk inclination is reduced in all nursery workers using the tools. The amount of trunk inclination reduction is large, ranging from 38 degrees to 17 degrees. In virtually all cases, the maximum trunk inclination required when using the handle tools is in the range of 45 degrees. In an industry where "stooped" posture can reach or exceed 90 degrees this is important progress.
One of the nursery industry's most serious and unique ergonomic risk factors is the tight finger pinch grip required to handle plant containers. Use of the handle tools completely eliminates this grip, substituting a whole hand "power" grip. While finger pinch is eliminated, the hand must still exert considerable force to grip the tool handle. However, as is demonstrated in the analysis, maximum voluntary grip capacity exertion as measured here, is reduced by large amounts (78% on three I gallon containers, and 52% on one 5 gallon container). These are important improvements with respect to hand, arm, and other upper extremity MSDS. The development of the curved handle is important in extending practical use of the handle tools to the task of unloading trailers in the field. This is a continuous and high frequency task and the use of the handle tool in it means involved workers are relieved of important risk factors for hand, arm and upper extremity MSDS. Other variations of the handle tool further extended the risk factor exposure reduction to workers performing ground-to-ground spacing and loading of trailers at the canning line.
The other important ergonomic risk factor addressed is the nature and loading of the repetitive lifting (and lowering) movements involved in moving plant containers. Results from the NIOSH Lifting Equation suggest that lifts are indeed much improved. As the equation's Recommended Weight Limit increases for the changed conditions created by the handle tools, the effective body loading of the lift is reduced. Further, the magnitude of the RWL increases indicate important changes in related risk exposure, ranging from 37% to 137%. These are very large magnitude changes and are achieved without changing the weights involved, without changing the workstation layout for most tasks, without changing the shift period, and without changing work pace.
All of these ergonomics risk factor reductions are important, both individually and in combination. However, for this work to have an industry-wide impact it is important that the interventions be low-cost, fit with predominant work practices, and at least not reduce current productivity levels. The handle tools appear to meet all of these added constraints. They are of low complexity and involve little material or production costs. They fit well with current container handling practices and current equipment investments. Early observations of productivity suggest that they have small to moderate increases in productivity once workers become familiar with them. Finally, and perhaps most importantly for eventual field impact is the fact that workers like them. While their effective use does require a short period of familiarization (1-2 weeks), once achieved, workers report clear preference for handling containers with the tools.
These are important results. They tell us that agricultural field jobs which involve serious ergonomics risk factors for musculoskeletal disorders can be effectively addressed using accepted ergonomics approaches. Further, they suggest an under-realized opportunity for intervention in these jobs using small tools. For the past several decades, engineering development in agriculture has concentrated on large machines, leaving small tools used throughout the industry largely untouched and unconsidered. It is time to take another look at many of the jobs and tasks which are routine in agriculture and which are largely taken for granted as immutable.
REFERENCES