Journal of Community Health Vol. 17, No. 4, August 1992
WORK-RELATED INJURIES OF THE HAND" DATA FROM AN OCCUPATIONAL INJURY/ ILLNESS SURVEILLANCE SYSTEM Denise M. Oleske, PhD; Jerome J. Hahn, MD
ABSTRACT: Administrative and clinical data from a network of occupational medicine clinics were combined to evaluate the utility of these data in the surveillance o f non-fatal occupational injuries. Incident cases of work-related h a n d injuries were characterized to evaluate that process. In 1988, h a n d and finger injuries were found to be a m o n g the most c o m m o n (n = 4,120) o f all occupational injuries recorded in the system. H a n d / f i n g e r injuries accounted for 30.0 percent o f all episodes o f work-related injuries treated, with the incidence of these decreasing with increasing company size. H a n d injuries were found to be potentially severe with nearly 20 percent resulting from a crushing motion and nearly 10 percent being fractures or amputations. Hands being caught in machines or struck by metal items or h a n d tools accounted for 36.2 percent of the injuries. A surveillance system based u p o n ambulatory care data can be a feasible method for identifying priority areas for the prevention o f work-related injuries.
Injuries account for over 90% of all the reported work-related health problems in the United States. ~ In 1983, 4.7 million job-related injuries occurred in the private sector, representing rate of 7.5 injuries per 100 full-time workers) By 1989, 7.8 million job-related injuries were reported, yielding a rate of 8.6 per 100. 3 Noteworthy increases in the rate of disabling occupational injuries have occurred, resulting in an increase in the lost workdays rate of 57.2 per 100 in 1983 to 78.7 per 100 in 1989. 3 Rises in the rate of fractures, strains/sprains, cuts, lacerations and punctures have also been documented during this time. 4 Yet, while it is recognized that effective injury surveillance systems are needed for the planning and evaluation of injury control programs, "a standardized case-ascertainment or reporting system tor occupational Denise M. Oleske, PhD, is Assistant Professor in the Department of Health Systems Management and the Department of Preventive Medicine, Rush University, Chicago, IL; Jerome J. Hahn, MD, is Medical Director, Rush Occupational Health Centers, Chicago, IL. The research described herein was supported by a grant from the Centers for Disease Control #R49CCR502360 "Occupational Injury Surveillance by Health Centers." Requests for reprints should be addressed to: Denise M. Oleske, PhD, Department of Health Systems Management, Rush University, 1653 W. Congress Parkway, Chicago, IL 60612. © 1992 Human Sciences Press, Inc.
205
206
JOURNAL OF COMMUNITY HEALTH
injuries does not exist in the U.S. TM A National Academy of Sciences Committee on National Statistics reported in 1987 that the Bureau of Labor Statistics (BLS) data are inadequate for providing the Occupational Safety and Health Administration the data that it needs as the data collected by the BLS are seriously underreported? Underreporting is not unique to the US. Sass echoes this concern in Canada, and further remarks that incorrect theories about workplace injuries handicap efforts to reform workplace safety. 7 Current sources of work-related injury data have also been criticized because of the lack of uniformity in classifying cases with respect to their work-relatedness. 6 In the effort to control the incidence of injury/illness in the workplace, particularly those of a disabling nature, whose rates have been steadily increasing, high quality surveillance data are required. Recognizing the limitations associated with existing data sources, it is important to investigate alternative approaches to the surveillance and control of work-related conditions. Occupational medicine clinics have been identified as potential sources of occupational health data useful for surveillance purposes? 9 Data derived from these clinics could help to address the reported need for injury surveillance at the local level, ~°'~l particularly with regard to the non-fatal injury experience of small firms--a group which presents a significant challenge to health professionals. I'~ Under a grant from the Centers for Disease Control (No. R49CCR502360), the Rush-Presbyterian-St. Luke's Occupational Health Centers (ROHC) in Chicago have developed a model for the surveillance of occupational injuries by occupational medicine clinics. In evaluating the utility of a core data set prior to its implementation at five occupational health centers, Oleske, et al., utilized the core data set to identify that manual material handling practices were the probable causative factors contributing to the excessive injury rates, particularly of strain/sprains, in an automotive parts manufacturing company.L3 With interventions based upon knowledge of risk patterns, injury rates were significantly reduced and the company received a substantial rebate in its insurance premiums. 14 We illustrate the application of the system utilizing non-fatal work-related hand injuries as a model. These were among the most common work-related injuries identified through an occupational injury/illness surveillance system operated by a network of occupational health centers? 5 Most of the current existing information concerning risk factors for hand and finger injuries is of limited value being derived from case studies, small clinical series, and surveys of injured workers with low participation rates. A review of the literature has implicated a number of micro-environmental factors in the etiology of this injury category:
Denise M. Oleske and Jerome J. Hahn
207
high force, repetitive motions, hydrofluoric acid, vibration, mechanical power presses improper tool design (e.g., lack of shock absorbency, grippability, weight, etc.), handling debris (e.g., glass, sharp objects, etc.) and the handling of heavy objects or heavy power tools such as chain saws and air hammers. ~6-25Occupations at risk are: cash-register workers, assembly line packers, forklift operators, investment casting workers, corn-pickers, machine operators, mechanics, butchers, sheet metal workers, textile workers, woodworkers, pneumatic chipping hammer operators, cabinet makers, carpenters, warehouse workers, and laborers. 252~ The source of injury category to which severe hand and finger injuries are most often attributable is fixed machinery, specifically, table saws and presses (other than printing)? 3 In a Bureau of Labor Statistics (BLS) survey, hand and finger injuries most commonly occurred with fixed machinery when hands/fingers were either caught in, under, or between machinery or objects, hit against moving machine parts, or struck by moving machine parts. 23'~ When machinery has been attributed to hand or finger injuries, less than 15% have been attributed to accidental activation? ° NIOSH special investigations have provided insight into how detailed operations of the job task may be associated with hand or finger injuries such as attempting to correct placement of pieces after the downstroke of a punch press has been initiated. 31 In addition to these micro-environmental factors, macro-environmental factors for work-related hand injuries have been identified. These include stress and tension in the workplace produced by machine pacing, production standards, incentive systems, 32 and small company size? ° Human risk factors for work-related hand injuries which have been identified are: age less than 25 years, nine or more hours of sleep in previous night, tobacco use, presence of cardiovascular disease, longer than average working day, less than one year of experience on the job, inadequate training for the task, and performance of task not part of usual job. 2~-~4'31'~3However, in a study of amputation cases, 97 percent of which were to the hand or fingers, 60.4 percent were performing routine tasks? ~ There is a growing body of evidence to suggest that employee knowledge, attitudes, and beliefs about safety and safety practices may play a major role in the likelihood of the occurrence of a work-related injury. In a BLS survey of workers with hand/finger injuries, the most commonly cited categories of conditions or events felt to have contributed to the injury were: an increased work pace, lack of awareness of time, distance, or severity of potential hazard, and sudden or unintended movement of work materials, tools, equipment or the hand itself. With respect to other behavioral risk factors, only 34 percent of the injured workers responding to the survey received informa-
208
JOURNAL OF COMMUNITY HEALTH
tion r e g a r d i n g the use of gloves or other h a n d protection on the job; only 26 percent reported use of h a n d protection at the time of the injury. O f those who reported having used some form of h a n d protection at the time of the injury, 85 percent reported that the seriousness of the incident was r e d u c e d by the protection. Although these data are useful, a major limitation of the BLS surveys are the low participation rates, with only 56 percent of workers responding to a survey of h a n d injuries. 29 Non-use of gloves was f o u n d in another study to be associated with a 2.6 fold increase in h a n d injury. '~7 In a series of 150 persons with work-related h a n d injuries evaluated in two "Accident and Emergency Units" in the U.K., only 40 percent of workers could report why their injury had o c c u r r e d ? ~ Given the frequency of h a n d and finger injuries, we sought to characterize work-related h a n d injuries in accordance with a model of occupational injury causation we had previously described. ~3 T h e computerized combining of clinical with administrative data obtained in an ambulatory care setting could facilitate the identification of additional risk factors and their distribution for background information in planning etiologic studies. This system may also be a vehicle for reducing the likelihood of work-related injury t h r o u g h the dissemination of information in a timely m a n n e r about potential risk factors. MATERIALS AND METHODS
The Rush-Presbyterian-St. Luke's occupational Health Centers (RPSLOHC) includes five outpatient occupational medicine clinics located throughout the Chicago metropolitan area. A total of 2,342 private sector companies, representing a workforce of approximately 200,000 persons, participate in the surveillance system. The Centers treat cases requiring urgent or episodic care. During 1988, the Centers saw a total of 15,374 new patients, 13,616 of which were from private sector companies, for work-related problems. Of the total number of cases, 92.1 percent were for treatment of a work-related injury. All individuals who receive care for a work-related injury or illness at any one of the five clinics are included in the occupational injury/illness surveillance system. For the purposes of this study, only injuries to the hand or finger (ANSI Body Part Nos. 330 and 340) which were from private sector companies and were evaluated by RPSLOHC physicians are reported herein. Criteria used to identify a work-related case are those published by the U.S. Department of labor. ~ Injury/illness information is coded in accordance with the American National Standards Institute (ANSI) Z-16.2 standard (1969 revisions)~5 with enhancements by the Bureau of Labor Statistics Supplementary Data System?6 RPSLOHC physicians and nurses are trained in the use of
Denise M. Oieske and Jerome j. Hahn
209
this system. For each new work-related injury/illness, physicians code the nature of injury and body part affected, nursing staff code the source of injury and injury type (mechanism of injury) using the ANSI system. Quality assurance procedure include the training of new clinic personnel, manual and computer checks of entered data, monthly review of a 6 percent random sample of charts for coding accuracy and completeness, and periodic in-service training programs for all clinic staff. The entry of initial treatment, and follow-up care, and medical charge are computer entered at the conclusion of each visit. Standardized reports are produced for each clinic every two weeks by the data processing unit for the ROHC. Job titles reported by the worker are computer-entered by a clerk upon admission. Job titles are subsequently coded in batch by a Research Assistant using the U.S. Census Bureau Occupation codes? 7 The Standard Industrial Codes (SIC) for the private sector companies served were obtained through the Illinois Department of Employment Security. For companies not found on the IDES data base, assignment of SIC code (four-digit) was made according to published criteria. ~7 Information on the number of company employees was obtained through a special survey of area employers conducted by the ROHC in 1988 with a response rate of 80.8 percent. Data were summarized and analyzed using the software package SAS on an IBM 3081 mainframe computer located at Rush-Presbyterian-St. Luke's Medical Center. Chi-square tests were performed to evaluate the significance of comparisons of categorical variables. Standard normal scores were used to evaluate the significance of the study sample means when contrasted to the means derived for the total worker population in the RPSLOHC occupational injury/ illness surveillance system.
RESULTS
A total o f 4,120 individuals with work-related h a n d a n d finger injuries f r o m private sector c o m p a n i e s were treated at the Rush-Presbyterian-St. Luke's Occupational Health Centers d u r i n g 1988, a n d as a category o f injuries c o m p r i s e d 31.7 p e r c e n t of all work-related injuries f r o m private sector companies. T h e m e a n age o f workers a n d h a n d / finger injuries was significantly y o u n g e r than the m e a n age of the p o p u lation o f i n j u r e d workers treated by the R P S L O H C (34.4 years vs. 35.6 years, p
WORK-RELATED INJURIES OF THE HAND" DATA FROM AN OCCUPATIONAL INJURY/ ILLNESS SURVEILLANCE SYSTEM Denise M. Oleske, PhD; Jerome J. Hahn, MD
ABSTRACT: Administrative and clinical data from a network of occupational medicine clinics were combined to evaluate the utility of these data in the surveillance o f non-fatal occupational injuries. Incident cases of work-related h a n d injuries were characterized to evaluate that process. In 1988, h a n d and finger injuries were found to be a m o n g the most c o m m o n (n = 4,120) o f all occupational injuries recorded in the system. H a n d / f i n g e r injuries accounted for 30.0 percent o f all episodes o f work-related injuries treated, with the incidence of these decreasing with increasing company size. H a n d injuries were found to be potentially severe with nearly 20 percent resulting from a crushing motion and nearly 10 percent being fractures or amputations. Hands being caught in machines or struck by metal items or h a n d tools accounted for 36.2 percent of the injuries. A surveillance system based u p o n ambulatory care data can be a feasible method for identifying priority areas for the prevention o f work-related injuries.
Injuries account for over 90% of all the reported work-related health problems in the United States. ~ In 1983, 4.7 million job-related injuries occurred in the private sector, representing rate of 7.5 injuries per 100 full-time workers) By 1989, 7.8 million job-related injuries were reported, yielding a rate of 8.6 per 100. 3 Noteworthy increases in the rate of disabling occupational injuries have occurred, resulting in an increase in the lost workdays rate of 57.2 per 100 in 1983 to 78.7 per 100 in 1989. 3 Rises in the rate of fractures, strains/sprains, cuts, lacerations and punctures have also been documented during this time. 4 Yet, while it is recognized that effective injury surveillance systems are needed for the planning and evaluation of injury control programs, "a standardized case-ascertainment or reporting system tor occupational Denise M. Oleske, PhD, is Assistant Professor in the Department of Health Systems Management and the Department of Preventive Medicine, Rush University, Chicago, IL; Jerome J. Hahn, MD, is Medical Director, Rush Occupational Health Centers, Chicago, IL. The research described herein was supported by a grant from the Centers for Disease Control #R49CCR502360 "Occupational Injury Surveillance by Health Centers." Requests for reprints should be addressed to: Denise M. Oleske, PhD, Department of Health Systems Management, Rush University, 1653 W. Congress Parkway, Chicago, IL 60612. © 1992 Human Sciences Press, Inc.
205
206
JOURNAL OF COMMUNITY HEALTH
injuries does not exist in the U.S. TM A National Academy of Sciences Committee on National Statistics reported in 1987 that the Bureau of Labor Statistics (BLS) data are inadequate for providing the Occupational Safety and Health Administration the data that it needs as the data collected by the BLS are seriously underreported? Underreporting is not unique to the US. Sass echoes this concern in Canada, and further remarks that incorrect theories about workplace injuries handicap efforts to reform workplace safety. 7 Current sources of work-related injury data have also been criticized because of the lack of uniformity in classifying cases with respect to their work-relatedness. 6 In the effort to control the incidence of injury/illness in the workplace, particularly those of a disabling nature, whose rates have been steadily increasing, high quality surveillance data are required. Recognizing the limitations associated with existing data sources, it is important to investigate alternative approaches to the surveillance and control of work-related conditions. Occupational medicine clinics have been identified as potential sources of occupational health data useful for surveillance purposes? 9 Data derived from these clinics could help to address the reported need for injury surveillance at the local level, ~°'~l particularly with regard to the non-fatal injury experience of small firms--a group which presents a significant challenge to health professionals. I'~ Under a grant from the Centers for Disease Control (No. R49CCR502360), the Rush-Presbyterian-St. Luke's Occupational Health Centers (ROHC) in Chicago have developed a model for the surveillance of occupational injuries by occupational medicine clinics. In evaluating the utility of a core data set prior to its implementation at five occupational health centers, Oleske, et al., utilized the core data set to identify that manual material handling practices were the probable causative factors contributing to the excessive injury rates, particularly of strain/sprains, in an automotive parts manufacturing company.L3 With interventions based upon knowledge of risk patterns, injury rates were significantly reduced and the company received a substantial rebate in its insurance premiums. 14 We illustrate the application of the system utilizing non-fatal work-related hand injuries as a model. These were among the most common work-related injuries identified through an occupational injury/illness surveillance system operated by a network of occupational health centers? 5 Most of the current existing information concerning risk factors for hand and finger injuries is of limited value being derived from case studies, small clinical series, and surveys of injured workers with low participation rates. A review of the literature has implicated a number of micro-environmental factors in the etiology of this injury category:
Denise M. Oleske and Jerome J. Hahn
207
high force, repetitive motions, hydrofluoric acid, vibration, mechanical power presses improper tool design (e.g., lack of shock absorbency, grippability, weight, etc.), handling debris (e.g., glass, sharp objects, etc.) and the handling of heavy objects or heavy power tools such as chain saws and air hammers. ~6-25Occupations at risk are: cash-register workers, assembly line packers, forklift operators, investment casting workers, corn-pickers, machine operators, mechanics, butchers, sheet metal workers, textile workers, woodworkers, pneumatic chipping hammer operators, cabinet makers, carpenters, warehouse workers, and laborers. 252~ The source of injury category to which severe hand and finger injuries are most often attributable is fixed machinery, specifically, table saws and presses (other than printing)? 3 In a Bureau of Labor Statistics (BLS) survey, hand and finger injuries most commonly occurred with fixed machinery when hands/fingers were either caught in, under, or between machinery or objects, hit against moving machine parts, or struck by moving machine parts. 23'~ When machinery has been attributed to hand or finger injuries, less than 15% have been attributed to accidental activation? ° NIOSH special investigations have provided insight into how detailed operations of the job task may be associated with hand or finger injuries such as attempting to correct placement of pieces after the downstroke of a punch press has been initiated. 31 In addition to these micro-environmental factors, macro-environmental factors for work-related hand injuries have been identified. These include stress and tension in the workplace produced by machine pacing, production standards, incentive systems, 32 and small company size? ° Human risk factors for work-related hand injuries which have been identified are: age less than 25 years, nine or more hours of sleep in previous night, tobacco use, presence of cardiovascular disease, longer than average working day, less than one year of experience on the job, inadequate training for the task, and performance of task not part of usual job. 2~-~4'31'~3However, in a study of amputation cases, 97 percent of which were to the hand or fingers, 60.4 percent were performing routine tasks? ~ There is a growing body of evidence to suggest that employee knowledge, attitudes, and beliefs about safety and safety practices may play a major role in the likelihood of the occurrence of a work-related injury. In a BLS survey of workers with hand/finger injuries, the most commonly cited categories of conditions or events felt to have contributed to the injury were: an increased work pace, lack of awareness of time, distance, or severity of potential hazard, and sudden or unintended movement of work materials, tools, equipment or the hand itself. With respect to other behavioral risk factors, only 34 percent of the injured workers responding to the survey received informa-
208
JOURNAL OF COMMUNITY HEALTH
tion r e g a r d i n g the use of gloves or other h a n d protection on the job; only 26 percent reported use of h a n d protection at the time of the injury. O f those who reported having used some form of h a n d protection at the time of the injury, 85 percent reported that the seriousness of the incident was r e d u c e d by the protection. Although these data are useful, a major limitation of the BLS surveys are the low participation rates, with only 56 percent of workers responding to a survey of h a n d injuries. 29 Non-use of gloves was f o u n d in another study to be associated with a 2.6 fold increase in h a n d injury. '~7 In a series of 150 persons with work-related h a n d injuries evaluated in two "Accident and Emergency Units" in the U.K., only 40 percent of workers could report why their injury had o c c u r r e d ? ~ Given the frequency of h a n d and finger injuries, we sought to characterize work-related h a n d injuries in accordance with a model of occupational injury causation we had previously described. ~3 T h e computerized combining of clinical with administrative data obtained in an ambulatory care setting could facilitate the identification of additional risk factors and their distribution for background information in planning etiologic studies. This system may also be a vehicle for reducing the likelihood of work-related injury t h r o u g h the dissemination of information in a timely m a n n e r about potential risk factors. MATERIALS AND METHODS
The Rush-Presbyterian-St. Luke's occupational Health Centers (RPSLOHC) includes five outpatient occupational medicine clinics located throughout the Chicago metropolitan area. A total of 2,342 private sector companies, representing a workforce of approximately 200,000 persons, participate in the surveillance system. The Centers treat cases requiring urgent or episodic care. During 1988, the Centers saw a total of 15,374 new patients, 13,616 of which were from private sector companies, for work-related problems. Of the total number of cases, 92.1 percent were for treatment of a work-related injury. All individuals who receive care for a work-related injury or illness at any one of the five clinics are included in the occupational injury/illness surveillance system. For the purposes of this study, only injuries to the hand or finger (ANSI Body Part Nos. 330 and 340) which were from private sector companies and were evaluated by RPSLOHC physicians are reported herein. Criteria used to identify a work-related case are those published by the U.S. Department of labor. ~ Injury/illness information is coded in accordance with the American National Standards Institute (ANSI) Z-16.2 standard (1969 revisions)~5 with enhancements by the Bureau of Labor Statistics Supplementary Data System?6 RPSLOHC physicians and nurses are trained in the use of
Denise M. Oieske and Jerome j. Hahn
209
this system. For each new work-related injury/illness, physicians code the nature of injury and body part affected, nursing staff code the source of injury and injury type (mechanism of injury) using the ANSI system. Quality assurance procedure include the training of new clinic personnel, manual and computer checks of entered data, monthly review of a 6 percent random sample of charts for coding accuracy and completeness, and periodic in-service training programs for all clinic staff. The entry of initial treatment, and follow-up care, and medical charge are computer entered at the conclusion of each visit. Standardized reports are produced for each clinic every two weeks by the data processing unit for the ROHC. Job titles reported by the worker are computer-entered by a clerk upon admission. Job titles are subsequently coded in batch by a Research Assistant using the U.S. Census Bureau Occupation codes? 7 The Standard Industrial Codes (SIC) for the private sector companies served were obtained through the Illinois Department of Employment Security. For companies not found on the IDES data base, assignment of SIC code (four-digit) was made according to published criteria. ~7 Information on the number of company employees was obtained through a special survey of area employers conducted by the ROHC in 1988 with a response rate of 80.8 percent. Data were summarized and analyzed using the software package SAS on an IBM 3081 mainframe computer located at Rush-Presbyterian-St. Luke's Medical Center. Chi-square tests were performed to evaluate the significance of comparisons of categorical variables. Standard normal scores were used to evaluate the significance of the study sample means when contrasted to the means derived for the total worker population in the RPSLOHC occupational injury/ illness surveillance system.
RESULTS
A total o f 4,120 individuals with work-related h a n d a n d finger injuries f r o m private sector c o m p a n i e s were treated at the Rush-Presbyterian-St. Luke's Occupational Health Centers d u r i n g 1988, a n d as a category o f injuries c o m p r i s e d 31.7 p e r c e n t of all work-related injuries f r o m private sector companies. T h e m e a n age o f workers a n d h a n d / finger injuries was significantly y o u n g e r than the m e a n age of the p o p u lation o f i n j u r e d workers treated by the R P S L O H C (34.4 years vs. 35.6 years, p
