CURRENT INTELLIGENCE BULLETIN 57

VIOLENCE IN THE WORKPLACE

Risk Factors and Prevention Strategies

U.S. DEPARTMENT OF HEALTH AND HUMAN SERVICES
Public Health Service
Centers for Disease Control and Prevention
National Institute for Occupational Safety and Health
Division of Safety Research

June 1996

DISCLAIMER

Mention of any company or product does not constitute endorsement by the National Institute for Occupational Safety and Health.

This document is in the public domain and may be freely copied or reprinted.

Copies of this and other NIOSH documents are available from

Publications Dissemination, EID
National Institute for Occupational Safety and Health
4676 Columbia Parkway
Cincinnati, OH 45226�98

Fax number: (513) 533�73
Telephone number: 1�0�朜IOSH (1�0�6�74)
E-mail: pubstaft@niosdt1.em.cdc.gov

To receive other information about occupational safety and health problems, call 1�0�朜IOSH (1�0�6�74), or visit the NIOSH Home Page on the World Wide Web at http://www.cdc.gov/niosh/homepage.html

DHHS (NIOSH) Publication No. 96�0

FOREWORD

The purpose of the Occupational Safety and Health Act of 1970 (Public Law 91�6) is to assure safe and healthful working conditions for every working person and to preserve our human resources. In this Act, the National Institute for Occupational Safety and Health (NIOSH) is charged with recommending occupational safety and health standards and describing exposures that are safe for various periods of employment, including (but not limited to) the exposures at which no worker will suffer diminished health, functional capacity, or life expectancy as a result of his or her work experience.

Current Intelligence Bulletins (CIBs) are issued by NIOSH to disseminate new scientific information about occupational hazards. A CIB may draw attention to a formerly unrecognized hazard, report new data on a known hazard, or disseminate information about hazard control. CIBs are distributed to representatives of academia, industry, organized labor, public health agencies, and public interest groups as well as to Federal agencies responsible for ensuring the safety and health of workers.

Each week in the United States, an average of 20 workers are murdered and 18,000 are assaulted while at work. These staggering figures should not be an accepted cost of doing business in our society梟or should death or injury be an inevitable result of one's chosen occupation.

This CIB reviews what is known about fatal and nonfatal violence in the workplace, defines research gaps, and recommends general approaches to workplace violence prevention. The document also summarizes issues that need to be addressed when dealing with workplace violence in various settings such as offices, factories, warehouses, hospitals, convenience stores, and taxicabs. No definitive strategy will ever be appropriate for preventing violence in all workplaces, but we must begin to change the way work is done in certain settings to minimize the risk to American workers. We must work together to address the research and prevention challenges posed by the complex issue of workplace violence. This document serves as the foundation for developing a comprehensive strategy for reducing violence in U.S. workplaces.

Linda Rosenstock, M.D., M.P.H.
Director, National Institute for Occupational Safety and Health
Centers for Disease Control and Prevention

ABSTRACT

This document reviews what is known about fatal and nonfatal violence in the workplace to determine the focus needed for prevention and research. The document also summarizes issues to be addressed when dealing with workplace violence in various settings such as offices, factories, warehouses, hospitals, convenience stores, and taxicabs.

Violence is a substantial contributor to occupational injury and death, and homicide has become the second leading cause of occupational injury death. Each week, an average of 20 workers are murdered and 18,000 are assaulted while at work or on duty. Nonfatal assaults result in millions of lost workdays and cost workers millions of dollars in lost wages.

Workplace violence is clustered in certain occupational settings: For example, the retail trade and service industries account for more than half of workplace homicides and 85% of nonfatal workplace assaults. Taxicab drivers have the highest risk of workplace homicides of any occupational group. Workers in health care, community services, and retail settings are at increased risk of nonfatal assaults.

Risk factors for workplace violence include dealing with the public, the exchange of money, and the delivery of services or goods. Prevention strategies for minimizing the risk of workplace violence include (but are not limited to) cash-handling policies, physical separation of workers from customers, good lighting, security devices, escort services, and employee training. A workplace violence prevention program should include a system for documenting incidents, procedures to be taken in the event of incide nts, and open communication between employers and workers. Although no definitive prevention strategy is appropriate for all workplaces, all workers and employers should assess the risks for violence in their workplaces and take appropriate action to reduce those risks.

ACKNOWLEDGMENTS

This document was prepared by Lynn Jenkins of the National Institute for Occupational Safety and Health (NIOSH). The author thanks the following NIOSH personnel for their review and comment: Jane Lipscomb, Ph.D.; Naomi Swanson, Ph.D.; Tim Pizatella; Nancy Stout, Ed.D.; Terry Wassell, Ph.D.; and William Halperin, M.D. The staff of the Injury Surveillance Section of the Division of Safety Research, NIOSH, also provided support and assistance to the development of this document.

The author gratefully acknowledges the work of the following reviewers: Bill Borwegen of the Service Employees International Union; Jordan Barab of the American Federation of State, County, and Municipal Employees; Jay Malcan, Ph.D., of the Virginia Department of Criminal Justice Services; and Pam McMahon, Ph.D., of the Division of Violence Prevention at the National Center for Injury Prevention and Control.

Editorial review and production assistance were provided by Anne Hamilton, Herb Linn, Joyce Spiker, Vanessa Becks, Jane Weber, Susan Feldmann, Rosemarie Hagedorn, Ann Stirnkorb, Dick Carlson, and Julie Tisdale.

CONTENTS

Foreword
Abstract
Acknowledgments
Public Health Summary

Introduction Purpose and Scope Homicide in the Workplace NIOSH Data Sex Age Race Geographic Distribution Method of Homicide Industry and Occupation Bureau of Labor Statistics Data Discussion Nonfatal Assaults in the Workplace Victimization Studies Estimated Magnitude of the Problem Discussion Risk Factors and Prevention Strategies Risk Factors Prevention Strategies Environmental Designs Administrative Controls Behavioral Strategies Developing and Implementing a Workplace Violence Prevention Program and Policy Responding to an Immediate Threat of Workplace Violence Dealing with the Consequences of Workplace Violence Current Efforts and Future Directions: Research and Prevention References Cited Related Reading

PUBLIC HEALTH SUMMARY

Violence in the Workplace

What are the hazards?

An average of 20 workers are murdered each week in the United States. The majority of these murders are robbery-related crimes. In addition, an estimated 1 million workers are assaulted annually in U.S. workplaces. Most of these assaults occur in service settings such as hospitals, nursing homes, and social service agencies. Factors that place workers at risk for violence in the workplace include interacting with the public, exchanging money, delivering services or goods, working late at night or during ear ly morning hours, working alone, guarding valuables or property, and dealing with violent people or volatile situations.

How can I be exposed or put at risk?

Anyone can become the victim of a workplace assault, but the risks are much greater in certain industries and occupations. For workplace homicides, taxicab drivers have the highest risk of any occupational group; for nonfatal workplace assaults, workers in health care, community services, and retail settings are at increased risk.

What recommendations has the Federal government made to protect my health?

A number of environmental, administrative, and behavioral strategies have the potential for reducing the risk of workplace violence. No single strategy is appropriate for all workplaces, but all workers and employers should assess the risk of violence in their workplaces and take appropriate action to reduce those risks. Collecting information about all incidents of workplace violence helps determine whether prevention strategies are necessary, appropriate, and effective.

Where can I get more information?

The references and related reading list at the end of this document provide a useful inventory of published reports and literature. A number of unions, employer groups, and professionals in occupational safety and health, human resources, and employee assistance have also developed materials regarding workplace violence. Any resource should be evaluated in light of the violence experienced in specific workplaces. Additional information from NIOSH can be obtained by calling the following number:

1�0�朜IOSH (800�6�74)

INTRODUCTION

Recently, violence in the workplace has received considerable attention in the popular press and among safety and health professionals. Much of the reason for this attention is the reporting of data by the National Institute for Occupational Safety and Health (NIOSH) and others regarding the magnitude of this problem in U.S. workplaces. Unfortunately, sensational acts of coworker violence (which form only a small part of the problem) are often emphasized by the media to the exclusion of the almost daily kil lings of taxicab drivers, convenience store clerks and other retail workers, security guards, and police officers. These deaths often go virtually unnoticed, yet their numbers are staggering: 1,071 workplace homicides occurred in 1994. These homicides included 179 supervisors or proprietors in retail sales, 105 cashiers, 86 taxicab drivers, 49 managers in restaurants or hotels, 70 police officers or detectives, and 76 security guards [BLS 1995]. An additional 1 million workers were assaulted each year. Thes e figures indicate that an average of 20 workers are murdered and 18,000 are assaulted each week while at work or on duty. Death or injury should not be an inevitable result of one's chosen occupation, nor should these staggering figures be accepted as a cost of doing business in our society.

PURPOSE AND SCOPE

The purpose of this document is to review what is known about fatal and nonfatal violence in the workplace to determine the focus needed for prevention and research efforts. This document also summarizes issues to be addressed when dealing with workplace violence in various settings such as offices, factories, warehouses, hospitals, convenience stores, and taxicabs.

Although no definitive strategy will ever be appropriate for all workplaces, we must begin to change the way work is done in certain settings to minimize or remove the risk of workplace violence. We must also change the way we think about workplace violence by shifting the emphasis from reactionary approaches to prevention, and by embracing workplace violence as an occupational safety and health issue. This document examines these issues and proposes new strategies for prevention.

Defining workplace violence has generated considerable discussion. Some would include in the definition any language or actions that make one person uncomfortable in the workplace; others would include threats and harassment; and all would include any bodily injury inflicted by one person on another. Thus the spectrum of workplace violence ranges from offensive language to homicide, and a reasonable working definition of workplace violence is as follows: violent acts, including physical assaults and threats of assault, directed toward persons at work or on duty. Most studies to date have focused primarily on physical injuries, since they are clearly defined and easily measured. But this document examines data from multiple sources and acknowledges differences in definitions and coverage to learn as much as possible from these varied efforts.

The circumstances of workplace violence also vary and may include robbery-associated violence; violence by disgruntled clients, customers, patients, inmates, etc.; violence by coworkers, employees, or employers; and domestic violence that finds its way into the workplace. These circumstances all appear to be related to the level of violence in communities and in society in general. Thus the question arises: why study workplace violence separately from the larger universe of all violence? Several reasons exi st for focusing specifically on workplace violence:

HOMICIDE IN THE WORKPLACE

NIOSH Data

Data from the National Traumatic Occupational Fatalities (NTOF) Surveillance System indicate that 9,937 workplace homicides occurred during the 13-year period from 1980 through 1992, with an average workplace homicide rate of 0.70/100,000 workers (Table 1) [NIOSH 1995]. Over the course of the 1980s, workplace homicides decreased; but in the 1990s, the numbers began to increase, surpassing machine-related deaths and approaching the number of workplace motor-vehicle-related deaths (Figure 1). Although the 199 2 figure was lower than that for 1991, it exceeded the 1990 figure and did not include 1992 data for New York City and the State of Connecticut. NTOF is an ongoing, death-certificate-based census of traumatic occupational fatalities in the United States, with data from all 50 States and the District of Columbia. NTOF includes infor-mation for all workers aged 16 or older who died from an injury or poisoning and for whom the certifier noted a positive response to the injury at work? item on the death certifi cate. For additional discussion of the NTOF system and the limitations of death certificates for the study of workplace homicide, see Castillo and Jenkins [1994].

Sex

The majority (80%) of workplace homicides during 1980� occurred among male workers. The leading cause of occupational injury death varied by sex, with homicides accounting for 11% of all occupational injury deaths among male workers and 42% among female workers [NIOSH 1995]. The majority of female homicide vicims were employed in retail trade (46%) and service (22%) industries (Table 2). A large number of male homicide victims were employed not only in retail trade (36%) and service (16%) industries but i n public administration (11%) and transportation/communication/public utilities (11%) (Table 2). Although homicide is the leading cause of occupational injury death among female workers, male workers have more than three times the risk of work-related homicide (Table 3).

Age

The age of workplace homicide victims ranged from 16 (the youngest reported in NTOF) to 93 during 1980�. The largest number of workplace homicides occurred among workers aged 25 to 34, whereas the rate of workplace homicide increased with age (Table 3). The highest rates of workplace homicide occurred among workers aged 65 and older; the rates for these workers were more than twice those for workers aged 55� (Table 3). This pattern held true for both male and female workers.

Table 1. Workplace homicides in the United States, 1980�*

Year    Number   Rate�

1980 929 0.96 1981 944 .94 1982 859 .86 1983 721 .72 1984 660 .63 1985 751 .70 1986 672 .61 1987 649 .58 1988 699 .61 1989 696 .59 1990 725 .61 1991 875 .75 1992 757 .64

Total 9,937 .70

Source: NIOSH [1995]. *Data not available for New York City and Connecticut. 哖er 100,000 workers.

Table 2. Workplace homicides by industry and sex桿nited States 1980�*

                                           Homicides (% of total)�
                                               Male    Female

Industry workers workers Retail trade 36.1 45.5 Services 16.0 22.2 Public administration 10.5 2.9 Transportation/communication/public utilities 10.6 3.8 Manufacturing 7.0 4.9 Construction 4.1 .6 Agriculture/forestry/fishing 2.7 .6 Finance/insurance/real estate 2.4 6.8 Wholesale trade 1.7 1.1 Mining .6 .1 Not classified 8.5 11.7

Source: NIOSH [1995]. *Data for New York City and Connecticut were not available for 1992. 哖ercentages add to more than 100% because of rounding.

Table 3. Workplace homicides by age group and sex桿nited States, 1980-92*, �/H4>

Male workers Female workers All workers Age group Number Rate Number Rate Number Rate 16� 242 0.55 102 0.25 344 0.41 20� 796 .87 285 .35 1,081 .62 25� 2,020 .89 591 .33 2,611 .65 35� 1,841 .99 423 .28 2,265 .68 45� 1,344 1.04 293 .29 1,637 .71 55� 1,055 1.22 191 .31 1,246 .84 65+ 620 2.59 115 .71 735 1.83

Total� 7,935 � 2,001 � 9,937 � Average � 1.01 � .32 � .70

Source: NIOSH [1995]. *Data from New York City and Connecticut were not available for 1992. 哛ates are per 100,000 workers. 嘥otals include victims for whom age data were missing (17 male workers) and 1 worker whose sex was not reported.

Race

Although the majority of workplace homicide victims were white (73%), black workers (1.39/100,000) and workers of other races (1.87/100,000) had the highest rates of work-related homicide (Table 4).

Geographic Distribution

During 1980�, the largest number of homicides and the highest rates per 100,000 workers occurred in the South (N=4,819; rate=1.02/100,000) and the West (N=2,278; rate= 0.79/100,000) (Table 5). Note that during the early years of the NTOF data collection effort, four States桳ouisiana, Nebraska, Oklahoma, and New York梬ere unable to provide data on work-related homicides. In addition, data for 1992 were unavailable from New York City and Connecticut.

Homicide was the leading cause of occupational injury death over the 13-year period in Alabama, Connecticut, the District of Columbia, Michigan, and South Carolina. Although complete data for the period are not available for New York, estimates and data for recent years indicate that homicide was also the leading cause of occupational injury death in that State.

In the document Fatal Injuries to Workers in the United States, 1980�89: A Decade of Surveillance [Jenkins et al. 1993], all occupational injury deaths were analyzed for 1980 through 1989. Geographic differences in the leading causes of death were examined by mapping the State-specific, cause-specific rates in relation to the average cause-specific rate for all States combined. This analysis revealed that most (N=45) States had workplace homicide rates within one standard deviation above or below the aver age workplace homicide rate [Jenkins et al. 1993].

Table 4. Workplace homicides by race桿nited States, 1980�*

Race/ethnicity of victims  Number  % of total      Rate�
White (includes Hispanic)  7,239      72.8         0.59
Black                      1,938      19.5         1.39
Other                        760       7.6         1.87

Source: NIOSH [1995]. *Data for New York City and Connecticut were not available for 1992. 哖er 100,000 workers.

Table 5. Workplace homicides by Bureau of the Census geographic region桿nited States, 1980�*

Region          Number  % of total      Rate�
North Central   1,797      18.1         0.50
North East      1,043      10.5          .35
South           4,819      48.5         1.02
West            2,278      22.9          .79

Source: NIOSH [1995]. *Data for New York City and Connecticut were not available for 1992. 哖er 100,000 workers.

Method of Homicide

Between 1980 and 1992, 76% of work-related homicides were committed with firearms, and another 12% resulted from wounds inflicted by cutting or piercing instruments (Table 6). During this period, the number of firearm-related homicides declined then gradually increased, with the number of firearm-related workplace homicides in 1991 exceeding that in 1980 (Figure 2). The number declined slightly in 1992, but data for this year are incomplete. Firearms accounted for an increasing percentage of the total workp lace homicides over the 13-year period: 74% in 1980 and 84% in 1991. Firearms were used in 79% of the workplace homicides in 1992, but data are missing for New York City and Connecticut for this year.

Table 6. Workplace homicides by method桿nited States, 1980�*

Method                          Number     % of total
Firearm                         7,590        76.4
Cutting or piercing instrument  1,231        12.4
Strangulation                     185         1.9
All other methods                 931         9.4

Source: NIOSH [1995]. *Data for New York City and Connecticut were not available for 1992.

Industry and Occupation

During the 13-year period 1980�, the greatest number of deaths occurred in the retail trade (3,774) and service (1,713) industries, whereas the highest rates per 100,000 workers occurred in retail trades (1.6), public administration (1.3), and transportation/communication/public utilities (0.94) (Table 7).

At the more detailed levels of industry (Table 8), the largest number of deaths occurred in grocery stores (N=330), eating and drinking places (N=262), taxicab services (N=138), and justice/public order establishments (N=137). Taxicab services had the highest rate of work-related homicide during the 3-year period 1990� (41.4/100,000). This rate was nearly 60 times the national average rate of work-related homicides (0.70/100,000). This figure was followed by rates for liquor stores (7.5), detective/protec tive services (7.0), gas service stations (4.8), and jewelry stores (4.7) (Table 8). The rates show an increase from the previously published rates for 1980� for taxicab services, detective/protective services, grocery stores, and jewelry stores. Rates decreased in liquor stores, gasoline service stations, justice/public order establishments, and hotels/motels; they remained the same in eating and drinking places.

Table 7. Workplace homicides by industry桿nited States, 1980�*

 Industry                                       Number  % of total    Rate�
Retail trade                                    3,774      38.0       1.60
Public administration                             889       8.9       1.30
Transportation/communication/public utilities     917       9.2        .94
Agriculture/forestry/fishing                      222       2.2        .50
Mining                                             45       0.5        .40
Service                                         1,713      17.2        .38
Construction                                      335       3.4        .37
Finance/insurance/real estate                     327       3.3        .35
Wholesale trade                                   155       1.6        .27
Manufacturing                                     650       6.5        .24
Not classified                                    910       9.1          �

Source: NIOSH [1995]. *Data for New York City and Connecticut were not available for 1992. 哖er 100,000 workers.

When detailed occupations were analyzed for 1990� (Table 9), the highest homicide rates were found for taxicab drivers/chauffeurs (22.7), sheriffs/bailiffs (10.7), police and detectives梡ublic service (6.1), gas station/garage workers (5.9), and security guards (5.5). Compared with previously published data for the 7-year period 1983�, these data indicate that rates increased more than two and a half times for sales counter clerks and nearly two times for motor vehicle and boat sales workers and sales w orkers in other commodities (includes workers in jewelry, food, sporting goods, book, coin, and other retail stores). Homicide rates for taxicab drivers and security guards were one and a half times higher during the early 1990s than they had been during 1983�. However, some rates decreased: for 1990�, the rate for hotel clerks was less than half the 1983� rate, and the rate for police and detectives was two-thirds the 1983� rate. During 1990�, an extraordinary number of homicides (N=372) occurred among sales supervisors and proprietors, who had double the number of deaths in any other single category during both periods.

Table 8. Workplace homicides in high-risk industries桿nited States, 1980� and 1990�*, �/H4>

                                       1980�           1990�
Industry                              Number  Rate      Number  Rate
Taxicab services                       287    26.9       138    41.4
Liquor stores                          115     8.0        30     7.5
Gas service stations                   304     5.6        68     4.8
Detective/protective services          152     5.0        86     7.0
Justice/public order establishments    640     3.4       137     2.2
Grocery stores                         806     3.2       330     3.8
Jewelry stores                          56     3.2        26     4.7
Hotels/motels                          153     1.5        33     0.8
Barber shops                            14     1.5         4       �
Eating/drinking places                 734     1.5       262     1.5

Source: NIOSH [1995] (data for 1980� from Castillo and Jenkins [1994]). *Data for New York City and Connecticut were not available for 1992. 哛ates are per 100,000 workers. 嘡ate was not calculated because of the instability of rates based on small numbers.

Table 9. Workplace homicides in high-risk* occupations桿nited States, 1983� and 1990��, �/H4>

                                         1983�           1990�
Occupation                             Number   Rate     Number   Rate
Taxicab driver/chauffeur                197     15.1      140     22.7
Sheriff/bailiff                          73     10.9       36     10.7
Police and detective梡ublic service     267      9.0       86      6.1
Hotel clerk                              29      5.1        6      2.0
Gas station/garage worker                83      4.5       37      5.9
Security guard                          160      3.6      115      5.5
Stock handler/bagger                    189      3.1       95      3.5
Supervisor/proprietor, sales            662      2.8      372      3.3
Supervisor, police and detective         12      2.2        0        �
Barber                                   14      2.2        4        �
Bartender                                49      2.1       20      2.3
Correctional institution officer         19      1.5        3        �
Salesperson, motor vehicle and boat      21      1.1       17      2.0
Salesperson, other commodities           98      1.0       73      1.7
Sales counter clerk                      13      1.2       18      3.1
Fire fighter                             18      1.4        8      1.3
Logging occupation                        4        �        6      2.3
Butcher/meatcutter                       11       .6       12      1.5

Source: NIOSH [1995] (data for 1983� from Castillo and Jenkins [1994]). *High-risk occupations have workplace homicide rates that are twice the average rate during one or both time periods. 咲ata for New York City and Connecticut were not available for 1992. 嘡ates are per 100,000 workers. ate was not calculated because of the instability of rates based on small numbers.

Bureau of Labor Statistics Data

Information from the Bureau of Labor Statistics (BLS) Census of Fatal Occupational Injuries (CFOI) Program identifies the same high-risk demographic and occupational groups as NIOSH NTOF data and allows description of the circumstances of workplace homicides for the period 1992�. According to the BLS data, 73% to 82% of the homicides occurred during a robbery or other crime, whereas only 9% to 10% were attributed to business disputes, and only 4% to 6% were attributed specifically to coworkers or former e mployees (Table 10). A shift occurred in the robbery and other crimes category with the creation of the new security guard in line of duty category, but the distribution of the circumstances has remained fairly stable during the 3 years in which data have been collected. The CFOI system uses multiple sources, including administrative documents from Federal and State agencies (e.g., death certificates, medical examiner records, workers' compensation reports, and regulatory agency reports) as well as news rep orts and followup questionnaires to business establishments [Windau and Toscano 1994].

The BLS described a number of the robberies as occurring while workers were locking up at night or making money drops or pickups, but these were not specifically quantified. Also, homicide appeared to be primarily an urban problem, with eight of the largest metropolitan areas accounting for nearly half of the workplace homicides in 1993 [Toscano and Weber 1995]. The self-employed accounted for 24% to 27% of the homicides documented by the CFOI program for 1992�, whereas this group accounted for only about 9% of the workforce during those years [BLS 1993, 1994b, 1995].

Discussion

Despite differences in data collection and the resulting total number of homicides reported by the NTOF and CFOI fatality surveillance systems, the ranking of high-risk industries and occupations is consistent, with taxicab drivers/chauffeurs, law enforcement and security personnel, and retail trade workers experiencing the greatest risks and the largest numbers of workplace homicides. Findings about the distributions by demographic characteristics are also remarkably similar [Windau and Toscano 1994; Tosca no and Weber 1995; Castillo and Jenkins 1994].

Differences in leading causes of occupational injury death by sex can be attributed at least in part to variations in employment patterns [Jenkins 1994]. For example, homicide is the leading cause of occupational injury death for female workers because they are exposed less frequently than male workers to hazards such as heavy machinery and work at elevations. The same is also true for differences among industries in leading causes of death. Workers in retail trade, services, and finance/insurance/real esta te are not exposed to the same kinds of hazards as workers in construction, agriculture/forestry/ fishing, mining, or transportation/communication/public utilities. These factors are extremely important to the future direction of occupational safety and health as employment patterns shift from traditional heavy industry to retail trade and service sectors. Workplace homicide must be addressed to continue the trends of decreasing numbers and rates of occupational injury deaths [Jenkins et al. 1993; Stout et al. 1996].

Elevated rates of workplace homicide among workers aged 65 and older may be attributable to a number of factors, including a decreased abilitiy to survive injury or the perception that such workers are softer targets [Jenkins et al. 1992].

Regional data for homicides in the general population show a similar pattern to those in the workplace, with crude homicide rates being highest in the South and the West [O'Carroll and Mercy 1989].

Table 10. Circumstances of Workplace Homicides桿nited States, 1992�

                                    Homicides (% of total)*
                                 1992        1993         1994
Circumstance                  (N=1,004)    (N=1,063)    (N=1,071)
Robbery and other crime           82          75           73
Business dispute/work associate    9          10            9
Coworker/former coworker           4           6            5
Customer/client                    5           4            4
Police in line of duty             6           6            7
Security guard in line of duty     �           5            7
Personal dispute/acquaintance      4           4            4

Source: BLS [1994b, 1995], Windau and Toscano [1994]. *Percentages add to more than 100% because of rounding. 員his category was not included in 1992.

The percentage of work-related homicides attributed to firearms (76%) is slightly higher than that found in the general population, where 71% of the 1993 murders with victims aged 18 or older were committed with firearms [FBI 1994].

Changes in the risk of workplace homicide in specific industry and occupation groups between the 1980s and the early 1990s may be attributable to a number of factors, including increased recognition and recording of cases as work-related, changes in training or other work practices, increased levels of crime in certain settings, and the distribution of resources in response to perceived levels of crime. The shift in risk for public police officers and private security guards is particularly noteworthy, as t he data indicate a decline in rates among public police officers and a dramatic increase among private security guards. We do not know the extent to which these findings are attributable to efforts among public police forces to reduce risks through training and use of protective equipment, the employment of private security guards by businesses and communities that had previously relied solely on public safety personnel, and the level of training and background of private security officers. However, further research is warranted.

The circumstances of workplace homicides differ substantially from those portrayed by the media and from homicides in the general population. For the most part, workplace homicides are not the result of disgruntled workers who take out their frustrations on coworkers or supervisors, or of intimate partners and other relatives who kill loved ones in the course of a dispute; rather, they are mostly robbery-related crimes.

NONFATAL ASSAULTS IN THE WORKPLACE

Victimization Studies

Limited information is available in the criminal justice and public health literature regarding the nature and magnitude of nonfatal workplace violence. The criminology literature contains a few victimization studies that include designation of victimizations that occurred at work. Using the 1982 Victim Risk Supplement to the National Crime Victimization Survey, Lynch [1987] used log linear modeling to examine workplace victimizations with regard to demographic variables as well as features of the workplace . Features of the workplace included exposure to and public access to the workplace, local travel, overnight trips, perceived dangerousness of the neighborhood and the workplace, and the frequency with which money was handled on the job. These analyses indicated that the risk of workplace victimization was related more to the task performed than to the demographic characteristics of the person performing the job. Factors related to an increased risk for workplace victimization included routine face-to-fac e contact with large numbers of people, the handling of money, and jobs that required routine travel or that did not have a single worksite. Using a 1983 crime survey in the metropolitan Washington, D.C., area, Collins and Cox [1987] found results similar to those of Lynch; the delivery of passengers or goods and dealing with the public were the factors associated with an increased risk for workplace assault. State-specific studies of workplace assaults using workers' compensation data have also been conduc ted, as have industry- and occupation-specific studies; a summary of these appears in Castillo [1994].

Estimated Magnitude of the Problem

A number of recent estimates have been made of the current magnitude of nonfatal assaults in U.S. workplaces. The first comes from the BLS Annual Survey of Occupational Injuries and Illnesses (ASOII). The ASOII is an annual survey of approximately 250,000 private establishments. This survey excludes the self- employed, small farmers, and government workers. These data indicate that 22,400 workplace assaults occurred in 1992; these represented 1% of all cases involving days away from work [BLS 1994d]. Unlike homicides, nonfatal workplace assaults are distributed almost equally between men (44%) and women (56%). The majority of the nonfatal assaults reported in the ASOII occurred in the service (64%) and retail trade (21%) industries. Of those in services, 27% occurred in nursing homes, 13% in social services, and 11% in hospitals. In retail trade, 6% occurred in grocery stores, and another 5% occurred in eating and drinking places (Table 11). The source of injury in 45% of the cases was a health care patient ( Figure 3), with another 31% described as other person and 6% as coworker or former coworker. The BLS coding system requires that the object or substance that directly inflicted the injury be coded as the source of the injury; thus 5% of the assaults are coded as structures and surfaces (these are likely events where workers were pushed into walls or to floors), and another 4% are categorized as tools (these include events in which knives or other weapons were used). Nearly half (47%) of the workplace assaul ts were described as incidents involving hitting, kicking, or beating; there were also cases of squeezing, pinching, scratching, biting, stabbing, and shooting, as well as rapes and threats of violence (Table 12). The median days away from work as the result of an assault was 5, but this figure varied by type of assault (Table 12).

Another estimate of the magnitude of nonfatal workplace assaults comes from a survey by the Northwestern National Life Insurance Company, which indicates that 2.2 million workplace assaults (defined as physical attacks) occurred between July 1992 and July 1993 [Northwestern National Life 1993]. This estimate and the findings from this survey must be used with caution, as the estimate for assaults was based on only 3% of the sample of 600, or 15 workers who reported having been attacked. In addition, the res pondents to this survey did not accurately represent the actual distribution of the workforce [Castillo 1994].

Table 11. Violent acts resulting in days away from work in 1992, by industry

Violent acts resulting in days away from work
Industry                                   (% of total)
Services                                        64
        Nursing homes                           27
        Social services                         13
        Hospitals                               11
        Other services                          13
Retail trades                                   21
        Grocery stores                           6
        Eating and drinking places               5
        Other retail                            10
Transportation/communication/public utilities    4
Finance/insurance/real estate                    4
Other                                            4
Manufacturing                                    3

Source: BLS [1994d].

Table 12. Violent acts resulting in days away from work梡rivate industry, 1992

                                                         Median days
                                                Number    away from
Type of violent act                            of cases     work
Hitting, kicking,beating                        10,425        5
Squeezing, pinching, scratching, twisting        2,457        4
Biting                                             901        3
Stabbing                                           598       28
Shooting                                           560       30
All other specified acts (e.g., rape, threats)   5,157        5

Source: BLS [1994c].

A final estimate of assaults in the workplace comes from the National Crime Victimization Survey (NCVS)梐n annual, national, household-based survey of more than 100,000 individuals aged 12 or older. NCVS data for 1987� indicate that each year, nearly 1 million persons were assaulted while at work or on duty (Table 13); this figure represents 15% of the 6.5 million acts of violence experienced by Americans [Bachman 1994]. Sixteen percent of workplace victimizations resulted in injuries.

When the Bureau of Justice Statistics (BJS) analyzed the relationship of the victim to the offender for these events by sex, female workers appeared to be most likely to be attacked by someone they knew, although only 5% of victimizations were attributed to an intimate (defined as a husband, ex-husband, boyfriend, or ex-boyfriend) (Table 14). Probably a customer, client, or patient with whom the victim had an ongoing professional relationship would have been coded to the acquaintance or well-known categorie s, so these findings by sex may be misleading and may reflect the distribution of the workforce in service sectors more than real sex differences in victimization.

BJS also analyzed workplace victimizations by type of work setting and found that 61% occurred in private companies, 30% occurred among government employees, and 8% of the victims were self-employed [Bachman 1994]. BJS points out in its report that government workers make up only 18% of the workforce and thus appear to be suffering a disproportionate share of the attacks; it should also be noted that risk factors such as dealing with the public and delivery of services are common among government employees. In addition, all local, State, and Federal police are included in this category.

A BJS analysis of the location where victimizations occurred includes both specific and generic categories (Table 15).

In 32% of the workplace assaults, the victim faced an armed assailant, whereas 62% of the perpetrators had no weapon; weapon use was not ascertained for 6% of the cases [Bachman 1994].

When individuals in the NCVS were asked whether this workplace victimization was reported to the police, 56% indicated that it was not. For 40% of respondents, the reason cited for not reporting to the police was that the event was believed to be a minor or private matter. Another 27% did not report to the police because the incident was reported to another official such as a company security guard [Bachman 1994].

The NCVS also solicits information about days away from work and lost wages due to the victimization. As a result of workplace victimizations, approximately half a million workers lost 1.75 million days of work annually (an average of 3.5 days per crime) and victims lost more than $55 million in wages, not including days covered by sick or annual leave. As a result of the 16% of victimizations in which injuries were incurred, 876,800 workdays were lost annually and $16 million were lost in wages, not includ ing days covered by sick or annual leave [Bachman 1994].

Table 13. Crimes of violence against persons at work or on duty, 1987�

                       Average annual number
Type of crime           Victims     Injuries
Simple assault          615,160      89,572
Aggravated assault      264,174      48,180
Robbery                  79,109      17,904
Rape                     13,068       3,438*
     Total              971,517     159,094

Source: Bachman [1994]. *Injuries are those in addition to the rape.

Table 14. Workplace victimizations by victim-offender relationship and sex桿nited States, 1987�

Victim-offender           % of workplace victimizations
 relationship            Male workers       Female workers
Stranger                      58                 40
Acquaintance                  30                 35
Well-known person             10                 19
Relative                       1                  1
Intimate (spouse, ex-spouse)   1                  5

Source: Bachman [1994].

Table 15. Workplace victimizations by location桿nited States, 1987�

Location where                       % of total
victimization occurred             victimizations
Office, factory, warehouse              14
Restaurant, bar, nightclub              13
Parking lot/garage                      11
School property                          9
Other commercial establishment          23
Public property (streets, parks, etc.)	22
Other                                    8

Source: Bachman [1994].

Discussion

Nonfatal assaults in the workplace clearly affect many workers and employers. Although groups at high risk for workplace homicide and nonfatal workplace assaults share similar characteristics such as interaction with the public and the handling of money, there are also clear differences. For example, groups such as health care workers are not at elevated risk of workplace homicide, but they are at greatly increased risk of nonfatal assaults. Castillo [1994] suggests that some of the distinctions between fat al and nonfatal workplace assaults can be attributed to differences between robbery-related violence and violence resulting from the anger or frustration of customers, clients, or coworkers, with robbery-related violence being more likely to result in a fatal outcome. The premeditated use of firearms to facilitate robberies is also likely to influence the lethality of assaults in the workplace.

RISK FACTORS AND PREVENTION STRATEGIES

Risk Factors

A number of factors may increase a worker's risk for workplace assault, and they have been described in previous research [Collins and Cox 1987; Davis 1987; Davis et al. 1987; Kraus 1987; Lynch 1987; NIOSH 1993; Castillo and Jenkins 1994]. These factors include the following:

Contact with the public
Exchange of money
Delivery of passengers, goods, or services
Having a mobile workplace such as a taxicab or police cruiser
Working with unstable or volatile persons in health care, social service, or criminal justice settings
Working alone or in small numbers
Working late at night or during early morning hours
Working in high-crime areas
Guarding valuable property or possessions
Working in community-based settings
Prevention Strategies
Environmental Designs

Commonly implemented cash-handling policies in retail settings include procedures such as using locked drop safes, carrying small amounts of cash, and posting signs and printing notices that limited cash is available. It may also be useful to explore the feasibility of cashless transactions in taxicabs and retail settings through the use of machines that accommodate automatic teller account cards or debit cards. These approaches could be used in any setting where cash is currently exchanged between workers and customers.

Physical separation of workers from customers, clients, and the general public through the use of bullet-resistant barriers or enclosures has been proposed for retail settings such as gas stations and convenience stores, hospital emergency departments, and social service agency claims areas. The height and depth of counters (with or without bullet-resistant barriers) are also important considerations in protecting workers, since they introduce physical distance between workers and potential attackers. Consi deration must nonetheless be given to the continued ease of conducting business; a safety device that increases frustration for workers or for customers, clients, or patients may be self-defeating.

Visibility and lighting are also important environmental design considerations. Making high-risk areas visible to more people and installing good external lighting should decrease the risk of workplace assaults [NIOSH 1993].

Access to and egress from the workplace are also important areas to assess. The number of entrances and exits, the ease with which nonemployees can gain access to work areas because doors are unlocked, and the number of areas where potential attackers can hide are issues that should be addressed. This issue has implications for the design of buildings and parking areas, landscaping, and the placement of garbage areas, outdoor refrigeration areas, and other storage facilities that workers must use during a w ork shift.

Numerous security devices may reduce the risk for assaults against workers and facilitate the identification and apprehension of perpetrators. These include closed-circuit cameras, alarms, two-way mirrors, card-key access systems, panic-bar doors locked from the outside only, and trouble lights or geographic locating devices in taxicabs and other mobile workplaces.

Personal protective equipment such as body armor has been used effectively by public safety personnel to mitigate the effects of workplace violence. For example, the lives of more than 1,800 police officers have been saved by Kevlar� vests [Brierley 1996].

Administrative Controls

Staffing plans and work practices (such as escorting patients and prohibiting unsupervised movement within and between clinic areas) are included in the California Occupational Safety and Health Administration Guidelines for the Security and Safety of Health Care and Community Service Workers [State of California 1993]. Increasing the number of staff on duty may also be appropriate in any number of service and retail settings. The use of security guards or receptionists to screen persons entering the workpl ace and controlling access to actual work areas has also been suggested by security experts.

Work practices and staffing patterns during the opening and closing of establishments and during money drops and pickups should be carefully reviewed for the increased risk of assault they pose to workers. These practices include having workers take out garbage, dispose of grease, store food or other items in external storage areas, and transport or store money.

Policies and procedures for assessing and reporting threats allow employers to track and assess threats and violent incidents in the workplace. Such policies clearly indicate a zero tolerance of workplace violence and provide mechanisms by which incidents can be reported and handled. In addition, such information allows employers to assess whether prevention strategies are appropriate and effective. These policies should also include guidance on recognizing the potential for violence, methods for defusing o r de-escalating potentially violent situations, and instruction about the use of security devices and protective equipment. Procedures for obtaining medical care and psychological support following violent incidents should also be addressed. Training and education efforts are clearly needed to accompany such policies.

Behavioral Strategies

Training employees in nonviolent response and conflict resolution has been suggested to reduce the risk that volatile situations will escalate to physical violence. Also critical is training that addresses hazards associated with specific tasks or worksites and relevant prevention strategies. Training should not be regarded as the sole prevention strategy but as a component in a comprehensive approach to reducing workplace violence. To increase vigilance and compliance with stated violence prevention polici es, training should emphasize the appropriate use and maintenance of protective equipment, adherence to administrative controls, and increased knowledge and awareness of the risk of workplace violence.

DEVELOPING AND IMPLEMENTING A WORKPLACE VIOLENCE PREVENTION PROGRAM AND POLICY

The first priority in developing a workplace violence prevention policy is to establish a system for documenting violent incidents in the workplace. Such data are essential for assessing the nature and magnitude of workplace violence in a given workplace and quantifying risk. These data can be used to assess the need for action to reduce or mitigate the risks for workplace violence and implement a reasonable intervention strategy. An existing intervention strategy may be identified within an industry or in similar industries, or new and unique strategies may be needed to address the risks in a given workplace or setting. Implementation of the reporting system, a workplace violence prevention policy, and specific prevention strategies should be publicized company-wide, and appropriate training sessions should be scheduled. The demonstrated commitment of management is crucial to the success of the program. The success and appropriateness of intervention strategies can be monitored and adjusted with continued da ta collection.

A written workplace violence policy should clearly indicate a zero tolerance of violence at work, whether the violence originates inside or outside the workplace. Just as workplaces have developed mechanisms for reporting and dealing with sexual harassment, they must also develop threat assessment teams to which threats and violent incidents can be reported. These teams should include representatives from human resources, security, employee assistance, unions, workers, management, and perhaps legal and publ ic relations departments. The charge to this team is to assess threats of violence (e.g., to determine how specific a threat is, whether the person threatening the worker has the means for carrying out the threat, etc.) and to determine what steps are necessary to prevent the threat from being carried out. This team should also be charged with periodic reviews of violent incidents to identify ways in which similar incidents can be prevented in the future. Note that when violence or the threat of violence oc curs among coworkers, firing the perpetrator may or may not be the most appropriate way to reduce the risk for additional or future violence. The employer may want to retain some control over the perpetrator and require or provide counseling or other care, if appropriate. The violence prevention policy should explicitly state the consequences of making threats or committing acts of violence in the workplace.

A comprehensive workplace violence prevention policy and program should also include procedures and responsibilities to be taken in the event of a violent incident in the workplace. This policy should explicitly state how the response team is to be assembled and who is responsible for immediate care of the victim(s), re-establishing work areas and processes, and organizing and carrying out stress debriefing sessions with victims, their coworkers, and perhaps the families of victims and coworkers. Employee a ssistance programs, human resource professionals, and local mental health and emergency service personnel can offer assistance in developing these strategies.

Responding to an Immediate Threat of Workplace Violence

For a situation that poses an immediate threat of workplace violence, all legal, human resource, employee assistance, community mental health, and law enforcement resources should be used to develop a response. The risk of injury to all workers should be minimized. If a threat has been made that refers to particular times and places, or if the potential offender is knowledgeable about workplace procedures and time frames, patterns may need to be shifted. For example, a person who has leveled a threat agains t a worker may indicate, "I know where you park and what time you get off work!" In such a case, it may be advisable to change or even stagger departure times and implement a buddy system or an escort by security guard for leaving the building and getting to parking areas. The threat should not be ignored in the hope that it will resolve itself or out of fear of triggering an outburst from the person who has lodged the threat. If someone poses a danger to himself or others, appropriate authorities should be notified and action should be taken.

Dealing with the Consequences of Workplace Violence

Much discussion has also centered around the role of stress in workplace violence. The most important thing to remember is that stress can be both a cause and an effect of workplace violence. That is, high levels of stress may lead to violence in the workplace, but a violent inci- dent in the workplace will most certainly lead to stress, perhaps even to post-traumatic stress disorder. The data from the National Crime Victimization Survey [Bachman 1994] present compelling evidence (more than a million workda ys lost as a result of workplace assaults each year) for the need to be aware of the impact of workplace violence. Employers should therefore be sensitive to the effects of workplace violence and provide an environment that promotes open communication; they should also have in place an established procedure for reporting and responding to violence. Appropriate referrals to employee assistance programs or other local mental health services may be appropriate for stress debriefing sessions after critical inci dents.

CURRENT EFFORTS AND FUTURE DIRECTIONS: RESEARCH AND PREVENTION

Although we are beginning to have descriptive information about workplace violence, a number of research questions remain:

What are the specific tasks and environments that place workers at greatest risk?
What factors influence the lethality of violent incidents?
What are the relationships of workplace assault victims to offenders?
Are there identifiable precipitating events?
Were there any safety measures in place?
What were the actions of the victim and did they influence the outcome of the attack?
What are the most effective prevention strategies?

These questions should also be addressed in developing violence prevention strategies for specific workplaces.

A number of these questions were raised in 1990 at a workshop convened by NIOSH. They continue to require attention through the collaborative research and prevention efforts of public health, human resource, and criminal justice professionals. A number of other recommendations were made by a panel of experts in interpersonal violence on directions for NIOSH in this area [NIOSH 1992]. These recommendations have been implemented or initiated and include efforts to

In the fall of 1993, NIOSH released an Alert on preventing homicide in the workplace [NIOSH 1993] and encouraged employers, workers, unions, and others with a vested interest to look at their workplaces and take immediate action to reduce the risk for workplace homicide. In related efforts, NIOSH responded to numerous requests from the media, resulting in print, radio, and television coverage of the data and the NIOSH prevention message: Although no single intervention strategy is appropriate for all workpl aces and no definitive strategies can be recommended at this time, immediate action should be taken to reduce the toll of workplace homicide on our Nation's workforce. This message still holds true and applies not only to workplace homicide, but to all workplace violence. Clearly, violence is pervasive in U.S. workplaces, accounting for 1,071 homicides in 1994 and approximately a million nonfatal assaults each year. NIOSH continues to pursue research and prevention efforts to reduce the risk of workplace vi olence for the Nation's workers. The murder of an average of 20 workers each week is unacceptable and should not be considered the cost of doing business in our society.

REFERENCES CITED

Bachman R [1994]. Violence and theft in the workplace. In: U.S. Department of Justice Crime Data Brief. Washington, DC: U.S. Government Printing Office, NCJ�8199.

BLS [1993]. Employment and earnings. Washington, DC: U.S. Department of Labor, Bureau of Labor Statistics, January issue.

BLS [1994a]. Annual survey of occupational injuries and illnesses. Washington, DC: U.S. Department of Labor, Bureau of Labor Statistics. Unpublished database.

BLS [1994b]. National census of fatal occupational injuries, 1993. Washington, DC: U.S. Department of Labor, Bureau of Labor Statistics, BLS News, USDL��4.

BLS [1994c]. Violence in the workplace comes under closer scrutiny. Issues in labor statistics. Washington, DC: U.S. Department of Labor, Bureau of Labor Statistics, Summary 94�.

BLS [1994d]. Work injuries and illnesses by selected characteristics, 1992. Washington, DC: U.S. Department of Labor, Bureau of Labor Statistics.

BLS [1995]. National census of fatal occupational injuries, 1994. Washington, DC: U.S. Department of Labor, BLS News, USDL��8.

Brierley B [1996]. Personal communication on February 7, 1996, between B. Brierley of the IACP/Dupont Kevlar Survivors' Club and Lynn Jenkins, Division of Safety Research, National Institute for Occupational Safety and Health, Centers for Disease Control and Prevention, Public Health Service, U.S. Department of Health and Human Services.

Castillo DN [1994]. Nonfatal violence in the workplace: directions for future research. In: questions and answers in lethal and non-lethal violence: proceedings of the Third Annual Workshop of the Homicide Research Working Group. Washington, DC: National Institute of Justice.

Castillo DN, Jenkins EL [1994]. Industries and occupations at high risk for work-related homicide. J Occup Med 36:125�2.

Collins JJ, Cox BG [1987]. Job activities and personal crime victimization: implications for theory. Soc Sci Res 16:345�0.

Davis H [1987]. Workplace homicides of Texas males. Am J Public Health 77:1290�93.

Davis J, Honchar PA, Suarez L [1987]. Fatal occupational injuries of women, Texas 1975�84. Am J Public Health 77:1524�27.

FBI [1994]. Uniform crime reports for the United States, 1993. Washington, DC: U.S. Department of Justice, Federal Bureau of Investigation.

Jenkins EL [1994]. Occupational injury deaths among females: the U.S. experience for the decade 1980 to 1989. Ann Epidemiol 4(2): 146�1.

Jenkins EL [1996]. Workplace homicide: industries and occupations at high risk. Occup Med State of Art Reviews 11(2):219�5.

Jenkins EL, Layne LA, Kisner SM [1992]. Homicide in the workplace: the U.S. experience, 1980�88. AAOHN J 40:215�8.

Jenkins EL, Kisner SM, Fosbroke DE, Layne LA, Stout NA, Castillo DN et al. [1993]. Fatal injuries to workers in the United States, 1980�89: a decade of surveillance; national profile. Washington, DC: U.S. Government Printing Office, DHHS (NIOSH) Publication No. 93�8.

Kraus JF [1987]. Homicide while at work: persons, industries, and occupations at high risk. Am J Public Health 77:1285�89.

Lynch JP [1987]. Routine activity and victimization at work. J Quantitative Criminol 3: 283�0.

NIOSH [1992]. Homicide in U.S. workplaces: a strategy for prevention and research. Morgantown, WV: U.S. Department of Health and Human Services, Public Health Service, Centers for Disease Control, National Institute for Occupational Safety and Health, DHHS (NIOSH) Publication No. 92�3.

NIOSH [1993]. NIOSH Alert: request for assistance in preventing homicide in the workplace. Cincinnati, OH: U.S. Department of Health and Human Services, Public Health Service, Centers for Disease Control and Prevention, National Institute for Occupational Safety and Health, DHHS (NIOSH) Publication No. 93�9.

NIOSH [1995]. National Traumatic Occupational Fatalities (NTOF) Surveillance System. Morgantown, WV: U.S. Department of Health and Human Services, Public Health Service, Centers for Disease Control and Prevention, National Institute for Occupational Safety and Health. Unpublished database.

Northwestern National Life [1993]. Fear and violence in the workplace: a survey documenting the experience of American workers. Minneapolis, MN: Northwestern National Life.

O'Carroll PW, Mercy JA [1989]. Regional variation in homicide rates: why is the West so violent? Violence and Victims 4(1):17�.

State of California [1993]. Guidelines for security and safety of health care and community service workers. Sacramento, CA: Division of Occupational Safety and Health, Department of Industrial Relations.

Stout NA, Jenkins EL, Pizatella TJ [1996]. Occupational injury mortality rates in the United States: changes from 1980 to 1989. Am J Public Health 86:73�.

Toscano G, Weber W [1995]. Violence in the workplace. Compensation and working conditions. Washington, DC: U.S. Department of Labor, Bureau of Labor Statistics.

Windau J, Toscano G [1994]. Workplace homicides in 1992.

Compensation and working conditions, February 1994. Washington, DC: U.S. Department of Labor, Bureau of Labor Statistics.

RELATED READING

AFSCME (American Federation of State, County, and Municipal Employees) and AFL-CIO [1995]. Hidden violence against women at work. Women in Public Service 5(fall):1�

Alexander BH, Franklin GM, Wolf ME [1994]. The sexual assault of women at work in Washington State, 1980 to 1989. Am J Public Health 84:640�2.

Amandus HE [1994]. Status of NIOSH research on prevention of robbery-related intentional injuries to convenience store workers. In: questions and answers in lethal and non-lethal violence: proceedings of the Third Annual Workshop of the Homicide Research Working Group. Washington, DC: National Institute of Justice.

Amandus HE, Hunter RD, James E, Hendricks S [1995]. Reevaluation of the effectiveness of environmental designs to reduce robbery risk in Florida convenience stores. J Occup Environ Med 37(6):711�7.

Barab J [1995]. Workplace violence: how labor sees it. New Solutions, spring issue.

Bell CA [1991]. Female homicides in United States workplaces, 1980�85. Am J Public Health 81:729�2.

Bell CA, Stout NA, Bender TR, Conroy CS, Crouse WE, Myers JR [1990]. Fatal occupational injuries in the United States, 1980 through 1985. JAMA 263(22):3047�50.

Bensley L, Nelson N, Kaufman J, Silverstein B, Kalat J [1993]. Study of assaults on staff in Washington State Psychiatric Hospitals. Olympia, WA: State of Washington Department of Labor and Industries.

BJS [1994]. Criminal victimization in the United States, 1992: a national crime victimization survey report. Washington, DC: U.S. Department of Justice, Bureau of Justice Statistics, NCJ�5125, p. 79.

Block R, Felson M, Block C [1984]. Crime victimization rates for incumbents of 246 occupations. Sociol Soc Res 69:442�1.

Borwegen W [1995]. Violence as a preventable occupational hazard: a labor perspective. New Solutions, spring issue.

Bowie V [1989]. Coping with violence: a guide for the human services. Sydney, Australia: Karibuni Press.

Cabral R [1995]. Workplace violence: viable solutions under collective bargaining. New Solutions, spring issue.

Carmel C, Hunter M [1989]. Staff injuries from inpatient violence. Hosp Community Psychiatry 40:41�.

CDC (Centers for Disease Control and Prevention) [1994]. Occupational injury deaths of postal workers桿nited States, 1980�89. MMWR 43(32):587, 593�5.

Chapman SG [1986]. Cops, killers and staying alive: the murder of police officers in America. Springfield, IL: Charles C. Thomas.

Chelimsky E, Jordan FC, Russell LS, Strack JR [1979]. Security and the small business retailer. Washington, DC: National Institute of Law Enforcement and Criminal Justice, Law Enforcement Assistance Administration, Department of Justice.

Civil Service Employees Association [1993]. A matter of life and death: worksite security and reducing risks in the danger zone. Albany, NY: Civil Service Employees Association, Inc.

Crime Prevention Division [1990]. Safety tips for the taxi driver and the for-hire vehicle driver. New York, NY: City of New York Police Department.

Crow W, Bull JL [1975]. Robbery deterrence: an applied behavioral science demonstration. La Jolla, CA: Western Behavioral Sciences Institute.

Erickson R [1991]. Convenience store homicide and rape. In: Convenience store security: report and recommendations. Alexandria, VA: National Association of Convenience Stores.

FBI [1992]. Killed in the line of duty: a study of selected felonious killings of law enforcement officers. Washington, DC: U.S. Department of Justice, Federal Bureau of Investigation.

Friedman RJ, Framer MB, Shearer DR [1988]. Early response to post-traumatic stress. EAP Digest, September/October issue.

Goetz RR, Bloom JD, Chenell SL, Moorhead JC [1991]. Weapons possession by patients in a university emergency department. Ann Emerg Med 20:8�.

Goodman RA, Jenkins EL, Mercy JA [1994]. Workplace-related homicide among health care workers in the United States, 1980 through 1990. JAMA 272(21):1686�88.

Hales T, Seligman P, Newman SC, Timbrook CL [1988]. Occupational injuries due to violence. J Occup Med 30:483�7.

Jama 100 years ago: assaults upon medical men [1992]. JAMA 267(22):2987.

Keep N, Gilbert P [1992]. California Emergency Nurses Association's informal survey of violence in California emergency departments. J Emerg Nurs 18:433�2.

Lavoie FW, Carter GL, Danzi DF, Berg RL [1988]. Emergency department violence in United States teaching hospitals. Ann Emerg Med 17:1227�33.

Levin PF, Hewitt JB, Misner ST [1992]. Female workplace homicides: an integrative research review. AAOHN J 40:229�6.

Lipscomb JA, Love CC [1992]. Violence toward health care workers: an emerging occupational hazard. AAOHN J 40:219�8.

Liss GM, Craig CA [1990]. Homicide in the workplace in Ontario: occupations at risk and limitations of existing data sources. Can J Public Health 81:10�.

Lusk SL [1992]. Violence experienced by Nurse's aides in nursing homes: an exploratory study. AAOHN J 40:237�1.

Mahoney BS [1991]. Victimization of Pennsylvania emergency department nurses in the line of duty [dissertation]. State College, PA: Pennsylvania State University.

Nelson N [1995]. Violence in Washington workplaces, 1992. Olympia, WA: Washington State Department of Labor and Industries, Technical report 39𢴏995.

Norris D [1990]. Violence against social workers: the implications for practice. London, England: Jessica Kingsley Publishers.

Office of the Attorney General [1991]. Study of safety and security requirements for "at-risk businesses." Tallahassee, FL: Office of the Attorney General, State of Florida, Department of Legal Affairs.

Oregon Department of Consumer and Business Services [1994]. Violence in the workplace, Oregon, 1988 to 1992: a special study of workers' compensation claims caused by violent acts. Salem, OR: The Department Information Management Division.

Public Law 91�6: Occupational Safety and Health Act of 1970.

Reeves FL, Hall JA [1990]. Taxicab operators' safety program. Atlanta, GA: Department of Public Safety, Bureau of Taxicabs and Vehicles for Hire.

Rowett CR [1986]. Violence in social work: a research study of violence in the context of local authority social work. Cambridge, England: University of Cambridge Institute of Criminology.

Safety Shields Task Force, Manitoba Taxicab Board [1990]. Report and recommendations: taxi safety shields. Manitoba, Canada: The Task Force.

Schreiber FB [1991]. 1991 national survey of convenience store crime and security. In: Convenience store security: report and recommendations. Alexandria, VA: National Association of Convenience Stores.

Seligman PJ, Newman SC, Timbrook CL, Halperin WE [1987]. Sexual assault of women at work. Am J Ind Med 12:445�0.

Shepherd E, ed. [1994]. Violence in health care: a practical guide to coping with violence and caring for victims. New York, NY: Oxford University Press.

Stout NA [1992]. Occupational injuries and fatalities among health care workers in the United States. Scand J Work Environ Health 18(Suppl 2):88�.

Sullivan C, Yuan C [1995]. Workplace assaults on minority health and mental health care workers in Los Angeles. Am J Public Health 85(7):1011�14.

Thomas JL [1992]. Occupational violent crime: research on an emerging issue. J Safety Res 23:55�.

Toscano G, Windau J [1994]. The changing character of fatal work injuries. Monthly Labor Review 117(10):17�.

Uchida CD, Brooks LW, Koper C [1987]. Danger to police during domestic encounters: assaults on Baltimore county police, 1984�86. CJPR 2:357�1.

Zimmering FE, Zuehl J [1986]. Victim injury and death in urban robbery: a Chicago study. J Leg Studies 15(1):1�.

FIGURES

Figure 1. Leading causes of occupational injury deaths桿nited States, 1980�. Data were not available for New York City and Connecticut. (Source: Jenkins [1996].)

Figure 2. Work-related homicides by method and year. Data were not available for New York City and Connecticut. (Source: NIOSH [1995].)

Figure 3. Violent acts resulting in days away from work, by source of injury桿nited States, 1992. (Source: BLS [1994a].)

Back to TAXI-L Home Page