2
Assessment of Programmatic Relevance and Impact

This chapter constitutes the committee’s response to the first part of the charge to assess the National Institute of Occupational Safety and Health (NIOSH) Traumatic Injury (TI) Research Program’s contribution—through occupational safety and health research—to reductions in workplace hazardous exposures, illnesses, or injuries through

  • An assessment of the relevance of the program’s activities to the improvement of occupational safety and health; and

  • An evaluation of the impact that the program’s research has had in reducing work-related hazardous exposures, illnesses, and injuries.

This chapter reviews the inputs, activities, outputs, and outcomes of each of the TI Research Program’s eight goals. Much of the evidence derives from the evidence package prepared by TI Research Program staff. While the committee had information on extramural TI research grants, the information in the evidence package was presented with a focus on the work of the staff in the Division of Safety Research (DSR). Thus, the committee’s evaluation is primarily of the TI Research Program’s intramural research and related efforts. The committee considered the TI Research Program under review to include only those activities, outputs, and outcomes included in the evidence package and evaluated the program on that basis. The committee did not attempt to independently verify the facts in the package (for example, a statement of how many brochures were mailed was believed to be ac-



The National Academies | 500 Fifth St. N.W. | Washington, D.C. 20001
Copyright © National Academy of Sciences. All rights reserved.
Terms of Use and Privacy Statement



Below are the first 10 and last 10 pages of uncorrected machine-read text (when available) of this chapter, followed by the top 30 algorithmically extracted key phrases from the chapter as a whole.
Intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text on the opening pages of each chapter. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.

Do not use for reproduction, copying, pasting, or reading; exclusively for search engines.

OCR for page 43
2 Assessment of Programmatic Relevance and Impact T his chapter constitutes the committee’s response to the first part of the charge to assess the National Institute of Occupational Safety and Health (NIOSH) Traumatic Injury (TI) Research Program’s contribution—through occupational safety and health research—to reductions in workplace hazardous exposures, illnesses, or injuries through • An assessment of the relevance of the program’s activities to the im- provement of occupational safety and health; and • An evaluation of the impact that the program’s research has had in reducing work-related hazardous exposures, illnesses, and injuries. This chapter reviews the inputs, activities, outputs, and outcomes of each of the TI Research Program’s eight goals. Much of the evidence derives from the evidence package prepared by TI Research Program staff. While the committee had informa- tion on extramural TI research grants, the information in the evidence package was presented with a focus on the work of the staff in the Division of Safety Research (DSR). Thus, the committee’s evaluation is primarily of the TI Research Program’s intramural research and related efforts. The committee considered the TI Research Program under review to include only those activities, outputs, and outcomes included in the evidence package and evaluated the program on that basis. The committee did not attempt to independently verify the facts in the package (for example, a statement of how many brochures were mailed was believed to be ac- 43

OCR for page 43
t R au m at I C I n j u Ry R e s e a R C h nIosh 44 at curate), nor did the committee seek to expand the evidence by searching outside the evidence package. Although the committee was aware of some NIOSH-sponsored work that did not appear in the evidence package but that the committee would have thought part of the TI Research Program, the committee decided to evaluate only what was presented to them. The committee did use its expert judgment and knowledge of the field to evaluate claims in the evidence package about the role of NIOSH-funded work in intermediate and end outcomes. The committee did not attempt to read and evaluate the quality of each research project or each dissemina- tion product. The committee’s evaluation was informed by the evidence package, committee expertise in the field, stakeholder input, and published material. This chapter begins with a section describing the external factors that impact the work the TI Research Program undertakes, as well as the outcomes one might reasonably expect. This is followed by a section describing issues, methods, and challenges in occupational injury surveillance, which cuts across all eight program goals. EXTERNAL FACTORS WITH BROAD IMPACT ON THE TI RESEARCH PROGRAM External factors “may be considered as forces beyond the control of NIOSH that may affect the evolution of the program” (see Appendix A, p. 137). They may influence NIOSH research at any phase. External factors identified by the com- mittee that broadly impact NIOSH research activities, including those of the TI Research Program, include budgetary issues (such as congressional funding and earmarks), lack of complete occupational injury surveillance data, inadequate action by regulatory agencies such as the Occupational Safety and Health Admin- istration (OSHA), the sharing of research responsibility with other agencies, and industry challenges; each of these is discussed below. External factors also impact the TI Research Program’s work in specific goal areas. Some of these are discussed in the sections that appear later in this chapter outlining each of the eight goals. Budgetary Issues NIOSH was appropriated $286 million in fiscal year (FY) 2005 to carry out its mission as the sole federal agency responsible for conducting research for the prevention of occupational injury and illness (NIOSH, 2007a, p. 17). The total TI Research Program budget for all eight research goals in FY2005 was $17.2 million (see Table 1-2), an amount the committee finds to be inadequate given the scope of the TI Research Program’s mandate. More recently, the NIOSH budget has de- creased; it was appropriated $255 million and $253 million for research in FY2006 and 2007, respectively (NIOSH, 2007a, p. 17). The FY2006 decrease reflected a new

OCR for page 43
p R o g R a m m at I C R e l e va n C e I m paC t 45 and Centers for Disease Control and Prevention (CDC) budget structure that draws from program allocations to support operational (overhead and indirect) costs. According to the evidence package, in 2006 Congress redirected $35 million from the NIOSH budget appropriation to CDC for business support services (NIOSH, 2007a, p. 17). Congressional funds received by the TI Research Program are sometimes ear- marked for a specific purpose, such as research that targets a specific industry or population. Such mandates have led to TI Research Program initiatives and ensured sustained research funding in several specific areas such as children in agriculture, workers in Alaska, firefighters, and workplace violence (NIOSH, 2007b, p. 47). Congress or the Department of Health and Human Services (HHS) can also assign NIOSH the task of leading or participating in research on evolving public health issues (NIOSH, 2007a, p. 17). For example, after the September 11, 2001, terrorist attacks, Congress and HHS directed NIOSH to evaluate the health of workers and volunteers at Ground Zero. These targeted areas can be productive, but the strategic decision of where to focus funds and effort is sometimes external to NIOSH and its own strategic thinking and analysis. Incomplete Occupational Injury Surveillance Data As will be discussed in greater detail in the following section, surveillance data on occupational injuries is incomplete, particularly for nonfatal injuries, due to factors such as gaps in the scope of surveillance and poor reporting of injuries in national surveys. Far more nonfatal occupational injuries than fatal occupational injuries occur each year (BLS, 2008), and risk factors for fatal and nonfatal injuries are not necessarily the same. Fatal injuries are not surrogates for nonfatal injuries. Yet the TI Research Program has had to rely primarily on fatality data to inform its research efforts. Key variables missing from data sets also interfere with the TI Research Pro- gram’s ability to do research in particular goal areas. Examples cited in the evi- dence package include lack of work relationship information in current crash data systems, which impedes the identification of risk factors for occupational crashes (NIOSH, 2007c, p. 58) and the lack of accurate data on the number of active firefighters (career and volunteer) and the number of hours of exposure, which precludes calculation of accurate firefighter fatality rates (NIOSH, 2007d, p. 154). Inadequate Regulatory Action The impact of TI Research Program activities is limited by NIOSH’s mandated inability to regulate and enforce measures that impact the occurrence of occupa-

OCR for page 43
t R au m at I C I n j u Ry R e s e a R C h nIosh 46 at tional injuries. According to the Occupational Safety and Health Act (OSH Act) of 1970, NIOSH is responsible for conducting research and making recommendations for occupational safety and health standards. Criteria Documents are developed to provide the basis for comprehensive occupational safety and health standards. These documents generally contain a critical review of the scientific and techni- cal information available on the prevalence of hazards, the existence of safety and health risks, and the adequacy of methods to identify and control hazards. Special Hazard Reviews (SHRs) and Occupational Hazard Assessments are written to complement NIOSH recommendations for standards. These documents assess safety and health problems and recommend appropriate methods for control and monitoring.1 Although OSHA’s work is informed by NIOSH research and NIOSH research can affect change in the workplace absent OSHA regulation, OSHA—in the De- partment of Labor (DOL)—is the agency ultimately responsible for the develop- ment and enforcement of workplace safety and health regulations.2 Regulatory inaction on the part of OSHA can delay the translation of TI Research Program findings into workplace policies. Additionally, depending on the research area, inaction or underachieved expectations on the part of regulatory agencies other than OSHA (e.g., the Department of Transportation or the U.S. Department of Agriculture [USDA] or other agencies) can also stall translation of TI Research Program findings. In the absence of regulations, employers may be less likely to adopt safety measures. Exemptions from federal occupational safety and health regulations—such as the exemption from the 1970 OSH Act of employers of public-sector workers, the congressional exemption for farms employing fewer than 11 people, and the fam- ily farm exemption in federal child labor laws—are also barriers to application of some of NIOSH’s research findings. Sharing Research Responsibility with Other Agencies Garnering support for research where there is a shared responsibility can be complicated when the occupational component represents a relatively small pro- portion of the overall societal problem. For example, this has been true for NIOSH in the area of injuries due to motor vehicle incidents; although NIOSH focuses solely on occupation-related crashes, it still shares research responsibility with the 1 During the time period of the committee’s evaluation, NIOSH has issued four Criteria Documents and one SHR. The last Criteria Document was issued in 2006; the previous was issued in 1998. Only the 1997 SHR, Child Labor Research Needs, addressed a TI concern. 2 OSHA and the Mine Safety and Health Administration have delegated approval of respirators to NIOSH.

OCR for page 43
p R o g R a m m at I C R e l e va n C e I m paC t 47 and National Highway Traffic Safety Administration (NHTSA). Occupation-related crashes represent only about 3 percent of the total number of motor vehicle crashes (NIOSH, 2007c, p. 58), a fact that has made it difficult for the TI Research Program to synergistically and strategically engage the broader traffic safety community and also results in the exclusion of some occupational risk factors from national policy (e.g., exclusion of ambulances from crashworthiness consideration). To some extent the same has also been true for research on workplace violence, because violence that occurs in the workplace is a comparatively small part of the overall problem of violence. Sharing research responsibility with other agencies can also make causal at- tribution of end outcomes to the TI Research Program’s research findings more complicated (e.g., decreases in occupationally related motor vehicle deaths and workplace homicides can be reflections of an overall change in motor vehicle deaths and homicides rather than occupational initiatives). Industry Challenges Some challenges to conducting research are related to the nature of the indus- tries targeted by the TI Research Program—for example, the mobile nature of the agriculture and construction industries and the small-business composition of the retail industry. It is difficult to enlist employer participation in studies to prevent rare events. There are also barriers to corporate investment in injury prevention technologies that are related to there being no accepted way of predicting how improvements in workplace safety will reduce the risk of worker injury. ORGANIZATIONAL DESIGN Strategically and organizationally, the TI Research Program supports a larger NIOSH mission. In terms of organizational design, the TI Research Program has been an instrumental part of the crosscutting NIOSH Program Portfolio Matrix Management Initiative (Table 1-1). NIOSH’s formal organizational structure de- picts DSR as part of a product/service organizational structure with geographic distribution. Its Matrix Management Initiative has recently been reorganized to be industrial-sector programs × crosscutting programs,3 of which the TI Research Program is a component. The traditional benefits of matrix management include the flexibility derived from resourcing from both axes of the matrix. The TI Research Program goals 3As described in Chapter 1, the coordinated emphasis areas have been moved into the cross- cutting programs.

OCR for page 43
t R au m at I C I n j u Ry R e s e a R C h nIosh 48 at represent a mix of risk factors in different occupations, various work environments, and different vulnerable populations. Since goals are in crosscutting categories and NIOSH is organized in a matrix, the assumption is that the structure (matrix) will support the strategy of crosscutting research, but previous users of matrix organi- zational structures caution that the structure can also be a source of overlap and inefficient use of resources. As an example, “fall from heights,” a major source of occupational injury and fatality, is a major problem across sectors (construction falls from rooftops, surface mining falls from heavy equipment, falls from towers in telecommunications, etc.). Falls may be experienced disproportionately by some worker groups (e.g., Latinos) as well as in certain firms (e.g., small). Given the relatively modest budget for occupational health and safety research in general, the organizational structure should support information sharing and efficient use of resources in such cases, rather than having multiple, parallel efforts to address a common issue. Of particular concern is whether the crosscutting programs and the sector programs are maximizing the efficient use of resources and information. Matrix management is an attempt to provide flexibility in terms of personnel assignment and reassignment and in a sense, an attempt to allow the benefits of the two axes while promoting synergistic additional benefits. In terms of injury prevention, there is great potential. To illustrate the potential of crosscutting and true matrices, consider Haddon’s matrix (Haddon, 1968, 1972). Combining his medical and engineering expertise, Haddon provided a “matrix” for analyzing injury based on the person injured, the cause of the injury, and the environment and for identifying temporal factors across the pre-event, event, and post-event life cycle. To further investigate the root cause of falls from heights discussed earlier, kinetic energy is closer to the root cause and, in fact, is the most common agent that causes occupational injury (e.g., tissue tearing, crushing with bleeding, organ failure and death from car crashes, falls, gunshot wounds). While prevention, modification, or attenuation of the release of this energy is common across all of the sectors that NIOSH has created, the TI Research Program (as well as many or most others) may be stuck between sectors, simply because of the way NIOSH is organized. One potential solution—if reorganization is not desired or practical—is to use something like Haddon’s matrix as what is called in the organizational field an “integrating mechanism” that facilitates the organization and integration of common information. In summary, there are common pitfalls with matrix management that should be noted as the TI Research Program and NIOSH continue to evolve. Internally, there can be confusion over lines of authority for personnel attached to multiple programs or units within the organization. Although this classic challenge was not observed in the committee’s review, the related issue of project “ownership” and associated information sharing was observed. That is, it was sometimes difficult to fully un-

OCR for page 43
p R o g R a m m at I C R e l e va n C e I m paC t 49 and derstand which projects fell under the purview of the TI Research Program, because of its positioning within a matrix structure. The committee is an external body to which NIOSH needed to communicate its framework structure and processes, and committee members were occasionally confused about the TI Research Program’s responsibilities within that framework. As NIOSH continues to work externally, for example, in sharing responsibility with other agencies, it will have to commu- nicate effectively how it is structured organizationally (i.e., product/service), how it is managed (matrix), and how it operates (logic model), and more importantly, it should effectively share and manage its portfolio of injury prevention knowledge in a resource efficient manner (e.g., effective integrating mechanisms). SURVEILLANCE Surveillance of traumatic injuries provides the empirical basis for setting research priorities and for evaluating the impact of research, knowledge transfer, and intervention activities. Although substantial strides have been made over the past 20 years in the development of a national surveillance system for fatal occu- pational injuries, there is widespread evidence of substantial disparities between the number of nonfatal occupational injuries that are reported and the actual number that occur. A series of studies beginning in the 1980s demonstrated that both nonfatal and fatal injuries are underreported (Azaroff et al., 2002; Stout and Bell, 1991). Based on these studies—many of which were performed or supported by NIOSH—the TI Research Program developed the NTOF (National Traumatic Occupational Fatality) Surveillance System. This system provided much improved reporting of injuries resulting in death and contributed to the creation of the Census of Fatal Occupational Injuries (CFOI) in 1992. Improved surveillance of fatal occupational injuries is an impressive achievement of the TI Research Program and is currently an important input into the TI Research Program’s planning process. The TI Research Program also expanded fatality surveillance by establishing the Fatal Assessment Control and Evaluation (FACE) Program.4 This program, con- ducted in collaboration with state partners, provides in-depth information about 4 The NIOSH FACE Program concentrates on investigations of fatal occupational injuries. It has two components—an in-house program that began in 1982 in which participating states voluntarily notify NIOSH of traumatic occupational fatalities resulting from targeted causes of death (currently, deaths associated with machinery, deaths of youths under 18 years of age, and street/highway construction work zone fatalities) and a state-based program that began in 1989 in which nine states with coopera- tive agreements with NIOSH conduct surveillance, targeted investigations, and prevention activities using the FACE model (NIOSH, 2008). The number of state-based FACE programs has declined in recent years due to lack of funding despite a growing interest and expertise on the part of the states.

OCR for page 43
t R au m at I C I n j u Ry R e s e a R C h nIosh 50 at targeted types of deaths. This information—collected through onsite, research- oriented field investigations of fatal incidents—is used to develop recommenda- tions to prevent future deaths. These recommendations are in turn disseminated widely to industry and labor professionals, manufacturers, policy makers, and health and safety professionals. Although substantial strides have been made in the surveillance of fatal oc- cupational injuries, many types of traumatic workplace injury rarely result in death, and fatal injuries are not good surrogates for nonfatal injuries. Planning and evaluation also depend on accurate surveillance of nonfatal injuries. It is widely recognized that the current major source of national surveillance data on nonfatal occupational injuries—the Bureau of Labor Statistics (BLS) Annual Survey of Injuries and Illnesses—is very incomplete. National estimates of injuries derived from this survey exclude approximately 22 percent of the U.S. workforce, includ- ing workers in the public sector and the self-employed (Leigh et al., 2004). There is also mounting evidence of injury underreporting on the OSHA-required injury logs maintained by employers on which the survey is based (Conway and Svenson, 1998; Rosenman et al., 2006). Moreover, completeness of reporting likely varies by type of injury, employer characteristics, and worker characteristics, although due to the nature of the problem the extent of variability is unknown. Gaps in surveillance and incomplete reporting result in substantial underesti- mates of the national burden of occupational injuries. As a consequence, decisions based on these data may allocate fewer resources to research and interventions designed to reduce workplace hazards. Secular changes in reporting over time can (if reporting improves) suggest that successful programs have failed or (if reporting declines) suggest that ineffective programs have succeeded. In addition, differen- tial reporting by condition, subpopulation, or employer type can lead to focusing prevention efforts in the wrong areas. Injuries for which we have particularly poor measures of incidence may receive inadequate public health attention. The TI Research Program has carried out substantial surveillance activities and research in order to provide more reliable surveillance data, including data on nonfatal injuries. Perhaps the most innovative of these efforts has focused on ag- riculture and young workers. The Child Agricultural Injury Program, for example, has developed several new surveillance activities designed to provide an ongoing picture of the incidence of injuries among both adults and youths. For adult farm workers, Occupational Injury Surveillance of Production and Agriculture (OISPA) surveys farm operators to estimate the number and incidence of occupational inju- ries to adults working on farms. In addition, NIOSH has developed a farm worker injury module to supplement data already collected from workers by the DOL in the National Agricultural Workers Survey (NAWS). This survey includes both adult and young workers. Using ongoing survey relationships of the National Agricultural

OCR for page 43
p R o g R a m m at I C R e l e va n C e I m paC t 5 and Statistics Service (NASS), NIOSH has collected information from farm operators about agricultural injuries to youth in 1998, 2001, and 2004 (the Childhood Agri- cultural Injury Surveys [CAIS]). A parallel survey of minority farms, the Minority Farm Operator Childhood Agricultural Injury Surveys (M-CAIS), was conducted in 2000 and 2003. These surveys include a project that ascertained the number of agricultural injuries incurred by ninth to twelfth graders in Minnesota as reported by emergency departments. Since the early 1990s, NIOSH has had an interagency agreement with the Consumer Product Safety Commission to collect data on nonfatal occupational injuries through the National Surveillance of Nonfatal Occupational Injuries Using the National Electronic Injury Surveillance System (NEISS). NEISS collects data on injuries resulting in emergency department visits from a sample of U.S. hospitals. Not all injuries are seen in emergency departments, but the agreement has allowed NIOSH to collect data on nonfatal injuries for all industries and occupations and has allowed for better understanding of risk factors, particularly for those injuries that are more likely to require emergency department visits. Through the agree- ment, NIOSH has also been able to conduct follow-up surveys of injured workers to collect more detailed information on their injuries. To describe risks to working youth, NIOSH has used NEISS and developed new surveillance sources both by building on existing surveys conducted by other agencies and by funding new surveillance activities. These include new state- based surveillance activities in Wisconsin and Massachusetts focused on working youth. Despite these activities, there remains a paucity of high-quality national data on nonfatal injuries. This may be one reason that the TI Research Program frequently seems to focus on fatality data in setting program priorities. Although fatal injuries should carry substantial weight, too much emphasis on fatalities can lead to an underinvestment in research and transfer activities relevant to injury types that are very rarely fatal yet impose significant human and economic costs. Notably, occupational injury surveillance is dependent upon having valid information about the number of injuries that occur. It also requires information about populations at risk (denominator data) necessary to generate injury rates that allow for identification of disproportionate risks among segments of the population and subsequent priority setting. Better information on employment and hours at work (as well as injuries) is also needed, especially for the vulner- able populations such as youth and immigrant and minority workers whose work experience may not be well captured in traditional sources of employment data. Optimally, information on worker exposure to hazards would be available to allow for exposure-based estimates of risk. NIOSH, in concert with other agencies, can work to develop better systems for capture of denominator data.

OCR for page 43
t R au m at I C I n j u Ry R e s e a R C h nIosh 52 at In summary, NIOSH has engaged in a range of activities, both intramural and extramural, that have made strong contributions to improving surveillance. Still, these projects do not appear to be part of a coordinated interagency strategy to improve national surveillance of traumatic nonfatal occupational injuries. For example, there is little research funded to identify the strengths and weaknesses of current national systems or to find ways of combining data to generate better estimates of injury incidence. NIOSH has used population-based surveillance data, but more work could be performed to identify the strengths and weaknesses of new survey-based systems or to attempt to compare or combine data. In its future TI Research Program activities, NIOSH would benefit from an overall strategy of sur- veillance research and implementation designed to lead to a better national nonfatal injury surveillance program. This might be considered as a separate goal. GOAL 1: REDUCE INJURIES AND FATALITIES DUE TO MOTOR VEHICLES NIOSH and the TI Research Program’s involvement in motor vehicle injury research began in the late 1990s in response to data demonstrating a lack of progress in reducing work-related crashes and fatalities. At this time, TI Research Program staff also recognized the problem of poor data capture tying crash incidents to oc- cupational injuries in crash data systems. No one had yet taken the lead in making improvements in this area (NIOSH, 2007c, p. 53). The committee supports NIOSH’s decision to become a key player in motor vehicle injury research. There is a niche for NIOSH to evaluate aspects of motor ve- hicle injuries and workplace intervention strategies that are not already researched by other agencies. NIOSH developed two subgoals for research aimed at reducing injuries and fatalities due to motor vehicles: 1.1. Reduce occupational injuries and fatalities due to highway motor vehicle crashes5 1.2. Reduce occupational injuries and fatalities due to motor vehicle incidents in highway and street construction work zones Planning and Production Inputs Planning inputs were comprehensive and supported the need for research in both subgoal areas. Inputs were both qualitative and quantitative. 5Subgoal 1.1 specifically targeted (1) professional truck drivers and (2) workers who drive or ride in motor vehicles during work-related travel but who are not professional drivers (NIOSH, 2007c, p. 54).

OCR for page 43
p R o g R a m m at I C R e l e va n C e I m paC t 53 and Subgoal 1.1 inputs included fatal and nonfatal crash injury data from the BLS, which allowed identification of important crash risk factors for truck drivers. A qualitative source of input was feedback from stakeholders at conferences and meetings, which revealed agreement on a need for additional surveillance and risk factor research for truck drivers. The TI Research Program’s research on workers who drive or ride in motor vehicles during work-related travel, but who are not professional drivers, was initiated to address a research and regulatory gap (NIOSH, 2007c, pp. 54-55). An important input into the TI Research Program’s work on subgoal 1.2 was a review of BLS data, which showed that most fatalities occurring at road con- struction sites between 1995 and 2002 were reported to be transportation-related incidents (NIOSH, 2007c, p. 62). There had also been concern among stakeholders that increases in worker injuries would occur in coming years due to increases in road construction and structural changes in the industry, making this a potentially emerging issue (NIOSH, 2007c, p. 62). Combined intramural–extramural funding for motor vehicle research for the period FY1997-2005 totaled $12,657,927. Only the back injury program had lower total funding for this period. With the exception of a small funding decline in FY2003, intramural funding for motor vehicle research increased each year from FY1997 to FY2005, reflecting the growth of this program over time. However, the proportion going to extramural research began to decline in FY2003, and in FY2005, it accounted for less than 1 percent of the total funding for motor vehicle research (see Table 1-2). Full-time equivalents (FTEs) increased from a low of 4.3 in FY1998 to 14.8 in FY2005. Activities Activities for subgoal 1.1 included surveillance research as well as research to identify crash risk factors and interventions. Perhaps owing to the fact that NIOSH and the TI Research Program are relatively new to motor vehicle safety research, activities for subgoal 1.1 were rather limited given the broad scope of the goal. The committee agrees with the TI Research Program that it is continu- ing to develop a niche within transportation safety research. Appropriately, the TI Research Program has been working with partners such as the American Society of Safety Engineers and the Network of Employers for Traffic Safety to learn how it can best serve employers and employee drivers through research and outreach (NIOSH, 2007c, p. 55). Having focused motor vehicle safety research goals (rather than the broader goal of reducing injuries and fatalities due to highway motor ve- hicle crashes) will be important for informing the TI Research Program’s activities going into the future.

OCR for page 43
t R au m at I C I n j u Ry R e s e a R C h nIosh 00 at Program statistics on injuries among youth working on farms in the proposed Children’s Act for Responsible Employment (CARE) of 200556 (CARE Act, H.R. 3482). This act proposed changes to child labor laws in agriculture and identified the TI Research Program youth farm injury data as a source of data for an an- nual report on injuries to youth working on farms in the United States. The 2002 NIOSH report that provided recommendations to ESA on changes to child labor laws has been used by others advocating for changes in child labor regulations in the agriculture industry (NIOSH, 2007k, p. 183). The TI Research Program also notes an end outcome for subgoal 8.2—namely, a decline in the number of injuries and the rates of nonfatal injuries among agri- cultural workers under age 18 between 1998 and 2004. The number of work-related injuries decreased 50 percent over this period (from 11,970 in 1998 to 5,740 in 2004), and the rates of nonfatal, work-related injuries decreased 30 percent (from 10.5 per 1,000 working youths in 1998 to 7.3 per 1,000 working youths in 2004; NIOSH, 2007k, p. 184).57 The committee believes that the contributions of the TI Research Program, in particular through outreach activities done in collaboration with partners, likely played some role. As an intermediate outcome for subgoal 8.3, young worker safety efforts have continued for several state-based teams. Work of investigators with community- based demonstration projects in Oakland, California, and the Northeast Young Worker Resource Center has been extended through funding from OSHA (NIOSH, 2007k, p. 189). The Teens at Work surveillance project in Massachusetts resulted in a technological intervention. The project identified hot coffee and slurry from cof- fee brew baskets as a common source of occupational injury among Massachusetts teens. In response to the finding, the corporate headquarters of a national bakery chain in 2001 began to require that store owners purchasing new equipment in- stall brew baskets and shields to prevent spillage. A retrofit kit was designed and made available to store owners who are not purchasing new equipment (NIOSH, 2007k, p. 189). Stakeholder interest in the TI Research Program’s work in this area has been evidenced by requests for additional copies of TI Research Program- developed young worker safety and health materials; incorporation of materials into school-based occupational safety and health classes; and citation of TI Re- search Program statistics, findings, and prevention recommendations by the press and in safety programs (NIOSH, 2007k, p. 190). 56 Theact was submitted in the House of Representatives in July 2005, but never passed. 57 Thediscrepancy between the rate and the absolute number derives from decreasing numbers of youth working in agriculture.

OCR for page 43
p R o g R a m m at I C R e l e va n C e I m paC t 0 and Discussion The TI Research Program’s activities for young workers have demonstrated the significance of innovative, as well as continuing, surveillance activities. The value of such surveillance data to focused approaches cannot be overemphasized in terms of the benefits likely to accrue. In other areas of research activity, however, it is clear that many activities are better described as research to practice (r2p). While the committee endorses the TI Research Program’s efforts to work with various constituencies to implement products and programs, it appears that there is a significant danger that actual research into the causes of traumatic injury is underemphasized relative to implementation. The choice to add youth to the FACE Program in 1999 was a good one, and the committee supports continuation of a scaled-down youth FACE Program in the future if new interventions are no longer being identified. The young worker activity also focused on the development of legislative ap- proaches to reducing injuries. The moderate nature of its success apparently arises from a significant lack of interest and “pickup” by the various bodies responsible for promulgation and, perhaps more importantly, enforcement. The ancillary activity of partnering with groups representing the constituency at risk, however, appears to have been much more successful and demonstrated a “pull” for these activities, be they r2p or research, rather than the “push” that NIOSH is (appropriately) try- ing to avoid. Activities resulted in a number of relevant outputs. Particularly strong was the development of new surveillance sources that add to the knowledge base of risk factors for occupational injury among youth. More comprehensive data will help NIOSH and the TI Research Program in the evaluation of existing efforts as well as the targeting of future efforts toward youth. A number of products developed and tested through the demonstration projects seem to have been well received by stakeholders, as evidenced by requests for additional copies and incorporation of products into youth occupational safety programs. Finally, the TI Research Program’s goal on working youth appears to have resulted in several modest yet important intermediate outcomes. Outputs, such as surveillance data and the 2002 NIOSH report “National Institute for Occupation Safety and Health (NIOSH) Recommendations to the U.S. Department of Labor for Changes to Hazardous Orders”58 summarizing recommendations for updating child labor laws, have continued to be used by groups both in the government and in the private sector. Through outreach activities with partners, the TI Research Program in all probability had some part in the end outcome for subgoal 8.2, 58 Document can be found at: http://www.cdc.gov/niosh/docs/NIOSHRecsDOLHaz/pdfs/ DOL-recomm.pdf.

OCR for page 43
t R au m at I C I n j u Ry R e s e a R C h nIosh 02 at namely a decline in injuries among young agricultural workers between 1998 and 2004. EVALUATION OF RELEVANCE The guidance for assessing relevance in the Framework Document is “whether the program appropriately sets priorities among research needs” and “how engaged the program is in appropriate transfer activities” (see Appendix A, p. 137). A rel- evance score of 5 signifies a conclusion that the research is in high-priority areas and the program is significantly engaged in transfer activities, whereas a score of 4 signifies that the research is in priority subject areas and the program is engaged in appropriate transfer activities (see Box 2-1). The committee reviewed the work supporting the 8 specific goals (including the 19 subgoals) that constitute the TI Research Program. The goals represent a mix of long-standing safety concerns (e.g., agricultural injuries), newer or emerging areas of emphasis for the TI Research Program (e.g., falls from telecommunications towers), and attention driven by congressional directive (e.g., the AFS). Three goals represent specific worker populations identified by location (Alaska), age (youth), or sector (emergency response). NIOSH has clearly driven a national sensitivity to some specific safety problems. For example, the committee concludes that NIOSH attention to workplace violence has shined a light on a previously neglected area. The committee concludes that for the most part, the goals are appropriate and relevant to the burden of traumatic injury in the workplace. The burden of injury BOX 2-1 Scoring Criteria for Relevance 5 = Research is in high-priority subject areas and NIOSH is significantly engaged in appro- priate transfer activities for completed research projects or reported research results. 4 = Research is in priority subject areas and NIOSH is engaged in appropriate transfer activities for completed research projects or reported research results. 3 = Research is in high-priority or priority subject areas, but NIOSH is not engaged in appropriate transfer activities; or research focuses on lesser priorities but NIOSH is engaged in appropriate transfer activities. 2 = Research program is focused on lesser priorities and NIOSH is not engaged in or plan- ning some appropriate transfer activities. 1 = Research program is not focused on priorities and NIOSH is not engaged in transfer activities.

OCR for page 43
p R o g R a m m at I C R e l e va n C e I m paC t 03 and represented by the eight major goal areas is certainly high, although the committee did not attempt to independently assess the burden of injuries in all occupations or work sites in the country as part of its review. Rather, the committee understands the challenges NIOSH faces in prioritizing research with such restricted resources and concludes that given this limited budget and based on committee members’ expert judgment, the TI Research Program has made overall appropriate selections of general areas to pursue. While the committee concluded that many of the goal areas were high priority— such as Alaska and falls from elevations—it identified gaps, such as falls from same elevation, several sectors within workplace violence, and a narrow focus within machines. There were also areas in which TI Research Program research was “stuck” in its approach (e.g., a focus on engineering solutions to tractor rollovers, instead of a switch to policy research to accomplish transfer to the field). As described in previous sections of this chapter, the TI Research Program engages in appropriate transfer activity within some, but not all, of the goal areas. TI Research Program staff has participated in the ANSI Z15 committee (subgoal 1.1), an important collaborative transfer activity. The transfer activities in goal 2, if successful, are likely to lead to the introduction of TI Research Program- dependent safety innovations, primarily engineering-based, but also developed from FACE investigations. TI Research Program staff has urged ASABE to develop a performance standard for the AutoROPS (subgoal 4.1), it has contributed to a DOL-OSHA committee that resulted in OSHA guidelines for nursing homes (subgoal 5.1), it has engaged in significant transfer activities related to legislative changes to protect young workers (subgoal 8.1), and it has provided recommen- dations for revisions in federal child labor laws (which became effective in 2005). The TI Research Program-funded Youth at Work program was instrumental in the passage of revised child labor laws in Massachusetts. These transfer activities are appropriate and related to the intermediate outcomes demonstrated for some of the program activities. Score for Relevance In summary, the committee notes impressive work, including transfer, in priority goal areas. The committee assigns a score of 4 for the relevance of the TI Research Program. This score is based on the fact that research is mostly in priority and some high-priority subject areas and has a range of involvement in transfer activities.

OCR for page 43
t R au m at I C I n j u Ry R e s e a R C h nIosh 04 at BOX 2-2 Scoring Criteria for Impact 5 = Research program has made major contribution(s) to worker health and safety on the basis of end outcomes or well-accepted intermediate outcomes. 4 = Research program has made some contributions to end outcomes or well-accepted intermediate outcomes. 3 = Research program activities are ongoing and outputs are produced that are likely to result in improvements in worker health and safety (with explanation of why not rated higher). Well-accepted outcomes have not been recorded. 2 = Research program activities are ongoing and outputs are produced that may result in new knowledge or technology, but only limited application is expected. Well-accepted outcomes have not been recorded. 1 = Research activities and outputs do not result in or are NOT likely to have any application. NA = Impact cannot be assessed; program not mature enough. EVALUATION OF IMPACT The directions for assessing impact on end outcomes or well-accepted interme- diate outcomes in the framework seem to hinge on a distinction between “major contributions” for a score of 5 and “some contributions” for a score of 4 (see Box 2-2). The committee understands that a determination of major contributions does not “imply that the NIOSH program was solely responsible for observed improve- ments in worker health and safety”; the committee is aware of the challenges in assigning such responsibility and has not attempted to do so. Intermediate Outcomes The Framework Document describes intermediate outcomes as “important indicators of stakeholder response to NIOSH outputs. They reflect the impact of program activities and may lead to the desired end outcome of improved work- place safety and health.” Intermediate outcomes fall into the major categories of public policy impact, training and education, and self-reported use or repackaging of NIOSH material by stakeholders. The committee found evidence of effects on well-established intermediate outcomes in all of the eight goal areas and in all types of outcomes. These have been documented for each goal and subgoal in previous sections of this chapter. This section briefly summarizes some exemplars by the type of intermediate outcome rather than by goal.

OCR for page 43
p R o g R a m m at I C R e l e va n C e I m paC t 05 and Intermediate outcomes related to public policy impact refer to the production (by those outside NIOSH) of guidelines or regulations based wholly or partly on NIOSH research. The TI Research Program has been associated with the develop- ment of both voluntary guidelines and regulations in several areas (goal 1: motor vehicles; goal 2: falls from elevations; goal 3: workplace violence; goal 4: machines; goal 5: back injuries; goal 6: Alaska; and goal 8: working youth). Approval of the ANSI Z15.1 standard, Safe Practices for Motor Vehicle Opera- tions, is a success for goal 1. This filled the gap left by the Federal Motor Carrier Safety Regulations, which apply only to commercial drivers. The ANSI standard will help protect all employees who operate a motor vehicle as part of their job. Another very important intermediate outcome was the incorporation of a NIOSH recommendation related to HO No. 2 (motor vehicle occupations) into a final rule published by the DOL in 2004. The DOL published a final rule59 on the HO prohibiting all workplace driving on public roadways by 16-year-old workers and placed restrictions on driving by 17-year-old workers. The TI Research Program’s work on falls from telecommunications towers supported the development of two important policy documents—namely, the first statewide telecommunications tower standard (in North Carolina), which addresses safety procedures to be used during tower construction and maintenance, and the OSHA Telecommunications Tower Task Force Compliance Directives, which ad- dress inspection procedures. OSHA published recommendations for workplace violence prevention programs in 1998 and guidance documents for preventing workplace violence to healthcare and community service workers in 2004. Notable intermediate outcomes in goal 4 (machines) are the revisions to ANSI Z245.5 Stan- dards Committee to require a security switch. This will serve to protect workers at municipal and commercial recycling centers. An amendment to the FLSA restricts workers under the age of 18 years from loading balers that do not meet the ANSI standard and prohibits them from operating balers (HO No. 12).60 The research in patient handling has been cited as an important source of evidence leading to the passage of safe patient handling legislation in several states (Texas, Washington, Rhode Island, Ohio, Hawaii, and New York). As discussed above, BLS data indicate a decrease in rates of injuries likely related to patient handling from the early 1990s to the present. Other changes in public policy related to TI Research Program findings are the USCG Dockside Pre-season Boarding 59 Regulationsimplementing the act permit the Secretary of Labor to prohibit the employment of youth in occupations declared “particularly hazardous for the employment of children . . . or detri- mental to their health or well-being” (29 USC 201 Sec. 3(1)). These prohibited activities are referred to as Hazardous Occupations Orders (HOs). The minimum age, by statute, for HOs in nonagricul- tural occupations is 18; the minimum age in agricultural occupations is 16. 60 HO No. 12 (balers), 29 CFR Part 570.63.

OCR for page 43
t R au m at I C I n j u Ry R e s e a R C h nIosh 06 at Program and the helicopter safety guidelines related to goal 6, and changes in the FLSA in 2005 related to goal 8. Intermediate outcomes related to training and education refer to programs sponsored by other organizations. These have been documented in previous sec- tions of this chapter and are summarized only briefly here. The NIOSH document “Building Safer Highways and Work Zones”61 has been used in an OSHA training course and by other organizations for safety training. TI Research Program work on falls from telecommunications towers supported the development of OSHA train-the-trainer programs as well as comprehensive NATE manuals. TI Research Program staff and their research related to workplace violence have influenced OSHA recommendations and guidelines as well as guides on workplace violence prevention by the Office of Personnel Management and the American Society for Industrial Security. TI Research Program findings on safe patient handling have been used in professional organization training programs and textbooks. Work in Alaska aviation has been used extensively in FAA training programs, including public service announcements and training videos Another category of intermediate outcome is self-reported use and/or repack- aging by stakeholders. Many NIOSH reports have been picked up by stakeholders and disseminated. For example, the TI Research Program-authored report “Build- ing Safer Highway Work Zones: Measures to Prevent Worker Injuries from Vehicles and Equipment”62 has been distributed by OSHA offices and reproduced as part of the Laborers’ Health and Safety Fund of North America 2003. The blind area equipment reports have been requested by several stakeholders. The committee includes in this category stakeholder-initiated changes in safety technology and workplace environments.63 The FACE Program can be particularly useful in illuminating these kinds of risks. A FACE investigation of a scalping accident and the subsequent identification of other cases led to a retrofit by the manufacturer of the hay baler involved in all five cases. The manufacture of sealed agricultural plow frames revised its manufacturing process to avoid the scrap metal that appeared to be responsible for ignition hazards identified in FACE investiga- tions. Although the original ROPS work occurred outside the period of committee evaluation, manufacturers have changed the design of tractors (although uptake of the technology is quite slow). Equally important to the research that led to improvements in equipment design is research showing that a technology is not effective. The research on back 61 NIOSH. 2001. Building safer highway work zones: Measures to prevent worker injuries from vehicles and equipment. HHS (NIOSH) Publication No. 2001-128. 62 See footnote 61 for reference. 63 This is distinct from public policy impact, which would force stakeholders to make changes.

OCR for page 43
p R o g R a m m at I C R e l e va n C e I m paC t 07 and belt use for materials handling demonstrated no effect on injuries, and two major retail chains no longer provide or require back belt use. Research demonstrating no effect of an intervention is as important as research that does show an effect, because discontinuance of an ineffective intervention allows more promising in- terventions to be studied and introduced, which might not have occurred had the workers and their management assumed the technology to be effective. Other safety technology introductions based on the research of the TI Research Program, although limited, include those involving patient compartments, such as the Star of Life Ambulance Specifications, changes in the patient compartment in some ambulances in Florida, and restraint systems for a concept ambulance. End Outcomes NIOSH reports improvements in end outcomes for five of the eight goals: workplace violence (a very modest decrease in fatalities compared to societal ho- micides), machines, acute back injury in patient handling, high-risk occupations in Alaska, and nonfatal agricultural work-related injuries in youth (but not non- agricultural work-related injuries in youth). As noted in this list, the committee greets with some reservation the reports of an impact on workplace violence fatalities given that societal homicides decreased similarly over the same period of time and these trends likely are inextricably linked. The committee discusses important contributions to intermediate out- comes in a previous section. NIOSH reports a steady decrease in fatalities caused by machines, plant and in- dustrial powered vehicles, and tractors since 1992. The number of deaths decreased 16 percent and the fatality rates declined 30 percent. The committee agrees with the NIOSH conclusion that it is difficult to quantify the contributions of the TI Research Program to these decreases. The probable source of improvement in end outcomes related to goal 4 (machines) is primarily due to the FACE investigations (subgoal 4.3), not the other recent efforts on ROPS or the limited work on paper balers. The committee is supportive of the FACE Program and recognizes its con- tributions. It raises this merely to make the point that the end outcomes related to the machines program derive from one program and not from other interventions supported in subgoals 4.1 and 4.2. There is little doubt that the successes in the Alaska program are numerous and important. The program benefited from a dedicated source of funding (a congres- sional directive), a reasonably narrow focus, and locus as a state-based program. Having TI Research Program staff in close physical proximity to the many impor- tant stakeholders was likely influential in the success of the program. This is not a feasible strategy in most areas of TI Research Program work, however. As an aside,

OCR for page 43
t R au m at I C I n j u Ry R e s e a R C h nIosh 08 at the committee hopes that the TI Research Program can extend the AFS successes in Alaska and across the country. The combination of important partnerships, focus, sufficient resources, and a state-based program with specific attention to the public health model could be applied to other high-risk industries in Alaska as well as similar high-risk industries in other parts of the country. The AFS model can serve as a model of success for the program as a whole. Surveys from farm operators suggest a marked decrease in nonfatal work- related injuries among youth between 1998 and 2004. There is not, however, evi- dence of decreased injuries of working youth in other sectors. The committee views the changes in end outcomes with some reservations, un- sure of the TI Research Program contributions to the results. However, the commit- tee favorably views any progress in reducing occupational traumatic injuries and is confident of some degree of NIOSH contributions. Given the very limited resources under which the TI Research Program operates, results regarding end outcomes are admirable. Other important external factors that likely restrain success include OSHA inaction, the difficult nature of some of the industries in which injuries are high (e.g., the culture of individualism and risk acceptance in the agricultural and logging industries), and the difficulties of getting small- to medium-sized en- terprises to invest in safety technologies. The TI Research Program might benefit from efforts to increase its capacity and expertise in policy research, in behavioral and social science research, and in directed efforts to understand how to impact small- to medium-sized enterprises. Chapters 3 and 4 address these issues. Score for Impact The committee commends the TI Research Program for its contributions toward reducing occupational traumatic injuries. As documented in the previous section, the TI Research Program is associated with impact on either intermediate or end outcomes in each major goal. The committee recognizes that external fac- tors, specifically severely limited resources and inaction on the part of OSHA, can be significant barriers to the TI Research Program’s progress in some goal areas. However, the committee notes (1) the lack of demonstrated effect on end outcome data in three goal areas and in some subgoals in the other five goals, (2) the inability to determine what degree of responsibility the TI Research Program bears for the documented improvements in end outcomes or for the intermediate outcomes, and (3) a lack of significant intermediate outcomes for some subgoals. The committee assigns a score of 4 for the impact of the TI Research Program.

OCR for page 43
p R o g R a m m at I C R e l e va n C e I m paC t 09 and REFERENCES ASIS (American Society of Industrial Security). 2006. Guidelines for WPV. http://www.asisonline. org/guidelines/guidelineswpvfial.pdf (accessed July 23, 2008). Azaroff, L. S., C. Levenstein, and D. H. Wegman. 2002. Occupational injury and illness surveillance: Conceptual filters explain underreporting. American Journal of Public Health 92(9):1421-1429. BLS (Bureau of Labor Statistics). 2008. Injuries, illnesses, and fatalities. http://www.bls.gov/iif/home. htm (accessed July 23, 2008). Bobick, T. G., J. L. Belard, H. Hsiao, and J. T. Wassell. 2001. Physiological effects of back belt wearing during asymmetric lifting. Applied Ergonomics 32(2001):541-547. Conway, H., and J. Svenson. 1998. Occupational injury and illness rates, 1992-1996: Why they fell. Monthly Labor Review November:36-58. DOL (Department of Labor). 2008. OSH Act of 1970 Sec. 5 duties. http://www.osha.gov/pls/ oshaweb/owadisp.show_document?p_id=3359&p_table=OSHACT (accessed July 31, 2008). DOT (Department of Transportation). 2008. Glossary. http://ops.fhwa.dot.gov/publications/evac_ primer/26_glossary.htm (accessed July 24, 2008). Giorcelli, R. J., R. E. Hughes, J. T. Wassell, and H. Hsiao. 2001. The effect of wearing a back belt on spine kinematics during asymmetric lifting of large and small boxes. Spine 26(16):1794-1798. Haddon, W. 1968. The changing approach to the epidemiology, prevention, and amelioration of trauma: The transition to approaches etiologically rather than descriptively based. American Journal of Public Health 58:1431-1438. Haddon, W. 1972. A logical framework for categorizing highway safety phenomena and activity. Journal of Trauma 12(1):193-207. Hallman, E. M. 2005. ROPS retrofitting: Measuring effectiveness of incentives and uncovering inher- ent barriers to success. Journal of Agricultural Safety and Health 11(1):75-84. Hendricks, S. A., E. L. Jenkins, and K. R. Anderson. 2007. Trends in workplace homicides in the U.S., 1993-2002: A decade in decline. American Journal of Industrial Medicine 50:316-325. IOM (Institute of Medicine). 1998. Protecting youth at work: Health, safety, and development of working children and adolescents in the United States. Washington, DC: National Academy Press. Jenkins, E. L. 1996. Current Intelligence Bulletin 57, Violence in the workplace: Risk factors and prevention strategies. HHS (NIOSH) Publication No. 96-100. Leigh, J. P., J. Marcin, and T. R. Miller. 2004. An estimate of the U.S. government’s undercount of non- fatal occupational injuries. Journal of Occupational and Environmental Medicine 46(1):10-18. Merchant, J., and J. Lundell. 2001. Workplace violence intervention research workshop. American Journal of Preventative Medicine 20(2):135-140. Myers, J. R., and K. A. Snyder. 1995. Roll-over protective structure use and the cost of retrofitting tractors in the United States, 1993. Journal of Agricultural Safety and Health 1(3):185-197. NIOSH (National Institute of Occupational Safety and Health). 2006. NIOSH awards funding to ag research centers for initiative to reduce tractor deaths, injuries. NIOSH Update. March 21, 2006. http://www.cdc.gov/niosh/updates/upd-03-21-06.html (accessed July 24, 2008). NIOSH. 2007a (unpublished). Overview of NIOSH. In the evidence package provided to the Com- mittee to Review the NIOSH TI Research Program. NIOSH. NIOSH. 2007b (unpublished). Overview of the TI Research Program. NIOSH. NIOSH. 2007c (unpublished). Current TI research goals and sub goals: Reduce injuries and fatalities due to motor-vehicles. In the evidence package provided to the Committee to Review the NIOSH TI Research Program. NIOSH.

OCR for page 43
t R au m at I C I n j u Ry R e s e a R C h nIosh 0 at NIOSH. 2007d (unpublished). Current TI research goals and sub goals: Reduce injuries and fatalities to emergency responders. In the evidence package provided to the Committee to Review the NIOSH TI Research Program. NIOSH. NIOSH. 2007e (unpublished). Current TI research goals and sub goals: Reduce injuries and fatalities due to falls from elevations. In the evidence package provided to the Committee to Review the NIOSH TI Research Program. NIOSH. NIOSH. 2007f (unpublished). Appendix : Supporting evidence for the TI goals/subgoals. In the evidence package provided to the Committee to Review the NIOSH TI Research Program. NIOSH. NIOSH. 2007g (unpublished). Current TI research goals and sub goals: Reduce injuries and fatalities due to workplace violence. In the evidence package provided to the Committee to Review the NIOSH TI Research Program. NIOSH. NIOSH. 2007h (unpublished). Current TI research goals and sub goals: Reduce injuries and fatalities due to machines. In the evidence package provided to the Committee to Review the NIOSH TI Research Program. NIOSH. NIOSH. 2007i (unpublished). Current TI research goals and sub goals: Reduce acute back injury. In the evidence package provided to the Committee to Review the NIOSH TI Research Program. NIOSH. NIOSH. 2007j (unpublished). Current TI research goals and sub goals: Reduce injuries and fatalities among workers in Alaska. In the evidence package provided to the Committee to Review the NIOSH TI Research Program. NIOSH. NIOSH. 2007k (unpublished). Current TI research goals and sub goals: Reduce injuries and fatalities to working youth. In the evidence package provided to the Committee to Review the NIOSH TI Research Program. NIOSH. NIOSH. 2008. Fatality Assessment and Control Evaluation (FACE) Program. http://www.cdc.gov/ niosh/face/ (accessed July 23, 2008). OPM (Office of Personnel Management). 1998. Dealing with workplace violence: a guide for agency planners. OWR-09, February 1998. Pana-Cryan, R., and M. L. Myers. 2000. Prevention effectiveness of roll-over protective structures— Part III: Economic analysis. Journal of Agricultural Safety and Health 6(1):57-70. Pana-Cryan, R., and M. L. Myers. 2002. Cost effectiveness of roll-over protective structures. American Journal of Industrial Medicine 42(S2):68-71. Pratt, S. G., D. E. Fosbroke, and S. M. Marsh. 2001. Building safer highway work zones: Measures to prevent worker injuries from vehicles and equipment. Cincinnati, OH: U.S. Department of Health and Human Services, Public Health Service, Centers for Disease Control and Preven- tion, National Institute for Occupational Safety and Health. HHS (NIOSH) Publication No. 2001-128. Rosenman, K. D., A. Kalush, M. J. Reilly, J. C. Gardiner, M. Reeves, and Z. Luo. 2006. How much work-related injury and illness is missed by the current national surveillance system? Journal of Occupational and Environmental Medicine 48(4):357-365. Stout, N., and C. Bell. 1991. Effectiveness of source documents for identifying fatal occupational injuries: A synthesis of studies. American Journal of Public Health 81(6):725-728. Wassell, J. T., L. I. Gardner, D. P. Landsittel, J. J. Johnston, and J. M. Johnston. 2000. A prospective study of back belts for prevention of back pain and injury. Journal of the American Medical As- sociation 284(21):2727-2732.