Highway Worker Safety (2017)

Below is the uncorrected machine-read text of this chapter, intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text of each book. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.

31 chapter three Agency PrActices And PersPectives on HigHwAy worker sAfety introduction In the United States, the primary highway system owners are state DOTs, and they maintain signifi- cant autonomy and authority in this function. Their involvement and influence extend beyond the physical roadways, which the state DOTs design, construct, operate, and maintain, to other aspects of transportation system ownership. Because constructing and maintaining these facilities predomi- nantly require personnel to be located in potentially hazardous work sites close to active travel ways, state DOTs have a fiduciary and moral responsibility to promote the safety of their workers perform- ing tasks within these work sites. Although all state DOTs generally have the same role and responsibility of managing their state highway systems, each state has a unique structure for its safety program. States usually are responsi- ble for the construction and maintenance of primary routes, but some state DOTs have a more super- visory role; others have a more involved role, even on local and county roads. Because factors such as size, geography, weather, and agency organization and purview vary widely from state to state, there is not a one-size-fits-all approach when it comes to safety programs. This chapter describes the safety program elements currently included in state DOT safety programs. The elements described provide examples that other state DOTs may elect to include in their safety programs. However, it is not the goal of this chapter to construct a “template” for a successful safety program that can be used by all of the states. The goal of this chapter is to examine how safety programs are developed and operate in the state DOTs. Particular emphasis is placed on how incident and safety data are acquired, managed, and used. The chapter highlights sources of data that are more readily available, as well as data that may not be as available nationally. It is beneficial to understand in greater detail how agencies use the data for their safety programs. It is hoped this chapter will provide a national overview of safety programs in a predominantly quantitative format so that the practices and trends can be visualized in figures and tables. Additional exploration of specific safety programs by state DOTs that have the potential to be transferred and implemented in other states is included in chapter five. The primary mechanism to achieve these goals was a survey questionnaire developed and distrib- uted to state DOTs. The methodology section of this chapter outlines the process for conducting the survey and the procedure for gathering responses. This chapter is organized according to the same outline as the questionnaire and contains the following sections: • Methodology, • Demographics, • Incident reporting, • Data collection, and • Data utilization. The five sections of the questionnaire and this chapter have two purposes. The first is to create a logi- cal format and progression of items within the questionnaire document that are easily comprehended

32 by survey participants. The second is to have effective questions that solicit the following information about state DOT safety programs and practices: • Data sources that are available to state DOTs, how the sources are archived, and their robustness; • How data are analyzed and used by the state DOTs; • Policies and practices that have been implemented to mitigate highway worker safety risk; and • Agency perspectives on safety policies and practices. The survey questions were not limited to understanding the state DOT practices. A subsequent chap- ter explores concepts beyond these points. However, the focus of the survey questions was improved understanding of national trends related to these aspects of state DOT policies and practices. The acqui- sition of information directly from state DOT employees allowed for the collection of firsthand informa- tion about safety programs from individuals who regularly work with such programs and are invested in their successful implementation. However, this also resulted in the possibility of data about safety programs being based on estimates or personal impressions by one or more state DOT employees. MetHodology The goal of the survey was to obtain useful and detailed information about the safety programs cur- rently in place within state DOTs across the United States. A critical consideration in conducting the survey was to balance the ability to obtain the desired information with keeping the questionnaire short and simple enough to navigate and understand to achieve a high response rate from state DOT safety offices. For the data gathered from the survey to be useful in summarizing national trends, a target response rate of 80% (40 of 50) of the states was established. The survey questionnaire was drafted and underwent several internal revisions. The final draft of the survey questions was sent to the NCHRP Synthesis Topic 47-16 review panel for feedback. The questionnaire was edited based on the feedback. The survey questions were coded into the Qualtrics online survey software to provide an easy-to-navigate and common interface for the participants in the survey. A link to the Qualtrics-formatted questionnaire was sent to the panel to pilot test the question- naire to ensure that the question progression logic in the software was programmed correctly. At this point, after the panel’s review, the survey questionnaire was ready for nationwide distribution. The complete survey questionnaire that was distributed to the state DOTs is included in Appendix A. With the assistance of NCHRP personnel, a link to the Qualtrics-formatted questionnaire was dis- tributed to safety representatives in each state DOT. The questionnaire initially was distributed using an e-mail list of the members of the NAATSHO, an organization whose purpose is to communicate and distribute information regarding health and safety policies among state DOTs. The distribution of the results of this synthesis to state DOTs will help to fulfill NAATSHO’s organizational objective “to promote and improve the efficiency and effectiveness of the occupational safety and health pro- grams designed for the Highway and/or Transportation Departments in the United States, Canada and Mexico” (NAATSHO 2006). NAATSHO members are volunteers, most of whom are employees in the safety departments of state DOTs. For some states, the NAATSHO e-mail list contained multiple employees in the state. For these cases, e-mails were sent to all of the employees in the state on the list. As a result, e-mails containing the link to the survey questionnaire were sent to 63 NAATSHO mem- bers. In the initial distribution, two e-mails were returned as undeliverable. After the initial e-mail and a reminder e-mail from NCHRP, the total number of survey responses received was between 15 and 20. More reminder e-mails were sent over the course of 2 weeks, and the total survey responses increased to approximately 30. To increase the survey representation to achieve the 80% response rate target, follow-up phone calls were conducted with and e-mails sent to NAATSHO members. In addition, colleagues at state DOTs who had not yet responded were contacted to find appropriate safety personnel who would be willing to take the survey. After 2 weeks of these approaches, the survey response target (at least 40 of the 50 states) was met. In total, 41 (82%) states had a representative respond to the survey questionnaire. This number of responses exceeded the initial goal of 40 (80%) states.

33 To preserve the anonymity of the respondents, the states, and the state safety programs, no state is specifically identified with a response unless permission has been granted by a representative of that state DOT. States may be identified as part of broad national regions. It is not the purpose of this chapter to identify specific state programs but instead observe general trends common to state DOTs across the nation. State DOT size is a characteristic that can meaningfully differentiate states. To preserve the anonym- ity of the states and still provide general trends based on the characteristic of state DOT size, a categorical scheme was established. FHWA reported 2013 state DOT disbursements (total expenditures of each state DOT in 2013), and this measure was selected as a reasonable proxy for state DOT size. The following four ranges of disbursements were defined (FHWA 2013b): • Less than $1 billion in disbursements, • $1 to $2 billion in disbursements, • $2 to $4 billion in disbursements, and • Greater than $4 billion in disbursements. These four ranges were selected to have an approximately equal number of states per category. Several of the questionnaire responses were sorted by these categories. These are included in the tables in Appendix B. In those tables, each code (e.g., A1, B3, C5, etc.) corresponds to a state. The “A” states are in the less than $1 billion in disbursement category and randomly sorted within the category. This method was used for the other categories as well. These categories allow trends between state sizes to be highlighted without compromising the anonymity of the questionnaire respondents. Some of the qualitative responses in Appendix B were edited to preserve state anonymity. deMogrAPHics The questionnaire’s first section asked the respondents basic demographic questions regarding their role in worker safety and general information about their agency. This information is important to understanding the background, experience, and job roles of the individuals completing the survey questionnaire. In addition, because all state DOTs are structured and function differently, it was necessary to try to quantify the diversity among the states. The questions relating to the state DOT representative responding to the survey questionnaire included requesting his or her job title and the number of years worked with the current agency. Figure 5 shows a distribution of the number of years of experience of the respondents. In total, the 41 respondents had a combined 625 years of experience at their current agency, with an average of 15.2 years per respondent and a median of 15 years. The minimum was 1 year, and the maximum was 30 years. FIGURE 5 Frequency of responses by respondent’s years of experience (n = 41).

34 The job titles of the respondents varied widely and included such diverse titles as “safety and risk manager,” “statewide safety security and emergency coordinator,” and “highway safety inspector.” Table 1 provides a summary of how frequently various safety and managerial-related words appeared in the respondents’ job titles. There will be greater confidence in the survey results if the survey respon- dents have roles within the state DOT that provide accurate knowledge and perspectives of the state DOT safety programs and give them access to information about safety programs and incident report- ing practices. The high frequency of the words in Table 1 serves as partial justification for the validity of the results of the survey. Many of the job titles include multiple key words listed in Table 1. For example, the job titles “risk management safety officer” and “occupational safety and health training manager” contain three of the words in Table 1. It is not surprising that the word “safety” is well represented in the job titles. Of the 41 state representatives who responded, only five had titles that did not include the word “safety.” Of these five, some contain the word “risk” or are generic “program director” or “program administrator” titles. To demonstrate further that the respondents to the survey are knowledgeable about their state safety programs, the questionnaire asked the respondents how often they “work with worker injury claims and prevention programs.” Figure 6 is a histogram distribution showing the frequency that the respondents work with injury claims and prevention programs. The most striking aspect of Figure 6 is that a majority of the respondents (63%) work with injury claims and prevention programs on a daily basis. As a follow-up question to the frequency that they work with these programs, the respondents were asked to describe their specific role with these programs. The respondent who selected “Very rarely (yearly)” for the frequency ques- tion (see Figure 6) followed up that the particular “office oversees the Departments [sic] Safety program, but has no role in processing or overseeing claims.” So even respondents who may not work frequently with claims and prevention programs still are invested in some aspect of the safety program in their agency. Key Word Number of Titles Percentage of Titles Safety 36 88 Risk 5 12 Health 7 17 Manager/Coordinator 17 41 Director/Administrator 9 22 n = 41. TABLe 1 JOB TITLeS OF ReSPONDeNTS THAT CONTAIN Key WORDS FIGURE 6 Frequency that respondents work with injury claims and prevention programs (n = 41).

35 In response to the question asking the respondents to describe their role with worker injury claims and prevention programs, two general trends were present across the descriptions. The respondents are managers of these claims and programs. Their roles are to collect the information and manage the claims and programs based on the data collected. The second trend is that many of the respondents also are involved specifically in the implementation of safety programs, particularly the training aspects of the programs. The other questions in this section of the questionnaire referred to the characteristics of the state DOT that the respondent was representing in the survey response. The questionnaire asked the fol- lowing two questions regarding the characteristics of the agency: • What is the approximate size of your agency in total number of employees? • What is the approximate percentage of your agency’s employees who are regularly on construc- tion and/or maintenance sites? Figures 7 and 8 are histogram distributions presenting the frequency of each response. These questions aimed to demonstrate not only the diversity of the state DOTs but also the number of workers who are experiencing higher safety risk by spending time in highway work sites as a part of their job. Based on these two questions, a rough estimate of at least 75,600 state employees nation- wide are regularly on construction or maintenance sites. This underscores the volume of human life that is at risk of injury in work sites and further justifies the importance of this synthesis report. FIGURE 7 Distribution of state DOTs by number of employees (n = 41). FIGURE 8 Distribution of state DOTs by percentage of employees who are regularly on work sites (n = 41).

36 incident rePorting The second section in the questionnaire related to how the agency responds when injury incidents occur with their employees. The focus of these questions is to get a sense of the national practice among state DOTs for the process followed when an injury incident occurs on a jobsite. The questions specifically relate to the process for reporting and archiving incidents that occur with state highway employees in highway work sites. Postincident steps The first question in this section provided the respondent with a list of potential steps that could be taken by the state DOT after an incident has occurred in a highway work site. These work site inci- dents were separated into three types: • Public automobile—incidents involving vehicles owned by the public that enter highway work sites. • On-site vehicle/equipment—incidents involving vehicles and equipment that are allowed in the highway work site and are typically used to conduct the work. • On-site hazard—incidents in the highway work site that do not involve a vehicle or equipment. Figure 9 shows a summary of the frequency of each postincident step for each of the three types of work site incidents as defined by this synthesis report. FIGURE 9 Frequency of postincident steps by type of incident (n = 41).

37 As seen in Figure 9, many of the steps listed in the questionnaire are taken by the state DOTs after a work site incident. The figure reveals that the steps have similar clusters around the three types of incidents. Generally, each step received a similar number of responses across the three incident types. This result indicates that the reporting procedures for the state DOTs are similar regardless of the type of incident. The notable exception is the “inform law enforcement” step, which is more fre- quently taken in public automobile incidents. This step appears far less common after the occurrence of the two other types of incidents. For all the incident types, “Upload findings to ‘lessons learned’ database” and “Inform federal OSHA” are the least employed of the steps listed. To gauge the effectiveness of the steps that state DOTs identified, respondents were asked to rate the effectiveness of each step their agency took after an incident at a work site on a roadway. This effectiveness is judged according to how well the step contributes to the success of the agency’s safety program. The respondents were asked to rate the effectiveness of each step using a scale of 1 to 5 (1 = not effective, 5 = very effective). The overall average effectiveness rating for the public automobile incident steps and on-site hazard steps was 3.7; for on-site vehicle/equipment incident steps, it was 3.8. Communicating lessons learned agencywide was indicated by the respondents as a “very effective” step across all three of the incident types (average rating = 4.0). The highest average effectiveness rating (4.3) was associated with informing federal OSHA for on-site vehicle/equipment incidents, and the lowest average (2.9) was related to informing law enforcement about incidents resulting from on-site hazards. Tables B1a through B1d in Appendix B present a summary of the average effectiveness of four selected postincident steps according to the self-reported respondent per- spectives. These steps include communicating lessons learned agencywide, return to work initiatives, reviewing or modifying policies and procedures, and postincident investigations. Tables B1a through B1d divide responses based on the four disbursement categories and report the average effectiveness based on the categories for the three types of incidents. Archiving Process Three questions in this section of the questionnaire related to the state DOT’s incident report archiving process. The first of these questions asked which departments at the agency are responsible for com- piling and archiving incident reports. Figure 10 shows the frequency with which respondents identi- fied particular departments as having a role in compiling and archiving incident reports. Respondents could select all departments that applied. As is shown in Figure 10, the safety offices (regional or statewide) are the most frequently cited location for the compiling and archiving of incident reports. Many state DOTs have multiple depart- ments across disciplines contributing to and maintaining the archive of incident reports. Some of the “Other” responses for this question included emergency operations and the claims office. FIGURE 10 Frequency of agency departments that have archive responsibilities.

38 The second question asked what format is used to archive incident reports at state DOTs. Fig- ure 11 shows the frequency with which respondents identified particular formats used to maintain the archive of incident reports. Respondents could choose more than one format if their incident reports are archived in multiple ways. The third question related to archival processes of incident reports asked respondents to identify how incident reports are categorized. Figure 12 presents a summary of the frequency with which respondents cited particular categorization types for the incident reports that state DOTs compile and archive. It is encouraging that many state DOTs use multiple means of categorization to archive their inci- dent reports. Of the 18 respondents who selected “Other,” many indicated their data sources are also categorized by the date on which the incident occurred. Table B1d in Appendix B indicates that the largest states (with disbursements greater than $4 billion) are more likely to have an “Other” type of categorization. In fact, 78% of these states used an “Other” type of categorization, whereas between 30% and 38% of the states with $4 billion or less in disbursements used an “Other” type of categori- zation for incident reports. Maintaining several methods of categorization can allow the reports to be easily queried to find a specific incident or by using data to implement and evaluate safety initiatives. near Miss reporting Several of the questions in this section of the questionnaire referred to near miss incidents that occur on work sites. Only 44% (18 of 41) of the respondents indicated their agency has a system in place to report a near miss incident. FIGURE 11 Frequency of format of archived incident reports. FIGURE 12 Frequency of categorization type for archived incident reports.

39 Of the 18 states that have a near miss reporting system, nine use the same reporting process for a near miss that is used for other incident types. Among the nine states that have a different reporting process for a near miss, some of the differences included the following: • Similar forms to actual incidents but less detailed information for near misses, • Near miss incident reporting is optional, and • Different administrative communication tracks for near miss incidents. Descriptions of the near miss reporting processes for states that have them are included in Tables B2a and B2d. One state DOT provided a particularly detailed account of its near miss reporting process. When a near miss incident occurs, the foreman reports the incident to the county manager. This manager reports the event to the state DOT’s district safety coordinator, who completes a spe- cial notification form and submits it to an employee safety group. This group distributes details of the incident to the executive staff at the agency. This entire process takes only a few hours. For states that do not have a reporting system for near misses, the questionnaire prompted the respondents to select from a list of choices the reason the agency does not have a near miss reporting system. Figure 13 shows a distribution of the reasons agencies do not have such a system in place. The primary reason for not having a near miss reporting system was the lack of a clear definition of a near miss. Additional research and education efforts within state DOTs and the construction and maintenance industry more broadly could standardize the definition of a near miss so that the report- ing of these incidents can become more common and standardized among state DOTs. A consistent definition of near miss incidents across state agencies could help identify national trends regarding such occurrences. In addition, educating state DOTs about the value in tracking near misses to over- all safety performance and the connection of near misses to injury incidents would help promote the implementation of near miss reporting systems. dAtA collection The third section of the questionnaire sought to determine the sources of data that are available and used by the state DOTs regarding highway worker safety. The first question in this section asked respondents to indicate if a particular data set is available to their agency. In addition, respon- dents were asked whether the available data sets are used by their agency. Figure 14 shows, for each of the data sets, how many state DOTs have the data available and how many actually use that data set. FIGURE 13 Distribution of reasons agencies do not have a near miss reporting system.

40 The results shown in Figure 14 are indicative of how state policies and/or practices might restrict information from safety personnel because of confidentiality concerns. For example, the three data sets with the least availability to the state DOTs are “worker annual performance review,” “medical records,” and “contractor safety records.” In some cases, it is possible that individual employee records (performance or medical) may be protected by state policy and cannot be used in the devel- opment of safety programs. In addition, for each data set, some state DOTs have access to the data but the agencies do not currently make use of that particular data set. According to the survey responses, there were approxi- mately six states with access to any one particular data set that were not using the data. Leveraging all pertinent and readily available data is one approach to improving data-driven safety programs. If pertinent data are being collected and archived, the costs associated with integration into a safety program are measurably reduced. Respondents were asked, for data that was available or used by their agency, how complete the data are based on a scale of 1 to 5 (1 = very incomplete, 5 = very complete). For details on the results of this question and to see the data categorized by state DOT size, refer to Tables B3a through B3d in Appendix B. Table 2 shows the average completeness rating each data set received. FIGURE 14 Frequency of availability and usage of data sets (n = 41). TABLe 2 AVeRAGe COMPLeTeNeSS RATING FOR eACH DATA SeT Data Set Average Completeness Rating Incident report 4.0 Police citation report 3.9 Worker insurance claim 4.2 Worker annual performance review 3.7 Safety training records 3.6 Contractor safety records 2.9 Medical record 3.3 Fatality/njury datai 4.1 Roadway design 3.9 Roadside design features 3.9 1 = very incomplete; 5 = very complete.

41 Three of the data sets received an average rating of 4.0 or better in terms of their completeness: “Incident report,” “Worker insurance claim,” and “Fatality/injury data.” These data sets contain some of the core safety documents used for developing safety initiatives, and the documents are likely required by law in their state. Therefore, it is not surprising these data sets received the highest completeness ratings. The overall average completeness rating for all of the data sets is 3.7. This result is encouraging and indicates that, when the state DOTs have access to a particular data set, the data are reasonably complete and therefore would be beneficial in agency decision making regarding safety programs for workers in work sites. Another important characteristic of the data sets is the length of time after an incident the data are available for use by the state DOT. The results showed that internal state DOT structure and processing have the greatest impact on the time after an incident that the data becomes available for use. No one data set generally takes more or less time than the others. For example, of the 41 survey responses, 21 state DOTs have access to all of their available data sets within 3 months. In contrast, ten state DOTs must wait more than 1 year to have access to all of their available data sets. This disparity indicates that no single data set takes a certain amount of time to become available. Table B4c provides a summary of the time required for data to be available to state DOTs. This table indicates that the smallest states (those with less than $1 billion in disbursements) may have access to data the soonest. eighty-four percent of the data available to the smallest states are available within 1 month. No other size group has that level of data availability within that period. Further analyzing the processes of state DOTs that have quick turnarounds with their data sets would be beneficial for states that experience longer latencies between the occurrence of an incident and the availability of data sets. Having data available to state safety personnel soon after an incident allows the statewide safety strategies to be updated and adjusted based on current trends. These adjustments ensure that safety strategies are based on current trends and thus are most effective at protecting state workers in high- way work sites. The ways in which state DOTs analyze available data to promote highway worker safety are also of interest. Respondents were asked if their agency has conducted “any research or data analysis regarding highway worker safety in work sites on roadways and workers’ compensation related to injuries.” Only 39% (16 of 41) of respondents stated that their agency conducts this type of research. For agencies that conduct this analysis, the primary method involves examining past incident sta- tistics and other internal reviews of historic data. Some agencies partner with outside organizations, such as FHWA or an in-state university, to perform safety data analysis and research. Thirteen of the respondents identified data sources not currently available to their agency but that they believe would be beneficial. Some of these desired data sources include data from other state DOTs and other government agencies (federal, municipal, etc.). Other respondents indicated they would like their agency to have more integration of its current databases to more effectively categorize and understand incidences. Tables B5a and B5b in Appendix B list the descriptions of data sources that respondents thought would be beneficial for their state. The following three responses exemplify types of desired data: • An accident database with information about the types of vehicle crashes with state DOT equipment, • Information from private road management companies working for the state and from local municipalities, and • More detailed worker compensation data, including lost and restricted time. Despite the presence of state DOT research and analysis regarding highway worker safety, it is important to make strides toward equipping state DOTs with the data that can be the most beneficial to state DOT safety programs. In addition, it is vital that state DOTs be able to conduct internal research and partner with research institutions to extract the most value from collected data.

42 dAtA utilizAtion The fourth section of the questionnaire continued with querying respondents about data sources available to state DOTs. However, the questions in this section focused on how state DOTs imple- ment data available to them as part of safety policies and practices. To help determine the frequency that data are used, one question in the survey provided a list of policies and practices that could be developed from the data (either internal or external) and asked the respondents to indicate whether data were used in the development of the policy/practice. Table 3 is a summary of the list of policies and practices presented in the question and the number of respondents who indicated that their state DOT had used data to develop the listed policy or practice. Respondents were able to select all policies/practices that applied. Additional training options were the most prevalent forms of data-driven implemented practices. For the four respondents who selected “Other,” two cited programs relating to PPe. In addition, two respondents indicated that data have made their agency consider a safety incentive program, and one respondent used data to revise existing policies and procedures. Table B6a separates the responses to this question regarding the use of data to develop pro- grams or policies at state DOTs. For the “Driver awareness programs” and “Drug/Alcohol abuse programs,” the smallest states (those with less than $1 billion in disbursements) more than any other group used data to implement such programs. The percentage of implementation was more than 20% higher than the implementation percentages for the two programs from state DOTs with disburse- ments of $1 billion or more. Another question asked to what extent the state DOT shares the information it has collected with other organizations. The collection of data can be time consuming and expensive, so the more that existing information is shared, the more efficient and effective other organizations can be in terms of their safety programs. Table 4 presents a summary of the list of organizations presented in the question and the number of respondents who indicated that their state DOT shares data with that organization. The list of organizations was designed to include various levels of government agencies and nongovernmental entities. Respondents were able to select all agencies/organizations that applied. Sharing data with other state DOTs was the most common response. This response is encouraging because state DOTs have similar roles and needs for similar forms of information. The respondents Policy/Practice Number of Responses Percentage of Responses Additional training for workers 37 90 Additional training for supervisors 34 83 New standards for work site traffic control plans 28 68 Driver awareness programs 27 66 Worker behavior assessment programs 13 32 Safety incentive programs 10 24 Drug/alcohol abuse programs 18 44 Other 4 10 None 0 0 n = 41. TABLe 3 POLICIeS/PRACTICeS PReSeNTeD IN QUeSTION AND NUMBeR OF AGeNCIeS INDICATING THey USeD DATA TO DeVeLOP THe POLICy/PRACTICe

43 who selected “Other” described an assortment of types of organizations. Some agencies share data with professional organizations, such as the American Traffic Safety Services Association. Other state DOTs share information with other state agencies or in-state research universities. A large per- centage (22%; nine of 41) of the respondents indicated their state DOT does not share information with any other organization. All of these groups can benefit from sharing data with one another to make highway work sites safer. The final question of the survey solicited ideas from the respondents regarding recommenda- tions for other state DOTs to implement successful safety programs. The following are paraphrased examples of these recommendations: • encourage involvement in safety programs from all levels of the agency to ensure that manage- ment supports safety programs demonstrated to be effective for employees. • Use the latest technology and keep the programs updated. Take advantage of online training. • Share safety policies and practices with other states so that the successful programs can be shared among the state DOTs. conclusions The approach of gathering information regarding state worker safety policies and practices through a survey captures some understanding of current state DOT safety programs. In addition, the survey allows for capturing circumstances and challenges experienced by state DOTs with respect to their safety programs. As described in this chapter, state DOTs are diverse. each state DOT experiences its own set of issues but remains committed to improving the safety of its employees in highway work sites. With this diversity comes opportunity. This synthesis obtained survey responses from 41 state DOTs, and the results capture much of the national diversity. The survey respondents represent states from all regions in the United States. Some state DOTs have invested time and money in a new safety program, but others have invested in other programs and ideas. From a nationwide highway worker safety perspective, sharing the research and program methods is an economically efficient way to potentially improve worker safety nationwide. Although institutional limitations may prevent some state DOTs from follow- ing the model of other state DOTs that have a successful safety program element, understanding fellow state DOT safety programs can be useful in adapting broad safety ideas to a new organiza- tional context. There are distinct limitations of the survey approach to gathering this information and making generalized conclusions with confidence. To make the questionnaire as user friendly as possible, the questions were limited predominantly to numerical and multiple-choice responses. This format may have limited the depth of some responses and directed the respondents’ thinking in a way that would not have been the case if open response questions had been used. However, it was neces- sary to make the questionnaire simple and quick to complete to achieve the required response rate. Ten open response questions were included in the survey. Reporting results from such questions Organization Number of Responses Percentage of Responses Federal agencies 19 46 Other state DOTs 23 56 County/municipal governments 7 17 Private organizations 6 15 Other 10 24 None 9 22 n = 41. TABLe 4 ORGANIzATIONS WITH WHICH STATe DOT SHAReS DATA

44 and maintaining the anonymity of the states means the qualitative results can be reported only in broad terms. Improvements in data sharing and data availability are helpful in allowing states to make data- driven decisions for their safety programs. It is the hope that this chapter provides a better under- standing of national trends relating to state DOT data gathering and utilization practices for highway worker safety. This chapter highlights general opportunities for improvement that can be imple- mented at the state DOT level. The implementation of some of these opportunities can lead to more consistent nationwide safety practices and raise awareness of safety issues experienced by state employees working in highway work sites.