Supporting Statement A (1220-0133)

Census of Fatal Occupational Injuries

Supporting Statement

Part A. Justification

1. Circumstances necessitating data collection

The Bureau of Labor Statistics (BLS) was delegated responsibility by the Secretary of Labor for implementing Section 24(a) of the Occupational Safety and Health Act of 1970. This section states that "the Secretary shall compile accurate statistics on work injuries and illnesses which shall include all disabling, serious, or significant injuries and illnesses..."

Prior to the implementation of the Census of Fatal Occupational Injuries (CFOI), BLS generated estimates of occupational fatalities for private sector employers from a sample survey of about 280,000 establishments. Studies showed that occupational fatalities were underreported in those estimates as well as in those compiled by regulatory, vital statistics, and workers' compensation systems. Estimates prior to CFOI varied widely, ranging from 3,000 to 10,000 fatal work injuries annually. In addition, information needed to develop prevention strategies was often missing from these earlier systems.

In the late 1980s, the National Academy of Sciences study, Counting Injuries and Illnesses in the Workplace, and another report, Keystone National Policy Dialogue on Work-Related Illness and Injury Recordkeeping, emphasized the need for BLS to compile a complete roster of work-related fatalities because of concern over the accuracy of using a sample survey to estimate the incidence of occupational fatalities. These studies also recommended the use of all available data sources to compile detailed information for fatality prevention efforts.

In 1989 and 1990, BLS, in conjunction with the Texas and Colorado Departments of Health and the Texas Workers’ Compensation Commission, tested the feasibility of collecting fatality data in this manner. These studies, which can be found in Appendix A: Development of the Census of Fatal Occupational Injuries of BLS Report 845, April 1993, concluded that:

1. Multiple sources are needed to yield a comprehensive count of fatal occupational
   injuries;

2. Collecting data from multiple sources is feasible;

3. Detailed information is commonly available from administrative sources; and

4. Follow-up verification was necessary for instances where only one source document
   identified the death as work related.

The resulting CFOI program was implemented in 32 States in 1991. National data--covering all 50 States and the District of Columbia--have been compiled and published for 1992-2006, approximately eight months after each calendar year.

2. Purpose of data collection

The Census of Fatal Occupational Injuries compiles comprehensive, accurate, and timely information on work-injury fatalities needed to develop effective prevention strategies. The system collects information about the worker, the employer, and the fatal incident. (Data elements collected are listed in Attachment 1A.)

Data are used by employers and various public and private safety and health organizations to:

• develop employee safety training programs;

• assess and improve workplace safety standards;

• develop safety equipment;

• conduct research for developing prevention strategies; and

• compare fatalities between States to evaluate safety programs.

In addition, States use the data to publish State reports, to identify State-specific hazards, to allocate resources for promoting safety in the workplace, and to evaluate the quality of work life in the State.

CFOI data are also used by the Department of Health and Human Services to measure workplace fatalities in their Healthy People 2010 initiative. The National Safety Council and other organizations have adopted the CFOI program as the sole source of a comprehensive count of fatal work injuries for the U.S. The National Institute for Occupational Safety and Health (NIOSH) now uses CFOI data for the majority of its fatality research. This official systematic, verifiable count mutes controversy over the various counts from different sources. If this information were not collected, the confusion over the number and patterns in fatal occupational injuries would continue, thus hampering prevention efforts.

During 2006, BLS Washington staff responded to almost 1,400 requests for CFOI data from various organizations, including manufacturers, insurance companies, engineering firms, security services, universities, business associations, trade unions, government organizations, and news services. (This figure excludes requests received by the BLS regional offices and States.) In addition, the CFOI page of BLS Website averaged about 5,000 users per month in 2006.

The CFOI research file, made available to safety and health groups under letters of agreement to protect confidentiality, is currently being used by approximately 15 organizations to conduct studies on specific topics. (Attachment 1B lists data elements included on the research file.) Study topics include fatalities by worker demographic category (young workers, older workers, Hispanic workers); by occupation or industry (construction workers, police officers, landscaping workers, workers in oil and gas extraction); by event (heat-related fatalities, fatalities from workplace violence, suicides, and falls from ladders); or other research such as safety and health program effectiveness and the impact of fatality risk on wages.

A list of research articles and reports that include CFOI data can be found in BLS Report 2587, “Fatal Workplace Injuries in 2004: A Collection of Data and Analysis,” Appendix I, dated September 2007. Most of the articles listed in that Appendix can be found in the annual BLS Reports numbered 988, 977, 970, 961, 954, 934, 922, 913, 908, 891, 845, and 870.

3. Use of improved information technology to reduce burden

To reduce employer burden, the CFOI program primarily uses existing documents (death certificates, workers' compensation reports, Occupational Safety and Health Administration investigation reports, news sources, etc.) to identify and compile information on occupational fatalities. States are encouraged to use as many existing data sources as are available and necessary to compile a complete count of occupational fatalities. Instead of the administrative form itself, some agencies provide computer files or listings of the fatality information. In addition, participating State agencies make use of various Internet sites to gather information on fatal work injuries. The lack of standardization in data collected and coded between data sources and between States impedes additional computerization.

The followback questionnaire, BLS CFOI-1 (Attachment 2A), is used to substantiate work relationship when only one source document is available and to obtain missing data. The questionnaire is used to gather additional information on about 20% percent of the fatalities for which a report is received. The BLS CFOI questionnaire is designed so that only those sections pertinent to the respondent will be completed. The respondent only need complete the questions for the missing information. Because the respondent ordinarily completes only one section of the questionnaire, computerizing the form would not be cost-efficient.

The participating State and territorial agencies collect and process data for an average of about 115 occupational injury fatalities per year (or about 145 fatalities including illnesses). Data are uniformly coded by States and electronically transmitted to BLS for validation of coding and publication of results.

4. Efforts to identify duplication

During the development of the CFOI program, BLS met with representatives from the Occupational Safety and Health Administration (OSHA), the National Institute for Occupational Safety and Health (NIOSH), the National Center for Health Statistics (NCHS), and the Mine Safety and Health Administration (MSHA). The Bureau keeps these and other agencies informed of CFOI data collection and availability.

On September 9, 1992, a CFOI National Conference was held in Washington, DC to 1) introduce the program to potential data users; 2) solicit suggestions on how to improve the CFOI program; and 3) encourage the active participation of data users and providers. Representatives from various Federal and State agencies attended the one-day conference. The conference summary is available on request. Research papers presented at the conference were published in BLS Report 845, dated April 1993.

The Bureau also keeps abreast of developments in this area by reviewing current literature; publishing articles; and actively participating in national safety and health conferences and meetings. Staff members regularly make presentations at the annual National Safety Council Congress and American Public Health Association meeting as well as other selected safety and health conferences.

As mentioned previously, the CFOI program has been adopted by virtually all organizations involved in occupational health as the sole source of a comprehensive count of fatal work injuries for the U.S. The CFOI program is far more comprehensive and timely than previous national fatal work injury surveillance system. The only other major national occupational fatality surveillance system, the National Traumatic Occupational Fatality (NTOF) program operated by NIOSH, relied solely on information contained on death certificates marked “at work” and excluded fatalities occurring to individuals under 16 years of age. That system has been discontinued.

Each year a file of the job-related fatalities is created and made available, upon written request, to universities and safety and health organizations that conduct research in workplace injury prevention. Data are masked to the extent possible to protect confidentiality of decedents, companies, and data sources without losing details needed for research in injury prevention projects. Although State codes and personal identifiers have been removed from the file, it may be possible to identify decedents because job-related deaths are often unique and reported by the news media. To safeguard against unauthorized use of the file, each research organization requesting the file must agree in writing to adhere to BLS confidentiality policy. (See Attachment 1B for a list of data elements made available to researchers.)

The National Institute of Occupational Safety and Health (NIOSH) has currently over 60 staff members registered to use the CFOI Research File. NIOSH has requested additional data elements beyond those available on the current Research File for their use in conducting State-specific research and developing prevention programs. NIOSH has requested the following additional data elements: State codes, date of birth, date of death, death certificate identification number, and the narrative industry and occupation description. Providing this additional information was discussed with State representatives, who agreed that the additional information could be provided to NIOSH for meeting NIOSH's education and research mission. The additional data elements are provided to NIOSH under an agreement that requires adherence to the BLS confidentiality policy.

BLS has made arrangements to receive fatality reports from the following agencies: the Occupational Safety and Health Administration (OSHA), the Longshore and Harbor Workers and Federal Employee Compensation Divisions of the Employment Standards Administration (ESA), the National Transportation Safety Board, the Department of Transportation (Federal Railroad Administration), and the Coast Guard. Fatality reports from other Federal agencies, such as the Mine Safety and Health Administration (MSHA), NIOSH, and Federal Emergency Management Administration (FEMA), are accessed by the participating State agency from the Federal agencies’ Website. States have also made arrangements to obtain source documents, such as death certificates, workers’ compensation reports, medical examiner reports, and highway fatality reports, within their States.

Thus, the CFOI system is based primarily on administrative records received from Federal, State and local governmental agencies as well as some private sector organizations. Employers are not required to report fatalities to both the regulatory agency and to the State agencies operating the CFOI program. Employers or individuals are contacted only to clarify data or obtain missing information.

Using multiple data sources and followback techniques, the BLS CFOI program:

• compiles the most complete count of work-related injury fatalities in the United States;

• substantiates work-relationship using a consistent definition;

• includes work-related fatalities occurring to those of all ages;

• identifies the industry, occupation, and circumstances related to the fatal injury;

• provides detailed information on a timely basis about the event for developing and implementing prevention strategies; and

• allows policymakers, researchers, and other users to conduct research using a single database to assist in identifying relationships in serious workplace hazards that may not have otherwise been recognized.

5. Minimizing small employer burden

The burden on small businesses is minimal because the CFOI program relies on existing data sources for much of the information contained on the fatality database. Employers are not the sole source to contact to substantiate work-relationship or obtain required data elements missing from source documents. Alternate sources include the informant named on the death certificate and the attending physician or coroner or news reports. Approximately 1,200 of the total followback respondents are private sector employers. Data are not compiled to determine how many of these employers are small businesses.

When necessary, the followback to small businesses is to ensure that fatal injuries occurring to workers in smaller establishments are included in the count of fatal work injuries and that information concerning these deaths is correctly included on the CFOI database.

6. Consequences of less frequent data collection

Data collection is ongoing to compile and publish data on a current basis, usually 8 months after the reference year. By providing timely occupational fatality data, the CFOI program provides safety and health professionals the information necessary to respond to emerging workplace hazards. A number of organizations use CFOI data annually in their publications, in their research, and for strategic planning or performance measurement activities.

In 2006, 5,703 workers lost their lives as a result of injuries received on the job. This official systematic, verifiable count mutes controversy over the various counts from different sources. If this information were not collected, the confusion over the number and patterns in fatal occupational injuries would continue, thus hampering prevention efforts. Collecting data using a single data source and without verifying work-relationship would compromise the integrity of CFOI data.

The CFOI program has also been able to provide work-related fatality data for unexpected multiple-fatality events, such as the Oklahoma City bombing and the September 11 terrorist attacks, which would not be possible if data were collected less frequently.

7. Special circumstances relating to 5 CFR 1320.5

BLS is asking that followback respondents complete and return the CFOI questionnaire as soon as possible to reduce the number of nonrespondents due to relocation and to minimize recall problems for respondents.

Survey studies have shown that one's retention of information about an event declines over time. (See Judith T. Lessler, “Reduction of Memory Errors in Survey Research,“ Proceedings of the International Association of Survey Statisticians, 1989, International Statistical Institute, pp. 303-322, for a discussion of recall of information in survey studies.) Timely followback is, therefore, essential for collecting accurate fatality data.

8. Federal Register notice and outside consultation

Federal Register notice

BLS received comments from two parties in response to the Federal Register notice published in Volume 72, Number 192, October 4, 2007. The first commenter supports the continuation of the Census of Fatal Occupational Injuries program since it provides information for industry benchmarking.

The second commenter also supports the continuation of the CFOI program, regarding it as an extremely beneficial collection activity. This commenter also had several suggestions for increasing the utility of the CFOI data. Most of the comments pertained to the CFOI-1 Form (Attachment 2A), which is the questionnaire that the participating State agencies use to substantiate work relationship or to obtain missing information for a case. The form is used for approximately 20% of the cases received.

The first set of comments addresses Section II of the form “Employment Information.” The first comment pertained to question 1, which asks the respondent to check the appropriate item describing the decedent’s employment status at the time of the incident. The commenter questioned whether all the items were needed since the publication tables combine many of these categories and provided an alternative schema with definitions. BLS prefers to keep these items separate rather than combining them so that the detailed information is available for researchers interested in studying the particular class of worker. For example, the category for Active duty, Armed Forces assisted in identifying fatalities among the military for the article, “Fatal Occupational Injuries to Members of the Resident Military, 1992-2003” by Stephen M. Pegula, which appeared in the CWC Online, on April 25, 2005, and is available on the BLS Website at http://www.bls.gov/opub/cwc/sh20050421ar01p1.htm. The category for working in the family business has been used to study fatalities among young workers. (See “Occupational injuries among young workers,” by Janice Windau and Sam Meyer, which appeared in the October 2005 issue of the Monthly Labor Review and is available on the BLS Website at http://www.bls.gov/opub/mlr/2005/10/art2abs.htm.) BLS is adding additional wording to the self-employed and working in the family business categories to clarify these items.

The commenter also recommended an additional data item indicating whether the deceased was a member of a labor organization. Because adding such an item would entail a significant change to the CFOI computer system, BLS will consider this for a future revision. BLS will also need to consult with its State partners to see whether this information is commonly available from other sources. If not, the followback questionnaire, which is currently used for 20% of the fatalities, would have to be sent to additional respondents in order to obtain the information on a consistent basis.

The commenter also suggested that BLS provide employment data on tables showing fatalities by industry. This would assist data users in comparing detailed industries within major sectors, such as various types of construction industries, as well as when comparing between industry sectors, such as construction versus manufacturing. BLS does provide occupational fatality rates by industry, occupation, and demographic characteristics. The employment figures used in calculating the rates are included on this table. These rates are currently calculated using employment data from the BLS Current Population Survey, which provides more complete coverage of workers, such as the self-employed, than establishment-based employment data. CPS employment breakouts are not available for all the industry categories listed in the CFOI tables by detailed industry, however. For example, CPS has only one category for construction. Employment breakouts for residential vs. nonresidential construction and electrical vs. plumbing and heating are not available.

The commenter further recommended that the questionnaire list North American Industry Classification System (NAICS) categories in questions 2 and 7, which ask about the decedent’s occupation and industry respectively. Because occupation and industry are separate concepts with separate coding structures, BLS prefers not to list NAICS categories in question 2, which asks about occupation. The industry information provided by the respondent for question 7 will be available for State coding of occupation for those instances where the coding is dependent on the industry (for example, a farm laborer is assigned a different occupation code from a construction laborer). With respect to the suggestion for adding a list of NAICS categories for question 7, during our upcoming revision process we will ask our State partners whether adding such a list to the questionnaire will assist with industry coding, and if so at what level of detail.

The commenter questioned the relevance of Questions 9 and 10, which asks about the decedent’s longest held occupation and industry. BLS will assess the demand for these two items in deciding whether to retain them on the questionnaire.

The remaining comments pertained to Section III of the form “Information about the Incident.” Most of the comments pertain to items in questions 5 and 6 (which relate to the location and activity of the decedent at the time of the incident). In addition to assisting in determining the work relationship of the incident, these questions assist in coding the location and worker activity. The commenter mentions that some of the items listed do not appear to be under the employer’s control. The CFOI includes fatal injuries that occur while the decedent was working, regardless of whether the employer has direct control of the environment. The purpose of the CFOI program is not to assign fault to the employer or other party, but to compile a complete accounting of fatal injuries that occur at work. While including fatalities occurring to a worker while a passenger on a commercial airline does not appear to have a benefit in preventing worker fatalities, including such fatalities is necessary in determining costs of occupational fatalities and fatality risks of various occupations. BLS is modifying the wording to clarify that the first item in question 6 refers to one’s normal commute between home and one’s usual work location.

The final comment asked about the usefulness of question 9. The purpose of this question is to assist in classifying the incident as either a traumatic injury or an illness. This determination is necessary because the CFOI program only includes injuries in its annual counts.

Consultation outside the agency

BLS meets annually with the State agencies receiving grants to conduct the fatality census and has met semiannually with the Business and Labor Research Advisory Committees to solicit their respective views on data availability, frequency of collection, data elements collected, and recommendations for program enhancements.

In addition, BLS meets periodically with representatives from the following user groups:

• Occupational Safety and Health Administration and their advisory committees

• Mine Safety and Health Administration

• U.S. Department of Transportation

• United States Coast Guard

• Centers for Disease Control and Prevention

• National Institute for Occupational Safety and Health (NIOSH)

• National Center for Health Statistics

• Environmental Protection Agency

• Association of Vital Records and Health Statistics

• Council of State and Territorial Epidemiologists

• International Association of Industrial Accident Boards and Commissions (IAIABC)

• Medical Examiners

• The Hispanic Forum

• Center to Protect Workers Rights

• American Federation of Labor and Congress of Industrial Organizations

• American National Standards Institute (ANSI Z16.2) Committee

• Occupational Safety and Health State Plan Association

During the past several years, Bureau staff participated in numerous national conferences and presented CFOI data to the following major organizations: National Safety Council, American Public Health Association, Occupational Safety and Health State Plan Association, American National Standards Institute Z16 Committee, Voluntary Protection Programs Participants, American Society of Agricultural Engineers, International Safety Equipment Association, National Academy of Sciences, and the NIOSH National Occupational Injury Research Symposium.

Because of the usefulness of CFOI data in developing injury prevention programs, CFOI staff are often asked to participate in National workgroups such as the NIOSH National Occupational Research Agenda task force and the Hispanic Forum.

These exchanges provide BLS with input from data users to improve the CFOI program, promote the use of CFOI data in injury prevention programs, and provide BLS with information about other data systems. Researchers who use the CFOI research file also frequently provide feedback for improving information on the research file for use in developing injury prevention programs.

9. Payments to respondents

Followback respondents are not paid to complete the questionnaire. However, some agencies providing source documents, such as the State vital statistics agencies, require a nominal fee to cover copying costs.

10. Confidentiality provisions

The Commissioner's Order No. 1-06 "Confidential Nature of BLS Statistical Data," outlines BLS policy regarding the confidentiality of BLS statistical data. The cooperating State agencies must adhere to this policy, which states, "In conformance with existing law and Departmental regulations, it is the policy of the BLS that respondent identifiable information collected or maintained by, or under the auspices of, the BLS for exclusively statistical purposes and under a pledge of confidentiality shall be treated in a manner that will ensure that the information will be used only for statistical purposes and will be accessible only to authorized persons."

In addition, BLS and the cooperating State agencies agree to comply with the confidentiality provisions of the source agencies such as the vital statistics agency, workers' compensation agency, and other organizations that supply information.

The reporting form (BLS CFOI-1) used to substantiate occupational fatalities and solicit missing information about the circumstances of a fatal incident carries the statement, "The Bureau of Labor Statistics, its employees, agents, and partner statistical agencies, will use the information you provide for statistical purposes only and will hold the information in confidence to the full extent permitted by law. In accordance with the Confidential Information Protection and Statistical Efficiency Act of 2002 (Title 5 of Public Law 107-347) and other applicable Federal laws, your responses will not be disclosed in identifiable form without your informed consent."

The basis for the wording of this standard pledge is the Confidential Information and Statistical Efficiency Act of 2002 (Title 5 of Public Law 107-347), Commissioner's Order No. 1-06, "Confidential Nature of BLS Statistical Data," and advice received from the Department of Labor, Office of the Solicitor.

Those requesting the CFOI research file must sign a letter of agreement stating that they will comply with the Bureau’s confidentiality and computer security provisions. Data elements on the research file are masked to reduce the risk of identifying individual decedents or companies.

11. Sensitive questions

No questions of a sensitive nature are asked.

12. Estimated burden hours and costs

The burden of providing information on fatal injuries falls on two primary groups: federal, state, and local government agencies that provide the majority of source documents, and employers and individuals who are requested to fill out questionnaires when additional information is required. Federal, state, and local agencies may also be requested to provide information in a followback questionnaire if the fatally-injured person is a government employee. Because of the varying arrangements among these agencies to provide fatality reports, the total number of source documents and followback questionnaires received is being used for the number of responses on the Information Collection (IC) Forms.

The CFOI program requires that two or more independent source documents be collected to substantiate work-relationship for each fatality. When only one source document has been obtained for a fatality, the State agency mails a followback questionnaire (Attachment 2A) to a business or person named on one of the source documents. The followback questionnaire is used to substantiate work relationship or to obtain missing information for approximately 20% of the cases received.

Source agency burden

During program startup, BLS provided a template (Attachment 2B) to the States for use in setting up arrangements to obtain these source documents from other agencies within the State. States occasionally use this template to confirm these arrangements—particularly during periods of staff turnover. Significant modifications to the letter are discussed with the BLS regional office beforehand. Some States use the fax form (Attachment 2C) to obtain a source document for a fatality that has been identified by another source.

There are approximately 8,000 potentially work-related fatalities (injuries and illnesses combined) identified by the State agencies participating in the CFOI program. Source agencies (vital statistics agencies, workers' compensation agencies, State and Federal regulatory agencies, etc.) provided about 22,000 source documents annually for reference years 2004-2006. This total excludes media reports, which the participating State agencies collect themselves, and therefore do not result in public burden.

The burden to the source agencies includes retrieving and reproducing the necessary documents. We estimate this burden to be about 10 minutes per case, depending on how the data are stored. Note that some source agencies require a nominal fee to cover copying costs and some provide computerized files or microfiche. The burden hours and associated costs shown should, therefore, be considered a maximum estimate.

Burden to followback respondents

The followback questionnaire (BLS CFOI-1) is used to determine whether fatalities identified by only one source document are indeed work related or to gather additional information about the fatality. About 20% of the fatalities need a followback questionnaire to be mailed to a respondent. Some cases require that a followback questionnaire be sent to more than one respondent. Based on previous years’ counts, the total number of questionnaires for all respondents combined should not exceed 1,720 annually for all respondents combined.

The BLS CFOI-1 has been designed so that only those sections pertinent to the respondent will be completed. The respondent completes only those questions with missing information. The respondent burden can, therefore, vary from about 10 to 30 minutes, depending upon the information already available. This burden will average around 20 minutes and should total about 573 hours. (1,720 respondents x 20 minutes per response).

a. Federal Government

(Explanation of Part II, Attachment B, Paperwork Reduction Act Submission Worksheet)

Number of respondents

About nine Federal agencies provide BLS with source documents for work-related fatalities under their jurisdiction each year. Six of these agencies submit fatality reports on an electronic file or provide assistance with their Websites. In addition to submitting source documents, Federal agencies may be requested to complete a followback questionnaire pertaining to a fatally injured employee of the agency. There are about 20 Federal agency responses annually to the followback questionnaire for a total of 29 Federal agency respondents.

Number of responses

Federal agencies submit source documents for approximately 3,330 work-related fatalities under their jurisdiction. In addition, they complete about 20 followback questionnaires for fatalities to their employees for a total response of about 3,350 annually.

Annual time and cost burden

About 3,230 of the 3,330 Federal agency reports (97%) are submitted electronically or accessed from a Website. We estimate that it takes about 10 hours a year for each of the six agencies to extract the records from the administrative database and send the information to BLS, resulting in a burden of 60 hours a year. We estimate that the typical Federal employee performing this work to be a GS 12/1, with an hourly wage of $26.98. The cost burden for this portion is therefore $1,619.

The remaining 100 reports are submitted manually. The burden for submitting these reports is estimated to be 17 hours (100 x 10 minutes) for a GS 6/1 employee with an hourly wage of $13.69. The Federal burden for submitting reports manually is estimated to be $233. The total Federal cost burden for submitting source documents is estimated at $1,852.

Responding to the followback questionnaire results in an additional Federal agency burden of about 7 hours (20 x 20 minutes). We estimate that the typical Federal employee performing this work to be a GS 12/1, with an hourly wage of $26.98. The burden for this portion is therefore $189. The total Federal cost burden is estimated to be 84 hours, costing $2,041.

2. Individuals or Households

Participating State agencies are instructed to pursue other avenues for obtaining the necessary information before contacting individuals or households. If, however, these attempts fail, the State sends a followback questionnaire to an individual listed on the source document that has been received, although some States have policies against contacting the next-of-kin or other individuals. There are typically no more than 350 followback questionnaires sent to individuals and households annually. The burden hours for this response are estimated to be 117 hours (350 x 20 minutes), and the annual cost burden is $2,257. This estimate is based on an hourly wage rate of $19.29, which is the average wage for civilian workers (National Compensation Survey, Summary 07-03).

2. Private Sector

Private sector employers (including businesses, nonprofit institutions, and farms) respond to approximately 1,200 followback questionnaires annually. The burden for this response is estimated to be 400 hours (1,200 x 20 minutes), and the cost for this burden is estimated to be $10,460 annually. This estimate is based on an hourly wage of $26.15—the wage for a human resource specialist (National Compensation Survey, Summary 07-03).

2. State, Local, or Tribal Governments

State and local government agencies provide source documents for potentially work-related fatalities to the CFOI program. Some agencies require a nominal fee to cover copying costs and some provide computerized files or microfiche. Because these arrangements vary from State to State, it is difficult to adjust the burden estimates for reports sent electronically as was done for reports submitted by the Federal agencies. The burden hours and associated costs shown should, therefore, be considered a maximum estimate.

Number of respondents

There are approximately 220 State and local government agencies supplying the CFOI program with source documents (about 4 source agencies per State x 55 States and territories participating in the CFOI program). In addition, there are approximately 150 followback questionnaires completed by State or local government agencies for fatally injured employees. There are therefore a total of about 370 State, local, and tribal government respondents annually.

Number of responses

State and local governments provide a total of about 18,670 source documents and respond to about 150 followback questionnaires annually for an estimated total of 18,820 responses.

Annual time and cost burden

The burden hours associated with providing source documents to the State agencies participating in the CFOI program is estimated to be 3,112 hours (18,670 documents x 10 minutes). To estimate the cost of this burden, we used a wage rate of $15.05 (correspondence clerk, National Compensation Survey, Summary 07-03). The cost for this portion of the burden is estimated to be $46,836.

The State and local government burden for completing the followback questionnaires is estimated to be about 50 hours (150 questionnaires x 20 minutes). The wage rate of a human resource specialist, $26.15, was used to estimate the annual cost of $1,308.

The total burden for State, local, and tribal governments is estimated to be 3,162 hours, costing $48,144.

e. Information Collection Request (ICR) Summary of Burden

The following table summarizes the estimated annual burden for the various types of respondent.

Type of respondent	Number of respondents	Number of responses	Burden hours	Burden cost
Federal government—source documents	9	3,330	77	$1,852
Federal government—CFOI-1	20	20	7	189
Individuals—CFOI-1	350	350	117	2,257
Private sector—CFOI-1	1,200	1,200	400	10,460
State and local governments—source documents	220	18,670	3,112	46,836
State and local governments—CFOI-1	150	150	50	1,308
Totals	1,949	23,720	3,763	$62,900

13. Costs resulting from information collection

There are no additional costs to respondents or recordkeepers resulting from the information collection.

14. Estimated annual costs to the Federal government

Collection costs for the CFOI are funded on a 50/50 Federal-State matching basis. The Federal share of collection costs in Fiscal Year (FY) 2007 is $1.5 million, an amount that is being matched by participating States. An additional $2.0 million in FY 2007 non-collection expenses is being incurred by the BLS, for the purpose of providing and maintaining the computers, telecommunications capacity, and software needed to capture the data, as well as retaining the staff who refine the scope of the fatality census, maintain data quality standards, produce and distribute counts and rates to the States, and publish the national data.

15. Changes in burden

The decrease in the number of burden hours results from a change in the methodology used to estimate burden. The fact that most of the Federal agencies reporting to BLS send data files of administrative records was factored into the burden estimates. In addition, the estimate of the number of source documents received annually was reduced based on the actual number of reports received for reference years 2004-06. The source document count used in these estimates now exclude media reports for which there is no public burden.

16. Plans for tabulation and publication

Data collection for each reference year begins January 1 and continues through May of the following year. Participating State agencies code and key enter the data. BLS reviews the coding and data to validate the accuracy of State data collection. Approximately eight months after the end of the year, BLS combines data from all 50 States and the District of Columbia, generates various tabulations, and publishes reference year data for the entire nation. Once proper safeguards are in place to ensure confidentiality of individuals and companies, BLS creates a data file for researchers, which is available through a letter of agreement to protect confidentiality.

Annual BLS products of the CFOI program include:

• Press release;

• brochure;

• research articles on special topics published in national publications;

• the CFOI limited-access research file; and

• a compendium called "Fatal Workplace Injuries in [year]: A Collection of Data and Analysis," which includes research papers and data aggregations of selected worker and injury characteristics.

The press release for a reference year contains preliminary data for that year and revised figures for prior reference years. Similarly, the CFOI limited-access research file contains the preliminary data for the most recent reference year and updated data for the previous reference years. Beginning with the compendium for reference year 2002, the detailed tables in the appendices include the revised figures.

Participating State agencies also publish CFOI data in various formats similar to those listed above. Published data are distributed to individuals attending various State and national safety conferences, those requesting a specific publication, and those included on the CFOI mailing list. Press releases, charts, tabulations (both BLS- and user-generated), and research articles are available through the Internet. An electronic data retrieval system called the CFOI Profile System is also available to users. This system generates national and state profile and publication tables of occupational fatalities based on user selections.

In addition to published tables, BLS and participating State agencies produce various data aggregations to fulfill data requests. In 2006, BLS Washington alone responded to almost 1,400 requests for published and unpublished CFOI data. This figure excludes requests received by the BLS regional offices and participating State agencies.

17. Display of expiration date

The expiration date for OMB approval will be displayed on BLS CFOI-1.

18. Exception to certification statement

There are no exceptions to the certification statement.

13