Census of Fatal Occupational Injuries (CFOI)
The Census of Fatal Occupational Injuries (CFOI) provides a comprehensive count of workplace fatalities, collecting detailed data on various aspects such as occupation, industry, and demographic information. Despite challenges in data protection, CFOI continues to play a crucial role in understanding workplace safety and implementing improvements.
Download Presentation
Please find below an Image/Link to download the presentation.
The content on the website is provided AS IS for your information and personal use only. It may not be sold, licensed, or shared on other websites without obtaining consent from the author.If you encounter any issues during the download, it is possible that the publisher has removed the file from their server.
You are allowed to download the files provided on this website for personal or commercial use, subject to the condition that they are used lawfully. All files are the property of their respective owners.
The content on the website is provided AS IS for your information and personal use only. It may not be sold, licensed, or shared on other websites without obtaining consent from the author.
E N D
Presentation Transcript
Census of Fatal Occupational Injuries (CFOI) Hypercube Ellen Galantucci Research Mathematical Statistician Office of Compensation and Working Conditions Bureau of Labor Statistics November 2, 2021 1 U.S. BUREAU OF LABOR STATISTICS bls.gov
Outline What is the Census of Fatal Occupational Injuries (CFOI)? Previous CFOI disclosure rules and challenges Logic for how to improve CFOI disclosure Results of disclosure updates Other attempts to protect CFOI data Future work 2 U.S. BUREAUOF LABOR STATISTICS bls.gov
What is CFOI? Produces a complete count of all fatalities that occur during a calendar year as the result of workplace injuries Detailed data are collected such as occupation, industry, demographic information about the decedent, cause of the fatality There were 5,333 workplace fatalities in 2019 3 U.S. BUREAUOF LABOR STATISTICS bls.gov
CFOI Data Collection Data are collected from a variety of sources Some are public, such as news reports Some are confidential, such as coroners reports and OSHA reports Information is often pieced together from multiple sources to fill in information for a single case 4 U.S. BUREAUOF LABOR STATISTICS bls.gov
Why is CFOI so hard to protect? Many of the data are public The counts are small It s a census so there s no question about who is in it We want to produce a large number of statistics about a small number of cases 5 U.S. BUREAUOF LABOR STATISTICS bls.gov
CFOI Disclosure Rules Any cells in which all data are collected from public sources are considered public and not subject to disclosure rules For cells with all confidential data or a mix of public and confidential data, we use a sliding scale to determine disclosure risk based on: Number of incidents Number of cases Until RY2019, no secondary disclosure was applied within a single year s publication 6 U.S. BUREAUOF LABOR STATISTICS bls.gov
Disclosure Example in CFOI Private sector wage and salary workers Government workers Self-employed workers Fatal injuries Industry NAICS code Architectural, engineering, and related services Architectural services 5413 29 26 1 54131 3 3 Engineering services 54133 14 12 1 Building inspection services 54135 4 3 Surveying and mapping (except geophysical) services Testing laboratories 54137 4 4 54138 3 3 TABLE A-3. Fatal occupational injuries to private sector wage and salary workers, government workers, and self-employed workers by industry, all United States, 2018 7 U.S. BUREAUOF LABOR STATISTICS bls.gov
Secondary Disclosure in CFOI Private sector wage and salary workers Government workers Self-employed workers Industry NAICS code Fatal injuries Architectural, engineering, and related services Architectural services 5413 29 26 2 1 0 1 54131 3 3 0 Engineering services 54133 14 12 1 Building inspection services 54135 4 3 1 0 Surveying and mapping (except geophysical) services Testing laboratories 54137 4 4 0 0 0 54138 3 3 0 TABLE A-3. Fatal occupational injuries to private sector wage and salary workers, government workers, and self-employed workers by industry, all United States, 2018 8 U.S. BUREAUOF LABOR STATISTICS bls.gov
Secondary Disclosure in CFOI Private sector wage and salary workers Government workers Self-employed workers Industry NAICS code Fatal injuries Architectural, engineering, and related services Architectural services 5413 29 26 - - 0 - 54131 3 3 0 Engineering services 54133 14 12 - Building inspection services 54135 4 3 - - Surveying and mapping (except geophysical) services Testing laboratories 54137 4 4 0 0 0 54138 3 3 0 TABLE A-3. Fatal occupational injuries to private sector wage and salary workers, government workers, and self-employed workers by industry, all United States, 2018 9 U.S. BUREAUOF LABOR STATISTICS bls.gov
Secondary Disclosure in CFOI Violence and other injuries by persons or animals Exposure to harmful substances or environments Contact with objects and equipment Transportation incidents Fires and explosions Falls, slips, trips Industry NAICS code Fatal injuries Architectural, engineering, and related services 5413 27 1 17 - 3 3 3 54131 Architectural services 3 - 3 - - - - 54133 Engineering services Building inspection services Surveying and mapping (except geophysical) services Testing laboratories 13 - 7 - - - 3 54135 3 1 - - - - - 54137 4 - - - - - - 54138 3 - 3 - - - - TABLE A-9. Fatal occupational injuries by event or exposure for all fatal injuries and major private industry sector, all United States, 2018 10 U.S. BUREAUOF LABOR STATISTICS bls.gov
How to implement secondary disclosure? Applying secondary disclosure within each published table individually would be easy but problematic Table differencing Data users want more data than just the tables on our website A hypercube would protect all data 11 U.S. BUREAUOF LABOR STATISTICS bls.gov
Hypercube Goal: to create an estimate for all possible combinations of all variables available in CFOI Each combination screened for primary and secondary confidentiality and suppressions applied Any suppressed cell would result in all children of that cell being suppressed 12 U.S. BUREAUOF LABOR STATISTICS bls.gov
Hypothetical Example Fatal injuries (Calculated) Fatal injuries (Published) Industry NAICS code 11 543 543 Agriculture, forestry, fishing, and hunting 21 135 135 Mining, quarrying, and oil and gas extraction 22 19 S Utilities 23 1,103 1,103 Construction 31-33 320 320 Manufacturing 54 92 92 Professional, Scientific, and Technical Services 55 2 P Management of Companies and Enterprises 13 U.S. BUREAUOF LABOR STATISTICS bls.gov
Hypothetical Example Fatal injuries (Calculated) Fatal injuries (Published) Event Event code Violence or other injuries by persons or animals 1 192 192 2 707 707 Transportation incidents 3 57 57 Fires and explosions 4 454 454 Falls, slips, trips Exposure to harmful substances or environments 5 301 301 6 482 482 Contact with objects and equipment 7 21 21 Overexertion and bodily reaction 14 U.S. BUREAUOF LABOR STATISTICS bls.gov
Hypothetical Example Contact objects or equip Fatal injuries (Published) Falls, slips, trips Exposure to substances NAICS code Violence Transportation Fires Overexertion Industry 192 707 57 454 301 482 21 Total Agriculture, forestry, fishing, and hunting 11 543 48 265 16 30 35 143 6 Mining, quarrying, and oil and gas extraction Utilities 21 135 7 68 12 2 (P) 8 34 4 22 19 (S) 1 (X) 5 (X) 1 (X) 6 (X) 2 (X) 4 (X) 0 (X) 23 1,103 70 249 18 352 200 205 9 Construction 31-33 320 38 79 10 51 48 94 0 (X) Manufacturing Professional, Scientific, and Technical Services 54 92 27 41 0 (X) 12 8 2 2 Management of Companies and Enterprises 55 2 (P) 1 (X) 0 (X) 0 (X) 1 (X) 0 (X) 0(X) 0 (X) 15 U.S. BUREAUOF LABOR STATISTICS bls.gov
Hypothetical Example Contact objects or equip Fatal injuries (Published) Falls, slips, trips Exposure to substances NAICS code Violence Transportation Fires Overexertion Industry 192 707 57 454 301 482 21 Total Agriculture, forestry, fishing, and hunting 11 543 48 265 16 30 35 143 6 Mining, quarrying, and oil and gas extraction Utilities 21 135 7 68 12 2 (P) 8 34 4 (S) 22 19 (S) 1 (X) 5 (X) 1 (X) 6 (X) 2 (X) 4 (X) 0 (X) 23 1,103 70 249 18 352 200 205 9 Construction 31-33 320 38 79 10 51 48 94 0 (X) Manufacturing Professional, Scientific, and Technical Services 54 92 27 41 0 (X) 12 8 2 2 Management of Companies and Enterprises 55 2 (P) 1 (X) 0 (X) 0 (X) 1 (X) 0 (X) 0(X) 0 (X) 16 U.S. BUREAUOF LABOR STATISTICS bls.gov
Hypothetical Example Contact objects or equip Fatal injuries (Published) Falls, slips, trips Exposure to substances NAICS code Violence Transportation Fires Overexertion Industry 192 707 57 454 301 482 21 Total Agriculture, forestry, fishing, and hunting 11 543 48 265 16 30 35 143 6 Mining, quarrying, and oil and gas extraction Utilities 21 135 7 68 12 2 (P) 8 34 4 (S) 22 19 (S) 1 (X) 5 (X) 1 (X) 6 (X) 2 (X) 4 (X) 0 (X) 23 1,103 70 249 18 352 200 205 9 Construction 31-33 320 38 79 10 (S) 51 48 94 0 (X) Manufacturing Professional, Scientific, and Technical Services 54 92 27 41 (S) 0 (X) 12 8 (S) 2 (S) 2 (S) Management of Companies and Enterprises 55 2 (P) 1 (X) 0 (X) 0 (X) 1 (X) 0 (X) 0(X) 0 (X) 17 U.S. BUREAUOF LABOR STATISTICS bls.gov
Hypothetical Example Contact objects or equip Fatal injuries (Published) Falls, slips, trips Exposure to substances NAICS code Violence Transportation Fires Overexertion Industry 192 707 57 454 301 482 21 Total Agriculture, forestry, fishing, and hunting 11 543 48 265 16 30 35 143 6 Mining, quarrying, and oil and gas extraction Utilities 21 135 7 68 12 2 (P) 8 34 4 (S) 22 19 (S) 1 (X) 5 (X) 1 (X) 6 (X) 2 (X) 4 (X) 0 (X) 23 1,103 70 249 18 352 200 205 9 Construction 31-33 320 38 (S) 79 10 (S) 51 48 94 0 (X) Manufacturing Professional, Scientific, and Technical Services 54 92 27 41 (S) 0 (X) 12 8 (S) 2 (S) 2 (S) Management of Companies and Enterprises 55 2 (P) 1 (X) 0 (X) 0 (X) 1 (X) 0 (X) 0(X) 0 (X) 18 U.S. BUREAUOF LABOR STATISTICS bls.gov
End result Original goal was to produce all combinations of all variables Estimated it would take about 6 months to complete all of the computation Instead, focused on all estimates that can be turned into LABSTAT codes for publication 76% of LABSTAT matched to hypercube 24% publishable after suppressions 19 U.S. BUREAUOF LABOR STATISTICS bls.gov
Are there other options for protecting CFOI? One method being used more frequently for disclosure limitation is differential privacy Because of the small number of fatalities and the number of them that are public, differential privacy would require a huge amount of noise to be protected A simple attempt at adding noise to tabular data resulted in significant problems 20 U.S. BUREAUOF LABOR STATISTICS bls.gov
CFOI with noise added Some results from my attempt to add noise to 2018 data: -8 fatalities in NAICS 3259 (Other chemical product and preparation manufacturing) Fatalities in NAICS 325 (Chemical manufacturing) dropped from 18 to 9 Industries with 0 fatalities have up to 5 fatalities with the noise added 21 U.S. BUREAUOF LABOR STATISTICS bls.gov
Current work Adding estimates at additional levels of detail Determining if there are some additional estimates that can be released without increasing risk Matching more files to LABSTAT Testing our success at protecting our data BLS had a Civic Digital Fellow spend the summer attempting to reconstruct CFOI microdata from our publicly released data 22 U.S. BUREAUOF LABOR STATISTICS bls.gov
Contact Information Ellen Galantucci Research Mathematical Statistician Office of Compensation and Working Conditions Bureau of Labor Statistics galantucci.ellen@bls.gov 23 U.S. BUREAUOF LABOR STATISTICS bls.gov
