About the Data
This calculator is powered by five years of federal crash investigation data from the NHTSA Crash Investigation Sampling System (CISS) — the most granular, nationally representative dataset on real injury outcomes from motor vehicle crashes available to the public. Unlike calculators built on editorial assumptions or small samples, every estimate here is derived from actual crash records collected in the field by trained federal investigators.
WHERE THE DATA COMES FROM
The NHTSA Crash Investigation Sampling System (CISS) is a probability-based national survey of police-reported motor vehicle crashes conducted by the National Highway Traffic Safety Administration. CISS replaced the earlier National Automotive Sampling System (NASS) Crashworthiness Data System (CDS) beginning with the 2017 survey year and expanded the scope, sample size, and data variables captured.
CISS data is collected by teams of trained crash investigators stationed at sites across the country. When a crash is selected for investigation, investigators physically go to the crash scene or hospital, examine the vehicles, interview occupants, and review medical records. The resulting data is far more detailed than what appears in police reports alone.
The CISS occupant-level file (OCC.csv) contains a WORKDAYS variable that records the number of workdays missed by each injured occupant as a result of the crash. This is self-reported by the occupant or obtained from medical and employment records gathered during the investigation. It is this variable that powers the lost wages estimates in this calculator.
All CISS data files are publicly available for download at no cost from NHTSA’s file download portal at nhtsa.gov. The OCC.csv and INJURY.csv files used in this analysis can be downloaded directly and the figures verified independently. We encourage attorneys and researchers to do so.
HOW THE DATA WAS ANALYZED
Step 1 — File Selection
Two files were used from each survey year: OCC.csv (occupant-level data, containing the WORKDAYS variable and overall injury severity MAIS code) and INJURY.csv (injury-level data, containing the specific body region REGION variable and injury-level AIS severity code). The files join on CASEID, VEHNO, and OCCNO fields.
Step 2 — Filtering
Records were filtered to employed occupants with known work absence data. WORKDAYS code 97 (not employed) and code 99 (unknown) were excluded. Valid records have WORKDAYS values from 0 through 62. Records with MAIS codes 9 (unknown) or 99 (not applicable) were also excluded from the severity analysis.
Step 3 — Pooling Across Years
Data from all five survey years (2020, 2021, 2022, 2023, and 2024) were pooled into a single dataset to maximize sample size and statistical stability, particularly for less common injury type and severity combinations. Pooling produced 34,502 valid occupant records and 65,766 valid injury records.
Step 4 — Stratification
Records were stratified by body region (REGION variable from the INJURY file, which uses the CISS region coding scheme) and injury-level AIS severity (AIS variable from the INJURY file). For each region-severity cell, median days missed, mean days missed, 25th percentile, 75th percentile, and sample size were computed. Cells with fewer than 20 records are flagged as statistically thin in the underlying data.
Step 5 — The MAIS 1 Distribution Confirmation
For minor-severity injuries (MAIS 1 / AIS 1), a specific analysis was conducted to confirm that the 5.5-day mean is a real distribution finding and not an artifact of the dataset’s 62-day recording ceiling. 52% of workers with minor injuries show zero missed days. Only 5.3% hit the 62-day cap. The distribution is genuine: most minor injuries resolve without significant work absence, and those who do miss work average 5.5 days. This finding is reported honestly in all calculator outputs for AIS 1 injuries.
THE AIS INJURY SEVERITY SCALE
The Abbreviated Injury Scale (AIS) is an internationally standardized anatomical scoring system developed and maintained by the Association for the Advancement of Automotive Medicine (AAAM). It classifies individual injuries by body region and severity on a scale from 1 (minor) to 6 (maximum, unsurvivable). The AIS is used by trauma surgeons, crash investigators, insurance actuaries, and legal experts worldwide as the standard framework for injury severity classification.
In the CISS data, each individual injury sustained by a crash occupant is coded to a specific AIS code (AISCODE variable) which maps to a region and severity level. The calculator uses the injury-level AIS severity (AIS variable) paired with the body region (REGION variable) to match the user’s inputs to the correct data cell.
AIS 1 — Minor: Soft tissue injuries, contusions, mild strains, minor lacerations
AIS 2 — Moderate: Simple fractures, disc herniations, moderate sprains
AIS 3 — Serious: Complex fractures, significant nerve or organ involvement
AIS 4 — Severe: Spinal cord involvement, major organ injuries, severe fractures
AIS 5 — Critical: Life-threatening injuries with uncertain survival
AIS 6 — Maximum: Injuries currently considered unsurvivable
THE 62-DAY RECORDING CEILING
The NHTSA CISS WORKDAYS variable records work absences from 0 to a maximum of 62 days. Any worker who missed 62 or more workdays is recorded as 62. This right-censoring of the data means that for injury cells where most workers hit the ceiling, the median will equal 62 — but 62 is a floor, not an estimate of the actual number of days missed.
For serious injuries (AIS 3 and above), the rate of hitting the 62-day ceiling ranges from 68% to 100% across injury regions. The calculator presents these cells as confirmed floors and uses the mean rather than the median as the primary estimate, since the mean — while also pulled upward by the ceiling — reflects the distribution more accurately than a censored median.
The one exception in the dataset is Spine/Back AIS 1 (minor spinal strain), which has a true, uncensored median of 3 days backed by 566 records and a 22% ceiling rate. This is the only cell where the median is a precise, defensible point estimate rather than a floor value.
KNOWN LIMITATIONS
Self-Report Bias
WORKDAYS data is collected from occupant interviews and medical records. Self-reported work absence data may understate actual missed time, particularly for occupants who returned to light duty, worked reduced hours, or had their work restrictions inadequately documented.
Right-Censoring at 62 Days
As described above, the 62-day ceiling means that actual work absence for serious injuries is understated in the data. The true median for AIS 3+ injury cells is higher than 62 days for most workers.
National vs. Florida-Specific Data
CISS is a national dataset. Florida-specific injury patterns, employer characteristics, and return-to-work factors may differ from the national distribution. The data is used here as a floor and reference point, not as a Florida-specific prediction.
Thin Cells at High Severity
Some injury region and severity combinations — particularly AIS 4–6 for less common injury regions — have small sample sizes even after five-year pooling. These cells are statistically less stable. The calculator falls back to the national MAIS-level distribution for cells with fewer than 5 records.
Questions About the Methodology?
Alan Sackrin reviews every inquiry personally. If you are an attorney or researcher with questions about the data, we welcome the conversation.
Data source: NHTSA Crash Investigation Sampling System (CISS), OCC.csv and INJURY.csv files, survey years 2020–2024. Downloaded from nhtsa.gov/file-downloads/CISS/. Analysis conducted in Python using pandas. All computations are reproducible from the public source data.