LODD Causal Factor Matrix

Structured analysis of the complete NIOSH FFFIPP corpus. 7,520 contributing factors across 742 unique investigated incidents spanning 1998-2025. 4,581 normalized recommendations. 1,110 victim records. 795 USFA cross-linked records. Companion data appendix to The Lottery of Death.

The 27-Year Recommendation Record

Every NIOSH FFFIPP recommendation from 27 years of investigations has been structurally extracted and normalized into canonical topics. The result is definitive proof of institutional stasis: the same recommendations have appeared continuously since the late 1990s.

Recommendation TopicRecommendationsFirst AppearedLast AppearedSpan
Wellness / fitness program3151997202528 years
Annual medical evaluation2921997202529 years
Written SOPs / policies2021998202527 years
Physical ability evaluation1651999201920 years
Exercise stress testing1591998202325 years
Size-up / risk assessment1591998202527 years
Physician clearance1471998202527 years
SCBA medical clearance1361997202225 years
Incident command system1191997202528 years
Benchmark communications1031996202529 years
Incident safety officer871998202527 years
Rapid intervention team811998202527 years
Personnel accountability system801996202529 years
Crew integrity742000202424 years
Pre-incident planning701996202529 years
Situational awareness training672001202524 years
Code enforcement641996202529 years
Coordinated ventilation631998202527 years
Mayday training482002202523 years
The Definitive Finding
NIOSH has recommended wellness and fitness programs 315 times, in 306 separate investigation reports, over 28 years. Annual medical evaluation has been recommended 292 times (278 reports) over 29 years. Exercise stress testing 159 times over a quarter century. Eight of the top twenty recommendation topics are medical or wellness items that map directly to NFPA 1582 and 1583. The fire service has received the same prescription 315 times and has not filled it. NFPA 5th Needs Assessment (2021): 28% of U.S. fire departments have any fitness program; 61% do not provide NFPA 1582-compliant medical evaluations.

Matrix 1: Causal Fingerprint by Nature of Death (n=1,110 victims, 7,520 factors)

For each manner of death, how many contributing factors of each category does NIOSH identify per victim? Reading across a row shows the causal fingerprint of that death type. Reading down a column shows which death types are driven by each kind of failure.

Intensity (avg factors per victim):
0
0.1-0.5
0.6-1.0
1.1-2.0
2.1-3.5
3.6-5.0
5.0+
Nature of Death Victims Procedural Medical Equipment Communication Training Structural Human Staffing Environmental
Cardiac 308 2.92 4.14 0.28 0.13 0.33 0.05 1.02 0.22 0.34
Trauma 303 2.02 0.05 0.95 0.40 0.67 0.46 0.62 0.18 0.46
Burns 174 2.61 0.03 0.87 0.70 0.60 0.64 0.22 0.26 0.31
Smoke Inhalation 93 3.18 0.06 1.04 0.88 0.63 0.97 0.28 0.38 0.42
Asphyxiation 70 3.81 0.13 1.21 0.89 0.94 1.11 0.30 0.47 0.50
Crush 34 2.94 0.00 0.76 0.74 0.53 1.21 0.38 0.21 0.47
Drowning 16 3.63 0.50 2.50 0.88 1.19 0.25 0.31 0.31 1.00
Heat Stroke 16 3.25 2.44 0.56 0.63 1.25 0.19 0.63 0.50 0.75
Electrocution 9 2.11 0.11 0.67 0.78 0.78 0.11 0.56 0.33 1.00
27-Year Causal Fingerprint
Cardiac deaths are a medical-plus-procedural problem: medical 4.14 factors/victim (screening gaps, pre-existing conditions, cleared-for-duty failures) combined with procedural 2.92 (response protocols, resuscitation, return-to-duty policies). The earlier picture of cardiac as a purely medical problem was an artifact of less-complete v1 extraction.

All fireground deaths -- trauma, burns, smoke inhalation, asphyxiation, crush, drowning -- are dominated by procedural factors (2.0 to 3.8 per victim) with near-zero medical involvement. Asphyxiation and drowning also carry heavy equipment loading (1.21 and 2.50). These two causal fingerprints -- cardiac versus fireground -- require fundamentally different interventions and have for the entire history of the FFFIPP program.

Factor Severity and Preventability (n=7,520)

For each contributing-factor category: total count, severity distribution, and NIOSH's own preventability rating.

Factor Category Total Severity Distribution Critical Major % Preventable Priority
Procedural 2,620
794 1,166
100%		 HIGHEST
Medical 1,453
553 568
85%		 HIGHEST
Equipment 722
173 283
90%		 HIGH
Human 617
204 240
64%		 MODERATE
Training 559
70 283
100%		 HIGH
Communication 440
115 206
98%		 HIGH
Structural 419
173 152
29%		 LIMITED
Environmental 405
62 160
8%		 NOT PREVENTABLE
Staffing 236
23 110
96%		 HIGH
Procedural Overtakes Medical
With consistent v2 extraction across the full 27-year corpus, procedural factors are now the #1 category (2,620) -- surpassing medical (1,453). In v1, medical factors were over-coded and procedural factors were under-coded; the re-extraction under a single canonical prompt fixed the classification. The true picture: fireground procedure is the largest preventable failure surface in the entire dataset, at 100% preventability. Training (100% preventable), communication (98%), and staffing (96%) complete the set of categories that are within direct control of fire department leadership and are essentially always preventable by NIOSH's own ratings. Structural factors (29% preventable) and environmental factors (8% preventable) are the only categories where the fire service legitimately cannot control the outcome.

Recommendation Evolution by Decade

Top five normalized recommendation topics per decade. The shift in the 2020s is unmistakable -- and it reveals something about NIOSH's investigative selection, not about the underlying fatality pattern.

Decade#1 TopicCount#2 TopicCount#3 TopicCount#4 TopicCount#5 TopicCount
1990s Wellness/fitness39 Annual medical eval35 SOPs/written policies23 Exercise stress testing22 SCBA medical clearance18
2000s Wellness/fitness159 Annual medical eval146 SOPs/written policies109 Physical ability evaluation89 Exercise stress testing80
2010s Wellness/fitness102 Annual medical eval95 Physical ability evaluation75 Size-up / risk assessment62 Physician clearance54
2020s Size-up / risk assessment30 Incident command system24 SOPs/written policies20 Incident safety officer17 Benchmark communications17
The Decade Shift -- And What It Really Means
Medical/wellness recommendations dominated the 1990s, 2000s, and 2010s. In the 2020s, every medical/wellness topic drops out of the top five and is replaced by fireground-tactical recommendations (size-up, incident command, safety officer, benchmark communications). This is not because medical issues stopped mattering. Cardiac remains roughly 49% of non-COVID USFA-recorded LODDs in 2020-2025.

The shift is an artifact of selection. Post-2019, NIOSH's coverage of USFA LODDs collapsed to 5-15% per year -- and within that thinned sample, the program skewed harder toward structural fireground incidents with tactical learning value. The cardiac deaths are still happening. NIOSH just stopped investigating most of them. The 2020s column reveals what NIOSH chose to look at, not what kills firefighters.

The Investigator's Lens: NIOSH vs USFA Population

The NIOSH investigated cohort is not a representative sample of U.S. firefighter LODDs. The cross-link between NIOSH investigations and USFA fatality records makes the selection bias measurable for the first time.

Characteristic (USFA 1998-2025) NIOSH-investigated
(n=781)		 Not investigated
(n=2,327)		 Direction of bias
Mean age at death 42.2 49.8 NIOSH investigates 7.6 years younger
Cause: cardiac / overexertion 39.7% 45.0% Cardiac undercounted
Cause: caught, trapped, or collapse 26.4% 19.8% Structural entrapment 33% overrepresented
Cause: vehicle collision 6.3% 12.8% Vehicle deaths cut in half
Wildland fireline 1.5% 6.6% Wildland roughly quartered
NIOSH Coverage Collapse, 1998-2025
Across 1998-2025, NIOSH investigated 25.1% of USFA-recorded LODDs. Year-over-year coverage peaked in the late 1990s and early 2000s (30-50%), drifted down through the 2010s (25-40%), and fell off a cliff in 2020 (12.0%), 2021 (8.1%), 2024 (7.6%), 2025 (5.2%). Any reading of the fire service's recent LODD trend that relies on NIOSH reports is drawing on a single-digit percentage of the real population. The USFA view is the only honest recent-era view.
LODD Causal Factor Matrix -- 27-Year Dataset (v2.2) · Revised 2026-08-24
Derived from 7,520 contributing factors and 4,581 normalized recommendations across 742 unique NIOSH FFFIPP investigations (1998-2025, with 12 pre-1998 FACE-era records). Supplemented by 3,916 USFA fatality records (1990-2025) and 795 USFA-NIOSH victim cross-links. Extraction performed via Claude Opus 4.7 under a single canonical prompt with JSON-schema-validated output; per-field and per-incident confidence scores populated on every record (mean 0.92). All data from public government sources.
Companion to: The Lottery of Death: Why Firefighters Keep Dying the Same Preventable Deaths