PSIF stands for Potential Serious Injury and Fatality, and it is the framework that most serious-injury prevention programs now run on. The core idea is simple: classify events by what they could have done, not by what they did. An event that nearly crushed a worker is a PSIF event whether or not anyone was hurt, because the potential for a fatality was present. This guide explains the framework, the precursor pyramid that makes it practical, and how to implement it in a forklift operation. It sells none of these systems.
The problem with outcome-based safety
Traditional safety measurement counts outcomes: recordable injuries, lost-time incidents, days away from work. Those counts are dominated by minor events, and they arrive after the fact. A facility can reduce its minor-injury rate while leaving its fatality risk untouched, because the minor events and the fatal mechanisms are driven by different hazards. The PSIF framework was built to separate the streams: track the events that carried the potential for a serious injury or fatality, and manage those with their own controls.
The precursor pyramid
The practical engine of PSIF is the precursor pyramid. At the base are the everyday events: near misses, unsafe acts, and unsafe conditions. Above them sit the events with clear serious-injury potential, such as a pedestrian stepping into the path of a moving truck. At the top are the actual serious injuries and fatalities. The pyramid works because the base is far larger and far more visible than the top: for every fatality, there are many serious-injury events, and for every one of those, many more precursors. Act on the base and you change the top.
In forklift terms, the base is visible in the data. This site’s dataset records 5,154 forklift-involved severe injury reports for 2015-2025, dominated by struck-by and caught-between events, and every one of those was a near miss seconds earlier. The accident statistics guide lays out the distribution, and the SIF explainer covers the same framework without the potential-severity emphasis.
Common objections and honest answers
The framework attracts objections, and they deserve honest answers. “We already track injuries” misses the point: injury counts are the outcome stream, and PSIF tracks the precursor stream that leads to outcomes. “Near misses are too many to review” is the strongest objection, and the answer is technology: the automatic logging layer filters the stream by severity and location, so the review meets the data where it is. “This is just another safety initiative” is fair, which is why the framework must produce a monthly review with named owners and a visible measure, the precursor rate per shift, that the operation can watch fall. The zero-accident culture guide on this site covers how the measurement system, not the slogan, is what makes the goal real.
The fourth objection is the quiet one: “we are not ready for this.” It is worth naming because it usually means the reporting culture does not exist yet, and a precursor program built on a silent floor produces an empty log. The answer is to start with the culture, not the framework: a blame-free channel, a visible response to every report, and the automatic logging layer that does not depend on volunteers. The framework is the destination; the reporting channel is the road.
Implementation in five steps
Step one is identification: log every event with serious-injury potential, including near misses, through a blame-free reporting channel. Step two is automatic capture: proximity detection and telematics systems log alarm events with time, location, and identities, which fixes the under-reporting problem. Step three is analysis: review each precursor for the conditions and behaviors behind it, using the safety rules guide as the reference. Step four is control: apply the engineering and procedural measures that reduce both frequency and potential severity, from pedestrian detection to speed control. Step five is review: a monthly meeting with named owners over the precursor stream, measuring whether the rate is falling.
The starting point is to score the operation before building the program. Run the Risk Scorer to see where your exposure sits, compare the control families on the systems index, and when you are ready for vendor numbers, send your scope through the quote form. The methodology page documents how the scorer derives its inputs, and your state’s injury series shows the local context.
