For well over a century, the U.S. Bureau of Labor Statistics (BLS) has collected data and published reports on occupational injuries, illnesses, and fatalities. As a result, the safety profession has utilized this information to drive safety metrics. The reason is obvious – companies are mandated to collect injury data in a consistent manner and BLS serves it up on a silver platter to allow for benchmarking by industry and location.
This is both a blessing and a curse.
It is a blessing in that there is a breadth of knowledge across many industries with significant historical perspective. It is simultaneously a curse since the safety profession has not adopted any other widely adopted means of measuring safety at a universal level. The result: safety has been measured solely based on lagging data – injuries – and that pattern continues to this day. This is a problem due to the fact that low injury rates can occur despite working unsafely, thus providing workers and organizations a false sense of security. Are you good or are you lucky?
A case was made for years to adopt leading indicators to measure safety in addition to injury or lagging indicators. However, leading indicators are not set up with a universal formula like the lagging indicators of injuries. Most organizations are faced with hundreds of different leading indicator possibilities and forced to throw a proverbial dart at the board to select which ones to adopt.
Proactive & predictive
In a recent publication, Transforming EHS Performance Measurement Through Leading Indicators, the Campbell Institute defined leading indicators as being “proactive, preventative and predictive measurements that monitor and provide current information about the effective performance, activities, and processes of an EHS management system that drive the identification and elimination or control of risks in the workplace that can cause incidents and injuries.” Organizations should only seek to track indicators that have been vetted and can meet all of these criteria. This means the indicator metric has ranges developed on its strong correlation to injuries. Having a proven metric can be helpful, especially if a range of expectations has been established based on diligent testing and study.
In the white paper, Predictive Analytics in Workplace Safety: Four Safety Truths that Reduce Workplace Injuries, a series of leading indicators were discovered that allowed for the prediction and prevention of injuries. These metrics can be derived from four primary metrics:
- Inspections – an inspection is a collection of one or more observations
- Observations – a single instance of a behavior or condition (e.g. a worker wearing a hard hat)
- Safe and At-Risk – a single observation defined according to expectation – safe or at-risk
Truth in measurement
These four metrics are the basic building blocks of a more comprehensive safety observation program. Concurrently, they help develop key leading indicators for an organizations measurement of safety. Here is a summary of the correlated findings:
Safety Truth 1 – as inspections increase, incidents go down.
This is the easiest to measure and it is important to promote this activity. But simply doing this alone will solve nothing. The act of simply collecting more and more safety inspections, by itself, does very little. That would be like trying to lose weight by standing on the scale more often. It helps with providing information, but is simply the first step.
Safety Truth 2 – the probability of having an incident decreases as the number and diversity of the people performing inspection increases.
Simply sending the safety team out to conduct more inspections doesn’t help. In order for safety to improve, ownership by the team is essential. This means that everyone in the organization – from leaders to front-line supervisors to workers – has a part to play in identifying hazards, reporting them, and helping to mitigate the risk they pose but short-term and long-term.
Safety Truth 3 – too many 100-percent safe inspections is predictive of higher injury rates.
Invariably, a high number of inspections with no at-risk findings are seen on worksites with a relatively higher rate of injury. This is a very interesting and potentially counter-intuitive metric. One would think that as safety efforts improve, then fewer at-risk findings would be found. However, as long as humans are involved in the process, human error will be present. In addition, as one systemic issue is discovered and addressed, another is likely to surface that was virtually unseen before. Another potential issue with this metric is the potential negative connotation it can pose to those within an organization. The trick is to view this instead as an opportunity to improve a system or process or task before it leads to an injury.
Safety Truth 4 – too many at-risk observations is predictive of higher injury rates.
While this may seem counter to Safety Truth #3, this is a relative measurement. Finding at-risks is not the problem. Finding the same systemic issue repeatedly can be a problem. As an example, you find someone who is standing on the top of a ladder during a worksite inspection. As a conscientious person, you stop the work and make it safe. You discuss the issue with the worker and seek a safe resolution. The problem is averted and you move on. But how many times has this happened? What if your data indicated it happened across your organization many times in the last month. Simply finding and fixing the issue is a start, but only by addressing the causal factors (why it is happening over and over) will it result in an improvement.
Based on these studies, the resultant metrics have yielded a great “starter kit” for leading indicators.
The value in this starter kit is that these metrics are relatively universal across industries. While the actual hazards and the approach may vary, the resultant metrics can be compared quite easily across industries, as is done with injury rates. If enough organizations collect this information and share the metrics (not necessarily the specific findings), then benchmarking of leading indicators can begin to take a foothold and aid in guiding better observational approaches and improved data use plans for leading indicator collection.
Remember, though, that it is not the adoption of leading indicators or the collection of leading indicators that leads to improvement; it is the actions taken with the information that determine success. Bear that in mind when adopting any leading indicators – adopt ones that are actionable. When driving continuous improvement, it is the frequency and quality of the feedback generated from the findings that determine the level of success.