Here’s one that may interest (trigger) people. This poster presentation highlights the effects of behavioural observations on reported safety incidents the following week.
They looked at data over two years from a chemical manufacturing plant and oil refinery plant respectively.
I’m always a bit sceptical on studies which rely on behavioural observations, and more importantly, incident/injury data (which tend to be statistically rare and randomly distributed).
Also surprised they could find a statistically significant effect for such a small change – unfortunately it doesn’t report whether the relationship was statistically significant.
I only recognised one of the authors (Ludwig), but he’s published a lot on behavioural programs.

They reported:
· “While behavioral observation programs have been shown to reduce injuries in the aggregate, we investigated the impact of individual observations on the reduction in the probability of injury in the days following”
· “A rolling sum time-series logistic regression analysis was performed. This determined whether observations (normalized by work hours to reflect the number per 8-hour shift) over the previous seven days decreased the odds of an incident (dichotomized by if they occurred or not) occurring over the next seven days”
· “The analysis from the two-year sample (2022-2023) resulted in a 9.38% incident probability occurring over any seven-day span across five departments”
· “One additional observation per shift over a 3-day period was shown to reduce incident probability by 1.21%”
· “This was predicted to lower the number of incidents from 25 to 23 per year, resulting in two fewer incidents annually for the five departments studied. When extrapolated to the entire facility, with over 50 departments, one additional observation per shift could prevent at least 20 incidents annually”
· For the chemical plant, “each additional observation can reduce the odds of an incident occurring by 23% (Manufacturing Department) and 17% (Maintenance Department) over the next three days”
· “Over a year, observations could reduce total incidents by four incidents and 16 incidents, respectively”

Thoughts? Violent objections?
Ref: Rupert, N et al. (2024). Behavioral Observations Reduce the Probability of Injury for a Week. RCIO 2024.
Study link: https://scholar.utc.edu/rcio/2024/posters/30/
Poster: https://scholar.utc.edu/cgi/viewcontent.cgi?article=1358&context=rcio
My site with more reviews: https://safety177496371.wordpress.com