Lies, Damned Lies, Statistics and Loss Time Injuries

"There are three kinds of lies: lies, damned lies and statistics" - Mark Twain  

I am very impressed by the large number of construction and manufacturing companies and/or large-scale projects that are celebrating one, two, ten, twenty, thirty… even one hundred million worker hours without a Loss Time Injury (LTI).   Even more impressive is that most of these companies are located in the middle east and developing countries, where worker safety performance is often less than stellar.

Hmm, do we really believe that these reported LTI statistics are real or might there be some playing around with the statistics? 

I am certainly no expert in this topic (and frankly no expert in most topics that I write about!).  However, I am quite capable of critical thinking and reviewing published reports from experts in the field to see what they say.  In this blog post I explore the issue of LTIs, what they are, what is a realistic value, how they are (mis)reported, and what use, if any, they may have to the HSE professional.

What is a Loss Time Injury (LTI)?

Simply put, a loss time injury is a work-related injury (or illness) that occurred at the workplace or at an off-site location where an employee was carrying out work activities.  A Lost-time Injury Frequency Rate (LTIFR) is the number of lost-time injuries within a given accounting period relative to the total number of hours worked in the same accounting period, and typically expressed as number of LTIs per million hours worked. 

Every country seems to have their own terminology and specific definitions for LTI or a similar worker injury/illness statistic.  I'll point a few below but we do not need to be overly concerned about specific differences in definitions or application.  At the same time, while the definition for LTI is seemingly straightforward, there are many nuances to this definition. In particular, differences among definitions of  injury, illness, time and who is to be included in the LTI statistic.

For instance, the specifics as to what constitutes a recordable injury and illnesses are defined differently under OSHA in the US, WHS in Australia, HSE in UK, and the EU.  These differences can relate to the severity of the injury, the type of illness, the time frame in which the affected person is away from work (or moved to another position), and whether the person is an employee, contract staff or sub-contractor.

In contrast, in developing countries there are typically either very general or no specific requirements for reporting work-related injuries or illness.    

Do you think it might be just a coincidence that we rarely see reports of millions of hours without an LTI in developed/western countries where the reporting requirements are well defined?

What is a Realistic LTI?

How many of us have actually considered what it means to work one million man hours without a work-related loss time injury or illness? Let's examine this, first, with some critical thinking.

First consider yourself.   You are a safety/HSE professional and are (presumably!) much more informed about safety issues and (hopefully!) less likely to injure yourselves than workers at your project site or facility.  Do you believe that you are likely to work for say 40 years without one single work-related accident?  Even for those of us working predominately in an office setting this is no small feat!  This means never once stumbling and twisting an ankle, lifting a heavy load and tweaking a back muscle, or developing tendonitis from repetitive activity.   Remember, the LTI stat requires (typically at least) only a single day off work or that the worker would need to be moved to a position with light job duties to qualify as an LTI. 

Next, consider your workers, who are working day in and day out in a stressful, noisy environment with machinery and other hazards.   How many of them are likely to finish their career without a single LTI?  Few, if any?

Then, let's assume the average employee works 2,000 hours per year (that is eight hours per day, five days a week, 50 weeks per year).   In that case, one million man-hours is equivalent to 500 people working full-time for one year or 100 people working full-time for five years or 50 people working full-time for 10 years, etc.   Now, think of the chance that 500 people could work for one year at construction site or in a manufacturing facility without even one minor accident that would require a day or more loss time.  Even with the world's best safety program and the most highly trained workers imaginable, my guess is that it is close to nil (even in an office setting)!  Am I overly pessimist? What do you think?

Finally, if  you find it hard to imagine, as I do, the low likelihood of 1,000,000 man-hours without an LTI, then try to consider 2X, 10X, 20X, 30X or 100X this value and the probability of reaching these multiple million hours without an LTI!  I cannot imagine that these multi-million LTIs are even remotely possible.  If you disagree, please let me know where my logic is failing me.

Admittedly, this line of reasoning is simply based on my own biases and as I do not proclaim to be an expert.  So, let's see what is stated in published reports by academics.  There are a lot of report and statistics available, and the statistical parameters used for worker injury rates vary among the reports.  Nonetheless, they are still useful for our comparative purposes.   To facilitate our discussion, I have converted the reported values to the number of injuries we might expect in a hypothetical company of 500 people working for one year, which equates to the magical/mythical one million LTI level:

• Australia - Self-reported work-related illness and non-fatal injury rates over last three years ranges from 0.6 to 4.3 percent for information sector workers and agricultural, forestry & fishery sector workers, respectively. This means that in Australia for a typical company with 500 people, that we would expect from 9 to 25 work-related injuries or illnesses per year in information sector companies and agricultural, forestry & fishery sector companies, respectively
• New Zealand – A benchmarking study of multiple industries reported an average LTIFR of 2.0 per 200,000 hours for the companies participating. This means that in New Zealand for a typical company with 500 people, that we would expect from 10 work-related injuries or illnesses per year.
• Alberta, Canada –  Alberta uses a statistic for disabling injury (DI) rate, which measures the number of times per 100 person years worked that a work-related injury causes a worker to be away from work beyond the date of injury or requires the worker to have modified duties.  The DI injury Rate in 2009 was estimated at 3.08, which means that in Alberta for a typical company with 500 people, that we would expect approximately 15 disabling work-related injuries or illnesses per year.   
• United States – The incidence rates (the number of injuries and illnesses per 100 full-time workers per year) averages 3.4 in the United States (over all industry and government), which equates to 17 work-related injuries or illnesses per year for a typical company with 500 people.
• Europe – Reported non-fatal accidents across Europe that resulted in four or more days off work averaged about 1,700 incidents per 100,000 workers in 2012. This equates to 9 work-related injuries or illnesses per year for a typical company with 500 people.

Based on the above values, it seems reasonable that in developed countries to expect in the order of 10-20 loss time injuries/illnesses per 1,000,000 hours worked. 

By virtually any safety metric that can be imagined, developed countries have much higher levels of safety performance than developing countries.  For instance, in a previous blog post - An Overview of Occupational Health and Safety (OHS) Laws and Regulations in Asia and the Middle East - I presented ILO data on worker fatalities that showed fatality rates were 5-10 higher in developing countries compared to developed countries.  While I do not have specific data for injury rates, my guess would be that the ratio would even be higher.  This means I would expect a minimum of 100 loss time injuries/illnesses per 1,000,000 hours worked for a typical company in the developing world. 

Finally, I want to touch on the issue of under-reporting.  As noted in a report to the ILO: “Data regarding occupational injuries and illnesses are far less available in developing countries, and where records do exist they are generally unreliable. A critical problem is underreporting.”  Under-reporting LTI is clearly a problem everywhere.  For instance, studies done using safety data from Alberta, Ontario, and the Untied States all have identified that this is a serious issue and may lead to under-reporting by 20-70 percent.   Why under report?  There are many reasons  ranging from corporate pressure directed to HSE department to "improve" performance (e.g., enhanced public relations, avoid regulatory scrutiny, etc.), HSE department's handling of LTI statistics (e.g., gain performance bonus, concerns about job security) and worker's hiding injuries (e.g., fear of loss wages, cost of treatment, loss of  job, etc.).

The Value of Loss Time Injury Statistics

After the review above the obvious next question is: what is the value of loss time injury statistics?   A report prepared for Safe Work Australia provides a thoughtful overview of issues in the measuring and reporting of safety data.   They note that LTI rates have, over time, become the cornerstone of mainstream injury reporting and the benchmark against which organisational, industry and national comparisons are made.  

They also noted a number of important limitations of LTIs.  For instance, LTIs correlate poorly with both the human and financial consequences of work-related injury and illness  (i.e., damage) meaning they do not provide a valid or reliable measure of performance for the purposes of evaluating either the consequences of work health and safety failure or the success of work health and safety controls and initiatives.  Furthermore, growing anecdotal claims of individuals seeking to manage the measure, rather than to manage performance, have pointed to practices of deliberate manipulation and under-reporting of LTI data (as I have already discussed above).   

They go on to describe more meaningful alternatives to LTI.  In particular, they discuss the use of positive (leading) performance indicators (PPIs) as tools to better inform decisions regarding work health and safety risk and the effectiveness of hazard control initiatives.  I will perhaps discuss PPIs in a future blog post, but for now will leave it to the readers to educated themselves further if interested.

How to Join the Multi-Million House Zero LTI Club (and Gain Your Performance Bonus!)

Want great LTI performance, perhaps to enhance your end-of-year bonus?  This is simple.  Follow these five easy steps to guaranteed zero LTI performance!

Step One.  Throw your ethics out the window.  After all, you have money at stake so why worry if the statistics are not wholly accurate.

Step Two.  Make up your own definition of a reportable incident.  This is your opportunity to get creative.  I suggest that Mondays and Friday are not reportable.  Also any incidents that occur before or after noon hour.

Step Three.  Exclude all contract staff.  Talk to HR and make sure that all workers are outsourced and/or subcontractors used so they are not reportable if (when) an incident occurs.     If an employee does get injured, hide him in the lunch room until he gets better.

Step Four.  Learn to use the “zero” key expeditiously.  Use the “zero” key in the right upper corner of the keyboard when filling in your weekly incident report.  Enter "0" in all cells of the worksheet that ask for number of incidents.

Step Five.  Hold a celebratory party with your workers.  Why wait till year end when you can celebrate early?  Make a cake and a sign and invite your boss and celebrate your millions of hours without an LTI! 

Congratulations.  You have joined the multi-millions of hours LTI club!

Finally, I do recognize that there are many examples of outstanding safety performance in the developing world.  I do not mean to belittle those companies and safety professionals that have truly gained the impressive feats of high levels of LTI performance in their organizations.  Congratulations to you and your workers and please share your approaches for success with others!

Thanks for reading.  Keep safe.  Be healthy.  Respect your environment.

We hope that you will bookmark the blog, share it with your colleagues and visit the blog frequently because you find it informative and helpful.  We value your feedback and suggestions for future topics.

Please enter your email in the box at the top of the post and subscribe to our blog HSE Asia - our weekly blog will be emailed directly to you.

Next Week’s Blog Topic: Kudos to the HSE Professional Working in the Developing World

Photo Credits:  Ancient alchemy laboratory image courtesy of Adam Korzeniewski at  www.freemimages.com