The Real Problem With Zero-Incident Safety Programs
Journey to Zero safety programs sound promising, but they can encourage under-reporting.
Everybody in the safety field has heard about Journey to Zero (JT0) programs. Sometimes they're called Target Zero, Zero Harm, or any number of similar names. The label changes, but the goals and principles are the same across the board.
These programs inspire strong feelings. Very few safety professionals are neutral about them. If you work in safety, you either hate them or love them.
In this article, I will take a close look at them to determine which of those attiutdes is warranted - the love or the hate. This is a deep dive intended to show exactly what JT0 programs are, what they promise, and how well they deliver on that promise.
A Bit of Background
Regardless of the program, Journey to Zero is an artificial construct adopted by many organizations in the energy, construction, and manufacturing industries. It states, broadly, that all injuries are preventable.
This ethos extends beyond the company itself. Contractors working for organizations with JT0 programs have to subscribe to this idea and demonstrate a near-perfect safety record in order to bid and win contracts from them. These companies thus define safety as the absence of accidents and rate their contractors on lagging indicators, with the assumption that the lower the rate of injury a contractor has, the safer their employees are and the least likely that an injury will happen while working on their projects. Some prime contractors go as far as stating that the goal of their projects is to be injury-free and, as such, expect all their subcontractors to be injury-free as well.
(Learn about How Leading Indicators Can Make Your Workplace Safer)
It is unclear when the first JT0 program appeared, but Clements (2005) notes that Dollard was discussing the demoralizing effects of zero goals as early as 1939. More recently, Edward Deming analyzed the efficiency of zero-goal practices in his 1986 edition of Out of the Crisis.
JT0 programs have only grown since then and we can argue that their preeminence is directly correlated with the health and safety performance of various organizations coming under public scrutiny. This scrutiny is facilitated by the mediatization of workplace accidents, which can have devastating effects on organizational image and implicitly on their ability to gain contracts and on their stock value. From this perspective, JT0 programs are an offshoot of the corporate social responsibility. They crop up in organizations that are not willing to appear like they are shunning their moral responsibility toward their employees.
While there is no consensus among safety and performance management professionals about the efficiency of the JT0 programs, some of them consider a goal of zero recordable injuries as a possible organizational goal.
What isn't clear, however, is whether a zero-injury goal is good leadership practice and whether these ambitious targets really are conducive to reducing the number and severity of workplace injuries.
The Theory and Practice of JT0
The Theory Behind Zero-Injury Safety Programs
Judging by the reduced rate of injuries reported by companies that embrace JT0 programs and the several studies published by zero-incident proponents, it seems that they are conducive to improving safety performance even if they don't quite reach their ambitious zero-injury goals.
According to Herzberg’s two-factor theory, workplace safety is a hygiene factor and therefore contributes to employee satisfaction and retention. Indeed, Maslow considers safety to be among our most basic needs, just above our physiological ones. If not satisfied, the need for safety would impede employees' attempts to satisfy higher level needs that are desired and beneficial for the employer. It comes as no surprise, then, that employers have a vested interest in promoting an injury-free workplace.
(Learn about the Three Levels of Safety)
JT0 also seems favorable from an ethical standpoint. A zero goal might seem like the only ethically justifiable goal - anything less implies that companies accept injuring workers as a cost of doing business. In theory, then, zero injury seems like a great representation of ethical leadership.
Robert Krzywicki, global practice leader for employee safety at DuPont Sustainable Solutions told EHS Today in 2013 that DuPont, a company with more than 60,000 employees and dozens of plants, has been striving for zero injuries for over 200 years and that some plants have achieved zero for extended periods of time.
Halliburton has their own JT0 program and states on their website that all unplanned events can be prevented. They even boast that this is already happening in many areas within the company.
JGC Fluor BC LNG, a joint venture executing what they call Canada’s largest infrastructure project, state as part of their orientation process and in the employment fact sheet available on their website that despite the gigantic size of their project and the multitude of trades and contractors they will employ to execute it, their project will be “the Safest Project on Earth.”
Zero-Injury in Practice
While Krzywicki states that some of their plants have reached zero, the reality is that the vast majority of their plants have not, even for short periods of time. Despite DuPont’s 200 years focus on process improvement, the system is not replicable and zero results are likely a product of chance rather than a system that works consistently. Moreover, OSHA has placed the organization in the Severe Violator Enforcement Program after the death of four employees at one of their Texas plants.
Halliburton, the largest oilfield service company, despite its claim to zero-injury success, has the highest number of severe injuries in its industry group.
Despite their ambitious promise, JGC Fluor BC LNG has seen a number of significant injuries among its contractors since the start of the project. In fact, their page touting the Safest Project on Earth now returns a 404 message, likely due to the distance between the claim and the reality in the field.
The disconnect between the celebration of JT0 programs and the actual results raises an important question: how is success being measured? And does the goal of zero encourage contractors to claim a perfect record by being creative with their injury recording?
A major obstacle in assessing the success of zero-injury programs is the lack of a common definition across industries. Each organization is free to define its own program and success criteria. Moreover, Soraghan (2017) points out that until 2015 employers were required to report to OSHA only fatalities and accidents where three or more employees were hospitalized, making most workplace injuries opaque to the government and the industry.
These days, many safety professionals and researchers conceded that a zero-goal safety program incentivizes employees and employers to under-report injury data.
In fact, the problem was so prevalent in 2012 that Richard E. Fairfax, OSHA's Deputy Assistant Secretary, issued a memorandum to its regional administrators and whistleblower program managers highlighting the extent and methods used by employers to suppress reporting.
The suppression of the reporting is, among others, encouraged by the fact that the organizations promoting JT0 programs condition the awarding of contracts to the safety performance of their contractors. Since past results cannot be changed and performance cannot be improved fast enough to allow the companies to bid, the only way a company can survive in JT0 environment is by being creative with their recording.
In other words, organizations make their own decisions about what constitutes a recordable injury and what does not. This creates major discrepancies between the organization’s self-reported data and the Worker’s Compensation or state and provincial Labor department data for that organization. Deming (2000), speaking about zero targets, said that “Fear invites wrong figures. Bearers of bad news fare badly. To keep his job, anyone may present to his boss only good news “.
This embellishment of the data happens sometimes without the knowledge of the senior management of the organization. Most frequently, however, it takes place with senior management’s knowledge and encouragement in order to rip the benefits listed above or to promote a better public image with direct connection to stock value.
Under-reporting is obvious also when comparing the self-reported injury rate and the fatality rate. Many of the JT0 supporters have a lower than the industry injury rate and a higher than the industry fatality rate.
A JT0 Case Study
Journey to Zero (JT0) programs have been promoted down the supply chain. For this reason, it is helpful to look at an industry as a whole by examining one of its major players and a smaller supplier down the supply chain.
I will analyze a large oil and gas player in Canada (Organization A) and one of their contractors (Organization B). These are real organizations, but to avoid any naming and shaming, I have anonymized them and altered their statistical data altered with a multiplier (the ratios have been maintained).
A Snapshot of the Oil and Gas Industry
While fatality rates for the O&G industry are certainly high, some executives and researchers make the argument that the fatality rate is going down, despite the number of employees increasing during the boom periods. Though that is mathematically true, the variation is insignificant. What is lost is that when superimposing the fatality graph and the oil price fluctuation graph we notice that the “best safety record” matches almost perfectly the periods of low oil prices, when employment in the industry decreased dramatically.
This suggests that the reduced fatality rate has nothing to do with JT0 and other safety improvement programs and is strictly correlated to low employment numbers.
According to the BLS statistics from 2017, the non-fatal injury rate of the O&G industry was 1.5, while the same rate for the construction industry was 3.1, statistically making the O&G safer than construction by a factor of 2. However, the BLS statistics fatal injuries rate for O&G is 12.9, while the construction fatality rate is 9.5, making O&G performance worse by a factor of 1.4.
This anomaly is further accentuated by the fact that, when arranged based on their non-fatal injury rate, the industries list will be similar to the list resulted from arranging them based on the fatality rate. This supports Heinrich’s (1931) and later researchers findings that there is a ratio between minor injuries, severe injuries and fatalities.
The O&G is the one notable exception, which is explained by a chronic under-reporting of injuries. Using the same ratio of fatalities/injuries as in construction, we can calculate a more realistic injury rate for the O&G sector is around 4.2 - one of the worst among all industries and certainly far from the stated goal of zero.
Furthermore, the real disparity is certainly higher, since there are 21 states where companies report to state agencies rather than OSHA. This includes large oil and gas producers, such as California, New Mexico, Utah, and Wyoming.
The ISNetWorld statistics for 2017 show a Total Recordable Injury Rate (TRIF) for NAICS code 211, O&G extraction, is 0.53, suggesting that zero is achievable. However, the fatality rate is 15.71, even higher than reported by BLS, underscoring the under-reporting problem noted above and the unreliability of self-reported data.
The same database uses different codes for the O&G industry in Canada, making benchmarking impossible. Despite that, we can see the same low TRIF contrasting a very high fatality records for some O&G sub-industries.
According to self-reported performance, it seems that the TRIF for the organization is very close to zero, allowing Organization A to claim that JT0 is coming to fruition.
This is contradicted by the Alberta Labor data, which shows a disabling injury rate three to four times higher than Organization A’s reported TRIF (Alberta Labor, 2019). The difference between the self-reported LTIF and the LTIF shown on Alberta Labor website (2019) is even higher.
It is also notable that 2016, a year characterized by low oil prices, shows also a reduction in the injury rate, further strengthening the thesis that apparent improvements in safety performance can be credited to low employment rates in the industry.
According to self-reported performance, Organization B’s TRIF averages around 1, which is a number good enough to keep the company on the bidder list for most major clients in the oil and gas industry.
However, for a specific client, Organization B’s TRIF was still too high to be allowed to be awarded the contract. The client suggested they only report the injuries related to the scope of work to be executed. By ignoring the injuries from another side of the business, Organization B artificially reduced its injury rate. After doing that, they were awarded the contract. This reinforces the idea that the industry knows and accepts the manipulation of injury data and plays the “zero injuries” game with full knowledge of the system being gamed by their subcontractors.
While JT0 programs are designed to improve safety performance, this shows how they can, in practice, simply make data less accurate.
Alberta Labor (2019) numbers are aligned with Organization B’s internal reporting. This is obvious from the discrepancy between these numbers and the WCB recorded injuries for Organization B, where only one year, 2015, shows an alignment between WCB, Alberta Labor, and self-reported data. For all other years researched, it appears that the self-reported injury numbers and rates have been generally halved when compared to WCB numbers - presumably in order to maintain a low TRIF and keep the organization competitive when bidding for contracts. A third-party auditor stated that the organization has improperly categorized recordable injuries as non-recordable.
One motivator for not including all the injuries that appear in the WCB record into the self-reported numbers is the perpetuation of "the careless worker myth” by the Government of Alberta and other regulatory bodies. The organizations feel it would be unjust to lose bidding opportunities because of workers' carelessness, and, as such, accidents that are felt to be the worker’s fault are removed from the record. A third-party auditor, in the review of Organization B’s incident investigations, concluded that in certain cases injuries were removed from the record when employees have been blamed for the accidents.
The auditor noted that although Organization B had a target zero goal, no other health and safety corporate goals and supporting programs have been identified, highlighting a major shortcoming of JT0 programs - most of them are not programs at all but mostly a leadership vision. They do not focus on systemic fixes of the working environment, but instead place the responsibility for being safe on the worker’s shoulders.
Finally, a quick look at Organization B and Organization A’s externally verifiable data shows that there is not a gradual progression towards zero. Rather, performance varies up or down from one year to another. Fluctuations in the oil and gas prices is one explanation for these swings. The second explanation is that long-term zero-injury periods seem to be the result of simple randomness and data manipulation, rather than the efficacy of JT0 programs.
Conclusions and Recommendations
Zero injuries is a misleading goal. What it aims for in theory is a high bar for safety. The absence of injuries, however, does not prove that an organization is safe. This is especially so given the fact that injuries can simply go unreported or under-reported.
(For related reading, see Help Me Help You: 10 Ways to Get Employees to File Incident and Near Miss Reports)
Another factor is the buck-passing that happens within some organizations. Employers invest in occupational safety and often equate that investment with being safe. With that belief in place, resulting accidents are explained by blaming them on employees. The Alberta government's “Stupid” and “Bloody Lucky” campaigns, for instance, focus strictly on worker’s behavior and reinforce the idea that incidents wouldn't happen if workers were not careless. There is hardly an orientation program that doesn’t state that “safety starts with you” and that the organization will not tolerate unsafe workers. And there are plenty of examples of accident investigations that simply chalk up the event to "human error."
Management is also subject to a major role conflict. While they likely state that safety is their first priority, the reality is that they are responsible to shareholders for turning a profit. Management, then, are incentivized to take a cost-benefit approach to safety. Basically, the employers will go for what they perceive to give them the best bang for their buck, not necessarily the safest solution in the market.
Another problem is that zero-injury is not a SMART goal. It is not specific enough. Achieving "target zero" does not provide any direction on who should do what to get that result. Worse, there is no clear, incontestable evidence that it is achievable - nobody has achieved a zero injury rate for an extended period of time.
We can even argue that zero is not actually a goal. At best, "zero injuries" can be seen as a slogan, and slogans have a poor record of improving performance.
JT0 programs are conducive to hiding or ignoring recordable injuries, and the existence f these programs is not demonstrably correlated with an improved safety record.
What should employers and safety professionals do in light of this?
Organizations, employers, and governments should, first and foremost, stop using zero injuries as a target and retire its associated slogans.
We should recognize the fact that many accident causes are beyond the control of the employee. Organizations should, therefore, focus their energies on identifying weaknesses in the system, rather than admonishing workers for being careless. It is the organization and not the employee who needs to have a proper system to select and train workers, provide and ensure the use of PPE and appropriate tools, and implement engineering and administrative hazard controls.
(Find out Who Pays for Personal Protective Equipment?)
We also need to shift the focus back to leading indicators of safety.
JT0 programs are focused on an end result of zero injuries. An injury rate is a lagging indicator and basing a safety program on it transforms the organization into a spectator with less influence on the outcome. To become an active player, each organization needs to identify, collect, and analyze meaningful leading indicators that can be used to achieve lower injuries.
Generating this data, organizing it, and distributing the knowledge drawn from it is instrumental to becoming a learning organization. And that, in turn, is essential for being proactive, which is indicative of better safety performance. Frequent hazard assessments, safety meetings, toolbox talks, inspections, and audits with management’s involvement are required and should lead to continuous improvement.
To minimize cultural shortcoming (such as being individualistic and short-term oriented), organizations should stop rewarding perfect performance and reward proactive reporting to increase the organization’s opportunities to learn. Rewarding the employees for reaching the leading indicators (inspections, hazard assessments) and for providing suggestions for improvement of the safety system is a way better approach then rewarding people for not getting injured. Shifting to collective rewards could also increase internal accountability.
(Find out why Your Incentives Are Compromising Safety Culture)
Continuing the leading indicators idea - Regulatory bodies should standardize most common leading indicators for the industry and normalize them (such as number of inspections/site, frequency of certification inspections/equipment, number of audits/year, etc) to create equal frames of reference among organizations. These bodies should externally audit these organizations, based on this set standard, make these audits public to be used as selection criteria in bidding processes, to eliminate data manipulation and make evaluations fair and easy.
Linking claims cost to insurance premiums creates incentives for employers to hide or under-report accidents. To eliminate the incentives for under-reporting and dealing with injuries internally, WCB insurance premiums should be set per industry and not based on individual performance, since performance pricing did not result in better performance. However, regulatory bodies should substantially increase the fines for willful safety violations, to ensure the fine is higher than the cost of compliance, thus enticing employers to comply proactively, not after recording an injury. Injury databases, such as WCB, Alberta Labor, and so on, should be synchronized to eliminate data gaming by employers.
Finally, it would be beneficial for all injuries and fatalities to be investigated with the participation of an external body, as is the case in the aviation industry. The accident, findings and effective corrective actions should be published on an industry wide platform to transform them in explicit knowledge and to help all organizations learn and continuously improve.
More realistically, organizations would benefit from freely sharing their best practices – this would accelerate the spread of good safety practices and help other organizations improve without having to go through the traumatic experience of an injury. We all want our employees to be safe, so the secrecy that dominates many aspects of the corporate world should not extend to occupational health and safety.
We can summarize the state of JT0 programs considering that zero injuries is rather an abstract concept, more like the ”pot of gold at the end of the rainbow” - a desirable but unlikely outcome.
Having a Zero injury goal by itself it is not conducive to better safety performance and in the absence of a substantial leading indicators program it is likely conducive to a poor safety climate that is inviting the hiding and under-reporting of accidents, which can hinder organizational learning and continuous improvement.