In practice, companies use lagging indicators like accident rates to record safety incidents after they happen. Leading indicators allow them to keep track of the effectiveness of safety measures aimed at preventing incidents before they happen, with common examples including near-miss reporting, pre-shift meetings, counting the number of training days held or toolbox talks.
There is increasing evidence that lagging indicators are not enough to generate continuous improvements in safety. Andre Freitas, at the University of Manchester’s department of computer science, says leading indicators are proactive because they measure safety initiatives that provide an early indication of potential adverse events, which in turn allows management to initiate corrective steps. There is, however, limited knowledge to guide the implementation of leading indicators in terms of selection and use. As a result, this study conducted a literature review on leading indicators in construction. Freitas, Clara Man Cheung, senior lecturer (associate professor) at the department of mechanical, aerospace & civil engineering at University of Manchester and Jing Xu lecturer in enterprise management at the Bartlett School of Construction and Project Management provide more information on the 16 leading indicators and their descriptions (see box, p17).
Discovering Safety seeks to develop new techniques to analyse data and aggregate data sources, the key output being to help prevent future accidents from occurring. Funded by the Lloyd’s Register Foundation, the initiative also has input from Atkins, Ohalo and Wood.
The adoption of leading indicators was a key focus for FirstGroup’s injury prevention programme, which involved the participation of around 135,000 employees across FirstBus in the UK and US divisions like FirstStudent and FirstTransit and Greyhound.
The programme occurred when Gary Catapano, chief strategy and safety advisor at Magtec Products, worked at FirstGroup. He explains that the initiative allowed frontline staff to stop working at any point where they felt unsafe. This could include a scenario in which a bus driver was handed a set of keys for a bus that did not look safe.
Part of the scheme also consisted of carrying out risk assessments prior to work. “So, if you’re working in the shop and need to lift a bus up, we were asking them to spend time and look at how to do that work safely and creating standards for safe lifting. We had those pre-task risk assessments that were developed in conjunction with our frontline employees.”
Additionally, staff members were given a handbook in which they could note down any hazards they found en route and document safety conversations. Also, supervisors were encouraged to acknowledge the positive safety behaviours of individuals.
NOT BOX TICKING
“We tried to teach managers that this is not about box ticking or finding fault with people, but its more about recognising the good they are doing,” he continues. “We had to teach our managers that we were interested in pinpointing the behaviour and describing it in a way that helps the employee understand that what they did was good.”
Naveed Qamar, former FirstGroup safety director, says the introduction of the system saw an increase in the number of reporting of near misses while targets focused on the number of safety conversations and subsequent actions generated and then eventually closed.
What effect did this have on safety? “Personal injuries went down by around 40% in the first 12 to 18 months, while bus collisions were reduced by 25%. This downward trend continued for three years.”
Qamar – now head of safety, health, environment and quality at Aberdeen Harbour Board – believes that leading indicators must be followed up with supervision so that staff do not fall back into bad habits. “Supervision is important to ensure staff have understood the training. Turning a blind eye to things being done the wrong way is a recipe for something bad to happen – and it usually does.”
This is especially important for ‘Lock out, tag out’, a procedure that ensures dangerous equipment is shut off and not able to be started again prior to the completion of work.
“Lock out, tag out is important because someone could be killed if you don’t do everything properly, and training alone won’t guarantee people are doing it consistently,” says Judy Agnew, senior vice president safety solutions at ADI (Aubrey Daniels International), and a 2021 Sir Moir Lockhead safety leadership award winner. “So, then it’s a matter of measuring how many people were trained and carrying out spot checks to ensure people are following the correct steps and coaching them if they are not.”
Leading indicators also need to be accompanied with accountability, which could translate into making sure a supervisor prioritises follow-up actions that come out of a pre-shift meeting through positive reinforcement and coaching.
However, the biggest challenge for Agnew is getting senior leaders to favour leading indicators over lagging indicators like incident rates. “If you can get that shift to happen, and that accountability piece, then you’ve got it made. It’s also important to be flexible because there are no such thing as perfect measures, as some things won’t work so well, so be willing to adjust.”
Catapano concurs, adding that it is difficult to drive a safety culture change without involvement at the top. “If you don’t have senior leadership buy-in and you’ve got the best leading indicators in the world, that is not going to help go where you need to go.”
APPLYING DATA SCIENCE
For the Discovering Safety study, research lead Steven Naylor says the merits of including leading indicators are two-fold. “First off, they provide an opportunity to detect and resolve health and safety issues before accidents occur. Secondly, they help health and safety functions more directly in deciding on improvements that need to be made as part of efforts to drive continuous improvements in performance.”
After identifying leading indicators, HSE worked with several contractors to determine whether the data collected at different projects like near miss reports, hazard spots, inspections and investigation data could be used to assign measures to these broad topics. That led to a surprise.
“We carried out a deep dive into the data collection practices and we found it was a challenge to generate quantitative metrics around them,” Naylor continues. “The core health and safety data that projects generated for quite a few of these areas didn’t really lend themselves to being able to do it in an effective way.”
The programme is now entering a new phase that will involve the use of BIM (building information modelling) software and 4D models. “We plan to work with contractors to see if it is possible to bring together work schedule data and test that idea of generating metrics around risk control measures and use data science and digital tools to do that.”
A 3D BIM model provides a visualisation of a building, but Discovering Safety technical lead Gordon Crick explains that a 4D model shows the sequence of construction over time. “This turns out to be incredibly powerful for health and safety, because you get to show how temporary state A leads to temporary state B leads to temporary state C, and you can begin to see the problems which are in those transitions,” Crick continues. “In terms of leading indicators, we are talking about is being able to predict, even at the design stage, what the design is telling you about where the risks may be and what the indicators are that you may need to look at.”
For example, it can be difficult for heavy machinery to move around a site at the foundation stage of a building. A 4D model would allow an engineer to define an exclusion zone as a piece of the design model that needs to be monitored. “So, then it becomes a leading indicator derived from the model, and I think a lot of the issues around organisation commitment translate themselves [into] the planning and design of the operation to make sure the uncertainty is reduced.”
Technology aside, Crick recognises that there needs to be a culture shift in which people want to work with leading indicators. “I am certain that the adoption of leading indicators is going to be as much human as it is machine.”
BOX: SWEET 16
Leading indicators from Discovering Safety
- Organisation commitment: client, designer, principal contractor and subcontractor commitment to safety
- Safety auditing: the process of collecting independent information on the safety management system and making plans for preventive actions
- Training and orientation: improving skills, knowledge, attitudes and experiences of employees to manage safety
- Client engagement: is the client is engaged in construction safety throughout a project?
- Designer engagement
- Principal contractor engagement
- Supply chain and workforce engagement
- Safety design: the degree to which preventing accidents during construction is considered as one of the objectives of design
- Plan for safety: is safety in construction considered in the planning process?
- Hazard identification and control: the process and outcome of identifying and controlling hazards and risks in workplace
- Safety learning: learning from accidents, incidents and relevant experiences
- Recognition and reward: mechanisms to motivate the workforce to comply with safety rules and actively participate in safety improvement activities
- Site communication: familiarising operatives with a job, informing them of risks and improving task-specific competence
- Safety climate: employees’ perception of the priority an organisation and workgroup places on safety-related policies, procedures and practices
- Worker involvement: workers’ buy-in when establishing, operating, evaluating and improving safety practices
- Competence: ensuring that employees have the skills, knowledge, attitudes and experience to safely carry out assigned tasks