01 May 2008
Duncan Spencer

If it's a hazard, you should have a risk assessment covering it. Is that really true? A hazard is some "thing" with the potential to cause harm, so there are literally hundreds if not thousands in each workplace. But where do you start and, more importantly, when do you finish assessing the risk they pose?

Without such rules you end up with countless risk assessments, each requiring management action, monitoring and review; an impossible task.

The HSE's sensible risk management campaign in 2007 tried to draw us away from risk assessing everything. But while safety practitioners do their best to rationalise and reduce the numbers of assessments, there's always someone - a well-meaning employee, a safety representative, an enforcement officer - who insists another hazard should make the list. So how do you find a way through this paradox?

If you can't demonstrate what should be risk assessed and, more importantly, what shouldn't, then it's difficult to argue against someone else's demand to add entries to your list. The soundest approach is founded in understanding activity, rather than just listing hazards.

Does it matter?

In deciding where to stand on your judgement, you can look to legal arguments of reasonableness and significance for help. Reasonableness is a very important legal concept in safety, but sadly not always well taught. It's about benchmarking the individual who has to make the judgement with other similar people, but what do you compare?

The spirit and intent of safety legislation is to require organisations to have management systems that enable them to predict safety problems (that is, risk) and choose preventative measures. More precisely, it's managers who instruct employees to carry out work activities, and should therefore ensure they are competent to do so and have safe equipment, environments and systems of work. How far you should go in identifying problems and controlling them is down to reasonable judgement. Assessing whether someone is capable of reasonable judgement is down to the tests listed below.

Is this reasonable?

Before deciding whether someone is capable of reasonable assessment of the risk posed by a given hazard judgement, you should consider five factors:

• training and/or qualifications - which can be checked by looking at training records
• experience - which can be verified by looking at the person's curriculum vitae and is arguably the most important aspect of reasonableness
• knowledge - checked by asking the individual technical questions about their field of expertise
• due skill - skills appropriate to the role; a manager's ability to lead and communicate, for instance
• knowing their limits - how far they recognise the boundaries of their competency and when to ask more knowledgeable people for advice.

Reasonable judgement also has to be exercised in the context of the
organisation's operation and industry. Do we really think that office managers should be worried about a minor paper cut suffered when filing? (I hope you answered "no" there.) But what about pathology technicians who might get a paper cut and then handle diseased human tissue? Different contexts demand different control responses - a very important fact to grasp before arguing the toss over an enforcement officer's recommendation.

The Approved Code of Practice (ACoP) for the Management of Health and Safety at Work Regulations (L 21, available from www.hsebooks.com at £8) provides more guidance. In the explanation of how to interpret Regulation 3 (the requirement to risk assess), it says employers should assess risk and "should record the significant findings of that risk assessment".

This implies that if you carry out a mental risk assessment and there aren't any significant findings, then a risk assessment record isn't required. When it explains the phrase "suitable and sufficient", the ACoP states more openly that "insignificant risks can usually be ignored". This is great news. But the implication is that to avoid swamping ourselves in the needless documentation of trivial risk assessments, we need a system for defining what is significant and what is not. So we apply reasonable judgement, in the context of our organisation's operation, to decide what is significant and should be a concern, and what should not.

Floor-walking flaw

Try wandering around the organisation with a clipboard to identify hazards at 10am on a Monday morning. How certain are you that you have captured everything you need to know? In the main you will probably capture physical hazards (such as holes in the floor) and may miss transitory ones (reversing vehicles, water being walked through the warehouse when vehicles are washed down at the end of the day). You won't see shift differences, or identify significant seasonal variation.

Activity-led risk identification cuts through these limitations. If you identify activities, you are less likely to miss things or focus on things that don't matter.

Only by carrying out an activity will people come into contact with a hazard. If you never send people to work on the roof, why worry about people falling off it? As for likelihood, it's understanding the activity and how well it's controlled that enables the risk assessor to decide if the predicted accident is more or less likely to occur. In my experience, the added benefit of being activity-orientated is the way it engages managers. Talk to a manager about a hazard and they glaze over; it's safety technical stuff. Talk to them about the activities their people do and they make an immediate connection.
What do you do?

So the first step in the identification process is to identify the activities. What are the scary things that your workers do? At Waitrose we start by listing activities under four headings.

• Location: what activities do we do in the yard, for instance? (Reverse vehicles, move cardboard bales to storage.)
• People: what does the security guard or checkout person do? (Apprehend and search a suspected shoplifter, scan goods through a checkout.)
• Equipment: how do we use that baling machine; what do we do with it?
• Process: what is the sequence of events when taking delivery of stock and what are the activities at each stage? (Cages are offloaded onto a scissor lift, the lift descends, roll-cages are moved off and placed to one side, and so on.)

This approach gives a systematic view of what we do, where we do it, what we do it with and as part of what operational activity. Such a systematic approach makes it very unlikely that risky activity is missed. This brainstorming session produces a long list. But we then apply a three-stage filter.
- First: if the activity isn't significant, it doesn't even make the activity list (putting on a high-visibility jacket or securing a flap on a cage, for instance).
- Second: it may be an operationally significant activity, but the risk may be trivial. So it makes the activity list to show we thought of it, but it doesn't get assessed (use of a ticketing machine or barcode reader, for example).
- Third: activity titles can be rewritten to merge related activities and further reduce risk assessment numbers - so preparing chickens for cooking, placing them on rotisserie rods, loading and unloading ovens, becomes simply "preparing and cooking chickens" and would encompass all of the above.

Identifying activity titles begins the process. Next, we recognise that activities may have more than one hazard associated with them. At Waitrose, our hazard explanation shows we understand all the hazards associated with the activity. For the task "moving around café tables to clear away dishes and wipe down", we specify:
"While employees are moving around the café they can bump into fixtures, fittings and furniture. However, furniture legs and discarded bags provide a trip hazard, or spilt fluid a slip hazard. If the employee fell in this environment they would instinctively place out their hands to break their fall, potentially causing a fractured wrist or collar bone."

Tying it up

If we use our reasonable judgement to identify significant activities and which of those has significant risks associated with them, we achieve a focus. At Waitrose, by using these techniques, we have been able to reduce the number of recorded risk assessments for our large branches from 720 assessments (born of a hazard-orientated approach) to only 85. When these are shared between the different section managers (six to 10 each) it all becomes manageable.

If we are asked to list which risk assessments we should have, we now can. If we were asked which risk assessments were missing - we could answer that too. What's more, we have a system and process to back it all up. A system that can be reviewed and updated as our operational circumstances evolve.

So what do you do when someone suggests you have missed a hazard, or that something is unsafe? Ask first what the activity is, then check if it's on your activity list. If it's not, then consider its significance. If it's significant then add it, if not then discard it. But using these techniques, it's highly unlikely they will catch you out.

Duncan Spencer CFIOSH MIIRSM is safety manager for the Waitrose Division of the John Lewis Partnership.

Categories:
Risk assessment, Risk assessment, Safety, Article