You don't 'need' to report any. Do we assume you mean reporting to the HSE? It depends upon a number of factors. Your reporting criteria e.g. RIDDOR definitions Your own internal organisation criteria, if you have any. Non injury accidents that meet the criteria reporting That element of 'luck' that happens in many accidents/incidents which may or may not make an incident reportable. I don't believe its a simple as saying there is a 10x ratio etc

Hi Fisher quoted ratios for the top three incidents does anyone know the figure for non-reporting? As this will be the real problem area after tackling those that have been reported. Badger

The idea behind the theory is that by focussing appropriate resources on and reducing the number of non-injury events you will reduce the likelihhod of more serious events. Only the likelihood - in life there are no guarantees. Various models measured in various industries, but the theory and required intervention principles are the same.

The Heinrich (Bird) pyramid is allegorical and not based on any scientific evidence. The purpose of which is to make people aware of the link between different types of accidents and incidents. Hence the accident/near miss ratios are not to be taken literally.

Fisher BST (formerly Behavioral Safety Technology Inc) estimate that only 20% of minor injury accidents are precursors of a Serious Injury or Fatality (SIF) and consider that the causes and correlates of minor injury accidents are often different to those for SIFs. Heinrich is unhelpful in terms of managing risks associated with the low probability high consequence events as BP and others have found repeatedly. Ditto many occupational health risks, particularly those associated with chronic conditions. Heinrich defaults to the nearest "proximate" cause which is usually human error. However, two large scale studies in Pennsylania in 1953 and 1960 emphasise the need to look for unsafe conditions. Each looked at over 90,000 accidents (i.e. we are looking at numbers sufficiently large to start doing trend analysis) 1953 Fatals - Unsafe Acts 92%, Unsafe Conditions 94% Non-Fatals - Unsafe Acts 97%, Unsafe Conditions 93%

when lecturing on safety statistics I used to use the figures reported on accidents on the Paris ring road, "périphérique) for a typical year : * Number of accidents resulting in one or more fatalities - 12 (fatals) * Number of accidents resulting in serious injuries - 256 (lost time injuries) * Number of no injury accidents but vehicle seriously damaged - approx 4000 (first aid)(to the vehicle) I would then ask participants to estimate how many violations of road traffic laws had occurred in the same period. (dangerous behaviours/near misses) What's your guess ? And I do agree that Heinrich/Bird should be taken with a pinch of salt. Or a large glass of champagne. Merv

Westonphil is getting towards the other triangle; first (I think) proposed by Dom Cooper : Three factors and their interrelationships strongly influence the safety culture, and thus the results, of an organisation ; the technology - machines, installations etc, Management - attitudes and practices and the people - thinking and behaviour. Lay those three points out on a triangle and consider their interrelationships. Management factors influence the technology in use and the thinking and behaviours of people. People, employees and others, influence management thinking and the quality of the technology (do we maintain it correctly, do we safeguard safety equipment). The quality and appropriateness of the technology influences both people's and management's behaviour and attitudes ; "thats a crap machine. Next time it won't go, give it a clout", vs "there's too much downtime in your area. Do something !" I simplify. As usual. So, improving just one, or even two of those factors wont get you where you want to be. Train and motivate the employees but not the management ? You lose. Supply the best technology available without training and motivating employees or management ? You lose. Put every member of management and supervision through an IOSH course without giving equivalent training to employees, or continuing to use crap equipment ? You lose. Go round the circle (or triangle) I have found that improving the technology is the hardest and longest thing to do (ever heard of budget constraints ?) Take it gently, one step forward at a time. Train, motivate the management, then the employees. Meanwhile dropping hints about the technology, pushing gently (unless you have a "stop working" order from the HSE) for improvements in technology. Then have another go at management. And so on. Evolution, not revolution. My brain hurts. I'm going out for a choucrout and a bottle of geverztraminer. Are you with me ? Merv

merv, "Train and motivate the employees but not the management?" I understand your point of course, but for me all people employed by the company are employees and all employees should receive the appropriate training for their responsibilities. So when I speak about employee training I mean all employees and with regards to ensuring they are competent for their respective roles. However, as I said I understand your point and so am just explaining how I see things. I follow the points in your post and agree with what you are saying and so the differences between us will just be semantics. I understand the points about budget constraints and what we should do first, second etc. I think however that each company/business has some points in common with others and some points which are quite different and of course the amount of common and different will vary from firm to firm and industry type to industry type and so on. So with that in mind I do not think there is any one size fits all and no magic solution which fits all, be that a triangle, technology, training course, advisor, etc., and so each advisor must assess the hazards and evaluate the risks and then implement the appropriate precautions based on what they find, and of course taking into account the resources available. The problem is that when we discuss things in forums we cannot cover everything! :-) OK that was in reply to your points, thanks. This next bit is just me rattling on some additional points for discussion in relation to the topic in general. These are just points to maybe bring out some differing views. :-) If we start talking about near misses then let's say someone is not correctly trained for the job, then often when the person does that job it's a near miss. Let's say the management are not training the shop floor workers properly then when the management allow the persons not to be trained properly and do their jobs then again often it's a near miss. Let's say a fork truck drives past me and its 2 metres away and that is not a near miss and then it is 2mm away and it's a near miss, then at what distance does it turn from non near miss to near miss I ask myself. I could have umpteen near misses in a day and never really know because I was never fully aware of the danger. Let's say a lesser experienced engineer went into a live electrical panel to check a voltage for fault finding and a more experienced engineer did the same job. One person may argue that really the lesser experienced engineer was not quite competent enough for the job and it was a near miss whereas the engineer themselves did not really see it as such and so did not report it. How competent exactly does a person need to be for each and every job I ask myself and do we actually know that each and every person is competent for each and every job they do each and every day? When people carry out manual handling how many times do they use exactly the correct technique and how many times do people not use the correct technique, because every time they get it wrong then surely that is yet another near miss. Near miss reporting is in many cases a incorrect because no one can clearly define what a near miss is or else properly measure it. No company knows how many near misses they actually have. Let's take CE marking and the machinery directive, how many people have actually checked their machinery/plant against the machinery directive and found the provided manuals and training to be short of what is required for the machinery directive? Is that another near miss, each time the engineer works on the equipment and does not have the correct information? Should that engineer identify that they do not quite have the correct information and stick a near miss form in or should we just implement what is set out in PUWER and in which case there is no near miss? I go with the approach of raising H&S awareness across the board and then carrying out proper risk assessments and then implementing proportionate precautions in a priority which relates to the risk level. If that identifies design issues as the highest priority then so be it and if it identifies training issues then so be it and if it identifies management issues then so be it and if it identifies maintenance issues so be it. I think the Heinrich triangle can be used as a communication tool in order to sell a message but I will go with the risk assessment methodology in order to decide where to target resources. I tend to think we all follow more or less the correct approach because we all implement what is required for the business and conditions we find, albeit we may all talk about things differently. :-) Regards.