Individual, Group and Population Risk

Individual, Group and Population Risk Sharing Risk Risk may be distributed and shared amongst, or shared between a number of persons. This chapter discusses understanding the way risks may be different if you are an individual person, a small group or a whole population. For a quick example, consider an explosives disposal unit carrying out their 20 or so defuse operations per year. In the UK, these teams operate around the coastline dealing with washed-up unexploded or dangerous-looking devices. The single lead person who initially has to approach the explosives is taking the whole risk for the group – but if the team take it in turns to defuse each bomb, then they are sharing the personal risk. However, the whole group risk remains the same. Historical Incident A man has been discharged from hospital after being seriously injured while testing a rifle-fired grenade at a military range in West Wales. It is understood the employee was injured in his arm and abdomen by flying shrapnel at the Pendine Ranges in Carmarthenshire. The incident, which happened at 1345 BST on Tuesday 15th April 2003, is being investigated by the Ministry of Defence (MoD), the Health and Safety Executive and Dyfed-Powys Police[BBC 2003]. Individual Risk As we have seen, rates of incidents (targets or actual data) are often expressed as the likelihood of a particular outcome per ‘x’ number of events, even if those events are just days of operation i.e. in the time base. When a rate is expressed as the likelihood of a fatality per 100,000 worker days of labour, this has a number of meanings. For example, if you had 10 people working for 10,000 days (a 45 year working life) in a particular occupation where the historical fatality rate was one fatality per 100,000 worker days, this rate would be correct if one of the ten was killed during his working life, his death being caused by something at work (not an event away from work). However, if you had a company of 1,000 staff, it might be said that there would be one fatality every 100 days. This appears to be very shocking and far too high. In both cases the individual rate is still 1 per 100,000 worker days, but the group risk is very different – 1 per 45 years or 1 per 100 days. The data only makes real sense where the risks to the individuals are gained from the historical

data of risks to the whole industry population. Consider the following data from the US Bureau of Labor Statistics [US DoL 2005] and the UK Health and Safety Commission [HSC 2005a], which quote actual recorded data for workplace fatalities in a number of industry types;

Table 10.1 US Industry Fatality Statistics US Industry Number of fatalities

(Y2004) Workplace fatality rate per 100,000 workers (Y2004)

Manufacturing 459 2.8 Construction 1224 11.9 Mining 152 28.3 Government 526 2.5 Agriculture 659 30.1 Transport & Warehouse 829 17.8 Retail Trade 372 2.3 Leisure 245 2.1 Finance 115 1.2 ALL INDUSTRY 4.1

Table 10.2 UK Industry fatality statistics UK Industry Number of fatalities

(FY2004/05) Workplace fatality rate per

100,000 workers (FY2004/05)

Manufacturing 41 1.2 Construction 72 3.5 Extraction and utility 2 1.1 Service industries 63 0.3 Agriculture 42 10.4 ALL INDUSTRY 0.6 NB. As this data is for a 12-month period, the values also represent the fatality rate per worker-year, as the denominator of the equation would be 1. It should also be noted that each country does not necessarily define the industry sectors in the same way, however, the definitions are considered close enough to be useful for general comparison – particularly the all-industry figure. Group Risk Consider any one of these industry sectors, let’s pick on US construction. From the data we can see that there were 1224 fatalities in 2004 at a rate of 11.9 per 100,000 workers over the working year. This is roughly 0.1% or 1x10-3, so you can say that each individual in the US construction sector has a risk of fatality of

0.1% per year. For a large construction company employing a group of workers that numbers 1000 or so, the probability of a fatal accident during the year is getting close to 1, i.e. a certainty. One of the staff at work will be killed by an accident at work over the next year. So the risk exposure for the individual might be considered as fairly low, however, the risk exposure for all the workers exposed to the same risk is much higher. Now let me ask you, what do you think about these numbers? Is a group risk of fatality at 1 really tolerable? Discuss. After some thought, you might consider that these fatality rates are pretty much tolerable, after all nobody has stopped all construction work in these countries – the economic consequences would have been catastrophic (therein lies a classic example of risk vs. benefit in real life). However, legislators in these countries have indeed decided that these fatality rates are not tolerable – targets for reduction have been developed and published. In the UK a document was published in 2000 on the subject of revitalising health and safety in the workplace [Department for Environment, Transport and Regions 2000]. This gave targets for fatality reduction, which might be viewed as something approaching national safety targets for use in a hypothetical national safety case (!):

ƒ Reduce the number of working days lost per 100,000 workers from workrelated injury and ill-health by 30% by 2010

ƒ Reduce the incidence rate of fatal and major injury accidents by 10% by 2010

ƒ Reduce the incidence and rate of cases of work-related ill-health by 20% by

ƒ Achieve half the improvement by 2004. The published statistics on UK health and safety at work inform the measurement of progress against the targets for reducing work-related injuries, ill health and working days lost set in the ‘Revitalising Health and Safety’ strategy. Annual progress reports have been published each Autumn since then. (All these documents are on the HSE website at www.hse.gov.uk/statistics/targets.htm.) HSE statisticians’ latest assessments, at the mid-point of the strategy period, are as follows [HSC 2005b]:

Progress on fatal and major injuries The Revitalising Health and Safety target for 2004/05 is to reduce the incidence rate of fatal and major injury by 5% from 1999/2000. The available sources indicate no clear change since the base year in the rate of fatal and major injury to employees. The target has therefore not been met. Progress on work-related ill health incidence The Revitalising Health and Safety target for 2004/05 is to reduce the incidence rate of work-related ill health by 10% from 1999/2000. The evidence suggests that

incidence has fallen for most major categories of work-related ill health. Overall, the 10% target has probably been achieved. Progress on working days lost The Revitalising Health and Safety target for 2004/05 is to reduce the number of working days lost per worker due to work-related injury and ill health by 15% from 2000-02. There has been a significant fall in working days lost since the base period, possibly enough to meet the 15% target.