Fashion & Beauty

A statistical study on temporary work and occupational accidents: Specific risk factors and risk management strategies

DICCA Research Group
of 10
All materials on our website are shared by users. If you have any questions about copyright issues, please report us to resolve them. We are always happy to assist you.
Related Documents
  A statistical study on temporary workand occupational accidents: Specific risk factorsand risk management strategies Bruno Fabiano * , Fabio Curro`, Andrea P. Reverberi, Renato Pastorino Chemical and Process Engineering Department ‘‘G.B. Bonino ” , University of Genoa, via Opera Pia 15, 16145 Genoa, Italy Received 23 March 2006; received in revised form 26 April 2007; accepted 4 May 2007 Abstract Temporary work, supplied by temporary-help agencies and sometimes referred to as ‘‘job in leasing ” , was only recentlyintroduced in Italy, and has since spread considerably thanks to its flexibility and cost effectiveness. In this study, trends inthe rates of occupational injuries in different sectors of Italian industries in the period 2000–2004 are explored, contrastingdirect employment and temporary work. Data on occupational injuries from the National Organization for Labour InjuryInsurance as well as data directly obtained through a field survey in three large manufacturing firms were analysed to high-light the interaction between injury frequency index (FI) and the characteristics of the labour force. FI for temporaryworkers ranged between 89.23 and 94.10, i.e., between 136.4% and 175.2% more than the value found for direct employeesin the most hazardous industrial sector. Also accident severity (assessed on the basis of time lost due to injuries) is twice theoverall value of the severity index. The results from the field survey confirmed the trend: FI for direct employees rangesfrom 25.7 to 45.0 in respect of total hours of work in the range 1.99  10 6  –2.40  10 6 ; whereas FI in temporary workersranges from 86.2 to 163.5 in respect of total hours of work ranging from 1.25  10 5 to 1.28  10 5 . Given this evidence, anempirical statistical analysis, based on responses from injured temporary workers was carried out. The questionnaire wasintended to collect data on a variety of control variables relating to personal characteristics of respondents. The analysis of the questionnaire data using ANOVA and response surface methodology highlighted the interaction between short dura-tion of work assignments, inadequate training period and FI increase. Reasons can be traced to lack of experience in theactivity, insufficient specific knowledge (formal and informal knowledge) about a particular installation and to inadequatetraining period. Means of promoting safety include diffusion of information on regulatory provisions, management train-ing in safety, enhancement of specific training and formation.   2007 Elsevier Ltd. All rights reserved. Keywords:  Frequency index; Injury; Questionnaire; Response surface methodology; Temporary work; Training 0925-7535/$ - see front matter    2007 Elsevier Ltd. All rights reserved.doi:10.1016/j.ssci.2007.05.004 * Corresponding author. Tel.: +39 0103532585. E-mail address: (B. Fabiano).  Available online at Safety Science 46 (2008) 535–544  1. Introduction On a national level, Italy has enjoyed declining trends in the number and rates of occupational fatalities andinjuries for more than three decades. While these figures are encouraging, far too many preventable injuriesare still occurring.The factors influencing accident frequency can be divided into the following categories:   technical factors: low automation, multi-product industries, discontinuous operating cycles, non standard-ized production affect safety negatively, since they require a greater interaction between man and devices.On the other hand, a reduction in individual exposure to severe hazards was reported where mechanizedprocesses and equipment were introduced in the mining (Asogwa, 1988) and the logging industry(Laflamme and Cloutier, 1988);   economical factors, e.g., general economic situation (Saari, 1982), unemployment rate, labour and social-insurance legislation, (Blank et al., 1996);   labour organization, e.g., management system and performance, work practice, oversight, communicationstructure, etc.;   environmental conditions: about half the accidents in Italy are related to labour environment (Fabianoet al., 2001) and they could be prevented by rather simple lay-out and protection measures, which howeverprove extremely difficult or even unfeasible in small concerns, because of operating, economic and/or spaceconstraints (Fabiano et al., 2004);   human factors, both individual and inter-individual, e.g., workload, experience and training, competence,fatigue, etc.Current market conditions often make it necessary to resort to outsourcing to remain competitive, partic-ularly by employing external and precarious human resources. In fact, in the last 20 years there has been asignificant growth in labour governed by casual, part-time, subcontract or franchised arrangements, virtuallyin all OECD countries (less so in Italy, Canada and Luxenbourg) (Ferrie et al., 1999). Results of internationalresearch seem to point out that precarious labour is associated with increased fatalities, occupational injuries,illness, in various industry sectors across a number of countries (Kochan et al., 1994; Morris, 1999; Mayhewand Quinlan, 1999). Temporary work (i.e., a job arrangement whereby workers are hired out to firms by Tem-porary Work Agencies) is a sort of labour outsourcing and precarious work. This kind of work arrangement,sometimes referred to as ‘‘job in leasing ” , has only recently become widespread in Italy. Due to the flexibilityand cost effectiveness of this contract, in the last few years Italy has experienced constant growing rates of temporary work. The whole number of contracts provided by temporary work agencies soared from194,835 in 1999 to 744,438 in 2003, corresponding to an average increase of about 56.4% per year. Virtually,every industrial sector shared this trend, with an average daily percentage of temporary workers correspond-ing to 1% of ‘‘in-house employees ” , for all industrial sectors in the year 2003. Process, metal–mechanic andtransport sectors displayed the highest average daily percentage of temporary workers, amounting to about3% of direct employees. In the same year, the average EC daily incidence of temporary work was 1.5%, withhigher values in the UK (3.2%) and in the Netherlands (4.5%), while the daily incidence in the United Stateswas 2.5%. Despite the international diffusion of temporary work, in the open literature there is little informa-tion on possible safety implications and contributing factors of occupational injuries. Moreover, even if mostsafety-related regulations and safety management systems already incorporate provisions for the managementof contractors, a striking number of injuries occurs among ‘‘non in-house employees ” . In Spain, 60% of occu-pational injuries occurred to temporary workers, with a rate nearly three times that of those in permanent jobs(Artiles and Alos-Moner, 1999). A survey conducted among three manufacturing industries of the plastic sec-tor in the USA, showed that the trend of accident rate over the years 1993–1998 does not appear to be a steadyone, either for direct employees, or for temporary workers. However, the evidence obtained from suchresearch indicates that the injury rate for temporary workers is constantly higher (from 2.6 to 3.8 times) thanthe one recorded for permanent employees (Morris, 1999).This paper offers a perspective on the trends in the rates of occupational injuries in the main sectors of Italian industries, during the period 2000–2004, contrasting direct employment and temporary work and 536  B. Fabiano et al./Safety Science 46 (2008) 535–544  focusing attention on the human factor. Generally speaking, the different reasons for people failure can begrouped as follows: lack of training or instruction; lack of motivation; lack of physical or mental ability; slipsand lapses of attention (Kletz, 1993). On this basis, an investigation into the relationship between occupationalaccidents and temporary work is performed, adopting a questionnaire survey, for the definition of peculiarrisk factors and for setting priorities to improve safety standards in this context. 2. Materials and methods  2.1. Research design Raw data were obtained from the National Organization for Labour Injury Insurance (I.N.A.I.L. IstitutoNazionale per l’Assicurazione contro gli Infortuni sul Lavoro) (INAIL, 1970–2004). One of the main measuresof performance is the lost time accident (LTA), which however has only limited value and should be supple-mented by other measurements, such as the total accident rate, the cost of the damage caused by accidents andother dangerous incidents and, if possible, a numerical measure of the results of plant audits (Kletz, 1993). Thefrequency index (FI), allows to find out the incidence of accidents of workers exposed to risk and is defined asfollows:FI ¼ Number of total accidentsNumber of worked hours 10 6 :  ð 1 Þ As ‘‘worked hours ”  can be calculated regularly, this index may offer a proper instrument for evaluation of accidents.To estimate trends, we modelled the rate of injury frequency index as a function of the year, by adoptinga multiple regression model with a log transformed rate (Bailer et al., 1997; Fabiano et al., 2001), asfollowslnFI ¼ a þ b ð year i  year 0 Þ ;  ð 2 Þ where year 0  is the first year of the study, to which corresponds the baseline of injury rate. Data on temporaryworkers were collected directly from the 16 leading temporary-job agencies in Italy, covering nearly 90% of thewhole market.Data from INAIL and from direct field survey (temporary job agencies) were combined with data from theState Statistics Institute, in order to highlight the interaction between the trend of the injury frequency indexand the characteristics of the labour force, throughout the years from the introduction of the new workarrangements. The critical issue is whether there is a statistical relation between occupational injuries and tem-porary work, and whether or not such relation differs for different occupation groups. In order to confirm theimpact of new working regulations on the occurrence of occupational injuries, a detailed case study was madeon three large firms of the manufacturing sector, characterized by high proportion of temporary work. Finally,an empirical statistical analysis was carried out, based on responses by injured temporary workers to an‘‘ad-hoc ”  questionnaire specially designed for the purpose.  2.2. Questionnaire development The questionnaire was intended to collect data on a variety of control variables relating to personal char-acteristics of respondents. The main section of the questionnaire was designed to characterize technical, orga-nizational and individual factors of suffered injuries, as well as to find out how firms try to reduce theoccupational risk to which temporary employees may be exposed.The content of the questionnaire is described below. Demographic and work details : respondents were asked to give information on their sex, age, job title,employer, extent of their job experience and specific work task, and task rotation if any. Safety attitudes : respondents were asked to rate their job and safety training, extent of the temporary con-tract, whether or not they were accompanied by a job tutor, how much support they received from fellowworkers on the job. B. Fabiano et al./Safety Science 46 (2008) 535–544  537  Accident and near misses : respondents were asked to describe the accident and its severity based on lost timeaccident (LTA), time elapsed from starting of the contract until the accident, perceived accident cause (distrac-tion, work pressure, difficulty in job performing, lack of specific/safety training, insufficient preventive mea-sures in plant/machine/equipment, inadequate personal protective equipment, fatalistic event). Data fromthe structured questionnaire were coded and entered into a database for subsequent univariate analysis of var-iance (ANOVA). The independent variables whose effects on number of injuries and severity were evaluatedincluded: worker’s age, job position, training period, on-site experience, temporary contract duration, per-ceived accident cause. Significant results were analysed adopting response surface methodology (RSM), thislatter being a collection of mathematical and statistical techniques widely adopted to optimise different pro-cesses (e.g., Dasu and Panda, 2000), used in accident analysis for the first time, at least to our knowledge.  2.3. Sample The study population was identified from 16 staffing companies in Milan, Italy, covering nearly 90% of thewhole market of temporary work. The questionnaire was distributed to 700 injured workers resulting from theinvestigation. All ‘‘ in itinere ”  accidents (accidents occurred on the way to or from the workplace) wereexcluded from the sample. Respondents were informed that the interview data were to be treated as anony-mous and that only aggregated results would be produced.The response rate for this distribution was 43.3% ( n  = 303). The average age of all respondents was 29.2years (SD = 8.23) in the range from 18 to 45 years. 3. Results and discussion 3.1. Injury trends As shown in Fig. 1, the overall value of FI on a national level decreased, in the last three decades, from94.92 in 1970 to 22.54 in 2000, with a mean reduction per year corresponding to 5.48%, obtained from thefit of the regression model (Eq. (2)). This encouraging trend was common to every industrial sector, with somedifferences depending on sector. It must be pointed out that raw data from INAIL obviously do not includeinjuries in case of irregular or unregistered employment (e.g. moonlighting phenomena), as well as accidents Fig. 1. Overall injury frequency index (FI) over the period 1970–2000 and regression model with prediction intervals (95%).538  B. Fabiano et al./Safety Science 46 (2008) 535–544  involving less than three days absence from work, since, in Italy, they need not be reported. However, theselimitations concern not only the Italian reality, but generally the worldwide safety community; in fact, theproblem of improving the organizational memory and the need of a new approach to the collection and ana-lysis of accident data are becoming more and more important (Kletz, 1993). Table 1 shows the values of the national frequency index and mean values over five years, for different industrial sectors (in-house workers)and for temporary workers, over the period 2000–2004, i.e., since the introduction in Italy of temporary-workarrangements.Again, a general decline may be observed in the frequency of injuries, for direct employees, across all indus-tries, from 2000 until 2004, at an average estimated rate of decline of 4.29% per year, from a baseline of 22.54in 2000. Practically speaking, all sectors are reaching minimum values of the frequency index, which can beregarded as a statistically reinforced reference for the minimum risk levels per individual employed in the dif-ferent industrial activities. Obviously, this is only a statistical consideration, not to be regarded as an attemptto renounce the effort of reducing even the lowest injury frequency recorded. The trend for temporary work isnot decreasing, but shows some variations from year to year, mainly because of the different percentage dis-tribution of temporary workers in the various industrial sectors. However, it should be pointed out that the FIvalue for temporary workers is 2.65 times the mean value for the highest risk sector at national level (Buildingsector). It is worth noting that the average FI value of temporary workers over the time span considered iscomparable to the value of the national frequency index for direct employees recorded in 1971, i.e., 91.83(see Fig. 1).In order to offer a better insight on this observation, Table 2 shows FI values calculated for the Lombardiaregion, where nearly 63% of temporary workers were employed in the years considered.FI values in Lombardia for direct employees are constantly lower than the corresponding values at nationallevel, with mean differences ranging from   5.35% in the food sector to   20.68% in the process sector. Con-sidering this difference in the reference region, the injury trend declines constantly by 3.76% per year and FI of temporary workers is by far higher than FI of the most hazardous sector, to a percentage of 410.9%. Thesefindings are in good agreement with the few studies available in the international literature (Artiles andAlos-Moner, 1999). A case-study conducted over five years in a process industry in India concluded that acci-dent incidence, accident frequency and severity rates were significantly higher for temporary workers than forpermanent workers, although the two groups were similar in relation to age, level of education, habits andnature of work. Relative risk varied from 2.3 to 18.0 in case of time loss accidents and from 1.1 to 2.6 in caseof no time-loss accidents (Saha et al., 2004). Higher accident risk of temporary workers might have been dueto less effective experience and to lack of job security (Saha et al., 2005). The results presented here underlinethat temporary work may be considered a risk factor. Reasons are to be traced to its intrinsic characteristics:age of workers, lack of experience in the activity, insufficient specific knowledge, low level of experience in thesame firm, lack of a sufficient training period and heavy workload.In order to confirm the impact of innovative employment arrangements on the occurrence of occupationalinjuries, a detailed case study was carried out on three large firms of the manufacturing sector, characterizedby a high proportion of temporary work. The three firms belong to the metal and mechanic sector and theiremployees have similar characteristics: number of hours worked, percentage of temporary workers, position Table 1Injury frequency index (FI) calculated, on a national level, for direct workers in the different industrial sectors and for temporary workersMean Year2004 2003 2002 2001 2000Metal and mechanic 34.07 ± 3.07 30.09 32.08 34.15 36.52 37.51Process 19.39 ± 1.69 17.25 18.11 19.60 21.02 20.98Food 22.61 ± 1.49 20.62 21.90 22.60 23.38 24.56Services 20.18 ± 1.51 18.76 18.93 19.87 20.91 22.41Building 34.77 ± 2.78 32.02 32.76 34.19 35.93 38.95All sectors 20.76 ± 1.42 18.98 19.70 20.87 21.70 22.54Temporary work 91.63 ± 2.22 89.42 89.23 94.10 93.30 92.10 B. Fabiano et al./Safety Science 46 (2008) 535–544  539
Similar documents
View more...
Related Search
We Need Your Support
Thank you for visiting our website and your interest in our free products and services. We are nonprofit website to share and download documents. To the running of this website, we need your help to support us.

Thanks to everyone for your continued support.

No, Thanks