Home  |  Sitemap  |  Contact  |  Staff Mail webmaster for comment
Home > Services > SPEED

IRAS - Speed

logo University Utrecht logo Shell logo University of North Carolina
(1) Utrecht University, Institute for Risk Assessment Sciences, Environmental and Occupational Health Division, Utrecht, The Netherlands (2) Shell Nederland Raffinaderij B.V. / Shell Nederland Chemie B.V, Pernis, The Netherlands (3) University of North Carolina, Department of Environmental Sciences and Engineering, Chapel Hill, North Carolina, USA

Roel Vermeulen1, Hans Kromhout1, Wim Braun1,
Paul Rocchi2, Daphnis Breederode2,
S.M. Rappaport3, Rogelio Tornero-Velez3

Statistical Program for the Evaluation of Exposure Data

SPEED is a user-friendly computer based statistical Program for the Statistical Evaluation of Exposure Data and is based on the procedures described by Rappaport et. al. (1995) and Lyles et al. (1997). It is a generalization of the existing statistical methodology for assessing occupational exposures. The approach relies upon an intuitively reasonable model for shift-long exposures, and requires repeated measurements on at least some members of a random sample of workers from a job group. Furthermore, the programme is capable of storing and evaluating prospectively collected exposure data. As such, this programme can play an important role in health surveillance systems.

SPEED is as simple to use as Windows. Click on buttons or check boxes to perform special functions in windows and dialog boxes that are displayed. It can be used as a stand-alone application and in combination with an exposure database.

SPEED is built as an Microsoft Excel 2000 / XP / 2003 (UK-version) application. Therefore, it can only be used in combination with Microsoft Office 2000 / XP / 2003 (UK-version).

The program including a detailed manual can be purchased from University Utrecht through the contacts mentioned below. The costs of a single license agreement is €215,-- (Approx. $ 190,-) excl. VAT.

For further information or purchase of the program please contact us.

Hans Kromhout
Department of Environmental Sciences,
Environmental and Occupational Health Division
IRAS
PO Box 80176
3508 TD Utrecht
The Netherlands


Introduction

One of the primary tasks of an occupational hygienist is the evaluation of chemical exposures at the workplace. Several methods could be applied to evaluate the exposure ranging from estimating the exposure (qualitative assessment) to actual exposure measurements (quantitative assessment). Quantitative exposure assessment can be used for several objectives:

Several methods for determining whether exposures are acceptable relative to the Occupational Exposure Limits (OELs) are used. When using compliance testing every individual measurement value is compared with the OEL for the agent of interest (Rappaport, 1991). In this case the exposure will be declared acceptable only when all measurements are below the OEL. Unfortunately, this method discourages monitoring, as the probability that a single measurement will exceed the OEL depends strongly on the sample size (Rappaport, 1984). Furthermore compliance testing encourages the use of biased measurement strategies, like monitoring worst-case scenarios,resulting in data which are inappropriate for other purposes. Another often-applied method is the test on 'Exceedance'. This approach focuses on the probability that a worker might be exposed to concentrations above the OEL during a single work shift. These methods take into account the day-to-day variability of exposure and assume that the measured concentrations are part of a lognormal distribution function. However these methods ignore the fact that exposures can vary considerably between-workers within homogeneous exposure groups (Rappaport, 1993; Kromhout, 1993). Therefore, it is possible that in situations with high between worker variability some workers will have exposures which exceed the OEL frequently even though the exceedance is low for the qroup as a whole.

Since we are interested in securing the health of every individual worker, we developed a method which uses the observational (homogeneous) group approach but which also recognizes that exposure varies both within and between workers. In order to focus attention upon individual workers, we define 'overexposure' relative to the likelihood that a randomly selected worker's true mean exposure would exceed the OEL. Thus, our approach is only suitable for situations where we seek to evaluate and control long-term exposures which can give rise to chronic health effects. That is, it would be inappropriate to employ this approach to evaluate and control short-term exposures or scenarios where acute effects are likely. Furthermore the method suggests appropriate interventions when exposures are unacceptable.

The Statistical Program for the Evaluation of Exposure Data (SPEED) is based on the strategy described by Rappaport et al. (1995) and Lyles et al. (1997). The approach relies upon an intuitively reasonable model for shift-long exposures, and requires repeated measurements on at least some members of a random sample of workers from a job group. It is suggested that at least 20 shift long measurements be randomly collected from an observational (homogeneous) group for preliminary analysis.


Overexposure strategy

The overexposure strategy consists out of five levels of decisions arranged in a sequential structure.

Figure 1
Figure 1: Overexposure strategy for assessing long-term exposures to chemicals

In the first level the data is checked for log normality by applying a random effects ANOVA model. A graphical procedure based upon standardized estimates of the random (worker) effects is used to assess the fit of the model. If the fit of the model is considered acceptable, the observational group is regarded as a monomorphic group for subsequent testing. At the second level the statistical parameters describing the within- and between-worker (lognormal) distributions of the monomorphic group are used to determine whether the probability of overexposure is acceptably small. The three remaining levels are specific to observational groups for which the exposure is thought to be unacceptable. At level three, resampling is considered as an option to increase the power of the statistical test performed at level two. At level four another ad hoc test is used to determined whether the group is uniformly exposed, i.e., whether the worker-specific mean exposure levels are within a narrow range, At level five control options are considered. If the group is uniformly exposed then control is focussed upon engineering or administrative controls and, if not, individual personal environments (e.g., tasks, practices, locations) are investigated.

Figure 2
Figure 2: Flow chart Statistical Program for the Evaluation of Exposure Data (SPEED)

Literature Cited

Rappaport,S.M.
Assessment of long-term exposures to genotoxic and carcinogenic agents. Int. Archs. Occup. Environ. Hlth. 65,S29-S35
Rappaport, S.M. (1984)
The rules of the game: An analysis of OSHA's enforcement strategy. Am J. Ind. Med.6;291-303.
Rappaport S.M., Kromhout H., Symanski E. (1993)
Variation of expsoure between workers in homogeneous exposure groups. Am. Ind. Hug. Ass. J. 54;654-662.
Kromhout H, Symanski E, Rappaport S.M.
A comprehensive evaluation of within- and between-worker components of occupational exposure estimates. Am. J. Ind. Med. 12, 551-562
Rappaport, S.M. (1995)
An exposure-assessment strategy accounting for within- and between-worker sources of variability. Ann. Occup. Hyg. 39, 469-495.
Lyles R.H., Kupper L.L., Rappaport S.
A lognormal distribution-based exposure assessment method for unbalanced data. Ann. Occup. Hyg. 41;63-76.
Searle, S.R., Casella, G. and McCulloch, C.E. (1992)
Variance components. John Wiley & Sons, New York.
Dempster, A.P., and Ryan, L.M. (1985)
Weihted normal plots. J. Am. Stat. Ass. 80;845-850
Lange, N. and Ryan, L.M. (1989)
Assessing normality in random effects models. Ann. Stat. 17;624-642.