Dear Editor

We would like to comment on the study by Flodin et al.[1] We believe it has multiple methodological flaws and does not add to the understanding of the relationship of exposure to trace concentrations of waste anesthetic gases and the health of operating room workers.

First, the method of data collection, i.e, by an “appeal” (authors’ word)” in a trade journal for nurses for individuals with multiple sclerosis to identify themselves is bound to result in responder bias. Many of the early epidemiologic surveys examining the relationship of exposure to waste anaesthetic gases and adverse health effects were confounded by responder bias. An example of the problem was reported by Axelsson and Rylander[2] who reviewed all histories in one survey which claimed an increased risk of spontaneous abortion in nurse anaesthetists. Medical records were examined to determine the occurrence of abortion. When data from non- responders were included, there was no difference in adverse effects between the groups.

Another major flaw in the study is that dosage of waste anaesthetic gases to which nurse anaesthetists were exposed was not measured. In fact it is not known if nurse anaesthetists actually were exposed as scavenging of waste anaesthetic gases was implemented in operating rooms world-wide during the period of the study. When scavenging systems are in place, exposure to waste anaesthetic gases is minimal. Indeed, it is possible that some of the nurses worked only in scavenged theaters. For the study to have any relevance, exposure of affected (those with MS) and unaffected nurses must be reported. Also lacking is information regarding the number of air exchanges in the operating room where the nurses worked.

Another problem with the study is there is no reference to possible confounding factors, such as exposure to other environmental pollutants (tobacco smoke, methyl methacrylate from cement in orthopedic operating rooms), radiation, cytotoxic drugs,[3] and other variables thought to affect health, such as shift work and stress. If ill affects occurred, they may not have been due to exposure to trace levels of waste anaesthetic gases.

Finally, the authors do no mention the studies that have reviewed the epidemiologic data examining the purported link between exposure to trace concentrations of waste anaesthetic gases and adverse health effects. Analysis of these data by two independent groups of epidemiologists[4,5] showed that the reported differences in adverse health affects were small and within the range that might be due to bias or uncontrolled variables, i.e, flaws in study design especially data collection and responder bias as may have occurred in the study by Flodin et al.[1] More recently, a mortality study[6] showed that there was no difference between anesthesiologists and internists in death rates from chronic neurological disease. Finally, Spence and Maran[7-10] carried out a prospective study in the United Kingdom collecting health and anaesthetic exposure data from 11,500 female physicians aged 40 or less working in hospital practice. They did not find an increased incidence of neurological problems related to occupation.

In summary, we believe the conclusions in the report by Flodin et al.1 are incorrect because they are based on flawed methodology. Its publication has probably caused unnecessary anxiety in operating room workers. By our letter, we hope to allay these concerns.

Diana G. McGregor, M.B.B.S., FRCA
Department of Anesthesiology
Mayo Clinic
Rochester, MN 55902

Richard I. Mazze, M.D. Jeffrey M. Baden, M.B.B.S., FRCA, MRCP (UK)
Department of Anesthesiology
Stanford University
Stanford CA 94305

References

(1) Flodin U, Landtblom AM, Axelson O: Multiple sclerosis in nurse anaesthetists. Occup Environ Med 2003;60: 66-8.

(2) Axelsson G, Rylander R: Exposure to anaesthetic gases and spontaneous abortion: response bias in a postal questionnaire. Int J Epidemiol 1982; 11: 250-256

(3) Hirst M, Tse S, Mills DG, Levin L, White DF: Occupational exposure to cyclophosphamide. Lancet 1984; 1: 186-8.

(4) Buring JE, Hennekens CH, Mayrent SL, Rosner B, Greenberg ER, Colton T: Health experiences of operating room personnel. Anesthesiology 1985; 62: 325-330

(5) Tannenbaum TN, Goldberg RJ: Exposure to anesthetic gases and reproductive outcome. A review of the epidemiologic literature. J Occup Med 1985; 27: 659-668

(6) Alexander BH, Checkoway H, Nagahama SI, Domino KB: Cause-specific mortality risks of anesthesiologists. Anesthesiology 2000; 93: 922-30.

(7) Spence AA, Knill-Jones RP: Is there a health hazard in anaesthetic practice? Br J Anaesth 1978; 50: 713-719

(8) Spence A. How safe is anesthesia for you and your patient? Occupational risks of the operating room? Data from the U.K. ten year prospective study. Bulletin of the New York State Postgraduate Assembly 1985; Dec: 140.

(9) Spence AA: Environmental pollution by inhalation anaesthetics. Br J Anaesth 1987; 59: 96-103

(10) Maran NJ, Knill-Jones RP, Spence AA: Infertility among female hospital doctors in the UK. Br J Anaesth 1996; 76: 581P.

Dear Editor

Sorahan and Nichols,[1] writing in this journal, incorrectly understate the strength of evidence for work-related increased mortality among their cohort of production workers in the UK flexible polyurethane foam industry. Their study actually found “some” evidence for a work-related increase in all-cause mortality, respiratory disease mortality, and lung cancer mortality in this exposure circumstance, especially taking into account the healthy worker effect.[2] We are concerned to correct this error, because the UAW represents substantial numbers of workers exposed to this process, and the UK data provides the first evidence of a mortality hazard in this industry, in contrast to two previous, perhaps weaker studies.[3,4]

The authors observed an all-cause SMR among men of 107 (101 to 113), a respiratory disease SMR of 120 (101 to 141). Elevated mortality of similar magnitude from these causes was observed among the smaller number of women, and the SMRs for both genders combined were significantly elevated. Elevated SMRs for all-cause and respiratory disease mortality are hardly ever seen in occupational cohorts except for foundry and asbestos workers. Typically, the SMR for all-cause mortality is about 80, and SMR for most cancer causes about 90 in the absence of exposure to a carcinogen at the site.[5] We have observed SMRs for all-cause mortality as low as 60 in UAW vehicle assembly and stamping cohorts.[6] We are surprised that these authors mentioned a deficit for all-causes in the abstract of their previous study of this cohort, but make no mention of the excess in the present paper.[7]

For lung cancer, the authors noted a significant SMR of 181 (126 to 251) for lung cancer among women. They discount this partly because the SMR of 107 (90 to 227) among men was only slightly elevated compared to the general population, without also noting that the combined SMR was 117 (101 to 136) and statistically significantly elevated. The authors also fail to mention that the SMRs for pancreatic cancer were elevated to a similar degree in both genders, and the combined SMR was 147 (1.02 to 2.12) and significantly elevated. We believe that consistency in direction of effect is more important than statistical significance, especially in view of the healthy worker effect bias against seeing an effect if it were there.

These findings apply to an exposure circumstance with several suspect agents. The isocyanates are most prominently associated with non- malignant respiratory disease. Therefore, the increase in mortality from this cause is of distinct interest. In addition, pancreatic cancer was noted in gavage studies of toluene diisocyanate.[8] In our experience, the most substantial exposure with carcinogenic risk in foam molding is methylene chloride,[9] although brominated and chlorinated alcohol flame retardants, and formaldehyde are usually present in foam molding operations. Catalyst amines may also be absorbed through the skin in physiologically significant amounts.[10] These multiple exposures, often in different parts of the process, including off-gassing from stored foam, undermine the ability to see an effect of isocyanates alone.

We now turn to the exposure response portion of the study. Health related termination of exposure has previously been noted as an obstacle to finding an exposure response effect based on duration.[11] Those with highest exposure to isocyanates would be expected to be sensitized and migrate into lower exposure jobs; in any event, there were only 19 of 1652 deceased workers with more than 5 years in higher exposed jobs, no lung cancer victims and only 2 respiratory disease victims. In our view, the absence of an exposure response relationship in a cohort with such a small higher exposed group detracts little from our concern for occupational cause of an observed excess.

More damaging to the evidence of occupational causation is the absence of a monotonic increasing trend with latency from first exposure: The general trend of increased risk in exposure strata greater than 10 years latency, clearly significant for all-cause mortality, is not seen in those with greater than 30 years latency. However, we note that the all cause and respiratory disease SMRs are at unity or above for this long latency strata, itself a highly unusual observation. The confidence intervals overlap between strata, so while there is not a significant increase, there is no inverse latency response relationship. In addition, much the largest portion of this long latency group must have come from two of the eleven plants (Factory 3 and 4 in Table 1) where exposures may have contrasted to other 9 facilities.

In summary, this study has found a highly unusual and statistically significant elevation in all cause mortality and respiratory disease mortality in the cohort as a whole, consistently in both men and women. For women alone, and men and women combined there was significantly increased mortality for lung cancer, and for both genders combined, pancreatic cancer. This is certainly “some evidence” for work related mortality, respiratory disease mortality, and cancer mortality in the exposure circumstance of polyurethane foam production. The nature of the cohort gave little prospect for observing an exposure response relationship, if it were there. We also note that, unlike all the other papers in this edition of the journal, these authors have neglected to acknowledge their funding source.

Reference

(1) T. Sorahan and L. Nichols, "Mortality and cancer morbidity of production workers in the UK flexible polyurethane foam industry: updated findings, 1958-98." Occup Environ Med (2002) 59, 751-8.

(2) A. J. McMichael, "Standardized mortality ratios and the "healthy worker effect": Scratching beneath the surface." J Occup Med (1976) 18, 165-8.

(3) T. M. Schnorr, K. Steenland, G. M. Egeland, M. Boeniger, and D. Egilman, "Mortality of workers exposed to toluene diisocyanate in the polyurethane foam industry." Occup Environ Med (1996) 53, 703-7 .

(4) L. Hagmar, H. Welinder, and Z. Mikoczy, "Cancer incidence and mortality in the Swedish polyurethane foam manufacturing industry." Br J Ind Med (1993) 50, 537-43.

(5) R. M. Park, N. A. Maizlish, L. Punnett, R. Moure-Eraso, and M. A. Silverstein, "A comparison of PMRs and SMRs as estimators of occupational mortality." Epidemiology (1991) 2, 49-59.

(6) R. Park, J. Krebs, and F. Mirer, "Mortality at an automotive stamping and assembly complex." Am J Ind Med (1994) 26, 449-63.

(7) T. Sorahan and D. Pope, "Mortality and cancer morbidity of production workers in the United Kingdom flexible polyurethane foam industry." Br J Ind Med (1993) 50, 528-36.

(8) National Toxicology Program,"TR-251: Toxicology and Carcinogenesis Studies of Commercial Grade 2,4 (80%)- and 2,6 (20%)- Toluene Diisocyanate (CAS No. 26471-62-5) in F344/N Rats and B6C3F1 Mice (Gavage Studies)."; 86

(9) NIEHS, TR-306: Toxicology and Carcinogenesis Studies of Dichloromethane (Methylene Chloride) (CAS No. 75-09-2) in F344/N Rats and B6C3F1 Mice (Inhalation Studies), (1986).

(10) E. L. Baker, D. C. Christiani, D. H. Wegman, M. Siroky, C. A. Niles, and R. G. Feldman, "Follow-up studies of workers with bladder neuropathy caused by exposure to dimethylaminopropionitrile." Scand J Work Environ Health (1981) 7 Suppl 4, 54-9.

(11) K. Steenland, J. Deddens, A. Salvan, and L. Stayner, " Negative bias in exposure-response trends in occupational studies: modeling the healthy workers survivor effect." Am J Epidemiol (1996) 143, 202-10.

Dear Editor

We would like to comment on the paper by Satin et al, [1] which reports an update of a mortality investigation on two cohorts of petroleum refinery workers. The Authors claim that one of the major aims of their study was the assessment of “health risks relative to more contemporary levels of exposure and work environments”. Nevertheless, they explicitly admit that a previous investigation in such cohorts,[2] using the population of California as referent, found a strong “healthy worker effect” (i.e. a significantly lower than expected mortality risk from cardiovascular disease and lung cancer). In our opinion, this observation, along with some other drawbacks, only in part expressly acknowledged in the paper, might have biased most of the results obtained, leading the Authors to draw unreliable conclusions. We shall discuss this issue in detail, illustrating the main possible biases and how we believe they should have been interpreted.

Comparison bias
Exposure effects should be assessed in cohort studies by comparing the exposed cohorts with at least an unexposed one, as similar as possible in all relevant aspects.[3] The new results by Satin et al. have confirmed the occurrence of the “healthy worker effect” observed in the previous follow up. Such a finding may indicate a comparison bias concealing the associations, if any, between exposure and health risks.[2] In fact, occupational cohorts may differ from the general population in many features that have been associated with various risk factors, including socio-economic status and personal habits.[3] The presence of a comparison bias, at least in the Richmond refinery cohort, seems to be suggested by the risk for leukaemia in the sub-group with the shortest duration of employment (<5 years), which is more than 4 times lower than the referent population (and nearly 7 times lower than those of workers who worked the longest, i.e. more than 30 years). Finally, the lack of data on smoking, whose differential distribution is among the main factors known to be responsible for the “healthy worker effect”, should have suggested a more cautious interpretation of the results of analyses about diseases associated with such a risk factor, especially lung cancer. Owing to the quality of the data analysed, most of these limits are unavoidable. However, in our opinion, the Authors should have taken them into account in discussing their results. For instance, the low leukaemia risk observed, in particular, for workers hired after 1949, should not have been considered as evidence of a lack of effect of quite low doses of benzene.

Dilution effect
Petrochemical workers are likely to experience different kinds and levels of exposure by job category. As a consequence, results from an analysis carried out by pooling together different exposure categories may be affected by a dilution effect, i.e. an underestimation of the true mortality risk associated to exposure.[3] For example, Gennaro et al.[4- 6] highlighted an excess of lung cancer risk among petroleum workers exposed to asbestos in an Italian refinery, which became evident only by using an unexposed job category as an internal referent group. In this investigation, the most heavily exposed group (maintenance workers) was 38% of the whole cohort of employees, very similar to the proportion (36%) reported in a previous study on ten US refineries.[7] Furthermore, white collar workers constituted 22% and 21% of the workforce among the Italian and US refineries, respectively, suggesting that the composition of this kind of cohort tends to be similar, at least in Western countries. Unfortunately, the quality of the data in their possession prevented the Authors from carrying out risk analyses by job category, and they did not discuss about the possibility that the inclusion, if it occurred, of a notable proportion of workers scarcely or not at all exposed may have caused a significant lowering of the estimated risks.[3] Moreover, the inclusion in the present analysis of workers employed after 31 December 1980, thus inflating the at risk population estimates (person- years), could have further contributed to diluting the possible risks, in addition to preventing a precise comparison with the previous update. In fact, a long lag-time is expected between exposure and disease occurrence for most of the cancer sites considered. The mean time of follow up was roughly 33 years for workers hired before 1949 and only 23 for those hired after 1949, but the Authors have not provided any information about the group employed since 1981, making it impossible to estimate the true risks associated with prolonged exposures. Comparing the paper by Satin et al. to the previous follow up,[2] the number of workers enrolled after 1980 could amount to 3600 (i.e. 31% of the subjects hired after 1949) and the corresponding follow-up ranges from 1 to 15 years. However, these data do not allow the calculation of either the corresponding person-years at risk or the number of deaths occurred.

Statistical analysis
The Authors have indirectly evaluated the effect of exposures using the period of hiring (before vs. after 1949) and the number of years worked as factors. Due to the lack of more precise measures of the polluting concentrations, such substitute variables are of course necessary, even though in our opinion, the analysis for another cut-off after 1949 (e.g. 1969 or 1979) might have yielded some additional information about the variation of such risks over time. Furthermore, a possible confounding effect between the period of hiring and the other variables (e.g. length of exposure and latency) should have been taken into account, for instance, either by applying a multivariable statistical model, such as the Poisson regression, or by stratified analysis.[8]

Insensitive indicators
Mortality rates may be poor indicators of cancer risk for disease sites with a good prognosis, for example, leukaemia and larynx cancer.[3] For this reason, comparisons based on mortality rates might be affected by lack of statistical power. Moreover, the Authors admit that the potential inaccuracy of diagnostic information from death certificates may have caused misclassification between asbestosis and pneumoconiosis. However, this inaccuracy might have also affected the risk estimates for specific leukaemia types. While the Authors do not report any results for unspecified leukaemia, an other investigation on petrochemical workers [9] found surprisingly much higher risks for both “acute unspecified” and “cell type unspecified” leukaemia than those for each specified cell type, including acute myeloid (AML). In our view, when data come from death certificates, risk estimates for different leukaemia cell types must be interpreted with extreme care. In particular, the Author’s claim that “the lack of any increase in AML argues against benzene’s role in the increase of MM and NHL found here” is not justified.

Conclusion
Cohort studies based on mortality data and not including an internal group as a control may be affected by several biases. For this reason, the estimates of association between exposure to toxic chemicals and health risks obtained by these studies should be considered with caution. Moreover, the observed excess of risk, if any, should not be ignored simply on the basis of the lack of statistical significance. The need for further investigations for a better evaluation of such risk, for instance, through nested case-control studies, should be always suggested.

References

(1) Satin KP, Bailey WJ, Newton KL, Ross AY, Wong O. Updated epidemiological study of workers at two California petroleum refineries, 1950-95. Occup Environ Med 2002;59:248-56.

(2) Dagg TG, Satin KP, Bailey WJ, et al. An updated cause specific mortality study of petroleum refinery workers. Br J Ind Med 1992;49:203- 12.

(3) Hernberg S. “Negative” results in cohort studies – How to recognize fallacies. Scand J Work Environ Health, 1981;7:121-126.

(4) Gennaro V, Finkelstein MM, Ceppi M, Fontana V, Montanaro F, Perrotta A, Puntoni R, Silvano S. Mesothelioma and lung tumors attributable to asbestos among petroleum workers. Am J Ind Med 2000,37:275-82.

(5) Gennaro V, Montanaro F, Ceppi M, Fontana V, Perrotta A, Puntoni R, Finkelstein MM, Silvano S. RE: Mesothelioma and lung tumors attributable to asbestos among petroleum workers. Am J Ind Med 2000. 37:275-282. I. Reply to Tsai et al.'s Letter to the Editor and new evidencies (letter). Am J Ind Med 2001; 39(5): 517-521.

(6) Gennaro V, Montanaro F, Ceppi M, Fontana V, Perrotta A, Puntoni R, Finkelstein MM, Silvano S. RE: Mesothelioma and lung tumors attributable to asbestos among petroleum workers. Am J Ind Med 2000. 37:275-282. II. Reply to Bailey's Letter to the Editor (letter). Am J Ind Med 2001; 39(5): 522-523.

(7) Nelson NA, Barker DM, Van Peenen PFD, Blanchard AG. Determining exposure categories for a refinery retrospective cohort mortality study. Am Ind Hyg Assoc J 1985; 46: 653-57.

(8) Sahai H, Khurshid A (eds). Statistics in Epidemiology – Methods, Techniques, and Applications. CRC Press, New York, 1996.

(9) Divine BJ, Hartman CM, Wendt JK. Update of the Texaco mortality study 1947-93: part II. Analyses of specific causes of death for white men employed in refining, research, and petrochemicals. Occup Environ Med 1999,56; 174-80.

Dear Editor

We read with interest the article by Hoogendoorn et al. (2002) who examined the use of different approaches to analysing data from their prospective cohort study of work-related exposures and the future onset of low back pain.[1]

Exposures and outcomes are time dependent factors as they are subject to change over time. The strength of the relationship depends on the assumptions of time dependence (or independence) of exposures and outcomes. The effects of these assumptions can be investigated by adopting different modelling approaches to studies that have collected repeated measures of exposure and outcome data over time.

Hoogendoorn et al. (2002) have adopted such an approach in their study of work-related risk factors for low back pain. Information on work-related physical and psychosocial factors and low back pain outcome was collected at baseline and in three annual follow-ups. They demonstrated an increased risk of low back pain for work-related mechanical factors, when using two different generalised estimating equation (GEE) models compared to the standard logistic regression approach.[1] Conversely, for work-related psychosocial factors the association with low back pain was weaker when the GEE method was employed. Such an approach is enlightening and we agree that it is important to explore such analytical techniques in the investigation of work-related risk factors and musculoskeletal symptoms. Therefore further exploitation of this method of analysis seems appropriate.

We have recently conducted a prospective study of new onset low back pain in 1081 newly employed workers from twelve occupational settings.[2] We examined newly employed workers since studies conducted in well established work forces may be influenced by the healthy worker effect, whereby workers may have changed their job or certain aspects of their job as a result of musculoskeletal pain. In brief, at baseline subjects completed a questionnaire, including an assessment of pain status. A pre-shaded manikin was used to enquire about low back pain, defined as pain between the 12th rib and the gluteal folds, lasting at least 24 hours in the past month. Individuals free from low back pain at baseline were identified and followed up at 12 and 24 months. The detailed questionnaire also gathered information on a number of work-related mechanical and psychosocial exposures.

The models used for analysis were identical to those used by Hoogendoorn et al. (2002). The standard logistic regression model was used to examine the relationship between exposures and new onset low back pain at 12 or 24 months, in those free from low back pain at baseline. GEE models are used to analyse repeated measures data, by taking the within subject correlation into account, and providing a summary estimate over time. In GEE Model 1, the relationship between baseline exposures and new onset low back pain at 12 and 24 months was examined. In GEE Model 2 that relationship was examined for baseline exposures and new onset low back pain at 12 months, and 12 month exposures and new onset low back pain at 24 months.

The two models in which risk factors were assumed to be time independent (standard logistic regression and GEE Model 1) produced similar point estimates for developing new onset low back (Table 1 Part i, Part ii and Table 2), with narrower 95% confidence intervals for GEE. Model 1. In GEE Model 2, where risk factors are assumed to be time dependent, differences were noted for only a small number of variables (carrying on one shoulder, lifting at or above shoulder level, general health questionnaire). In addition, the 95% confidence intervals were narrower than those from the standard logistic regression and GEE Model 1. However, there was no consistent pattern of attenuation or growth noted in either the mechanical or psychosocial risk factors examined. (Table 1 Part i, Part ii and Table 2).

In summary, we agree that it is important to investigate different statistical techniques in an attempt to determine what effect the assumptions of time dependence (or independence) have on predictors of musculoskeletal pain. However, unlike the study by Hoogendoorn et al. (2002), our data show that the choice of model has relatively little influence on the magnitude of the results, although GEEs give more accurate estimates.

References

(1) Hoogendoorn WE, Bongers PM, de Vet HC, Twisk JW, van Mechelen W, Bouter LM. Comparison of two different approaches for the analysis of data from a prospective cohort study: an application to work related risk factors for low back pain. Occup Environ Med 2002;59:459-465.

(2) Nahit ES, Macfarlane GJ, Pritchard CM, Cherry NM, Silman AJ. Short term influence of mechanical factors on regional musculoskeletal pain: a study of new workers from 12 occupational groups. Occup Environ Med 2001;58:374-381.