Skip Navigation


Occupational Medicine Advance Access originally published online on March 17, 2009
Occupational Medicine 2009 59(4):255-263; doi:10.1093/occmed/kqp030
This Article
Right arrow Abstract Freely available
Right arrow FREE Full Text (PDF) Freely available
Right arrowOA All Versions of this Article:
59/4/255    most recent
kqp030v1
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Add to My Personal Archive
Right arrow Download to citation manager
Right arrow Disclaimer
Google Scholar
Right arrow Articles by McBride, D. I.
Right arrow Articles by Collins, J. J.
PubMed
Right arrow PubMed Citation
Right arrow Articles by McBride, D. I.
Right arrow Articles by Collins, J. J.
Social Bookmarking
Add to CiteULike   Add to Connotea   Add to Del.icio.us  
What's this?

© 2009 The Author(s).
This is an Open Access article distributed under the terms of the Creative Commons Attribution Non-Commercial License (http://creativecommons.org/licenses/by-nc/2.0/uk/) which permits unrestricted non-commercial use, distribution, and reproduction in any medium, provided the original work is properly cited.

Mortality in employees at a New Zealand agrochemical manufacturing site

David I. McBride1, Carol J. Burns2, G. Peter Herbison1, Noel F. Humphry3, Kenneth Bodner2 and James J. Collins2

1 Department of Preventive and Social Medicine, University of Otago, PO Box 913, Dunedin, New Zealand
2 Department of Epidemiology, The Dow Chemical Company, Midland, MI 48674, USA
3 Dow Chemical (Australia) Ltd, Altona, Victoria 3018, Australia

Correspondence to: David I. McBride, Department of Preventive and Social Medicine, University of Otago, PO Box 913, Dunedin, New Zealand. Tel: +64 3 479 7202; fax: +64 3 479 7298; e-mail: david.mcbride{at}otago.ac.nz


   Abstract
Top
 Abstract
Introduction
 Methods
 Results
 Discussion
 Funding
 Conflicts of interest
 References
 
Background Previous studies at the Dow AgroSciences (Formerly Ivon Watkins-Dow) plant in New Plymouth, New Zealand, had raised concerns about the cancer risk in a subset of workers at the site with potential exposure to 2,3,7,8-tetrachlorodibenzo-p-dioxin. As the plant had been involved in the synthesis and formulation of a wide range of agrochemicals and their feedstocks, we examined the mortality risk for all workers at the site.

Aims To quantify the mortality hazards arising from employment at the Dow AgroSciences agrochemical production site in New Plymouth, New Zealand.

Methods Workers employed between 1 January 1969 and 1 October 2003 were followed up to the end of 2004. Standardized mortality ratios (SMRs) were calculated using national mortality rates by employment duration, sex, period of hire and latency.

Results A total of 1754 employees were followed during the study period and 247 deaths were observed. The all causes and all cancers SMRs were 0.97 (95% CI 0.85–1.10) and 1.01 (95% CI 0.80–1.27), respectively. Mortality due to all causes was higher for short-term workers (SMR 1.23, 95% CI 0.91–1.62) than long-term workers (SMR 0.92, 95% CI 0.80–1.06) and women had lower death rates than men. Analyses by latency and period of hire did not show any patterns consistent with an adverse impact of occupational exposures.

Conclusions The mortality experience of workers at the site was similar to the rest of New Zealand.

Keywords      Chemical industry; cohort studies; dioxins; mortality; occupational exposure; risk assessment


   Introduction
Top
 Abstract
 Introduction
Methods
 Results
 Discussion
 Funding
 Conflicts of interest
 References
 
The former Ivon Watkins-Dow plant, now Dow AgroSciences New Zealand [1], has been in operation at the same site since 1962 making a diverse range of agrochemical products. Initially, the plant manufactured 2,4-dichlorophenoxy acetic acid and later 2,4,5-trichlorophenoxy acetic acid (2,4,5-T) and 4-chloro 2-methylphenoxy acetic acid. The 2,4,5-trichlorophenol feedstock for 2,4,5-T was manufactured from 1969, and smaller amounts of 2-methyl-4-chlorophenoxybutyric acid were manufactured from the early 1970s. In addition to phenoxy herbicides, numerous other pesticides and chemical raw materials were handled on the site: these included the sodium salt of 2,4,6-TCP, picloram, atrazine, simazine, amitrole and dichloropropionic acid. In 1987, the TCP plant closed and 2,4,5-T production ceased in 1988. The dioxin, 2,3,7,8-tetrachlorodibenzo-p-dioxin (TCDD), can be an unwanted contaminant in TCP and 2,4,5-T. Figure 1 provides the chemical structure of TCDD, TCP and 2,4,5-T. The current manufacturing activities are mostly centred around the formulation and packaging of insecticides, herbicides, fungicides and spraying oil.


Figure 1
View larger version (9K):
[in this window]
[in a new window]
[Download PowerPoint slide]
  		Figure 1. Chemical structures for TCDD, 2,4,5-trichlorophenol and 2,4,5-T.

 
Because of the potential exposure to TCDD, some workers at the plant were included in an international cohort of 18 910 phenoxy herbicide sprayers and production workers designed to evaluate health effects and exposure to TCDD. This study was carried out in 10 countries by the International Agency for Research on Cancer and the results, published in 1991, indicated that soft tissue sarcomas and non-Hodgkin's lymphoma were greater than expected among workers with exposure to 2,4,5-T [2].

The New Plymouth study subgroup included 1038 men and women who worked at least 1 month between 1969 and 1984, with vital status follow-up to 1987. Workers were classified according to the main study criteria as exposed to phenoxies, probably exposed, with ‘unknown exposure’ or unexposed. There were 782 employees who were considered exposed to phenoxy herbicides or trichlorophenol, with some workers involved in warehousing of products at other New Zealand sites. The unexposed workers were employed outside the production areas such as in administrative positions. This study found a small excess of all cancers combined but no increase in lung cancer, soft tissue sarcoma or non-Hodgkin's lymphoma, cancers which have been associated with TCDD exposure in some studies. The study was recently updated with follow-up to 2000 [3]. The findings were similar to the earlier study although there were more multiple myeloma deaths than expected among workers with potential dioxin exposure.

Given these reports about the health risks at the New Plymouth site, the principal aim of this study was to quantify the overall mortality hazard from employment there. The objectives were to identify and include all the employees who worked at the plant between 1969 and 2003, follow the vital status of these workers to the end of 2004 and determine if workers at the site have mortality levels different from the New Zealand population. This is the first study in a series which will examine the health of these workers.


   Methods
Top
 Abstract
 Introduction
 Methods
Results
 Discussion
 Funding
 Conflicts of interest
 References
 
Work history records were used to identify all the employees who worked for at least 1 day at the New Plymouth site between 1 January 1969 and 1 October 2003: the latter being the date that the work history records were collected. In the analysis, 23 individuals who had missing dates of birth were excluded, and 9 people lacking information on sex were assumed to be male. The total number of workers in the study was 1754.

Vital status for each study member was determined from 1 January 1969 to 31 December 2004. The search was based on the New Zealand Health Information Service (NZHIS) Mortality Collection [4], in which deaths registered in New Zealand from 1988 onwards are held in the electronic mortality database, and data from 1970 to 1987 are archived and are available on request. Deaths occurring between 1969 and 1987 were matched to work records using an algorithm which required an exact match on date of birth and ranked minor errors in the spelling of surnames and forenames, nine being a perfect match, with eight and seven being minor differences in surname or forename. All matches were examined, and matches of seven to nine proved acceptable. From 1988 onwards, searches were based on the National Health Index number on the mortality database, unique to each individual. We then carried out additional searches based on the last known address on the employee database, following name and address changes using the Electoral Roll and Habitation Index [5], Telephone Book, Companies Office [6], ‘Terranet’ property information database [7] and search engines on the Internet [8,9].

Additional deaths up to 1990 were sought from the Registrar-General's Index to Deaths. If a person was not known to be alive at the study end date (i.e. on the electoral roll), an additional search was made through births, deaths and marriages. Follow-up ended on 31 December 2004, at the date of death, or for those who had left New Zealand, or could not be traced, at the date they were last known to be in the country. NZHIS provided the coded underlying causes of death for all deaths in the study [10].

The expected numbers of deaths adjusted for age, gender and calendar year were calculated using the Occupational Mortality Cohort Analysis Program [11]. The New Zealand national mortality rates, obtained from NZHIS, were used as the reference population. Standardized mortality ratios (SMRs) and exact 95% confidence intervals (CIs) were calculated for major causes of death and cancer. Person-years for the study members were accumulated across 5-year age calendar year-specific categories from the date of entry into the study until the end of follow-up. Analyses were carried out by duration of employment, sex, latency and period of hire. For the duration of employment analyses, we divided workers into two groups. The first group included workers employed for ≤3 months to account for seasonal and short-term workers. The second group included those workers who worked for ≥3 month. Latency was defined as the period between the day last worked and the date of death, the end of study date and the date lost to follow-up or emigration, whichever came first [12]. Latency of less than and greater than 15 years and period of hire pre-1976 and post-1975 were chosen to examine possible exposure effects while having a reasonable number of person-years in each stratum.


   Results
Top
 Abstract
 Introduction
 Methods
 Results
Discussion
 Funding
 Conflicts of interest
 References
 
From the total of 1754 eligible persons, 247 deaths were observed. Twenty-two per cent of the cohort was lost to follow-up as shown in Table 1. This included 392 individuals, among which 156 (9%) were known to have emigrated before the study end date.


View this table:
[in this window]
[in a new window]

  		Table 1. Vital status follow-up of study group

 
Among all employees (Table 2), all cause mortality, SMR 0.97 (95% CI 0.85–1.10), and all cancer mortality, SMR 1.01 (95% CI 0.80–1.27), were close to expected levels. There were no statistically significant differences in any cause of death compared to the New Zealand population. Approximately, one-third of the deaths were due to heart disease and were close to expected levels (SMR 1.02, 95% CI 0.81–1.27). There were more than expected numbers of deaths due to cancers of the digestive system (27 observed versus 22.8 expected) and lymphohaematopoietic cancers (8 observed versus 6.6 expected). On the other hand, some specific cancers occurred at less than expected rates, such as lung cancer (15 observed versus 17.3 expected), breast cancer (2 observed versus 3.5 expected) and prostate cancer (3 observed versus 5.6 expected).


View this table:
[in this window]
[in a new window]

  		Table 2. SMR and 95% CI by duration of employment

 
The rates for those who had worked at the site for <3 months (short-term employees) were generally higher than those who worked for ≥3 months (long-term employees), short-term employees showing an all causes SMR of 1.23 (95% CI 0.91–1.62). This included a statistically significant excess of all cancers of the digestive system, SMR 2.52 (95% CI 1.15–4.78), reflected by an increased risk for all the subsites in this category except cancer of the large intestine. The increased risk was not only limited to malignant neoplasm but also occurred from heart disease, cirrhosis of the liver and motor vehicle accidents. Conversely, the long-term employees exhibited, in general, lower mortality rates with most causes at less than expected rates. These included all causes (SMR 0.92, 95% CI 0.80–1.06), all cancers (SMR 0.98, 95% CI 0.75–1.26), cancers of the digestive system (SMR 0.94, 95% CI 0.56–1.48) and heart disease (SMR 0.94, 95% CI 0.72–1.20). All eight lymphohaematopoietic cancers in the cohort were found in the long-term employees, producing a slightly elevated SMR of 1.21 (95% CI 0.52–2.39).

Those with ≥15 years of latency had lower mortality from all causes, all cancers and all heart disease as shown in Table 3. Deaths from all lymphohaematopoietic tumours remained raised in this group (7 observed versus 4.7 expected), mostly as a result of non-Hodgkin's lymphoma (SMR 2.09, 95% CI 0.57–5.35) and multiple myeloma (SMR 2.10, 95% CI 0.26–7.60). Accidents, including motor vehicle accidents, were also greater than expected, as were asthma, nephritis and nephrosis.


View this table:
[in this window]
[in a new window]

  		Table 3. SMR and 95% CI by latency

 
There were only 26 deaths in those employed after 1975, as shown in Table 4, with all causes, all cancers and all heart disease SMRs at 0.74 (95% CI 0.48–1.08), 0.62 (95% CI 0.25–1.28) and 0.60 (95% CI 0.20–1.41), respectively. Those employed prior to 1976 had SMRs close to expected levels for the same causes.


View this table:
[in this window]
[in a new window]

  		Table 4. SMR and 95% CI by date of hire

 
There were fewer deaths among females (27) than males (220) as shown in Table 5. Females had lower all causes mortality, SMR 0.68 (95% CI 0.45–1.00), than males, SMR 1.02 (95% CI 0.89–1.17). A decrease in female all cancer mortality was evident, with an SMR of 0.52 (95% CI 0.22–1.02), as compared to a male rate of 1.14 (95% CI 0.90–1.45). Cause-specific rates among the women were, however, imprecise due to the small numbers of observed and expected deaths.


View this table:
[in this window]
[in a new window]

  		Table 5. SMR and 95% CI by sex

 

   Discussion
Top
 Abstract
 Introduction
 Methods
 Results
 Discussion
Funding
 Conflicts of interest
 References
 
We found overall that workers in the study population had death rates similar to the rest of New Zealand. Employment duration accounted for the greatest differences in observed mortality rates: relative to long-term workers, those who were employed for <3 months had higher death rates for many causes including all causes, all cancers, ischaemic heart disease and all accidents. In addition, among these short-term workers, cancers of the digestive system were greater than expected especially for rectal cancer, oesophageal cancer and pancreatic cancer. In contrast, workers employed for >3 months had mortality levels consistent with the rest of New Zealand. These findings of higher death rates among short-term workers have been reported in other studies, attributed to the healthy worker effect [13] and lifestyle factors, especially alcohol use and tobacco [14–16]. Since the long-term workers had mortality and cancer rates similar to the New Zealand population, it is unlikely that brief workplace exposures contributed significantly to the mortality rates seen in the short-term workers.

Women working at the site had low mortality rates compared to women in the rest of New Zealand (SMR 0.68, 95% CI 0.45–1.00). These low rates among women at the site are attributable to low cancer and heart disease rates. Although there were few person-years of follow-up, these low rates for women in chemical plants have been seen in other studies [17–20].

One potential limitation of this study is the 22% loss to follow-up. However, our levels of loss are similar to other New Zealand studies. Much of the loss to follow-up is due to the high rate of emigration which occurs in New Zealand. We verified that at least 9% of our workers emigrated during the study period. The true emigration rate is likely to have been >9% since we only counted emigrants we could locate. Since emigrants do not contribute to the national mortality rates, their withdrawal from follow-up does not introduce a bias into risk estimation. The true number lost to follow-up of no more than 236 (or 13% of the cohort), and likely to be much less, is consistent with studies done in other countries. While it is possible that those lost to follow-up may have higher mortality rates than the rest of the study population, we feel that the loss is small enough to have little influence on risk estimates.

Another potential limitation of this study is size. While we have included all workers employed at this site from 1969 to near the end of 2003, there are some causes of death which we could not credibly evaluate because they are relatively rare. For example, we did observe 1 death from soft tissue sarcoma versus 0.4 expected. This cancer has been related to TCDD exposure in some studies. With the small number of observed and expected deaths from this cause in our study, it is not possible to confirm or refute an impact of workplace exposure on soft tissue sarcoma. Such an evaluation for rare causes of death will have to wait for larger studies with detailed exposure evaluations.

A previous study at this site examined mortality levels in a subset of the workers in our study who were potentially exposed to dioxins as contaminants of the 2,4,5-T production. The investigators reported that all causes mortality in production workers was not greater than expected; however, there were more cancers than expected including a statistically significant increase in deaths from multiple myeloma (3 observed versus 0.5 expected). In contrast, we found all cancers at expected levels and 2 deaths from multiple myeloma versus 1.2 expected. While the tracing methods used and numbers lost to follow-up were similar in both the studies, the differences in cohort definitions and subgroups studied could have accounted for the differences in observed deaths.

Due to the relatively large number of potential exposures at the plant, we did not classify the cohort with respect to particular chemicals or contaminants, but looked for effects through period of hire and latency. Period of hire is useful for isolating potential occupational exposures. The bulk of employment occurred in the decade following the inception of the study and the majority of those exposed to 2,4,5 TCP and 2,4,5,-T would have been included in the pre-1976 analysis. Most carcinogens are assumed to have long latency periods and analyses by latency are useful for examining potential workplace effects. We did observe higher total cancer rates in the pre-1976 period (SMR 1.08, 95% CI 0.84–1.37) than in the post-1975 period (SMR 0.62, 95% CI 0.25–1.28). However, the rates for total cancers were lower in the longer latency period (SMR 0.96, 95% CI 0.72–1.26) than in the shorter latency period (SMR 1.14, 95% CI 0.72–1.71). We also observed four deaths from non-Hodgkin's lymphoma in this study. Three deaths of non-Hodgkin's lymphoma occurred in the pre-1976 group (SMR 1.44, 95% CI 0.30–4.21) and one death in the post-1975 group (SMR 2.29, 95% CI 0.06–12.76). All non-Hodgkin's lymphoma deaths occurred in the longer latency group (SMR 2.09, 95% CI 0.57–5.36).

As a surveillance tool, the principal objective of this analysis was to quantify the potential hazard of employment at the site rather than of any specific chemical exposures. We observed rates for all causes, all malignant neoplasms and many other causes of death to be similar to the New Zealand general population rates, but higher than expected rates were consistently observed for short-term employees.

In summary, the results of our study indicate that there is no excess mortality risk for employees who worked at the New Plymouth site. The most critical step in addressing mortality is, however, to look at specific exposures in detail which will be the subject of future studies. This report establishes the basis from which this can be done.


Key points
  • Assessment of exposure to agrochemicals at this plant by duration of exposure, period of hire and latency showed that only short duration of employment (and potential exposure) was associated with increased risk of mortality.
  • It is unlikely that workplace exposures influenced the mortality risk of employees at this site.
  • The results of epidemiological studies which do not include specific exposure estimates must be interpreted with caution.

 


   Funding
Top
 Abstract
 Introduction
 Methods
 Results
 Discussion
 Funding
Conflicts of interest
 References
 
The Dow Chemical Company. Funding to pay the Open Access publication charges for this article was provided by The University of Otago.


   Conflicts of interest
Top
 Abstract
 Introduction
 Methods
 Results
 Discussion
 Funding
 Conflicts of interest
References
 
None declared.


   Acknowledgements
 
Past and present employees at the New Plymouth plant, Christine Fowler and the Clinical Coding team at the NZHIS.


   References
Top
 Abstract
 Introduction
 Methods
 Results
 Discussion
 Funding
 Conflicts of interest
 References
 

  1. Dow AgroSciences: New Zealand. About Dow Agrosciences (NZ) Ltd. http://www.dowagro.com/nz/about/ (26 February 2009, date last accessed).

  2. Saracci R, Kogevinas M, Bertazzi PA, et al. Cancer mortality in workers exposed to chlorophenoxy herbicides and chlorophenols. Lancet (1991) 338:1027–1032.[CrossRef][Web of Science][Medline]

  3. Mannetje A, McLean D, Cheng S, et al. Mortality in New Zealand workers exposed to phenoxy herbicides and dioxins. Occup Environ Med (2005) 62:34–40.[Abstract/Free Full Text]

  4. New Zealand Health Information Service. Health Statistics. http://www.nzhis.govt.nz/moh.nsf/indexns/stats (26 February 2009, date last accessed).

  5. Electoral Enrolment Centre CEO, and Electoral Commission. Elections New Zealand. http://www.elections.org.nz/ (11 June 2008, date last accessed).

  6. Ministry of Economic Development. New Zealand Companies Office. Search for a Company. http://www.companies.govt.nz/cms (26 February 2009, date last accessed).

  7. Terralink International Limited. Terranet Property Information Online. http://www.terranet.co.nz/terranet/ (11 June 2008, date last accessed).

  8. Google Inc. Google New Zealand. http://www.google.co.nz/ (11 June 2008, date last accessed).

  9. Yahoo!7. http://nz.yahoo.com/ (11 June 2008, date last accessed).

  10. New Zealand Health Information Service. Health Statistics. http://www.nzhis.govt.nz/moh.nsf/indexns/stats (26 February 2009, date last accessed).

  11. Marsh GM, Youk AO, Stone RA, Sefcik S, Alcorn C. OCMAP-Plus (1998) University of Pittsburgh Graduate School of Public Health Department of Biostatistics.

  12. Checkoway H, Pearce N, Crawford-Brown DJ. Research Methods in Occupational Epidemiology (1989) New York: Oxford University Press (USA).

  13. Langseth H, Kjærheim K. Mortality from non-malignant diseases in a cohort of female pulp and paper workers in Norway. Occup Environ Med (2006) 63:741–745.[Abstract/Free Full Text]

  14. Stewart PA, Schairer C, Blair A. Comparison of job, exposures, and mortality risks for short-term and long-term workers. J Occup Med (1990) 32:703–708.[Web of Science][Medline]

  15. Boffetta P, Sali D, Kolstad H, et al. Mortality of short term workers in two international cohorts. J Occup Environ Med (1998) 40:1120–1126.[CrossRef][Web of Science][Medline]

  16. Lamm SH, Levine MS, Starr JA, et al. Analysis of excess lung cancer risk in short-term employees. Am J Epidemiol (1988) 127:1202–1209.[Abstract/Free Full Text]

  17. Collins JJ, Riordan SG. Mortality surveillance and occupational hazards: the Solutia mortality experience 1980–94. Occup Environ Med (2000) 57:710–717.[Abstract/Free Full Text]

  18. Bond GG, McLaren EA, Cartmill JB, et al. Mortality among female employees of a chemical company. Am J Ind Med (1987) 12:563–578.[CrossRef][Web of Science][Medline]

  19. Pell S, O'Berg MT, Karrh BW. Cancer epidemiologic surveillance in the Du Pont Company. J Occup Med (1978) 20:725–740.[CrossRef][Web of Science][Medline]

  20. Burns CJ, Cartmill JB, Chau M. Cause-specific mortality among Michigan employees of a chemical company: 1940 to 1994. J Occup Environ Med (2002) 44:168–175.[Web of Science][Medline]


Add to CiteULike CiteULike   Add to Connotea Connotea   Add to Del.icio.us Del.icio.us    What's this?



This Article
Right arrow Abstract Freely available
Right arrow FREE Full Text (PDF) Freely available
Right arrowOA All Versions of this Article:
59/4/255    most recent
kqp030v1
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Add to My Personal Archive
Right arrow Download to citation manager
Right arrow Disclaimer
Google Scholar
Right arrow Articles by McBride, D. I.
Right arrow Articles by Collins, J. J.
PubMed
Right arrow PubMed Citation
Right arrow Articles by McBride, D. I.
Right arrow Articles by Collins, J. J.
Social Bookmarking
Add to CiteULike   Add to Connotea   Add to Del.icio.us  
What's this?