Article Text

Download PDFPDF

0407 An old trade with an unanswered question: does arc-welding fume exposure increase the risks of obstructive pulmonary diseases? First findings from the WELSHIP cross-sectional study
  1. Andrea Marongiu1,
  2. Omer Hasan2,
  3. Anila Ali2,
  4. Sharoon Bakhsh2,
  5. Bobby George2,
  6. Nabeel Irfan2,
  7. Cristina Canova1,
  8. Susie Schofield1,
  9. Joana Szram1,
  10. Cosetta Minelli1,
  11. Paul Cullinan1
  1. 1Imperial College London, London, UK
  2. 2Drydocks World Dubai, Dubai, United Arab Emirates


Objectives Increasingly, global manufacturing is shifting to emerging economies and with it the use of arc-welding for applications in different industries. The chronic respiratory adverse effects resulting from exposure to gases and ultrafine metal particles in welding fume are incompletely understood.

We aimed to measure the prevalence of arc-welding related pulmonary obstructive outcomes by analysing data collected in a shipyard in the Middle East.

Method Between January and December 2013, through cross-sectional survey, we collected spirometry data and behavioural, occupational and respiratory symptoms information from a random sample of male shipyard workers; 397 were exposed to welding fume and 127 were non-exposed. The sample was selected from a total population of about 8000 employees, by frequency matching for ethnicity and age relatively to full-time welders (‘highly’ exposed).

Results Of the 580 workers invited, 26 subsequently left their job; of the remainder, 95%(524) agreed to participate. The participants, from the Indian subcontinent (90%) or Philippines (10%), had a median age of 38 years. Ever smoking was reported by 37%, with full-time welders reporting the lowest proportion of current smoking, 18%(24/131).

Overall, 13% reported respiratory symptoms with a higher prevalence in the winter months. Post-bronchodilator spirometry data were available for 91% of workers. Mean values for FEV1 and FVC were 2.87L and 3.48L, with no statistically significant differences across exposure groups (p-values: 0.71 and 0.48).

Conclusions These preliminary results need to be explored further in relation to smoking, past and current occupational exposure. This population, it is hoped, will form the basis for a longitudinal study.

Statistics from

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.