Table 5 Cohort studies of refinery work and non-Hodgkin’s lymphoma (NHL)
Author, year (Reference)RRCIlowCIupNLocationExposureOutcomeAll cause SMRAdj RRAdj CIlowAdj CIup
Collingwood et al 1996510.710.092.572Paulsboro, New JerseyEmployed 30+ yearsLR SMR0.930.760.072.19
Consonni et al 1999522.120.684.955Northern ItalyTotal cohortNHL SMR0.872.460.785.08
Dagg et al 1992530.550.013.051Chevron, El SegundoEmployed 30+ yearsLR SMR0.730.750.002.94
Dagg et al 1992532.260.974.458Chevron, RichmondEmployed 30+ yearsLR SMR0.862.711.164.90
Divine et al 1999540.30.10.75Texaco Mortality StudyEmployed 30+ yearsLR SMR0.790.380.120.78
Gun et al 2006110.990.621.5022Health WatchTotal cohort (men)NHL SMR0.721.380.872.02
Honda et al 1995550.960.511.6413Illinois refineryHourlyLR SMR0.821.170.621.89
Huebner et al 2004561.460.862.318Exxon Mobile, Baton RougeEmployed 30+ yearsNHL SMR0.791.861.102.82
Huebner et al 2004560.660.271.367Exxon Mobile, BaytownEmployed 30+ yearsNHL SMR0.790.830.331.57
Kaplan 1986570.90.511.4616US refinery workers, TCA, APITotal cohortLR SMR0.781.150.661.79
Lewis et al 2000581.090.681.6522Canada, Exxon MobileTotal cohort (men)NHL SMR0.921.190.741.73
Pukkala 1998592.011.003.5911FinlandBlue collar workersNHL incidencena*nanana
Satin et al 1996600.270.030.972Gulf/Chevron, Pt ArthurEmployed 30+ yearsLR SMR0.880.310.030.87
Sorahan et al 2002281.220.971.51848 British oil refineriesTotal cohort (men)LR SMR0.881.391.111.70
Thomas et al 1982611.570.752.8910OCAWRetiredNHL PMRna*nanana
Tsai et al 1993621.210.154.372Shell, MartinezTotal cohortLR SMR0.891.360.133.90
Tsai et al 1996276.752.4814.696Shell, Deer ParkEmployed 30+ yearsLR SMR1.046.912.4913.55
Tsai et al 2003632.320.486.793Norro, LATotal cohortLR SMR0.743.150.597.71
Waxweiler et al 1983640.680.181.744Texas CityHourlyLR SMR0.80.850.221.89
Wong et al 2001a651.160.243.403Mobil, BeaumontEmployed 30+ yearsNHL SMR0.941.240.233.04
Wong et al 2001b661.140.243.353Mobil, TorranceHired 40+ years agoNHL SMR0.911.250.243.07
  • Adj, adjusted for the healthy worker effect; CIlow, lower 95 CI; CIup, upper 95 CI; exposure, the exposure category selected for this meta-analysis; LR, lymphosarcoma and reticulosarcoma; N, number of cases in the selected exposure category; PMR, proportionate mortality ratio; SMR, standardised mortality ratio; self-report, “Yes” means exposure history based primarily on self-reported work history.

  • *The healthy worker effect adjustment was not done because data were not available on all cause mortality (Pukkala 1998) or the study was a proportionate mortality study (Thomas et al 1982).