Pooled Analyses of Renal Disease Mortality and Crystalline Silica Exposure in Three Cohorts

Abstract

Respirable crystalline silica exposure has been associated with renal disease in recent literature. Approximately 2000000 people are occupationally exposed to silica in the USA, 100000 at more than twice the NIOSH recommended exposure limit of 0.05 mg/m 3 . We have examined renal disease mortality in three silica-exposed cohorts totaling 13382 workers; 51 died with renal disease (ICD codes 580–587, 9th revision) as the underlying cause and 153 others had renal disease as a contributory cause on their death certificates. All three of these cohorts had job– exposure matrices that enabled estimation of exposure over time; all three also had data on multiple cause mortality, particularly desirable for kidney disease, which may not be the underlying cause of death. Results from two cohorts have been published previously; follow-up for one has been extended by 6 yr. Using both underlying cause and contributory analyses (multiple cause), we compared renal disease in these cohorts with the US population and also conducted exposure–response analyses. We found excess mortality from renal disease [standard mortality ratio (SMR) 1.41, 95% CI 1.05–1.85] as the underlying cause and also using multiple cause analysis (SMR 1.28, 95% CI 1.10–1.47). Exposure–response analyses showed a monotonic increase in renal disease mortality with increasing exposure. Odds ratios by quartile of cumulative exposure were 1.00, 1.24, 1.77 and 2.86 (P = 0.0002) for multiple cause analyses and 1.00, 1.88, 1.96 and 3.93 (P = 0.03) for underlying cause analysis. Pooled analyses provide large sample sizes; these data represent the largest number of renal disease deaths analyzed to date in workers with well defined silica exposure and suggest a causal link between occupational exposure to silica and renal disease. Excess risk of death from renal disease (underlying cause) by age 75 due to a lifetime of occupational exposure at 0.1 mg/m 3 (the OSHA standard when respirable dust is 100% silica) is estimated to be 1.8% (95% CI 0.8–9.7%), above a background risk of 0.3%.