Table 6

Sensitivity analyses: The top panel shows the HRs using Cox proportional hazards models without frailty term, not adjusted for experience on the job. The bottom panel shows the HRs using incident rather than prevalent disease as the exposure

Prevalent diseaseHR95% CIp Value
Heart disease1.231.14 to 1.33<0.001
Hypertension0.980.94 to 1.030.444
Diabetes1.191.12 to 1.27<.001
Depression1.311.21 to 1.41<.001
Asthma1.141.04 to 1.240.003
Cox proportional hazards model. Results when all prevalent cases were included are unchanged from the main analysis
Incident diseaseHR95% CIp Value
Heart disease (n=80 881 jobs)1.341.17 to 1.52<0.001
Hypertension (n=74 613 jobs)0.980.92 to 1.060.667
Diabetes (n=79 896 jobs)1.181.06 to 1.310.003
Depression (n=81 049 jobs)1.281.12 to 1.47<0.001
Asthma (n=81 107 jobs)1.241.08 to 1.440.003
Here any employee-job with a disease diagnosis before the start of the job was excluded from analysis separately for each disease. All models are adjusted for calendar year, experience on the job, gender, ethnicity, whether the job was the first at the company, employee's age at the start of the job, whether the plant was a smelter, whether the plant was a union plant, whether the plant was an original company plant, and a frailty term at the employee level
Prevalent diseaseHR95% CIp Value
Heart disease1.211.10 to 1.34<0.001
Hypertension0.950.90 to 1.010.077
Diabetes1.171.08 to 1.26<0.001
Depression1.251.13 to 1.38<0.001
Asthma1.121.01 to 1.260.040
  • Model incorporating variables for presence of all five disease states in the same model.

  • All models are adjusted for calendar year, experience on the job, gender, ethnicity, whether the job was the first at the company, employee's age at the start of the job, whether the plant was a smelter, whether the plant was a union plant, whether the plant was an original company plant, and a frailty term at the employee level.