Cox proportional hazards models for incident disease diagnosis and mortality.

* p ** p Note: Sample includes individuals employed at the firm on January 1, 1996. To be considered a new diagnosis, the individual must have been free of the disease for the first two years of the study. For each of these conditions, individuals with one or more inpatient claims or two or more outpatient claims with a relevant ICD diagnosis code in a 365-day period are considered to have the disease in question. Each model includes dummy variables to control for age group at baseline (20–30 years old, 30–40 years old, etc.). Individuals were censored at the last date that they were active at the firm based on the personnel dataset. For mortality, individuals were censored at September 1, 2011, after which we do not have data on mortality.Cox proportional hazards models for incident disease diagnosis and mortality.