The performance of marginal structural models for estimating risk differences and relative risks using weighted univariate generalized linear models
Austin PC. Stat Methods Med Res. 2024; Apr 24 [Epub ahead of print].
Introduction — We examine colorectal cancer (CRC) survival for patients with and without severe psychiatric illness (SPI) to demonstrate the use of relative and absolute effects.
Methods — This included a retrospective cohort study of patients with CRC diagnosed between 01/04/2007 and 31/12/2012. SPI was defined as major depression, bipolar disorder, schizophrenia, and other psychotic illnesses occurring six months to five years preceding cancer diagnosis and categorized as inpatient, outpatient, or none. Associations between SPI history and death were examined using Cox proportional hazards regression (hazard ratios (HRs)) and Aalen's semiparametric additive hazards regression (absolute differences).
Results — A total of 24,507 patients with CRC were included. A total of 58.1% of patients with inpatient SPI history died, and 47.1% of patients with outpatient SPI history died. Patients with an outpatient SPI history had a 40% (HR 1.40, 95% confidence interval: 1.22–1.59) increased hazard of death, and patients with an inpatient SPI history had a 91% increased hazard of death (HR 1.91, 95% confidence interval: 1.63–2.25), relative to no history of a mental illness. Outpatient SPI history was associated with additional 33 deaths per 1000 person years, and inpatient SPI was associated with additional 82 deaths per 1000 person years.
Conclusion — We encourage future studies examining inequities with time-to-event data to use this method addressing both relative and absolute effect.
Mahar AL, Davis LE, Kurdyak P, Hanna TP, Coburn NG, Groome PA. Ann Epidemiol. 2021; 56:70-4. Epub 2020 Oct 23.
The ICES website uses cookies. If that’s okay with you, keep on browsing, or learn more about our Privacy Policy.