Adjusting Expected Mortality Rates Using Information From a Control Population: An Example Using Socioeconomic Status.

Hannah Bower,Michael J Crowther,Paul W Dickman,Therese M -L Andersson,Paul C Lambert,Mats Lambe

doi:10.1093/aje/kwx303

Abstract

Expected or reference mortality rates are commonly used in the calculation of measures such as relative survival in population-based cancer survival studies and standardized mortality ratios. These expected rates are usually presented according to age, sex, and calendar year. In certain situations, stratification of expected rates by other factors is required to avoid potential bias if interest lies in quantifying measures according to such factors as, for example, socioeconomic status. If data are not available on a population level, information from a control population could be used to adjust expected rates. We have presented two approaches for adjusting expected mortality rates using information from a control population: a Poisson generalized linear model and a flexible parametric survival model. We used a control group from BCBaSe-a register-based, matched breast cancer cohort in Sweden with diagnoses between 1992 and 2012-to illustrate the two methods using socioeconomic status as a risk factor of interest. Results showed that Poisson and flexible parametric survival approaches estimate similar adjusted mortality rates according to socioeconomic status. Additional uncertainty involved in the methods to estimate stratified, expected mortality rates described in this study can be accounted for using a parametric bootstrap, but this might make little difference if using a large control population.

Highlights

Expected mortality rates calculated from a reference population are used in occupational, actuarial and medical research, and are often used as a comparative measure for an exposed population of interest
For 3.7% of controls and 3.2% of cases socioeconomic status (SES) data were missing. 133 361 female controls were used for the adjustment of expected mortality rates and 26 913 female breast cancer cases for relative survival analyses after exclusions based on age at diagnosis
While the adjusted mortality rate for the medium SES group was similar to the unadjusted rate in those younger than 70 years, the adjusted mortality rate in the low SES was more similar to the unadjusted rate in those older than 70 years

Summary

Introduction

Expected mortality rates calculated from a reference population are used in occupational, actuarial and medical research, and are often used as a comparative measure for an exposed population of interest. Since both cancer-specific and all-cause mortality differ by SES [10,11], one must account for differences in mortality in the reference general population by SES if one wishes to estimate relative survival or similar measures by SES This problem has been addressed by creating stratified life tables by using individualized information for the whole population [12,13,14,15,16]. Others describe methods that create stratified life tables for such risk factors when individualized data are not available by using mortality information from subgroups of the reference population and prevalence of risk factors [4,12,18] We expand on these methods by presenting two models which use information from a control population to adjust expected mortality rates for covariates other than age, sex and calendar period. Using data on SES from a Swedish control population as an example, we describe how to account for additional uncertainty involved in such methods by using parametric bootstrapping

Methods

Results

Discussion

Conclusion