Estimating Model-Adjusted Risks, Risk Differences, and Risk Ratios From Complex Survey Data

Abstract

There is increasing interest in estimating and drawing inferences about risk or prevalence ratios and differences instead of odds ratios in the regression setting. Recent publications have shown how the GENMOD procedure in SAS (SAS Institute Inc., Cary, North Carolina) can be used to estimate these parameters in non-population-based studies. In this paper, the authors show how model-adjusted risks, risk differences, and risk ratio estimates can be obtained directly from logistic regression models in the complex sample survey setting to yield population-based inferences. Complex sample survey designs typically involve some combination of weighting, stratification, multistage sampling, clustering, and perhaps finite population adjustments. Point estimates of model-adjusted risks, risk differences, and risk ratios are obtained from average marginal predictions in the fitted logistic regression model. The model can contain both continuous and categorical covariates, as well as interaction terms. The authors use the SUDAAN software package (Research Triangle Institute, Research Triangle Park, North Carolina) to obtain point estimates, standard errors (via linearization or a replication method), confidence intervals, and P values for the parameters and contrasts of interest. Data from the 2006 National Health Interview Survey are used to illustrate these concepts.

Health consequences of the US Deferred Action for Childhood Arrivals (DACA) immigration programme: a quasi-experimental study

“…We adjusted for the complex survey design of the NHIS-which included use of sample weights (along with SEs robust to heteroscedasticity where appropriate)-in all analyses to recover nationally representative estimates. 33 This step is particularly important given the reported (potentially non-random) decline in participation in the NHIS and other US sample surveys, 34 which the weights are designed to take into account.…”

Comparison of HPV prevalence between HPV-vaccinated and non-vaccinated young adult women (20–26 years)

“…Adjusted prevalence ratios were obtained from average marginal predictions in the fitted logistic regression model. 31,32 We constructed 2 models with additional variables included in the subsequent model. Model 1 was adjusted for sociodemographic variables including age, race/ethnicity, education, income, smoking status, and marital status.…”

Discharge Timing, Outpatient Follow-up, and Home Care of Late-Preterm and Early-Term Infants

“…16 We adjusted for maternal age, education, race/Hispanic ethnicity, marital status, previous live birth, insurance status before pregnancy, method of delivery, and maternal length of hospital stay (calculated as discharge date minus admission date) on the basis of documented associations in the literature. All analyses were conducted by using SAS, version 9.2 (SAS Institute, Cary, NC), and SUDAAN, version 10.0.1 (RTI International, Research Triangle Park, NC), to account for selection and response probabilities of the survey design.…”