China
Africa
Explore chapters and articles related to this topic
Model Estimation and Evaluation
Published in Douglas D. Gunzler, Adam T. Perzynski, Adam C. Carle, Structural Equation Modeling for Health and Medicine, 2021
Douglas D. Gunzler, Adam T. Perzynski, Adam C. Carle
A covariate is a variable that is controlled for in the statistical model that may or may not be related to the outcome. A covariate is included in the model because the researcher deemed it is a potential confounder. If the covariate does affect the relationships between variables being studied, then it is indeed a confounder.
Individual Participant Data Meta-Analysis
Published in Christopher H. Schmid, Theo Stijnen, Ian R. White, Handbook of Meta-Analysis, 2020
In meta-analyses of clinical trials, there is often interest in whether a covariate can modify the effectiveness of a treatment. For example, the PARIS analysis examined whether the efficacy of antiplatelet therapy improved (or worsened) according to the age of the woman and whether she had diabetes or hypertension.
Strategies for Dealing with Intercurrent Events
Published in Craig Mallinckrodt, Geert Molenberghs, Ilya Lipkovich, Bohdana Ratitch, Estimands, Estimators and Sensitivity Analysis in Clinical Trials, 2019
Craig Mallinckrodt, Geert Molenberghs, Ilya Lipkovich, Bohdana Ratitch
Covariate adjustment is a common example of a strategy to compensate for imperfect experimental conditions. It is rare to observe perfect balance between arms, but covariates are often fit to estimate the treatment effect under the hypothetical scenario that the arms had equal proportions of patients for categorical covariates and equal means for continuous covariates.
Burden of illness associated with overweight and obesity in patients hospitalized with COVID-19 in the United States: analysis of the premier healthcare database from April 1, 2020 to October 31, 2020
Published in Journal of Medical Economics, 2023
Nina Kim, Abdalla Aly, Chris Craver, W. Timothy Garvey
Covariates for analysis included demographics (age at admission, gender, race/ethnicity), geography (state, region), comorbidities that could worsen outcomes of COVID-19 (Supplementary Table 1), and hospital characteristics (urban or rural population served, teaching status, bed capacity, level of care: acute care vs floor). Not all variables have been adjusted for in every model or were included in the final analysis. Clinical input and statistical tests were used to understand the relevance of each covariate in all the models. For some models, ICU stay or O2 supplementation was used as an explanatory variable and as an interaction term with BMI class variable to adjust for disease severity. Additional possible covariate interactions (e.g. age and BMI) were also explored. Finally, bivariate analysis between outcomes and covariates was used to test for associations that needed to be addressed in the final regression analysis.
Survival and epidemiology of amyotrophic lateral sclerosis (ALS) cases in the Chicago and Detroit metropolitan cohort: incident cases 2009–2011 and survival through 2018
Published in Amyotrophic Lateral Sclerosis and Frontotemporal Degeneration, 2023
Reshma Punjani, Theodore C. Larson, Laurie Wagner, Bryn Davis, D. Kevin Horton, Wendy Kaye
Cases were submitted to the National Death Index (NDI) to obtain vital status through 31 December 2018 and matched to the Chicago and Detroit cases by unique case identification numbers. We used Cox proportional hazards regression to model the effect of patient covariates on mortality (17). In all modeling, date of ALS diagnosis was the time origin and time on study the time axis. Subjects not determined to be deceased via the NDI were censored on 31 December 2018. Subject matter knowledge of the suspected and known sources of confounding guided covariate selection. We estimated hazard ratios (HRs) with 95% confidence intervals (CI) for the following covariates: sex, age at diagnosis (years, continuous), race/ethnicity (non-Hispanic White vs. other race or Hispanic), diagnosis type (definite or probable ALS vs. possible ALS or not classifiable), and time interval between symptom onset and diagnosis (years, continuous). Additionally, race/ethnicity was categorized as “White” to include white race and non-Hispanic ethnicity and “non-White” to be all minority non-white races and Hispanic ethnicity. In all models, incomplete records for patient covariates (i.e. listwise deletion) we excluded. A p value of less than 0.05 was deemed statistically significant. All analyses were conducted with SAS 9.4 (SAS Institute, Inc., Cary, NC). Data for this project were collected under a protocol approved by the Centers for Disease Control and Prevention Institutional Review Board.
Inference on moderation effect with third-variable effect analysis – application to explore the trend of racial disparity in oncotype dx test for breast cancer treatment
Published in Journal of Applied Statistics, 2022
Qingzhao Yu, Lu Zhang, Xiaocheng Wu, Bin Li
Other variables used in this study include demographic information for patients (e.g. age and insurance), cancer characteristics (e.g. tumor size and grade), and population characteristics at the county level (e.g. rural/urban and proportion of household below the federal poverty level). Besides race, year of diagnosis and the outcome variable (ODX test), we include 31 variables that can potentially explain the use and the racial disparity in the use of ODX test. In the analysis, we did not include factors on the first-course treatment because ODX test can guide the choice of treatment but not the reverse. Since we are interested in the trend of the racial disparity in ODX and how the effect of contributing factors in explaining the racial disparity varied with year, year-of-diagnosis was used as the moderator. A third-variable is defined as a variable that is significantly related to the exposure variable (race), and significantly related to the outcome (having an ODX test within one year of diagnosis). A covariate is a variable significantly associated with the outcome, but not with the exposure variable. We first tested each variable to identify potential third-variables and covariates. The significance level of tests is set at 0.05. Table 2 shows the potential third-variables and covariates as a result of tests.