27 Nov How to perform an explanatory analysis

Elicit the research hypothesis

All research requires hypotheses. For example, it may be: “Does our new hospital protocol reduce the number of readmissions?”. Sometimes this hypothesis is straightforward to elicit, sometimes it is more complex.

pvalue.io helps you to elicit it, by assuming the following: “X has an influence on Y”. It is up to you to determine the relevant X and Y. In the example on readmissions, X would be: the hospital protocol (coded for example in 2 classes: “former” and “new”), and Y the presence of a 30-day readmission (e. g. yes/no or 0/1). The hypothesis would therefore be: “The hospital protocol has an influence on 30-day readmission”.

However, there are many possible biases, so it is often necessary to use covariates. These covariates are the variables related to Y.

Select the covariates

Thus, pvalue.io first prompts you to select the variables known or assumed to be related to Y. These are the explanatory variables. Then, it automatically selects the covariates not known related to Y (extraneous variables). For explanatory analyses, we do not need to obtain estimates of the influence of the extraneous variables, only the explanatory variables.. If these variables are numerical, a natural cubic spline transformation can be performed.

Check that the conditions are met

This verification step is essential and uses automatic detection mechanisms (for example, the residuals normality check), or manual (for example, checking the linearity of X as a function of Y, or proportional hazards).

Check the robustness of the model

After a multivariable analysis, it is necessary to check the robustness of the model by removing the most influential variables from the statistical model. This procedure has not yet been implemented.

START THE ANALYSIS