Text this: Addressing the issue of bias in observational studies: Using instrumental variables and a quasi-randomization trial in an ESME research project.