Thanks again to Drs. Clark and Tolia for the in depth explanation about the Pediatrix Clinical Data Warehouse and BabySteps. Clearly, several unique features made CDW one of the ideal datasets to explore the question of changes in NICU utilitzation for infants with NAS.
Getting back to basics for trainees (and a nice refresher for everyone else), let's delve into the statistical analysis. Perhaps Dr. Bennett, the biostatistician on the paper, can help us with some of these questions:
What are some of the different types of data included in the dataset? How are continuous variables different from ordinal data, count data, and categorical variables? What is a Poisson distribution? How is it different from a normal distribution, and why would one use it to compare incidence of diagnoses in two time periods? What is a sensitivity analysis?
What are other questions that trainees out there have about the statistical analysis?