# Front Material

## 0.1 Technical Note

This document was created with $$\mathcal{R}$$ Bookdown, a type-setting language for creating HTML, PDF, and Kindle documents. For authors, Bookdown allows us to embed executable program code within the document, to generate plots and analyses ‘on the fly’. Later on in the lab sessions, you will also be able to copy and paste ‘live’ code from this text into the $$\mathcal{R}$$ command window; so the labs will also serve as a working template for your own projects

Please note that $$\mathcal{R}$$ was designed to create publication quality graphics, typically using functions in base $$\mathcal{R}$$ (the core program) or add-on libraries, such as lattice or ggplot. The latter represents a flexible “Grammar of Graphics”, a programming language created by Leland Wilkison for customized graphics. For this course, we will concentrate on making plots using the core graphics package in base $$\mathcal{R}$$.

## 0.2 Who Are We?

Even though this is a biostatistics primer, we are both pediatricians: Celia is a pediatric endocrinologist and Atul is a pediatric nephrologist. We mention this because we appreciate that it can sometimes be difficult to balance clinical and research interests, but we would also like to encourage you to persevere even when it seems tough, because it can be extraordinarily rewarding.

You may be wondering why we we distinguish ‘pediatric biostatistics’, since all biostatistics is presumably the same? This may be true, but we’ve taken great care to make sure the examples in the text and the computer lab are immediately relevant to pediatric trainees, who should be clinically familiar with the issues like low birthweight, maternal breast-feeding habits, and infant mortality. This isn’t chauvinism as much as recognizing that it’s easier to learn statistics if the examples are comprehensible. In our experience, learning is also enhanced by doing; hence, the emphasis on hands-on computer labs, where you will be asked to create descriptive statistics, explore your data graphically, and perform basic statistical analyses.

We are not laboring under the illusion that most of you are heading towards research careers. Rather, we’re trying to provide you with tools to facilitate the scholarly project mandated by the Royal College as part of your training. More importantly, even if you’re not interested in a full-time research career, this is the age of evidence-based-medicine (EBM), and these skills are essential to clinical practice. Whether it’s doing a critical review of the literature when you’re flumoxed by a complicated case or responding to a patient question, you need to understand the medical literature. In the internet era, your patients are also going to ask you to interpret new studies, and you no longer have the luxury of waiting for your professional college to issue new clinical practice guidelines. Unless you can comment promptly and intelligently, you will not inspire much confidence. In brief, for consumers of medical research literature like yourselves, these are essential clinical skills in day-to-day practice.

Since all faculty in the ASK curriculum were asked to comment on their personal career trajectories, our educational credentials follow:

Atul Sharma MD, FRCPC, MSc (Statistics), M.Stat

• MD, Dalhousie University
• Residency: Pediatrics, Montreal Children’s Hospital
• Chief Resident: Pediatrics, Montreal Children’s Hospital
• Fellow: Pediatric Nephrology, University of Minnesota
• MSc Statistics, Colorado State University

You will notice that my CV includes a year as chief resident at McGill. Many of you will already know my co-chief, Celia Rodd. So this workbook is clearly the result of a long-time collaboration :)

Celia Rodd MD, FRCPC, MSc (Epidemiology)

• MD, University of Toronto
• Residency: Pediatrics, Montreal Children’s Hospital
• Chief Resident: Pediatrics, Montreal Children’s Hospital
• Fellow: Pediatric Endocrinology, University of Toronto/Minneosta
• MSc Epidemiology, University of London School of Tropical Medicine and Hygiene

You should also note that we each spent 4 years in medical school, 4 years in residency, and 4 years as a clinical/research fellows, followed by 2 more years in grad school studying statistics and epidemiology, respectively. Despite our best intentions, we can’t possibly teach any of these subjects in an hour or a day. If you’re not comfortable with statistics, you may want to do some additional self study.

## 0.3 References for Self-Study

There are many good review texts in introductory biostatistics for clinicians, for skill levels ranging from neophyte to PhD. The Royal College has even produced a short research guide for trainees, which reviews basic statistical concepts in a concise and practical way. In terms of introductory texts, I’m a big fan of Martin Bland’s Introduction to Medical Statistics, and Celia likes Betty Kirkwood’s Essential Medical Statistics. Both are available in the NJM library, and you may need to consider reviewing relevant chapters to refresh your undegraduate statistics courses from a long time ago, in a galaxy far, far away.

## 0.4 Text Conventions

Since you’ve all studied biostatistics in in medical school at least, this text is intended as a review. For the most part, we will restrict ourselves to bullet points in manageable bites, confident that you can find library or other resources if you require more information. We will however occasionally mark important items for emphasis:

This is important!

In addition, we will mark some points that we would like you to make sure you have absorbed into your DNA before proceeding:

Be sure to understand this!

For those of you who don’t need more explanation, feel free to omit the optional sections marked by

This is an explanatory note.