Skip main navigation

Introduction: Statistical Application

Introduction: Statistical Application
Hello I’m so excited to be here and honored to talk to you about literature evaluation which has been my life’s passion for teaching. I try to get others to at least appreciate how important it is, even if you may not be as excited about doing it as I am. Now when I was told I was supposed to do these I was told that these were called MOOCs and so it’s kind of funny because you can see these are the animals that I own and actually one of them, her name is MOOC. So I actually had a Mooc before the MOOCs were actually online. I’d be remiss before I got started if I didn’t talk with you about my colleagues.
We work as a team to put these courses together. So I’d like to acknowledge them that the information wouldn’t be as accurate as complete if I didn’t work with a great team of people. So the goal of these sessions is to help you as I said to be able to appreciate literature evaluation and especially help you to understand that it’s not just something that’s painful to do but it can significantly add to patient care decisions to help you to make individualized decisions for patients.
And I’d like to also explain that these are primarily just hitting on some keys statistical points but you still have to always put it in the context of evaluating all parts I’m going to article to make sure that they’re safety efficacy cost and convenience and their role in clinical practice guidelines are also considered apart from these principles. so I’m gonna have to go through the statistical applications pretty quickly so I wanted to make sure you are aware of some other resources you could go to to take a look at other aspects of clinical trials that should be analyzed.
The most common one is the consort which is available online as you can see from the citation at the bottom and this provides checklist lists and really great examples of some good and bad studies in each area to help you to be able to analyze that better. and again you can’t take statistics totally in isolation, you have to evaluate that in in addition to all the other aspects of critiquing an article. So I found this humorous comic. It’s hard to find jokes about journal clubs but I thought this one was pretty good. And you know journal club is in your hospitals where clinicians get together and discuss the relevance.
It’s really hard to analyze a study totally on your own so you want to do it together. I thought this one was funny. it was talking about the different ways they were going to analyze, then the last guy pretty much said I like the pictures. So don’t be that person. Be the person who really wants to evaluate it. And the most important take-home message is if you don’t read the paper and contribute to, you’re really not learning anything about how to better analyze articles. So the learning objectives that we’re going to go through will be quite simple.
We’re going to look at alpha and beta errors and as it refers to superiority studies we will work on non-inferiority in another session. And then given a clinical trial we want to see the most common statistical tests. And then how using them correctly or incorrectly could potentially affect your interpretation of the results. Now,this is our ultimate destination and it looks strange until we actually talk about each part. I mean I think in very much flow diagrams and boxes and things like that.
So we’re going to learn about each one of these parts, and then fill out this box in totality to be able to then answer the questions about were they use appropriately the statistics in some excerpts that we have and how that might affect the how you interpret their results. So the first thing we have to look at this hypothesis testing and superiority studies it’s pretty much a statement of no difference. So you can really have only a couple of outcomes. You can either have look for example if we were studying drug A versus drug B. We could have A significantly better than B or B is significantly better than A.
And that means that the results that we saw were not due to chance alone according to hypothesis testing. Now the other one is no difference. And unfortunately, a lot of people think that no difference means that they’re not inferior equivalent and this is totally wrong. no difference means it’s an inconclusive result. and that’s why we’ll talk about non-inferiority studies later because you’ll often see investigators find no significant difference in a superiority study and they’ll claim non-inferiority and that is totally incorrect.

In this video, Prof. Mary Ferrill first introduces herself and her colleagues who work on this course as well.

The goal of these sessions is to provide an understanding of how to interpret and evaluate certain principles found in the medical literature.

CONSORT (Consolidated Standards of Reporting Trials) is significant in analyzing clinical trials, which can provide checklists and examples of good and bad studies in each area.

The most important take-home message is to be the person who really wants to evaluate the literature. Don’t be a person who doesn’t read paper. You will learn nothing.

There are two learning objectives for this course; one is to determine the possible statistical errors (alpha or beta) that can occur, and the other one is to determine which parameters are appropriate for use with the most common statistical tests used in the medical literature.

Finally, when we are evaluating results from hypothesis testing, what will be the possible outcomes for superiority testing? Please share your answers below.

This article is from the free online

Evidence-Based Medicine in Clinical Pharmacy Practice

Created by
FutureLearn - Learning For Life

Reach your personal and professional goals

Unlock access to hundreds of expert online courses and degrees from top universities and educators to gain accredited qualifications and professional CV-building certificates.

Join over 18 million learners to launch, switch or build upon your career, all at your own pace, across a wide range of topic areas.

Start Learning now