Lecture 3A: Finding the Evidence: Searching Principles

Loading...

En provenance du cours de Johns Hopkins University

Introduction to Systematic Review and Meta-Analysis

1199 notes

Johns Hopkins University

Introduction to Systematic Review and Meta-Analysis

1199 notes

We will introduce methods to perform systematic reviews and meta-analysis of clinical trials. We will cover how to formulate an answerable research question, define inclusion and exclusion criteria, search for the evidence, extract data, assess the risk of bias in clinical trials, and perform a meta-analysis. Upon successfully completing this course, participants will be able to: - Describe the steps in conducting a systematic review - Develop an answerable question using the “Participants Interventions Comparisons Outcomes” (PICO) framework - Describe the process used to collect and extract data from reports of clinical trials - Describe methods to critically assess the risk of bias of clinical trials - Describe and interpret the results of meta-analyses

À partir de la leçon

Searching Principles and Bias Assessment

In this module we will look at finding the evidence, as well as key sources, search strategy, and assessing the risk of bias.

Searching Principles and Assessing Bias 2:58

Lecture 3A: Finding the Evidence: Searching Principles5:48

Lecture 3B: Identifying Key Sources and Techniques for Searching17:34

Lecture 3C: Building a High-Quality Search Strategy17:07

Lecture 3D: Documenting Your Search and Conclusions8:00

Rencontrer les enseignants

Tianjing Li, MD, MHS, PHD
Assistant Professor, Epidemiology
Bloomberg School of Public Health
Kay Dickersin, PhD
Professor, Epidemiology
Bloomberg School of Public Health

[MUSIC]

Welcome to today's lecture, my name is Claire Twose.

I'm the Associate Director of Public Health and

Basic Science Informationist Services.

And I'm going to be talking with you about finding the evidence

to support a systematic review in meta-analysis.

We're covering Various searching principles involved.

This presentation is going to be in four parts.

I'm going to start with an introduction to various themes.

Spend the next section going over the variety of sources that

you need to use in order to do a comprehensive, unbiased search.

Then we're going to drill down and I'll show you how to build the kind

of search strategy that works in some of the large databases that you need to use.

And I'll end with a few words about Documentation and conclusions.

So, first part.

At this point in the process you now have an answerable research question.

What is it that you need to do next?

The next step is to develop the protocol.

In doing a systematic review in meta analysis,

you are doing a scientific experiment.

The only difference from other randomized trials is that

your subjects are the literature and the citations that you're going to find.

So, probably, foundational to the quality of everything you do from here on out,

is the search that you're going to perform.

Almost all of the researchers I work with are surprised, to put it mildly,

at the amount of effort and time that it takes to do a high quality search.

I'm going to be going over the various steps and

processes that you need to do this.

The other major piece is developing your inclusion.

The next exclusion criteria.

Both of the readings,

the Institute of Medicine Standards on developing systematic reviews.

Talk about the need to include an information

specialist if you where going to do a systematic review in the future.

An excellent resource that's going to go into more detail that I can cover in

this lecture.

Is chapter six from the Cochrane Handbook which is also available in your readings.

Another tip.

If you can start thinking from the beginning to document what you do

at every stage.

You will be saving yourself a lot of trial and heartache later on.

Don't underestimate the difficulty that you're going to find

in identifying the articles that you need to include in your study.

Kay Dickerson has already talked to you about publication bias and

the problems with data not being published.

The area I'm going to be covering is the challenges in finding the published data.

You might be asking yourself some questions at this point.

Why is it so important to do such a comprehensive, extensive search?

Wouldn't it be okay to go to PubMed?

PubMed is a huge database with 20 million citations.

You find the articles,

look in the bibliographies of the ones that are included.

Does doing a good search really matter that much?

I can give you several examples, about how important

the quality of the search is to your research, and to research in general.

One of the most heartbreaking examples happened right here at Hopkins.

Somewhat over 10 years ago a researcher undertook a study on asthma.

And used healthy subjects and

bronchial challenge to mimic the experience of asthma.

I think I'm referring to the hexmethonium trial,

in which a healthy 24 year old research assistant.

Who was part of that study,

died because of the adverse effects from the drug that was used.

Once this became public knowledge,

librarians in various places started doing searches.

And it became clear that evidence information that would have

flagged this as a dangerous procedure was actually easily available.

Had the researcher consulted with an information professional and

done an extensive search.

The quality of your search is going to certainly effect

the quality of the research you do.

It can even have Impacts on individual's lives.

How do you do the search?

You're going to develop a, sort of, sub-protocol for the search itself.

What you're going to need to do is identify and document the sources you use.

This will include Large developed databases of citations and

abstracts in various areas.

Will also include additional processes,

like hand-searching, which these days can be done electronically.

It means reviewing tables of content, it means reviewing the reference lists,

sometimes following citation tracks.

You're going to need to document how you did the searches,

the dates you did them, the exact strategies that you used in the databases.

You're also going to need to determine how you make decisions about what you

include and exclude.

For systematic reviews and meta analyses to reduce the possibility of bias.

Duplicate screening is used, that is two people independently review each

citation and abstract to see whether it's included or not.

So, in the next session I'm going to give you some more information about the key

sources you're going to use and the various techniques for searching.

Explorer notre catalogue

Rejoignez-nous gratuitement et obtenez des recommendations, des mises à jour et des offres personnalisées.

Coursera propose un accès universel à la meilleure formation au monde, en partenariat avec des universités et des organisations du plus haut niveau, pour proposer des cours en ligne.

© 2019 Coursera Inc. Tous droits réservés.

Télécharger dans l'App Store

Disponible sur Google Play