Pre-Capstone Guide for MPH Students

The phrase "systematic review" can actually refer to a number of different kinds of literature reviews, each of which has a different purpose and is performed differently. On the other hand, if you Google "systematic review," you're likely to find instructions and advice for conducting one very specific type of review. This is the type that academic journals usually publish, and it involves (among other things) more than one person reviewing the articles to decide if they're relevant to the topic. The article below describes other kinds of systematic literature reviews. You can discuss which type you're doing or should do with your faculty advisor, so that it will be easier for you to find instructions that are relevant to your project. 

Grant, Maria J., and Andrew Booth. "A typology of reviews: an analysis of 14 review types and associated methodologies." Health Information & Libraries Journal 26.2 (2009): 91-108.

This library guide, from Duke University, summarizes the types in the Grant and Booth article:

Systematic Reviews: Types of Reviews (Duke University)

This video, made by a librarian at the University of Southern California, explains the difference between a systematic review and a literature review, or systematic literature review, which is more similar to what you are likely doing for your Capstone project.

This article gives a good outline of the steps you'll need to take to complete a systematic literature review:

Pautasso, Marco. Ten simple rules for writing a literature review. PLoS Comput Biol. 2013;9(7):e1003149.

Depending on what kind of review you're doing, the requirements may be slightly different, but most systematic literature reviews will require the following: 

1. A formal research question: What exactly do you want to find out? 
2. Inclusion/exclusion criteria: Based on your research question, what kinds of articles will you include in your review? What kinds will you exclude, even if they turn up in your literature search? Some possible examples of inclusion/exclusion criteria are:
   1. Date: Decide how far back you want to go before you start searching. Do you think that articles published in 2001 are still relevant? Articles published in 2015? Some of this will depend on your topic, and how much research you can find on it. 
   2. Study type: Do you want to limit the articles you review to randomized clinical trials (RCTs)? Will you include qualitative or observational studies? 
   3. Language: Will you include studies in languages other than English in your review? 
   4. Is there a particular population/condition/intervention that is similar to what you are interested in, but not exactly the same? For example, you might decide that you will look at ADHD in children, but exclude articles that are about ADHD in adults. 
3. A search strategy: Decide what search terms you will use and which databases you'll search (you may want to contact your friendly MPH librarian for help). Most importantly, keep track of the terms and which databases you use them in. The idea is for other researchers to be able to at least roughly reproduce your search (it's not possible for the results to be exactly the same, since new articles are published every day). For information on how to construct a good search strategy, check the Search Strategy: Basic and Search Strategy: Advanced page on the Health Sciences Research Skills library guide. 
4. Citation management system: Any kind of systematic review involves searching more than one database, so you'll probably want some kind of citation management software to help you deal with duplicate articles. Zotero and Mendeley are good free options, and Endnote Desktop is a good paid option. (See the library guides for Mendeley and Endnote on this page). 
5. A flowchart showing which articles you found in which database, and how many were eliminated as not relevant or duplicates at each stage in the process. Keep track of these numbers as you go--they're really difficult to recreate after you've removed all the duplicate articles from your citation management system. The PRISMA chart here is the one most people use, and it is easy to fill out. Just download either the Word or PDF version and enter your numbers. 

The PRISMA diagram or flow chart is the standard way that researchers report their process of selecting articles to include in their systematic review. Below is a link to the actual diagram (when you submit a systematic review for publication in an academic journal, you're required to fill this out) and another link to academic articles that explain what it is and how it works.