A systematic review is one of the highest levels of evidence of evidence based practice. Other items that qualify as Level One evidence according to Evidence-Based Practice are meta-analyses and guidelines.
Do you know how sometimes multiple research studies on the same research question sometimes have differing results? One shows that an intervention is effective and another shows that it is not. Even when the researchers do everything correctly and control variables well, the results might vary. Unbeknownst to anyone, selected research subjects may have undiagnosed health conditions or an unidentified variable may be playing a role.
To compensate for the uncontrolled variation in the world, researchers may undertake a systematic review or meta-analysis. Researchers set out to find every relevant research study done on the identified research question. Then they group the results and compare the studies to see if, for example, 30% show that the intervention is effective or if 80% show the intervention is effective or if any trend is evident. If the collected studies are determined to be similar enough, then the results can be legitimately grouped and compared. It is statistically unlikely for all pooled results to display the observed difference, or greater, just by chance if the null hypothesis is true. It is unlikely that all studies would have the same errors or undefined variables IF their results were wrong. By pooling the results, it smooths out any local variables or circumstances that do not reflect the general population.
Because systematic reviews and meta-analyses require a search and identification of publications as a critical part of its methodology, librarians are often asked to act as reviewers for a systematic review to critique the search methodology in particular. The selected librarians have experience conducting systematic reviews themselves. And they often offer training to others in how to plan and execute an effective search for a systematic review.
A well planned and executed systematic review will minimize bias, retrieve all relevant empirical research, and increase the likelihood of publication.
Beware of using published systematic reviews as examples. Some papers use systematic review erroneously. Other journals--or even the same journal but with a new editor and reviewers--may decide not publish similar papers in the future.
The methodology for how to conduct a quality systematic review has not changed significantly over time. So older books are still good.
Now, your TOPIC may require you to consider recent changes and influences within your inclusion criteria...
Comprehensive Systematic Review for Advanced Practice Nursing, Second Edition by 
Research for Advanced Practice Nurses by 
Systematic Reviews to Answer Health Care Questions by 
How to Do a Systematic Literature Review in Nursing by 
Painless Evidence-Based Medicine by 
Systematic Synthesis of Qualitative Research by 
Systematic Reviews and Meta-Analysis by Because a systematic review is based on obtaining ALL relevant research studies and endeavors to provide the highest quality evidence to guide practice, the methodology of how those studies are sought and obtained and analyzed is VERY important. Your project will be evaluated based upon how well you follow best practices and control potential bias.
A systematic review project takes an average of 12-18 months to complete. Make sure you have the time to invest in it.
A team of a minimum of three people working on the systematic review project will minimize bias. Multiple people to select and evaluate the publications will minimize individual bias in the selection and critical appraisal process.
Not all questions are appropriate for a systematic review. Some better fit a scoping review or other type of review. Choose the methodology best suited to your research question.
Has enough research been conducted on your research question? If the question is too new or too narrowly defined, there may be few research publications for the SR.
Is your question too broad? Is it defined well enough that a search strategy can be designed that will retrieve studies on the topic that can be reliably compared and analyzed?
Are the published research projects similar enough in methodology so that the results can be compared? The question should be defined well enough in the research literature so that the research studies are comparable. A question that is too broad may be appropriate for another type of review project, such as a general review project or an evidence mapping project.
If your question has multiple steps or parts to its intervention that are not covered in each research publication, it may not be appropriate for a systematic review. A decision modeling model may be better suited.
Are you leaving out any part of your question? Do you have a completed PICO? Is it missing any parts?
Can you meet all of the requirements for a systematic review according to one of the systematic review organizations, like Cochrane or PRISMA?
The search will retrieve a very large number of studies. Can your institution and its library support getting the full-text? Is your team prepared to invest the time required to evaluate the records? To evaluate the narrowed list of relevant studies?
