Systematic Reviews (in the Health Sciences)

After assembling a team and selecting a general topic where there appears to be abundant evidence, the team needs to develop a precise, specific, and answerable research question. In evidence synthesis research, an "answerable" question is one where there is enough volume and similarity of basic research evidence to combine the results.

To focus and finalize your research question, you need to have identified several original research articles that examine your topic using similar research processes. Examine the research in five to ten papers and make a casual tally sheet noting very specific differences:

• Age, gender, geographic distribution, economic factors, disease stage or severity, co-existing conditions, and other relevant factors about the included patients. The exact relevant factors will change based on the question.
• What interventions (drugs, treatments, etc.) are being used? Note specific names of drugs and classes, specific surgical techniques, names of genes, etc., rather than generalities like "antidepressants" or "cardiac surgery."
• What kind of research is being done? Is it prospective or retrospective? Chart review? Randomized controlled trials? Observational studies?
• What scales or measurements are being used? Again, be specific: Beck Depression Inventory, Hamilton Depression Rating Scale, self-created questionnaire, etc.
• Over what time period did data collection take place?

Using this as a guide, try to determine which questions could be answerable based on what appears to have been collected. You may be initially interested in determining if providing antidepressants in the long term actually reduces depressive symptoms for all adults, but find that the studies do not allow you to answer this question.  Perhaps the characteristics of publications could allow you to focus on a specific part of this larger question:

• For adults who have been taking lithium or tricyclic antidepressants to successfully treat depression for at least 5 years, does continued treatment with lithium or tricyclic antidepressants reduce risk of relapse of depression as measured by the Hamilton Depression Rating Scale?
• For adults with postnatal depression, does 6 months of treatment with sertraline or paroxetine, or 6 months of weekly psychotherapy visits, have the same efficacy for reducing depression when measured 12 months post-birth?
• For adults newly diagnosed with major depression, is treatment with an SSRI alone, or treatment with an SSRI and benzodiazepine more effective at reducing depressive symptoms over 12 weeks of treatment?

This specific and focused research question will be used to guide all the further steps of this research process, so it is worthwhile to spend the time to ensure you have a question that reflects what is published, and is likely to be able to be completed.

The questions provided above are structured using a format called PICO. PICO is an acronym developed for evidence-based practice related work, to help succinctly and effectively communicate four required elements of clinical questions. Using PICO to help formulate and finalize your question can:

• ensure you have an answerable clinical question
• help support your search development
• ensure your final manuscript will communicate successfully with readers
• provide a short, succinct summary of your key interests, serving as a good reminder for the team while conducting the 12-18 month long process of the systematic review.

PICO was developed for clinical questions and quantitative research. It may not work as well for questions that are measured more qualitatively or using mixed methods research. Review these additional guidelines and readings for ways to use frameworks like PerSPEcTiF, SPIDER, PICOS, and more.