The first step in performing a Systematic Review is to develop your research question. Your question should be focused, i.e., not overly broad. To focus your question, use an approach called PICO.
What is PICO?
PICO is an acronym for:
P = Patient, Population, and/or Problem
I = Intervention
C = Comparison (not always applicable)
O = Outcome
Some researchers also add a "T" to the framework to define the Type of study that is best to answer the question that he or she is formulating. Or, alternatively, Time, which might look at variable lengths of time for the intervention outcome.
Other Frameworks: SPIDER & SPICE
SPIDER...
S = Sample. The group of people being looked at, because qualitative research is not easy to generalize, sample is preferred over patient.
P = Phenomenon of Interest - reasons for behavior and decisions, rather than an intervention.
D = The Design of research used, such as interview or survey.
E = Evaluation or outcome measures.
R = Research type, i.e., Qualitative, quantitative and/or mixed methods.
More information on the SPIDER method can be found at the link below.
Beyond PICO: The SPIDER tool for qualitative evidence synthesis Qual Health Res October 2012 vol. 22 no. 10 1435-1443
SPICE is used mostly in social science and healthcare research.
S = Setting: the location or environment relevant to your research (e.g. accident and emergency unit)
P = Population (or perspective): the type of group that you are studying (e.g. older people)
I = Intervention: the intervention/practice/treatment that you are evaluating (e.g. initial examination of patients by allied health staff)
C = Comparator: an intervention with which you compare the above comparator (e.g. initial examination by medical staff)
E = Evaluation: the hypothetical result you intend to evaluate e.g. lower mortality rates)
Clear and Present Questions: Booth 2004
The most important point is that you focus your question and design the question so that it is answerable by the research that you will be systematically examining. In the study mentioned before by Venekamp the population is Children who have Acute Otitis Media, the intervention is antibiotics, the comparator is placebo (i.e. no treatment), and the outcome is what the research found. In this particular study, Venekamp also looked at outcomes over time (24 hours, two to three days, four to seven days, and ten to twelve days -- any abnormal outcomes were extended into weeks).
Once you have your PICO defined, you can frame it as an answerable question: Should a medical practitioner prescribe antibiotics for children with Acute Otitis Media?
Once you've defined your question, you should run a few quick literature searches and check the systematic review registries to discover if your question has already been asked, or if there are other systematic reviews that are similar to that which you're preparing to do. If you know that there are no other systematic reviews on your subject of interest, then move on to Step 2: Registration.