Develop PICO Questions
Topic refinement is the first step to development. Key questions outline the specific populations, treatments, and outcomes that will be addressed in the practice parameter. Topic refinement and the literature search processes are “complementary”.[i]
Key Questions should be developed using a PICO(TS) framework (see Appendix D). Key Questions should be focused on the topic and are developed to inform the literature search strategy (see section 4 below).
PICO
- Patient, Population, or Problem: The "P" in PICO(TS) is a description of the patient(s) of interest. It includes the condition(s), populations or subpopulations, disease severity or stage, comorbidities, and other patient characteristics or demographics.
- Intervention or Exposure: The "I" in PICO(TS) refers to the specific treatments or approaches with the patient or population. It includes dose, frequency, method of administering treatments, and so on.
- Comparison: Comparisons of the interventions.
- Outcome: Outcome of interest. This includes short- and long-term outcomes, including patient-centered outcomes, such as quality of life, mortality, morbidity, and other complications.
- Timing (If Applicable): Duration of the outcome of interest.
- Setting (If Applicable): The "S" in PICO(TS) describes the setting or context of interest. Setting can be a location (such as primary, specialty, or inpatient care) or health policy that frames or restricts the important questions to be answered.
The participation and collaboration of a multidisciplinary group of Task Force members are essential to address all potential populations, treatments, outcomes, and comparisons of interest.
When developing Key Questions, Task Force members should use the PICO(TS) framework and expert Task Force members to develop clear and defined questions to guide the literature search.
A strong question is one that helps guide the research and can be addressed by a review of the evidence. Questions inappropriate for systematic reviews include those that involve clinical judgment, guidance for an individual patient, are vague or limited to a single procedure, or ask about general approaches to treatment. Practice parameters will not exceed one “chapter”.
Develop Literature Search Strategy
The Task Force should discuss a basic search strategy, including MeSH terms, inclusion and exclusion criteria (such as study type, date range for searchers, etc.), before the search is performed. Task Force members should inform the search by providing a list of suggested terms and/or work collaboratively with the medical librarian to develop the search strategy.
If a Task Force uses external medical librarian services, the Task Force Chair should communicate the preliminary search strategy to the Medical Librarian performing the search, and the final search strategy must be submitted to CNS staff. It is important that all final search strategies are kept on file for referral when a topic is reviewed/updated.
Literature Search
An extensive literature should be undertaken. At a minimum, the search should include the available English-language literature for a length of time appropriate to the subject using the computerized database of the National Library of Medicine. This will vary depending on whether the practice parameter is an original project or an update of a previous version. The search terms employed should reflect the clinical question in as much detail as is relevant to the key questions and PICO. The Task Force should also work collaboratively with the Medical Librarian to determine appropriate inclusion/exclusion criteria to guide the search. At a minimum, the search should only include studies done on humans, not involving “in vitro”, “biomechanical”, or results performed on cadavers. Criteria should also limit studies whose results are presented quantitatively and the study must be published in English. Systematic reviews, meta-analyses, or practice parameters developed by others are typically not considered as evidence to support practice parameters.
After the Task Force has reviewed and approved the search strategy, the Medical Librarian will run searches for each chapter using at least two electronic databases (the following are most common):
After the Medical Librarian completes the search, he or she will provide the chapter team members with a complete list of all abstracts yielded by the search. This list should be complete and “de-duplicated.”
Literature Search Results Screening and Article Retrieval
Abstracts are reviewed by the task force, and clearly relevant articles will be selected for full text evaluation. Team Members may request retrieval of full-text articles by CNS staff. The Task Force must record and submit the detailed search strategy and flow chart summary of article inclusion/exclusion to CNS staff.
Evidence Analysis
The purpose of practice parameters is to provide guidance in the absence of high quality evidence. Therefore, practice parameters will not include high quality evidence. However, when evidence is available it will be evaluated and inform “expert opinion”. When evidence does not exist and/or only low quality evidence exists, task force members will use expert opinion. Task Force members’ clinical and research experience are invaluable in assessing the literature and developing the practice parameters.
“Some outcomes, trials or studies cannot be or have not been performed and practice is informed only by expert opinion.” Expert opinion is defined as “opinion without explicitly critical appraisal, or based on physiology, bench research, or ‘first principles’.”[ii]
Statements using expert option will be “graded” as “Expert Opinion”. The Practice Parameters will also include the above definition and a disclaimer regarding the use of expert opinion.
[i] Agency for Healthcare Research and Quality. (October 2010) Methods Guide for Medical Test reviews, Paper 2- Developing the Topic and Structuring the Review: Utility of PICOTS, Analytic Frameworks, Decision Trees, and Other Frameworks. http://www.effectivehealthcare.ahrq.gov/ehc/products/248/560/Paper02_(Developing_Topic)_29_Oct_10.pdf, accessed March 28, 2016.
[ii] Ransohoff DF, Pignone M, Sox HC. How to decide whether a clinical practice guideline is trustworthy. Jama. 2013;309(2):139-140.