PICO Framework: Framing your research question
What is the PICO Framework?
The PICO framework — the acronym stands for Patient/Population, Intervention, Comparison and Outcome — is widely used in evidence synthesis and evidence-based medicine. It is a well-known tool to structure a clinical question and facilitate searching for a precise answer in the literature.
PICO has been widely adopted since its first introduction by Richardson in 1995 [1]. The Cochrane Handbook [2] prescribes the use of the PICO model at different stages in developing the review question for Cochrane Reviews. Although the PICO framework is mainly intended for therapy-related clinical questions, it can also be used for questions related to diagnosis or prognosis [3].
A question framework such as PICO allows you to delineate the scope of the review more precisely, reflect on sources and search methodology to use and serves as a starting point for setting up search methods and searches.
Is PICO required for the MDR?
The use of PICO is not mandatory for MDR-compliant literature reviews. In fact, the MDR does not even mention PICO. The MDR does, however, require literature reviews to be systematic and thorough, and the literature review methodology to be reproducible and documented. That is why using the PICO framework for MDR-compliant literature reviews is a good choice. With the additional advantage that PICO is well-known and widely accepted by notified bodies [4].
Advantages of using the PICO Framework
Using PICO allows you to:
- Use a structured approach to build literature searches
- Create transparent documentation of your search strategy
- Make your literature search method easy to read and understand
How to use PICO effectively in the clinical evaluation of medical devices
Use PICO to set up your literature searches in these 3 easy steps:
- Define the patient, intervention, comparison and outcome categories
- List appropriate keywords for each PICO category
- Use Boolean operators to build your query:
OR between the keywords of a PICO category
AND to combine the PICO categories into a single search
An example to get you started:
In case a full PICO search does not retrieve sufficient results, consider using only the P (patient) and I (intervention) keywords for your search, as the C (Comparison) and O (Outcome) elements of the PICO have been shown to result in a lower retrieval[5]. Alternatively, you could use PICO variants such as PICOT [6] or one of the many PICO alternatives (SPICE, SPIDER, PEO, PIRATE a.o.) described below to structure your review question.
PICO Variants
PECO
A PICO alternative for non-interventional studies, where the intervention item has been replaced by exposure.
PICOT
The T in PICOT stands for time frame, indicating that in the PICOT framework, the time frame over which the outcomes are assessed is added to the standard PICO elements [7,8].
PICOS
In the PICOS framework, the classical PICO elements are supplemented with study type (S). Adding the specific study types is useful to tailor your eligibility criteria to a specific set of study designs, geared towards the evidence levels needed in the review. Using PICOS is a valid option to reduce the number of articles retrieved by a search without losing relevant hits when time and/or resources are limited [9].
A PICOS framework could be used to restrict search results to clinical trials for high-risk devices in a highly publicized domain or limit the search for longitudinal studies, real-world evidence and registries for searches to retrieve data for post-market surveillance.
PICOC
The PICOC acronym adds the Context item to the standard PICO framework. The context item allows to tailor searches to the setting, circumstances, or conditions in which the intervention is applied. An example for such a context would be reintervention to replace (a part of) an implanted medical device, to gather evidence whether the replacement part and procedure are not adversely affecting the safety and performance of the device.
PICO Alternatives and when to use them
PICO was the initial question framework for systematic literature reviews and remains widely used, but a plethora of other question frameworks have been developed. The common thread these frameworks all share is their aim to conceptualize literature search and help the searcher to specify the inclusion and exclusion criteria of the review.
To illustrate the plethora of existing frameworks for systematic reviews: Booth et al. identified 38 different question frameworks for systematic literature reviews [10]. We highlight here the ones that are most useful for literature reviews on medical devices.
SPICE
|
Item
|
Definition
|
Example
|
|---|---|---|
|
Setting
|
The context for the question
|
Parkingson's disease
|
|
Perspective
|
The users, stakeholders or population involved
|
Patients, individuals, caregivers, healthcare professionals, neurologists, surgeons
|
|
Intervention
|
The intervention or action taken
|
Deep brain stimulation
|
|
Comparison
|
The outcomes used to determine the success of the intervention
|
Levodopa, stimulation off periods
|
|
Evaluation
|
The outcomes used to determine the success of the intervention
|
Bradykinesia, motor function, quality of life, symptom control
|
For MDR-compliant literature on medical devices, the SPICE framework can be useful for devices restricted to a very specific setting and user group. But keep in mind that for the clinical evaluation under the MDR manufacturers are required to express the clinical benefit of their device in terms of measurable clinical outcomes that are relevant to patients.
PEO
When conducting searches for qualitative, non-numeric outcomes, the PEO framework, consisting of the concepts population, exposure, and outcome is often used.
|
Item
|
Definition
|
Example
|
|---|---|---|
|
Population
|
The individuals, patients, users or group exposed
|
Children with type 1 diabetes
|
|
Exposure
|
The condition, intervention, risk or factor under study
|
New glucose sensing wearable vs similar devices already on the market
|
|
Outcome
|
Experiences, attitudes, feelings, improvement in condition, mobility, responsiveness to treatment, care, quality of life or daily living
|
Usability, adherence, quality of life
|
SPIDER
The SPIDER tool was specifically designed to identify qualitative and mixed-method studies [9,12]. In comparison to the PICO framework, using the SPIDER framework will yield less, but more precise and specific results [9]. This makes SPIDER a good method in domains where a lot of published data is available and time and/or resources are a limitation. When using the SPIDER framework, it remains important to address the risk of not identifying potentially relevant publications [9].
To build a search with the SPIDER framework, the keywords for each category are concatenated as follows [12]:
(S AND P) AND ((D OR E) AND R)
The sample category of the SPIDER framework maps more or less to the PICO population, whereas the phenomenon of interest can also be an intervention, and the evaluation item maps to the PICO outcome [8]. Where the PICO comparison item refers to the control intervention or placebo, the SPIDER design item is more tailored to naming the type of data collection and analysis to be used, especially in qualitative research which is mostly non-comparative [9].
|
Item
|
Definition
|
Example
|
|---|---|---|
|
Sample
|
Participants selection
|
Parkinson patients with a deep brain stimulator
|
|
Phenomenon of Interest
|
May include experiences, behavior or decisions in addition to an intervention
|
Quality of life Independence in activities of daily living
|
|
Design
|
Research method
|
Patient-reported outcomes questionnaires
|
|
Evaluation
|
Outcome parameters
|
Parkinson's Disease Questionnaire-39 Schwab & England Activities of Daily Living
|
|
Research type
|
Qualitative, mixed-methods or quantitative
|
Quantitative
|
PIRATE
The PIRATE framework is used to structure systematic reviews of diagnostic test accuracy [13], alongside its less elaborate (and frankly also less memorable) equivalent PIRD (cf. infra) [14]. It can be a suitable framework to structure literature reviews on in vitro diagnostic devices to meet the requirement of EU 2017/746 (IVDR).
|
Item
|
Definition
|
Example
|
|---|---|---|
|
Population
|
Population/participants
|
Patients with influenza like illness
|
|
Index test
|
Test under evaluation
|
Diagnostic test for Influenza A H1 N1 (swine flu)
|
|
Reference test
|
Comparator test
|
Viral culture
|
|
Accuracy methods
|
How will test accuracy be measured?
|
Sensitivity, specificity, predictive values
|
|
Test cut-off point
|
How will values be dichotomized into positive vs negative
|
Diagnostic cut-off
|
|
Expected test use
|
Role of the index test, e.g. diagnosis, additional test, triage and setting
|
Diagnosis of influenza Outpatient settings, emergency departments, clinical laboratories, community health centers
|
Need more acronyms?
Below is a non-comprehensive list of additional question frameworks for systematic literature reviews and for structuring clinical and health related questions.
|
Framework
|
Components
|
Usage
|
|---|---|---|
|
BeHeMoTh
|
Behavior of interest Health Context Models or Theories
|
Specific models or theories regarding health behavior
|
|
CIMO
|
Context Intervention Mechanisms Outcomes
|
Evaluation of management and policies
|
|
CoCoPop
|
Condition Context Population
|
Etiology or the association between risk or exposure and outcomes
|
|
ECLIPSE
|
Expectation Client group Location Professionals Service
|
Health policy and management information
|
|
PerSPECTIF
|
Perspective Setting Phenomenon/problem of interest Environment Comparison Time/Timing Findings
|
Clinical questions about the effectiveness of interventions
|
|
PIPOH/PIPOS
|
Population Intervention Professionals Outcomes Health care setting/context
|
Evaluation of practice guidelines
|
|
PIRD
|
Population Index test Reference test Diagnosis of interest
|
Diagnostic test accuracy
|
|
ProPheT
|
Problem Phenomenon of interest Time
|
Qualitative research assessing complex interventions
|
References and reading
- Richardson WS, Wilson MC, Nishikawa J, Hayward RS. The well-built clinical question: a key to evidence-based decisions. ACP J Club. 1995;123:A12-3.
- Higgins J, Thomas J, Chandler J, Al. E. Cochrane Handbook for Systematic Reviews of Interventions [Internet]. 2nd edn. John Wiley & Sons; 2019.
- Huang X, Lin J, Demner-Fushman D. Evaluation of PICO as a knowledge representation for clinical questions. AMIA Annu Symp Proc. 2006;359–63.
- Medical Devices Coordination Group (MDCG). MDCG 2020-13 Clinical evaluation assessment report template. 2020.
- Frandsen TF, Bruun Nielsen MF, Lindhardt CL, Eriksen MB. Using the full PICO model as a search tool for systematic reviews resulted in lower recall for some PICO elements. J Clin Epidemiol. Elsevier Inc; 2020;127:69–75. https://doi.org/10.1016/j.jclinepi.2020.07.005
- Rios LP, Ye C, Thabane L. Association between framing of the research question using the PICOT format and reporting quality of randomized controlled trials. BMC Med Res Methodol. 2010;10. https://doi.org/10.1186/1471-2288-10-11
- Thabane L, Thomas T, Ye C, Paul J. Posing the research question: Not so simple. Can J Anesth. 2009;56:71–9. https://doi.org/10.1007/s12630-008-9007-4
- Beck LL. The question: types of research questions and how to develop them [Internet]. Handb Des Conduct Clin Transl Surg. Elsevier Inc.; 2023. https://doi.org/10.1016/B978-0-323-90300-4.00107-5
- Methley AM, Campbell S, Chew-Graham C, McNally R, Cheraghi-Sohi S. PICO, PICOS and SPIDER: A comparison study of specificity and sensitivity in three search tools for qualitative systematic reviews. BMC Health Serv Res. 2014;14. https://doi.org/10.1186/s12913-014-0579-0
- Booth A, Noyes J, Flemming K, Moore G, Tunçalp Ö, Shakibazadeh E. Formulating questions to explore complex interventions within qualitative evidence synthesis. BMJ Glob Heal. 2019;4:1–7. https://doi.org/10.1136/bmjgh-2018-001107
- Cleyle S, Booth A. Clear and present questions: Formulating questions for evidence based practice. Libr Hi Tech. 2006;24:355–68. https://doi.org/10.1108/07378830610692127
- Cooke A, Smith D, Booth A. Beyond PICO: The SPIDER tool for qualitative evidence synthesis. Qual Health Res. 2012;22:1435–43. https://doi.org/10.1177/1049732312452938
- White S, Schultz T, Enuameh YAK. Synthesizing Evidence of Diagnostic Accuracy. 2011.
- Campbell JM, Klugar M, Ding S, Carmody DP, Hakonsen SJ, Jadotte YT, et al. Diagnostic test accuracy. Int J Evid Based Healthc. 2015;13:154–62. https://doi.org/10.1097/XEB.0000000000000061
