BACKGROUND/RATIONALE
FACULTY OF HEALTH
20 Credit Level 7 Research Module
SUMMATIVE ASSIGNMENT – PROTOCOL FOR A SYSTEMATIC REVIEW (3,000 WORDS).
The assignment for this module has been designed to give all students the opportunity to prepare a document demonstrating their ability to produce a coherent, theoretically based argument justifying secondary research within a specific field. The development of a protocol is a key requisite for systematic review work. The format for this assignment is informed mainly by Cochrane resources for preparing protocols and students are strongly encouraged to visit www.cochrane-handbook.org.
Remember this assignment is basically a statement of intent; the protocol outlines the plan for the review and should describe the rationale for the review; the objectives; and the methods that will be used to locate, select and critically appraise the studies, and to collect and analyze data from the included studies. Preparing a protocol for a review makes you stop and think about what you’re doing; it can act as a working document for the reviewer(s); it can prevent duplication; and very importantly it can minimise bias by being transparent about what you plan to do in advance. The following guidelines have been devised to help in the preparation of the assignment.
THE TITLE
The title has to provide enough information to help the reader decide if the review protocol is going to be relevant to them. The Cochrane Collaboration has decided on a standard format for titles which helps convey information as quickly as possible:
[Intervention] for [Problem] in [Category]
E.g. Topical negative pressure (TNP) for treating chronic wounds.
BACKGROUND/RATIONALE
You need to indicate the time span, languages and countries from which background information has been gained.
You need to present the background literature as a logical discussion, if necessary using sub-headings for clarification. It should include the size of the problem, uncertainty about dealing with it, why the intervention might work, and what it is supposed to achieve.
You need to critically analyse and appraise the background literature in two ways, consider firstly the data/theory proposed, secondly whether the methods used to gather the data were appropriate.
You need to conclude in a way that identifies the issues which arise from the background literature and leads into, or suggests the need for the proposed review. By the time the reader finishes your Background section they should be able to understand why you are asking the review question.
OBJECTIVES/REVIEW QUESTION(S)
In this section the main review question(s) to be addressed in your protocol needs to be stated. Getting the review question right is the most important step in doing your protocol. As well as telling others what the review is about, it will guide how you propose to search, select, appraise and analyse your studies. Make sure you spend time on this section. It is recommended that you structure your question using PICOD, PICO or PIO – this will depend on the type of question being asked.
P = Population i.e. the people affected by the intervention/exposure
I = Intervention/exposure under scrutiny
C = Counter intervention
O = Outcome(s) of interest
D = Design of the studies likely to yield the most valid data
E.g. To undertake a systematic review of all randomised controlled trials (RCTs) using TNP in the treatment of any patient with a chronic wound to determine:
a) If TNP is more effective than wound dressings in terms of improving healing rates, reducing cost, improving quality of life, minimising pain and providing comfort.
b) If there is an optimum TNP regimen in terms of foam type, degree of suction, continuous/intermittent suction, duration of suction.
METHODS OF THE REVIEW
This section is the formal description of what you plan to do once you have decided on your review question:
SEARCH STRATEGY
In your protocol the search strategy needs to be clearly described. PI(C)O(D) will help determine key words, Medical subject headings (MeSH), wildcards, acronyms, synonyms, transatlantic terms that will be used in your search strategy. You will need to describe how these will be linked with the appropriate Boolean operators (e.g. AND, OR, NOT) to develop a search strategy that will be used to search for primary studies from a variety of resources, which you need to state, such as electronic databases, journals, conference proceedings, reference lists, grey literature, research registers, researchers and manufacturers.
STUDY SELECTION CRITERIA AND PROCEDURES.
Following the search you need to clearly describe in your protocol the processes that will be used to decide if a primary study will be included or excluded from the review. This will initially depend on whether it fulfils the scope (PI(C)O(D))of the review protocol.
For types of participants you need to think about the health problem or population or setting.
E.g. You would consider a study fulfilling the scope of the TNP review protocol if it defined chronic wounds as…
For types of interventions with a medication for example, you need to think about drug preparation, route of administration, dose, duration, frequency. For non-drug interventions such as an educational intervention, defining the intervention can be a bit more difficult – you need to consider exactly what was done, how often it was done, who did it, were they trained, etc.
E.g. You would consider a study fulfilling the scope of the TNP review protocol if it described the intervention of interest (in this example TNP) as…
For types of counter interventions you need to decide whether you will be comparing the intervention group with a placebo, nothing, sham treatment or some other treatment.
E.g. You would consider any RCTs as fulfilling the scope of the TNP review protocol where topical negative pressure was compared with no treatment/sham/standard (which needs to be defined)/other experimental interventions.
For types of outcome measures you need to think about the primary outcome of interest and how that can be determined in a valid and reliable way. There may be secondary outcomes of interest e.g. cost, quality of life, pain, comfort and adverse effects and how they too can be determined in a valid and reliable way.
E.g. As there is no consensus as to the most valid and reliable means of measuring healing rates of wounds, you would consider a study fulfilling the scope of the TNP review protocol if it measured healing by some objective method such as the time to complete healing.
For types of studies you need to consider the design that will best answer the question and whether you will restrict studies on the basis of language, date or publication status.
E.g. To evaluate topical negative pressure for the treatment of any patient with a chronic wound – you would consider a study fulfilling the scope of the TNP review protocol, in the first instance, if it was a randomized controlled trial – and in the absence of any RCTs, controlled clinical trials would be considered.
STUDY QUALITY ASSESSMENT CHECKLISTS AND PROCEDURES
If the study fulfils the scope PI(C)O(D) of the review you need to clearly describe in your protocol the processes used to determine the methodological quality of the primary study to decide if it will remain included, or be excluded, from the review. It is best to refer to www.cochrane-handbook.org. in order to determine the best way of assessing bias in a study. The CONSORT statement also provides useful guidance for what ought to be reported in RCTS at www.consort-statement.org.
DATA SYNTHESIS
If the study fulfils the scope of the review, and is of high methodological quality, the results of this study need to be extracted and considered for data synthesis. Data synthesis involves collating and summarising the results of included primary studies You need describe in your protocol your proposed method of extracting and synthesising the results of primary studies and how this will depend upon heterogeneity of the studies identified. You need to consider whether your synthesis would provide an estimate of overall effectiveness of an intervention; review effectiveness in different studies, populations and settings; investigate differences; answer the review question.
Within your protocol you need to consider how a data extraction form might be designed to collect all the information needed to address the review question(s) which includes the name of the review, date of data extraction, publication details, the scope (PI(C)O(D)), methodological quality (specific factors), and results of the primary studies.
REFERENCES
All references should be appended to the protocol using the Harvard system.
Guidelines for preparing a protocol for a systematic review approach/level 7/Dr Debra Evans/September 2010
