Intro to systematic reviews

Question 1

What is a systematic review?

Answer 1

• An objective mechanism of summarising research evidence
• Rigorous and based on a protocol
• Can be applied to any type of litreature- epidemiological, randomised trials, observational studies
• Qualitative research, diagnostic tests, etc
• Useful for synthesising large volumes of info
• With the use fo stats techniques- becomes a meta analysis

Question 2

What are some differences between systematic reviews and traditional reviews?

Answer 2

Sys review:
- focused clinical question
- Explicit searc strategy of multiple databases
- Comprehensive sources
- Criteria based selection; uniformly applied
- Rigorous critical appraisal
- Summary = quantitative summary/ Also qualitative/narrative
- Inferences= based on available evidence
- All evidence = graded equally

Question 3

What are the procedures involced in a systematic review?

Answer 3

• Define the research question
• Develop a protocol for the review
• Determine the eligibility criteria
• Search the litreature
• Read, critique and assess the quality of the studies
• Summarise the evidence
• Interpret the findings

Question 4

What should you do when defining a research question?

Answer 4

• make sure it is specific
• do a preliminary search to assess wheter relevant litreature exists
• Decide what your primary outcome will be?
• Do you have secondary outcomes?
• Exactly what information are you looking for?

Question 5

What should you think about in advamce?

Answer 5

• developing inclusion and exclusion criteria based on relevant variables
• (Age, setting, types of participants, type of study, type of papers etc)

Question 6

What is a good search strategy?

Answer 6

• sensitive , specific and systematic
• search terms developed (abbrebiations , english vs american english etc)

Question 7

How to prepare for the project report?

Answer 7

• must keep accurate record using endnote?
• produce accurate flow diagram of the search
• show how many potentially relevant papers you found at each stage of the search and evaluation process
• show how may papers were rejected at each stage and why?

Question 7

What sould be included in a data extraction form?

Answer 7

Design a data extraction form (usually a table in excel) to record relevant
variables such as:
the publication date,
population, ethnicity (or species, strain etc..),
sample size (n=xx)
gender
methods used
setting
key results
limitations
notes (e.g., conflicts of interest etc..)
Data extraction should be guided primarily by the primary and secondary
endpoints/outcomes

Question 7

What is publication bias?

Answer 7

• thought to have occured when published research findings appear unpresentative of the populations studies
  -Reasons:
• tendency to publish only positive findings (scientists and editors can contribute to this
• Pharm industry -reluctant to publish negative/neuutral findings
• Financial interests/ social pressures
• Experimenter bias