 |
 |
Hi Sam
In case you haven’t already read chapter 12 of the Cochrane handbook, I’d recommend looking at the worked examples it provides. The example review looked at effects of midwife-led care and the overarching outcome was maternal satisfaction
of care, which combined various types of outcomes – it might give you inspiration about how valid it would be to combine all your outcomes.
https://training.cochrane.org/handbook/current/chapter-12
(If the webpage says a Cochrane account is required, it is free to register for a Cochrane account).
I’d be looking for the highest/broadest level that the outcomes could be legitimately grouped. If your review question is exploring whether health literacy interventions work/result in beneficial outcomes, then perhaps the studies could
all be combined. However, I wonder whether there is a difference between outcomes where health literacy has an impact on different types of health, and outcomes more closely related to health literacy ability (e.g. recognition of medical words). Please note,
I don’t have expertise in the health literacy field.
From your description of the outcomes, would it be valid to group the outcomes into ‘health outcomes’ and ‘knowledge related outcomes’ within the face to face and distance groups? Would there be enough data in each/most of these groups
to allow a meaningful synthesis? If the groups get so disparate that the synthesis is merely a report of the results of each individual study, this probably reduces the usefulness of the review for your end users.
Thinking about what will be useful for those who use your review, and being able to include a clear justification about why you grouped the outcomes as you did, are tactics that help me figure out how best to combine studies/data.
I hope this helps, and glad to hear any comments/opinions from others.
Best wishes
Mhairi
Mhairi Campbell
Investigator Scientist
MRC/CSO Social and Public Health Sciences Unit
University of Glasgow
Berkeley Square
99 Berkeley Street
Glasgow G3 7HR
From: Synthesis Without Meta-analysis for systematic reviews <[log in to unmask]>
On Behalf Of Samantha Belfrage (S.Belfrage.20)
Sent: 14 June 2021 10:18
To: [log in to unmask]
Subject: SWiM Support
Hello everyone,
I'm Sam a first year PhD student. I'm working on a systematic review exploring at the effectiveness of community-based health
literacy interventions for parents. I have a question regarding choosing the most appropriate standardized metric.
From my 11 included studies I have 10 different health literacy tools used to measure different aspects of health literacy (mental health, oral health, immunization knowledge,
recognition of medical words etc.).
All studies have a pre and post intervention measure of health literacy (mean), all but one study include the standard deviation.
The studies are going to be grouped by mode of delivery (face to face and distance).
I'm wondering if it's appropriate for me to calculate the effect size from the mean differences of the two groups or if that is inappropriate because the outcome measures are so
diverse?
Any thoughts or ideas greatly appreciated.
Many thanks
Sam
To unsubscribe from the SYNTHESIS-SWIM list, click the following link:
https://www.jiscmail.ac.uk/cgi-bin/WA-JISC.exe?SUBED1=SYNTHESIS-SWIM&A=1
To unsubscribe from the SYNTHESIS-SWIM list, click the following link:
https://www.jiscmail.ac.uk/cgi-bin/WA-JISC.exe?SUBED1=SYNTHESIS-SWIM&A=1