In my opinion

The Population (P)  - Adult obese patients with Type II diabetes

The Intervention (I) - Bariatric surgery or dietary approaches alone

Comparison groups (C)- The study will be comparing patients who have undergone bariatric surgery vs those who have had dietary approach alone

Outcome-(O)- Diabetic Remission

Hope this helps

I agree with the previous answer.

If you define that the "new approach"-new intervention is the bariatric surgery (and the "old approach"-the standard practice is the dietary alone), than the I-Intervention  would be the bariatric surgery and the C-comparison would be with the dietary alone.

To be very clear: If indeed bariatric surgery is the "new" approach:

• Population: Obese adults with type II diabetes
• Intervention: Bariatric surgery
• Comparator: Dietary approaches alone
• Outcome: Diabetic remission
• Time-frame: At 12 months (for example)

You may like to build a bit of flexibility into the time-frame (e.g. "10-12 months" or whatever is appropriate) as one often finds that investigators on different studies assess outcomes for the same interventions at different intervals. Of course, you would need to be cautious in deciding the width of that interval.

Another option could be to omit the "T" entirely (leaving only PICO) and include studies reporting diabetic remission at any reported follow-up. I don't know the diabetes literature, but we have often used only PICO in systematic reviews of interventions in HIV/AIDS and other areas.

Thank you for your answers!

Confusing for me was the fact that there is no RCT published within the time limit set (2013-2017) comparing bariatric surgery alone with dietary approach alone. They compared either bariatric + diet with diet alone, or bariatric + medical treatment/lifestyle changes with medical treatment/lifestyle changes alone.

Are those aforementioned studies also eligible to address my question? Should I thus change my C group?

I had a similar situation before.

First, that is a rather narrow range of years. If it is possible to do so, and you are most keen to know bariatric surgery's effect without co-interventions, you might consider expanding the search horizon.

Also, if you are currently considering to include only RCTs, you might expand that to include observational cohort studies.

Perhaps more usefully: you could consider doing a review of I: Any intervention or bundle of interventions that include bariatric surgery, with diabetic remission as an outcome; vs. C: Any intervention or bundle of interventions that do NOT include bariatric surgery, with diabetic remission as an outcome.

Studies may all be quite varied and heterogeneous, as you describe, but perhaps some will emerge that are strictly bariatric surgery vs. diet. It may be possible to meta-analyze those data.

If there are at least two studies with similar design, populations and combinations of interventions vs. comparators, you may be able to pool those too. In such cases, it is ideal when investigators report the specific impact of individual intervention components (thus potentially enabling analysis of bariatric surgery in isolation), but is often not done, and may not be possible.

Although I'm not familiar with this literature, it's possible that you would only find studies testing various combinations, no two quite the same. Whether or not any data are pooled, you would synthesize the evidence in a narrative analysis.

Hope all this helps!