| Methods | DESIGN: randomised controlled trial. BLINDING: double blind for probiotics/placebo, single blind for dietary intervention. UNIT OF COMPARISON: individuals. DURATION: supplementation with probiotic/placebo from early pregnancy until the end of exclusive breastfeeding. FOLLOW‐UP: 24 months postpartum. LOCATION: Finland. |
|
| Participants | TOTAL NUMBER: 256. No metabolic or chronic diseases. 7% of women were obese, 21% were overweight. |
|
| Interventions | PROBIOTIC: Lactobacillus rhamnosus GG, ATCC 53 103, Valio Ltd, Helsinki, Finland and Bifidobacterium lactis Bb12, Chr. Hansen, Hoersholm, Denmark, 1010 colony‐forming units/d each). PLACEBO: microcrystalline cellulose and dextrose anhydrate. DIETARY: intensive dietary counselling aiming to conform to currently recommended pregnancy diet. |
|
| Outcomes | PRIMARY: maternal glucose metabolism as measured by plasma glucose, blood HbA1c, serum insulin and HOMA and QUICKI indices at baseline, third trimester of pregnancy, 1, 6 and 12 months postpartum. | |
| Notes | NCT00167700 | |
| Risk of bias | ||
| Bias | Authors' judgement | Support for judgement |
| Random sequence generation (selection bias) | Low risk | Computer‐generated block randomisation of 6 women. The use of only 1 block size could make it possible to guess the randomisation of the dietary intervention of the last individuals of each block. However, since this randomisation was only blinded to the participants and not the study personnel, the selection bias risk is still considered to be low. |
| Allocation concealment (selection bias) | Low risk | Randomisation list generated by a non‐investigator statistician, sealed envelopes. |
| Blinding of participants and personnel (performance bias) All outcomes | Low risk | Placebo/probiotic allocation was blind to both participants and personnel, dietary therapy was not blinded to personnel. |
| Blinding of outcome assessment (detection bias) All outcomes | Low risk | All personnel who handled or analysed blood samples were blind to the intervention. |
| Incomplete outcome data (attrition bias) All outcomes | Low risk | Minimal loss to follow‐up by assessment of glucose tolerance. Total loss to follow‐up was 18.75% by 1 year postpartum. |
| Selective reporting (reporting bias) | Low risk | Reported all outcomes they intended to report. |
| Other bias | Low risk | No other biases detected. |
Hb: haemoglobin HOMA: homeostasis model assessment QUICKI: quantitative insulin sensitivity check index