| Ref ID | 1017 |
| First Author | K. Ginell |
| Journal | CHILDREN |
| Year Of Publishing | 2023 |
| URL | https://www.ncbi.nlm.nih.gov/pmc/articles/PMC10605774/ |
| Keywords |
• Endocrinology • Paediatrics • Error • Statistical |
| Problem(s) |
• Errors in effect estimate calculations or data synthesis • Errors in study inclusion or omission of relevant studies |
| Article Type | Editorial |
| Article Subtype | Comment/letter |
| First Author Country | United States |
| Aim | A letter to the editor regarding the conduct of a meta-analysis in a systematic review published in 2021 on the effectiveness of school-based interventions in changing BMI among primary school children. The authors aim to highlight flaws in the methods and to re-analyse the data in one of the figures. |
| Level of Investigation | Analytical |
| Summary of Findings | From a letter to the editor regarding the conduct of a meta-analysis in a systematic review published in 2021 on the effectiveness of school-based interventions in changing BMI among primary school children. The authors of the letter claim that examination of the meta-analyses of the effectiveness of school-based interventions in changing BMI among primary school children shows that: 1) The results in Figures 3 and 4 do not match the methods and figure captions. 2) Of the 48 studies qualitative analysis, the SR authors report excluding 10 studies from the meta-analysis but one of these studies was included in the meta-analysis. 3) The data used for one of the studies in Figures 3 and 4 do not match the report in the original study. 4) The re-analysed data contrasts with the review’s conclusion that BMI and BMI z-score were significantly reduced in the intervention group. |
| Number of systematic reviews included | 1 |
| Number of eligible systematic reviews assessed | 1 |
| Treatment impacted | Yes |
| Treatment impacted description | |
| Interpretation impacted | Yes |
| Interpretation impacted description | The re-analysis took out the study that should not have been included, and replaced the values for the second study that were incorrect, and used a fixed-effects model to re-calculate the data and results were no longer statistically significant. |