Table 2.
Exposure to breastfeeding compared to absence of breastfeeding in protection of children mental health. Grading of Recommendations Assessment, Development and Evaluation (GRADE) assessment
| Certainty assessment | Summary of findings | ||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|
|
Participants (studies) Follow-up |
Risk of bias | Inconsistency | Indirectness | Imprecision | Publication bias | Overall certainty of evidence | Study event rates (%) | Relative effect (95% CI) | Anticipated absolute effects | ||
| With absence of breastfeeding | With exposure to breastfeeding | Risk with absence of breastfeeding | Risk difference with exposure to breastfeeding | ||||||||
| Schizophrenia | |||||||||||
| 3 observational studiesa | Serious | Not serious | Not serious | Serious | Unclear |
⨁○○○ Very low |
261 cases 363 controls | OR 0.98 (0.57 to 1.71) | Satisfactory | ||
| 156 per 1,000 | 3 fewer per 1000 (from 61 fewer to 84 more) | ||||||||||
| Depressive disorders | |||||||||||
| 5 observational studiesb | Seriousc | Seriousd | Seriouse | Not serious | Unclear |
⨁○○○ Very low |
There is conflicting evidence regarding associations between breastfeeding and depressive disorders development with some studies showing small protective effect and others reporting no effect | ||||
| Anxiety disorders | |||||||||||
| 3 observational studies | Seriousf | Seriousg | Not serious | Not serious | Unclear |
⨁○○○ Very low |
There is conflicting evidence regarding associations between breastfeeding and anxiety disorders development with one cohort study (n = 3657) reporting no effect while one cross-sectional study (n = 98,702) and one case–control study (n = 450) demonstrating protective effect | ||||
CI confidence interval, OR odds ratio
aMcCreadi et al. 1997 study used siblings as a control group. Outcomes of the sensitivity analysis, which included this study did not differ (OR 0.89 (95% CI 0.58–1.38)) from the primary analysis. Two cohort studies provided conflicting evidence with Sorensen et al. 2005 demonstrating reduced risk and Leask et al. 2000 no effect
bDe Mola et al. 2016 study compared different durations of breastfeeding but did not provide “ever” vs. “never” comparison
cThree studies were ranked as “satisfactory” using NOS score, while one was rated as “good”
dThe direction and magnitude of effect varied across the studies. Overall, the results showed either small protective effect or no association between breastfeeding and depressive disorders
eWe judged the evidence to have potential serious indirectness due to substantial variability in the outcome measure used
fOne of two studies was rated as “unsatisfactory” as per NOS
gThe direction and magnitude of effect varied across the studies