Evaluation of the effectiveness of osteopathic treatment on the mother-newborn dyad in the event of painful breastfeeding in a maternity hospital despite the application of all usual aids: randomized interventional trial in two parallel arms without blinding

 

Critical Analysis of the Study on the Effectiveness of Osteopathic Treatment on the Mother-Newborn Dyad in Cases of Painful Breastfeeding

The study conducted by Elleau et al. aimed to evaluate the effectiveness of osteopathic manipulative treatment (OMT) as a complement to usual breastfeeding aids in reducing pain and improving breastfeeding rates in mother-newborn dyads suffering from significant pain. While the results report a significant improvement in breastfeeding rates in the treated group, several major methodological weaknesses compromise the robustness of these conclusions and considerably limit their scientific scope.

The extremely small sample size (n=23) constitutes the most critical flaw of this study. With only 13 dyads in the treated group and 10 in the control group, the statistical power to detect a real effect, even if it exists, is inherently low. The observed differences could easily be the result of chance or individual variations within the small groups rather than a true causal effect of OMT. This limitation makes any generalization of the results to a larger population highly speculative.

The lack of blinding introduces considerable bias. Knowledge of treatment allocation by both participants and researchers could subjectively influence pain perception, motivation to continue breastfeeding, and the way data were collected and interpreted. Performance bias is particularly concerning, as healthcare professionals might unconsciously offer differential support to the groups. Similarly, reporting bias is inevitable when participants know they have received a specific intervention, potentially leading them to report more significant improvements.

The recruitment difficulties and premature termination of the study raise important questions about the representativeness of the sample. The reasons for the low recruitment, notably the refusal of randomization and opposition to osteopathy, suggest a possible selection bias. The dyads who consented to participate may differ significantly from those who did not in terms of motivation, beliefs, and other factors likely to influence breastfeeding outcomes. The premature termination deprives the study of the opportunity to reach the initially planned sample size, exacerbating the problem of low statistical power.

The contamination of the control group by the three mothers who sought osteopathic care after discharge further complicates the interpretation of the results. Although these participants were not included in the treated group, their use of osteopathy could have influenced the breastfeeding rates observed in the control group, potentially underestimating the effect of OMT. The justification for not including them in the treated group, while statistically conventional, does not fully account for the reality of practice and the intention to treat.

The subjectivity of certain outcome measures, particularly pain intensity and responses to qualitative questions, is problematic in the absence of blinding. These measures are inherently sensitive to participant expectations and biases. The positive responses regarding the improvement of the mother-infant relationship and well-being, while interesting, are particularly vulnerable to confirmation bias when participants know they have received a treatment aimed at improving these aspects.

Although the study provides a detailed description of the osteopathic technique used, the lack of rigorous standardization between the two treating osteopaths could introduce uncontrolled variability in the intervention. Without a strict and verifiable treatment protocol, it is difficult to ensure that all dyads in the treated group received an equivalent intervention.

Finally, the general nature of the description of "usual breastfeeding aids" in the control group raises questions about the homogeneity of the reference treatment. Variations in the intensity and quality of this support could have influenced breastfeeding rates in the control group, making the comparison with the treated group less precise.

In conclusion, despite the reported positive results, this study suffers from significant methodological limitations, including an insufficient sample size, lack of blinding, recruitment problems, and possible contamination of the control group. These weaknesses seriously compromise the internal and external validity of the study, making it impossible to conclude with certainty about the effectiveness of osteopathic treatment in this context. The results should be considered preliminary and require confirmation by larger, rigorously designed, and blinded future studies to minimize bias. The interpretation of the current findings must remain cautious and cannot justify a widespread adoption of OMT as a routine intervention for painful breastfeeding without stronger evidence.