Aim: To analyze the effect of using Bach's emergency compound, Five-Flower essence, on the stress level of mothers with premature newborns.

Methods: The randomized, triple-blind, placebo-controlled clinical trial was conducted with 117 postpartum women. They were divided into intervention (with five-flower) and control groups. Their children were admitted to a Maternity Hospital in São Paulo, Brazil, from June 2019 to May 2021. The use of the Five-Flower essence was proposed during hospitalization (4 drops every 15 min four times a day) and at home (16 drops a day at pre-determined times) in the ten days after the first intervention. The pre-and post-intervention parameters were evaluated using the Instrument Parental Stress Scale: Neonatal Intensive Care Unit and cortisol analysis. Student's T, Mann-Whitney, Chi-square, Fisher's exact tests, and Generalized Estimating Equation models were used to analyze the data. The 5 % significance level was considered in statistical tests.

Results: The groups were homogeneous regarding sociodemographic, obstetric, and neonatal characteristics (p > 0.05). Mean stress scores were calculated for the control (3.75) and intervention (3.82) groups before intervention and a range was observed for the control (2.75-3.44) and intervention (2.55-3.54) groups after the intervention; the difference between groups was statistically significant (p < 0.05). Mean cortisol levels were calculated for the control (3.53) and intervention (3.45) groups before intervention and a range was observed for the control (3.38-2.75) and intervention (2.32-2.86) groups after intervention, showing distinct declines in the groups but without statistical difference (p > 0.05).

Conclusions: The use of the Five-Flower essence as a non-pharmacological therapy is effective in reducing the stress and cortisol levels in mothers of premature newborns admitted to neonatal and semi-intensive care units.

Declaration of competing interest The authors declare that there is no potential conflict of interest concerning the authorship and/or publication of this article.