Data and code templates for review: Reducing perceived stress and fostering mental health through breathing exercise interventions in healthy adults: a systematic review and multilevel meta-analysis

This repository contains the dataset and the R-code templates used in the multilevel meta-analysis. Contents:-DATA.sav: Dataset used in multilevel meta-analysis-R_coding.txt: Codetemplates for R coding Abstract of the related study Background: Breathing exercises have long been used to enhance well-being especially in Eastern cultures. However, evidence on their effects on overall subjective mental health, including both negative and positive dimensions, in healthy adults, as well as longer-term outcomes and factors influencing effectiveness, remains limited. This review aims to assess these effects and examines moderators influencing effectiveness, with a focus on longer-term outcomes. Methods: Four databases (PsycINFO, Scopus, CINAHL, PubMed) were searched in 08/2025 with an update search in 12/2025. To be included studies had to: 1) be RCTs or quasi-experimental studies, 2) include an intervention based on breathing solely, 3) measure self-reported mental health or subjective stress as an outcome, 4) include adult participants between 18–70 years of age and 5) be peer reviewed. Studies were excluded if they 1) included participants with diagnosed disease or medication, 2) included participants who were pregnant or used substances, 3) didn’t have a control group without breathing exercises, 4) used measurements only <24h from baseline with no longer follow-up or 5) used only breathing intervention which required special equipment. Risk of bias assessments were done by using ROB 2- and ROBINS-I V2 -tools. Data was analysed narratively and quantitively by using a multilevel meta-analysis. Results: The number of included studies in the review was 14 (1609 adults). Most studies (n=11) reported some effects in favour of the experimental group. A multilevel meta-analysis with 12 studies (1521 adults), contributing a total of 43 effect sizes, resulted in a pooled effect size of moderate magnitude (Hedges’ g=0.452, 95% CI 0.201–0.703, I²=65.63%) for overall mental health in favour of the experimental group. Subgroup analyses showed statistically significant effect sizes for stress (g=0.430, 95% CI 0.147–0.713), negative mental health outcomes (g=0.505, 95% CI 0.175–0.836) and positive mental health outcomes (g=0.435, 95% CI 0.076–0.794) in favour of the experimental group. Weekly exercise minutes were associated with intervention effectiveness (β=0.0037, 95% CI 0.0003–0.0071), with >75–100 minutes/week appearing sufficient according to sensitivity analyses. Discussion: Breathing exercises appear to have a moderate positive effect on overall subjective mental health in healthy adults. Weekly exercise exceeding 75–100 minutes seems sufficient to support longer-term outcomes. The certainty of evidence was moderate due to wide confidence intervals. Further large-scale studies with longer follow-up periods are needed to strengthen confidence of the findings. Other: Supported by the Finnish Nursing Education Foundation. PROSPERO registration: (CRD420251119911), 5/8/2025.