Effects of non-pharmacological interventions on depressive and anxiety symptoms in pregnant women: a systematic review and network meta-analysis

Abstract

Background: Given the distinctive physiological characteristics of pregnant women, non-pharmacological therapies are increasingly being used to improve depressive and anxiety symptoms. Our objective was to explore and compare the impact of various non-pharmacological interventions in improving depressive and anxiety symptoms, and to identify the most effective strategies for pregnant women with depressive and/or anxiety symptoms.

Methods: We conducted a systematic search of PubMed, Embase, the Cochrane Library, and Web of Science for randomized controlled trials (RCTs) that compared non-pharmacological interventions to usual care, from the inception of each database up to October 5, 2024. We included pregnant women with singleton pregnancies who, at baseline, exhibited early signs of depressive and/or anxiety symptoms but did not meet clinical diagnostic criteria or exceed the threshold for clinically significant symptoms. We excluded pregnant women diagnosed with schizophrenia, bipolar disorder, or severe acute psychiatric conditions, those with a history of substance abuse, and those undergoing in vitro fertilisation. We performed both pairwise meta-analyses and random-effects network meta-analyses (NMAs), calculating standardised mean differences (SMDs) with 95% credible intervals (CrI). We used the surface under the cumulative ranking probability curve (SUCRA) to estimate treatment ranking probabilities. The outcomes were assessed in two groups of participants: a high-risk pregnancy group, including pregnant women with depressive and/or anxiety symptoms and high-risk pregnancies (defined as having a history of miscarriage, pregnancy complications such as gestational hypertension, gestational diabetes mellitus, or preeclampsia, and advanced maternal age (i.e., over 35 years old); and a healthy group, including participants who exhibited depressive and/or anxiety symptoms only during pregnancy and did not have other high-risk pregnancy conditions or underlying health issues. This study is registered with PROSPERO, CRD42024523053.

Findings: We included 101 studies (92 RCTs and 9 quasi-RCTs) involving a total of 15,330 participants across 11 interventions (mindfulness, education, counseling, cognitive behavioral therapy, muscle acupoint therapy, relaxation, mind-body exercise, psychotherapy, foetal movement counting, physical exercise, and music). Among the studies included in this analysis, 73 studies exhibited a low risk of bias, 9 studies had an unclear risk of bias, and 19 studies demonstrated a high risk of bias. The results indicate that, for both high-risk pregnancy population and healthy populations, mindfulness therapy was found to be an effective non-pharmacological treatment for significantly improving depressive and anxiety symptoms in pregnant women compared with control groups. For pregnant women with depressive symptoms, mindfulness therapy (SUCRA = 80%; SMD = -0.86, 95% CrI = -1.2, -0.52; Nn = 598), cognitive behavioral therapy (CBT) (SUCRA = 65%; SMD = -0.69, 95% CrI = -1.0, -0.39; Nn = 712), and education therapy (SUCRA = 48%; SMD = -0.54, 95% CrI = -0.86, -0.23; Nn = 2285) all significantly improve depressive symptoms. In the subgroup analysis of healthy populations, muscle acupoint therapy (SUCRA = 77.17%; SMD = -0.89, 95% CrI = -1.55, -0.23; N = 99) and mind-body exercise (SUCRA = 47.54%; SMD = -0.53, 95% CrI = -0.88, -0.19; N = 352) also significantly reduce depressive symptoms. Subgroup analysis shows that, in addition to mindfulness therapy, mind-body exercises (SUCRA = 67.43%; SMD = -0.97, 95% CrI = -1.61, -0.33; N = 382) and cognitive-behavioral therapy (SUCRA = 52.60%; SMD = -0.74, 95% CrI = -1.38, -0.09; N = 480) may also be effective in alleviating anxiety symptoms among healthy pregnant women.

Interpretation: Our findings indicate that mindfulness therapy significantly reduces the risk of depressive and anxiety symptoms in both high-risk pregnancy population and healthy populations. Therefore, when selecting non-pharmacologic therapies for managing depressive and anxiety symptoms during pregnancy, it is recommended that this therapy be considered. We cannot overlook the limitations of this study. For example, some interventions, such as muscle acupoint therapy for depressive symptoms and relaxation therapy for anxiety symptoms, have limited literature support. Additionally, the diversity of conditions within the high-risk pregnancy population and the high heterogeneity observed in certain interventions are also issues that require attention. These factors may affect the accuracy of the data results. Although we have employed reliable methods to address these issues, the findings of this study should still be interpreted with caution.

Funding: None.

Keywords: Anxiety symptoms; Depressive symptoms; Network meta-analysis; Non-pharmacological intervention; Pregnant woman.

Fig. 1

PRISMA diagram. This is the flowchart of the literature selection process for this study， detailing the specific circumstances of each step, ultimately resulting in the inclusion of 101 articles in the research.

Fig. 2

Full Network Evidence Map for Depressive symptoms (A) and Anxiety symptoms (B). The width of the line in the graph represents the number of studies, this number is also given on each line. The circles represent the number of pregnant women participating in the study, with larger circles indicating a greater number of participants. The line connecting the two circles indicates a direct comparison between the two interventions.