Assessing the association between combined oral contraceptive use, selected biomarkers, and depressive episodes among premenopausal women: a protocol for a systematic review and meta-analysis

Abstract

Background

Combined oral contraceptives (COCs) are widely used among premenopausal women because of their contraceptive efficacy and favourable safety profile. However, concerns remain regarding their potential association with depressive symptoms and major depressive disorder (MDD), possibly mediated through alterations in neurochemical, inflammatory, and neurotrophic biomarkers. Hence, this study aims to provide a comprehensive synthesis of the available evidence on the link between COC and selected biomarkers associated with MDD risk and to assess whether the reported outcome is associated with the generational type of COC.

Methods

This systematic review protocol was prepared following the Preferred Reporting Items for Systematic Reviews and Meta-Analyses Protocols (PRISMA-P) 2015 statement. An extensive search will be conducted on Embase and MEDLINE through the Ovid and EBSCOhost interface. The Cochrane Library will be searched to augment relevant sources of information. Two reviewers will independently screen for eligible studies using a pre-defined criterion. The risk of bias and quality of included studies will be assessed via Cochrane risk of bias tools. The certainty of evidence will be assessed via Grading of Recommendations, Assessment, Development and Evaluation Assessment (GRADE) tool.

Discussion

The rate of depression among women of reproductive age remains alarming due to various factors related to childbearing, which may limit their economic contribution and full participation in societal development. As a result, some of these women may opt for the use of available contraceptive choices, such as COC, which are linked to the onset of MDD in high-risk individuals. More so, the change in hormonal milieu influences the secretion and physiologic role of histamine and other neurotransmitters, such as serotonin, among others, thereby contributing to an increased risk of MDD in premenopausal women. Thus, unravelling the role of COC in the secretions of these biomarkers and their contribution to the onset of MDD among premenopausal women is essential in filling the research gap that will provide insight and guidance when making informed decisions about the available contraceptive choices. The uniqueness of our review also extends to the focus on biomarker-based evidence of mood disorder risk in COC users, which may provide mechanistic insight and inform future risk-stratification research to assess the risk of MDD, both in and outside the clinical setting, among high-risk individuals.

Systematic review registration

PROSPERO CRD42021284745.

Supplementary Information

The online version contains supplementary material available at 10.1186/s13643-026-03197-8.

Introduction

Adult women account for about 5.0% of adults who suffer from major depressive disorder (MDD), which is a common mental disorder. In fact, MDD is projected to contribute to the global disease burden by 2030 [1]. Of note, aside from the formal diagnoses, most depressive symptoms are measured using surveys and screening questionnaires. One of the major caveats in using surveys and screening questionnaires is that susceptible individuals may feel uncomfortable sharing depressive symptoms in the survey [2]. Along with this, the degree to which people feel comfortable disclosing depressive symptoms may vary between countries as time progresses [3, 4].

Of all the challenges women face nowadays in the modern economy, the issue of childbearing is of priority, and it requires them to make contraceptive choices in order for them to fully participate in the economic development [5]. Notably, the impact of contraception on women’s socioeconomic status is well documented [6–8]. Previous evidence showed that contraceptive use reduces the rate of dropout of school among young women, thereby increasing occupational and economic opportunities among women [9]. Thus, as a woman’s socioeconomic status improves, stressful economic events can be avoided, leading to an improved mental health outcome [10].

Tentatively, COC is one of the most commonly used contraceptives among women of reproductive age, and COC exposure may aggravate the risk of depression [11] as shown previously in several case–control studies [12–14]. Furthermore, a recent study by De Wit et al. [15] showed that 32% to 60% of women discontinue COC use within 6 months for varying reasons, including mood changes. There were situations where COC improved the mood of participants, while some studies showed fewer mood swings among participants [16–18]. In contrast, some studies showed that COC worsens mood swings [19–21] which contrasts with the outcome of several other studies that showed evidence of no effect [22–25].

The prolonged use of COC can lead to increased change in body weight, predisposing an individual to the risk of being overweight or obese [20]. Interestingly, MDD is associated with perceived increased volumes of visceral fat resulting from metabolic syndrome [26]. However, the effect of COC on the body mass index (BMI) remains inconsistent, especially among women of reproductive age [27, 28]. It is noteworthy that uncontrolled BMI (overweight/obesity, i.e., body mass index above 25 kg/m²) and use of COC are independent risk factors for CVDs [29], which can increase the risk of depression among susceptible women [30] and vice versa.

Notably, a recent study reported that the observational associations between obesity and depression could be due to chronic systemic inflammation [31, 32], which can be regarded as a physiological consequence of obesity [33]. However, there is limited evidence about whether adiposity is a causal risk factor for psychiatric diseases and which component of higher adiposity (psychological/adverse social effect of excess weight, metabolic pathways or alternative pathways) causes the higher risk [33]. Evidence suggests a crosstalk between inflammatory pathways and the risk of depression [32] among women using COC [30, 34, 35]. For instance, C-reactive protein (CRP), an inflammatory marker of cardiovascular risk, was shown to be elevated in MDD among premenopausal women [30]. Whether the observed change in CRP level was due to COC usage type, dose, or duration is uncertain. Nevertheless, COC seems to exert more effect on the increased change in CRP level when compared with non-users [36].

Furthermore, increasing evidence suggests that histamine has a role to play in depression [31] due to its interaction with the release of serotonin ‘feel-good molecule’, in association with brain-derived neurotrophic factor (BDNF) and dopamine [37]. The functional role of histamine on the serotonergic and dopaminergic system is mediated by its receptor subtypes [38–40]. Modulation of histaminergic transmission may be a candidate target for antidepressant drugs [41]. Of note, histamine release from mast cells may be influenced by oestrogen and progesterone, as both sex hormones bind to mast cell receptors. Mechanistic evidence (in vitro and in vivo) suggests a regulatory role of these sex hormones on mast cell functionality and activity [42, 43]. Evidence suggests that increased mast cell numbers and histamine concentrations are predominant in the oestrous stage than in the proestrous and dioestrous stages [38]. However, there is a scarcity of evidence to elucidate the effect of exogenous sex hormones on histamine secretion and depressive episodes in humans.

Emerging evidence from previous studies in non-human primates demonstrated that exogenous estrogen in humans modulates brain serotonin systems at multiple sites, including synthesis, reuptake, and receptors [11, 44]. Emerging evidence from a population of healthy women indicates that COC reduces the binding capacity and levels of serotonin 4 receptor (5-HT4R) by 9–12% in the brain compared to non-users, with a predominant effect in the hippocampus [45]. Evidence also showed similarity in reduced 5-HT4R binding capacity between depressed COC users and non-users [46]. This evidence underscores the role of hormonal secretion on the selected biomarkers that are associated with the increased risk of MDD in high-risk individuals.

Given the preponderant role these biomarkers play in regulating mood changes and depressive episodes in the brain, it is essential to clarify the impact of COC on associated risk factors linked to MDD in premenopausal women. This clarification may provide insight when making informed decisions about the choice of contraception and provide prognostic value when evaluating the risk of manic-depressive episodes during consultation and follow-up in and out of clinical settings. It will also guide future experimental studies aimed at evaluating any mechanistic interactions between these risk factors. Hence, this present study aims to provide a comprehensive synthesis of the available evidence on the link between COC and the risk of MDD in premenopausal women (Fig. 1).

Fig. 1.

Integrated neuroendocrine–immunological framework linking combined oral contraceptive (COC) use and depressive outcomes in premenopausal women

Objectives

The objectives of the study are as follows:

1. To determine the prevalence of MDD in premenopausal women using COC when compared with non-users.

2. To determine the effect of COC on neurotransmitters, neurotrophic factors, and inflammatory biomarkers associated with the risk of depression in premenopausal women when compared with non-users.

Methods

The systemic review protocol was prepared according to the preferred reporting items for systemic review and meta-analysis protocol (PRISMA-P) and registered with the International Prospective Register of Systematic Reviews: PROSPERO (CRD42021284745).

Participants

The participants of the systematic review will mainly include adolescent and adult premenopausal women between the age of 15–40 years.

Exposure

The systematic review will include studies that report on all the generational types of COC (first, second, third, and newer-generation types). Briefly, COC pills contain an estrogen component combined with progestins from various generations, which differ in their androgenic and progestogenic properties. Thus, the criteria for selecting and differentiating COCs into generational types were based on the progestin component [47].

Comparators

Healthy adolescent and adult premenopausal women who are not on COC treatment.

Inclusion and exclusion criteria

Integrating multiple study designs into our review is essential to address our research questions regarding many facets of patient-relevant issues in healthcare interventions. Our systematic review and meta-analysis will include cross-sectional, cohort, and randomised controlled trials (RCTs). Studies that report on the use of COC as hormonal therapy, as well as those that report on the use of depression scales with clinical diagnosis and biomarker measurement as prespecified outcomes, will be included. There will be no language restriction. Reviews, books, letters to editors, and studies will be excluded.

Outcomes

The study outcomes will include the following:

Primary outcome

• i.Prevalence of MDD
• ii.
  Some selected biomarkers

  1. Neurotransmitters (serotonin, dopamine, histamine)
  2. Neurotrophic factors (BDNF)
  3. Inflammatory markers (CRP)

Peripheral biomarker measures will be considered and grouped separately from central measures during data synthesis. Peripheral measures can serve as useful indicators of systemic processes that interact with central neurobiological pathways; however, they may not directly reflect neurotransmitter activity within the brain due to compartmentalisation and blood–brain barrier constraints. As such, findings from peripheral measures will be interpreted as indirect proxies of central nervous system (CNS) function. We will identify and classify studies that reported on clinical and subclinical diagnoses of depression, based on the measurement tools used.

Information sources

A search strategy will be developed using medical subject headings (MeSH) words and conducted through the EBSCOhost and Ovid interface on the following databases: MEDLINE, Web of Science (WOS), Cumulative Index to Nursing and Allied Health Literature (CINAHL), African Journals Online (AJOL), APA PsycINFO. Furthermore, the Cochrane Central Register of Clinical Trials (CENTRAL) will also be searched to augment relevant sources of information. OpenGrey (System Information on Grey Literature in Europe) (www.open.eu) will be searched. In addition, the reference list of the selected studies will be scanned to identify relevant literature.

Search strategy

The search strategy will be developed using medical subheadings (MeSH) and keywords related to “oral contraceptives”, “depressive disorder”, “neurotransmitters” and “women”. The MeSH terms will include “oral contraceptive pills”, “birth control pills”, or “contraceptives”, “depression”, “depressive disorder”, “depressive symptoms” or “major depressive disorder”, “neurotransmitters” or “neuromodulators”, and “premenopausal women” or “adult women” (Supplementary Material 1).

Study selection

The studies will be screened by two independent authors (OAF and GPL) to avoid inconsistency in terms of eligibility of studies, objectivity enhancement, and prevent errors. Initially, studies will be screened by titles, abstracts, keywords, and synonyms, followed by identifying the full-text articles. Our prespecified inclusion criteria will be used to determine the relevance of all studies to the research question during the screening process. Should discrepancies ensue, PVD and BBN will independently screen such studies, and a consensus will be reached through agreeable discussion. The PRISMA flow diagram will be used to present the full results of our search and study selection as well as the inclusion process in the final systematic review [48]. Each study meeting the inclusion criteria will undergo data collection and a risk-of-bias assessment. The appropriate appraisal tools will be employed for this evaluation [49, 50].

Quality assessment and risk of bias in individual studies

The risk of bias and quality of included studies will be assessed using the Cochrane risk of bias tools. When the Cochrane risk of bias tool for non-randomised controlled trials becomes too difficult to apply for appraising the selected studies, we will consider other recommended appraisal tools, such as the Newcastle–Ottawa scale and/or the Joanna Briggs Institute (JBI) checklist for non-randomised controlled studies. OAF and GPL will make independent judgments based on the tools' domains and assign ratings. BBN will be consulted to arbitrate any disagreements that arise during appraisal.

Data item and management

The Mendeley desktop reference manager (version 1.19.4) will be used to archive relevant studies as well as remove study duplicates. Studies meeting the inclusion criteria will then be subjected to data collection while critical appraisal and risk of bias assessment are ongoing. Relevant data items will be retrieved using a structured data extraction form (Supplementary Material 1). Firstly, two independent reviewers will pilot the data extraction process with our data extraction form before the commencement of the main data extraction process in our review. The third reviewer will resolve any disagreements between the authors where necessary. The data to be extracted will include the author and year of publication, aim of study, the country, population (sample size), study design, setting, types, dosage and duration of contraceptive usage, age of participants, body mass index (BMI), socioeconomic variables, MDD assessment and outcome measured among oral contraceptive users and non-users. To avoid errors during data entry from selected studies, OAF and GPL will independently perform this process. Should discrepancies arise, BBN will be invited for arbitration. In case of insufficient data, the main authors of the studies will be contacted to obtain enough information. Importantly, the reference lists of included studies will be screened to ensure that no relevant studies are missed.

Data synthesis

Review Manager (RevMan) version 5.3 (The Nordic Cochrane Centre, The Cochrane Collaboration, 2014) will be used for statistical analysis. Clinical and methodological heterogeneity assessment will be conducted first, followed by examining statistical heterogeneity. A meta-analysis will be conducted if at least three studies report comparable outcomes with sufficient clinical and methodological homogeneity. The chi-squared (χ2) and I² statistic tests will be used to determine the level of heterogeneity across the included studies. An I² value of ≥ 50 will be considered moderate or substantial heterogeneity [51, 52]. Suppose studies demonstrate homogenous characteristics regarding the participants, intervention, comparisons, and outcomes. In that case, we will conduct a fixed-effect meta-analysis. Otherwise, a random-effects model will be employed where studies show substantial heterogeneity.

Furthermore, subgroup analysis and meta-regression will be conducted to explore the potential sources of heterogeneity within the included studies based on the overall risk of bias, study design, and study location. We will also perform a subgroup analysis based on the reported clinical and subclinical diagnoses of depression among included studies. The levels of the inter-rater agreement will be assessed using Cohen’s kappa [36]. We will minimize publication bias through an extensive search across multiple databases and grey literature while contacting study authors for unpublished data. The funnel plots will be used to assess publication bias. In the case of missing data, we will contact the original investigators to request missing data.

Where necessary, results extracted from study reports may need to be converted or transformed into a consistent, usable format for analysis, in line with the recommended Cochrane guidelines [53]. Differences in biomarkers will be expressed as standardised mean differences (SMDs) due to variation in assay methods and in the international system of units (SI) used for the outcomes. For adverse events, the odds ratio (OR) or risk difference with 95% CI will be used. Accordingly, OR < 1 indicates reduced odds with exposure, OR > 1 indicates higher odds of adverse events, and OR = 1 indicates the intervention is not associated with adverse events [54].

Where heterogeneity is high and cannot be explained through subgroup or meta-regression analyses, or where outcomes are conceptually or methodologically diverse, a narrative synthesis will be undertaken. The narrative synthesis will be presented in a table from which conclusions will be drawn. The results of the narrative synthesis will be grouped thematically based on the characteristics of the included studies, the methods used to evaluate MDD, the impact of COC on prespecified biomarkers, and reported outcomes linked to MDD. Given the biological interrelatedness of the biomarkers, correlation and conceptual overlap between biomarkers will be addressed by grouping biomarkers into predefined mechanistic domains. Furthermore, we will perform sub-group analysis or meta-regression using the Gini index of each country where the study was conducted to explore the role of geographic and demographic disparities on the reported outcome. The potential impact of missing data on the review's findings will also be addressed in the discussion section. Where no clear patient consent statement, Institutional Ethical Review Board, or study ethics certification number details are provided, we will incorporate them into the bias assessment report, conduct a sensitivity analysis, and highlight the implications for the reported outcome in our discussion.

Quality of cumulative evidence

The quality of evidence for primary outcomes will also be evaluated using the grading of recommendations assessment, development, and evaluation (GRADE) tool [55, 56]. This is important for drawing conclusions and making recommendations. The findings will be summarised and presented in the summary of findings (SoF) table (Supplementary Material 1). The SoF table will include crucial information such as absolute risks for both treatment and control groups, estimates of relative risk, and a quality ranking of the evidence. The SoF will cover outcomes and biomarkers related to the risk of MDD.

Sensitivity analysis

Studies with an I² ≥ 50 will be included to perform a sensitivity analysis to determine the influence of the COC formulation (estrogen dose, progestin type), duration of use, and participants’ age groups (adolescents vs adults) on the risk of MDD. To minimise healthy user and discontinuation biases, analyses will be stratified by exposure status (current, former, and new users) and duration of use. Sensitivity analyses will be conducted, excluding studies at high risk of selection bias and those limited to current users, to provide an estimate of the true effects. The outcome of our analysis will be visually presented using the bubble plot.

Discussion

Depression is one of the major health issues that can diminish the participation of most women of reproductive age towards economic contribution and the development of our modern society [5]. The role women play in our modern society cannot be overemphasised. According to available evidence, women of reproductive age are at risk of experiencing disruption either in their education or career trajectory because of the psychosocial need to give birth and raise children [57]. To mitigate this situation, some women may opt for different contraceptive measures to delay the possibility of childbearing. While the use of COC is associated with the risk of manic-depressive symptoms among premenopausal women, the impact of COC on specific biomarkers linked to MDD, such as histamine and some other neurotransmitters, remains inconclusive. Evidence suggests that cyclical changes in female sex hormone secretion influence the synthesis and functions of these biomarkers [58]. Previous evidence suggests an associative feedback mechanism between sex steroids, histamine synthesis and function [59, 60]. Alterations in the histaminergic system have been reported in several brain disorders, and these alterations play a significant role in their pathophysiology [61, 62]. Emerging evidence indicates that colocalization of estrogen and histamine receptors modulates histamine’s effects on dopamine in the striatum and substantia nigra of female mice [58]. Evidence also suggests that the effects of estrogen on the serotonergic system are usually mediated by its alpha (ERα) and beta (ERβ) receptors, thereby increasing serotonin binding to its 5-hydroxytryptamine (5-HT) receptors at the cellular level [63]. Increased levels of estrogen and progesterone activate ERβ, thereby upregulating 5HT_2A receptors and increasing their binding capacity [64]. Deactivation of ERα reduces 5HT_1A receptor expression [65]. The expression of the 5HT_2A receptor increases serotonin synthesis and action, while 5HT_1A receptors antagonise this effect [63]. Thus, unravelling the role of COC in the secretions of these biomarkers and their contribution to the onset of MDD among premenopausal women underscores the uniqueness of our review, which is essential in filling the research gap that will provide insight and guidance when making informed decisions about the available contraceptive choices. More so, focusing on biomarker-based evidence of mood disorder risk in COC users may provide mechanistic insight and inform future risk stratification research in assessing the risk of MDD, during consultation and follow-up in and out of clinical settings, among high-risk individuals.

Potential strengths and limitations

• This review is the first to offer a broad synthesis of the available evidence concerning the relationship between contraceptive use (COC) and some selected neurotransmitter functions during mood changes in women of reproductive age.

• The fact that this review will undergo a comprehensive search strategy means that there is a likelihood of identifying and retrieving all relevant articles pertinent to our research objectives that may guide the planning and construction of future experimental studies.

• Our systematic review protocol followed the PRISMA guidelines.

• The potential limitation will centre around the heterogeneity of the available studies in terms of study design, methodological approach, COC formulation, and the reported outcomes.

• Our data extraction template (Supplementary Material 1) will be available to aid reproducibility in subsequent studies.

• Furthermore, the findings of the systematic review will be disseminated through peer-reviewed publications and presented at national and international conferences.

Authors’ contributions

OAF conceptualised the study. OAF and BBN designed the study protocol. OAF drafted the protocol (both the first and final drafts). PVD and GPL helped review the drafted protocol for clarity and brevity. All authors reviewed and approved the final manuscript for submission. OAF and BBN are the guarantors of the review.

Funding

None.

Declarations

Ethics approval and consent to participate

The authors declare that they have no competing interests.

Consent for publication

Not applicable, as no patient-level data has been included in this manuscript.

Competing interests

The authors declare that they have no competing interests.