The literature search was conducted between January 2014 and July 15, 2025, in MEDLINE through PubMed (National Library of Medicine, Bethesda, Maryland, USA), Science Direct and Scopus through the Elsevier platform (Elsevier B.V., Amsterdam, The Netherlands), and PsycINFO through the American Psychological Association (APA). Web of Science provided by Clarivate Analytics (Philadelphia, Pennsylvania, USA). This approach was considered because the field of study is constantly evolving and this time period ensures that the review includes studies that respond to recent sociocultural changes and current theoretical paradigms.

The search terms used include those related to RPL (e.g., “recurrent pregnancy loss”, “recurrent miscarriage”, or “recurrent fetal loss”), constructs associated with mental health and well-being status (e.g., “anxiety”, “depression”, “stress”, “quality of life”, and “psychopathology”), and the study design (e.g., “observational”, “prospective”, and “case-control”) (see Supplementary Material in Appendix).

We included original studies that (i) assessed the impact of RPL on different constructs related to mental health (e.g., anxiety, depression, stress) and well-being status (e.g., psychopathology, quality of life) in women with RPL compared to women (or couples) without RPL, using standardized psychological assessment instruments; (ii) examined women or couples with RPL, defined as ≥ 2 gestational losses, unless subgroup analyses for RPL were reported separately; (iii) were published in English; and (iv) had an observational study design (e.g., case-control, cohort). In addition, only data from peer-reviewed articles were included, as defined on a journal’s website or based on the full text of the article.

We excluded studies that (i) only explored biomedical aspects associated with RPL; (ii) evaluated RPL as a single miscarriage; (iii) did not include measurement of constructs related to mental health and/or well-being status; (iv) did not have a design that could be included in a systematic review such as review articles and meta-analyses, case series, opinion articles or dissertations; or (v) lacked a full-text version.

Two researchers (EM-M and JVS-O) independently and masked proceeded to review the titles and abstracts of the non-duplicate articles, evaluating their eligibility according to the selection criteria. In the next step, the reviewers examined the full texts of the studies likely to be included in the review to identify eligible studies. In case of disagreement among the reviewers, discussion and consensus were reached with the third author (AB-P).

Relevant data from included studies were extracted using Microsoft Excel 2010. The information extracted from the studies included: publication details (authors, year and country), study type (where possible timing of assessment), characteristics of the population and sample groups (sample age, age, and number of RPL), and study outcomes (levels of anxiety, depression, stress, quality of life, and well-being measured and method of measurement). Descriptive analyses were made by comparing population characteristics and outcomes of the included studies. Validity and reliability measures reported in the studies were estimated using Pearsons r and derived correlations (e.g., biserial-point correlation).

The risk of bias and quality assessment of the included case-controls and cohorts studies regardless of whether they were cross-sectional or longitudinal were assessed using the Newcastle Ottawa scale (NOS; Wells et al., 2008). For cohort studies, this scale assigns points for representativeness of the exposed and control groups, adequate ascertainment of exposure, clarity of the absence of outcomes at study start, comparability of groups based on study design and analysis, blinded assessment or record linkage to confirm study outcomes, the sufficiency of follow-up duration to observe the outcomes of interest and reporting of a sufficiently low withdrawal rate that would not threaten a significant risk of bias to the study. For case-control studies, this scale assigns points for adequate case definition, representativeness of the cases, adequate selection and definition of controls, comparability of cases and controls based on the design or analysis, adequate ascertainment of exposure, same method of ascertainment for cases and controls, and same (non-response) rate for both groups. A maximum of nine points can be assigned.

The quality of the information included in the studies was assessed using the STrengthening the Reporting of OBservational studies in Epidemiology (STROBE) tool for observational cohort, case-control, and cross-sectional studies (von Elm et al., 2007). It consists of 22 items, and each item is rated as affirmative, negative, not available, or not applicable. The items cover various aspects of the articles, such as the title and abstract (point 1), the introduction (points 2 and 3), methodology (points 4-12), results (points 13-17), and discussion (points 18-21), as well as other relevant sections (point 22 on funding). Three categories were used to rate the quality of the studies: good, fair, and poor quality, according to criteria proposed by STROBE. In both assessments, disagreements were discussed until consensus was reached and, when needed, were resolved by a third author (AB-P). The STROBE checklist was used exclusively to assess the completeness of reports.

As detailed in Table 1, 30.4% (k = 7) of the studies were conducted in China, 17.3% (k = 4) in Iran, 13% (k = 3) in Denmark, 8.7% (k = 2) in Germany, whereas countries such as the Turkey, Kazakhstan, Lithuania, India, Romania, Nigeria, and USA contributed one study each (4.3%).

Most studies (k = 18) were cross-sectional, of which 61.1% (k = 11) (Abbaspoor et al., 2016; Gül & Kirka, 2021; He et al., 2019; Hedegaard et al., 2021; Kolte et al., 2015; Kulshreshtha et al., 2023; Liu et al., 2024; Van et al., 2023; Voss et al., 2020; T. Wang et al., 2023; Zhang et al., 2016) were comparative and correlational, the 27.8% were only comparative (k = 5) (R. Gao et al., 2022; Ghorbani et al., 2022; Kuhlmann et al., 2023; Nitu et al., 2023; Tavoli et al., 2018), and the other 11.1% were only correlational (k = 2) (L. Gao et al., 2020; Kukulskien & Žemaitien, 2022). The remaining studies were (k = 4) case-control studies, comparative, or correlational (Adib-Rad et al., 2019; Eleje et al., 2024; Issakhanova et al., 2023; Y. Wang et al., 2021), and another (k = 1) was a comparative prospective cohort study (Slot et al., 2022).

Overall, 12.759 were women aged between 19 and 47 years (X = 32.9, SD = 4.8) were included, of which 60.6% had RPL, 3.2% had 1PL, 5.2% were women couples, and 31% did not had RPL. As for the men, a total of 1419 aged between 20 and 49 years (X = 34.2, SD = 3.7) were included, of whom 83.4% were female couples with RPL and 16.6% without RPL.

Most of the studies assessed depressive symptoms (k = 12, 52.1%) (L. Gao et al., 2020, 2022; He et al., 2019; Hedegaard et al., 2021; Kolte et al., 2015; Kukulskien & Žemaitien, 2022; Kulshreshtha et al., 2023; Slot et al., 2022; Van et al., 2023; T. Wang et al., 2023; Y Wang et al., 2021; Zhang et al., 2016) using instruments such as the Edinburgh Postnatal Depression Scale (EPDS), Self-Rating Depression Scale (SDS), Major Depression Inventory (MDI), and Center for Epidemiologic Studies Depression Scale (CES-D). The 30.4% informed anxiety symptoms (k = 7) (L. Gao et al., 2020; Gül & Kirka, 2021; He et al., 2019; Kulshreshtha et al., 2023; T. Wang et al., 202; Y. Wang et al., 2021; Zhang et al., 2016) through the following tests Self-Rating Anxiety Scale (SAS), Visual Analogue Scale for Anxiety (VDAS), Generalized Anxiety Disorder-7 (GAD-7), and State-Trait Anxiety Inventory (STAI). Specifically, a study reported anxiety related to pregnancy (Kulshreshtha et al., 2023) measured with the Pregnancy-Related Anxiety Questionnaire – Revised (PRAQ-2), while another study informed mixed symptoms of anxiety and depression (Tavoli et al., 2018) through the Hospital Anxiety and Depression Scale (HADS).

Regarding stress-related outcomes, the 21.7% reported symptoms related to perceived stress (k = 5) (Hedegaard et al., 2021; Kolte et al., 2015; Kulshreshtha et al., 2023; Slot et al., 2022; Van et al., 2023) using the questionnaire Perceived Stress Scale (PSS). Post-traumatic stress was measured (k = 2), using the Impact of Event Scale – Revised (IES-R) test, in the 8.7% of studies (Kuhlmann et al., 2023; Kukulskien & Žemaitiené, 2022). Two study showed mixed results for anxiety, depression, and stress measured with the following instruments Depression Anxiety Stress Scales-21 (DASS-21) (Eleje et al., 2024; Issakhanova et al., 2023). Moreover, the 8.7% informed on coping styles (k = 2) (Kuhlmann et al., 2023; Nitu et al., 2023), using the Fertility Quality of Communication Inventory (FQCI) and Coping Inventory for Stressful Situations (CISS).

Approximately one-fifth (17.3%) of the studies reported general psychopathology (k = 4) (Abbaspoor et al., 2016; Adib-Rad et al., 2019; Nitu et al., 2023; T. Wang et al., 2023) using the following tests Symptom Checklist-25 (SCL-25), Symptom Checklist-90 (SCL-90), Symptom Checklist - 90 – Revised (SCL-90-R), and Psychological Profile Parameters (PPP). To a lesser extent, 13.4% of the studies (k = 3) reported on sleep quality (Gül & Kirka, 2021; Van et al., 2023; T. Wang et al., 2023) measured through Insomnia Severity Index (ISI) and Pittsburgh Sleep Quality Index (PSQI), 8.7% reported results related to quality of life assessed (k = 2) (Abbaspoor et al., 2016; Tavoli et al., 2018) using the Perceived Social Support Scale (PSSS) and another study (k = 1) (L. Gao et al., 2020) reported on perceived social support measured with Short Form - 36 Health Survey (SF-36). Moreover, the 8.7% of the studies reported mental health-related results (k = 2) (Kuhlmann et al., 2023; Voss et al., 2020) from a multidimensional perspective measured with the Screening Instrument for Subfertility Patients (ScreenIVF).

Finally, sexual health outcomes were explored in 26% of studies (k = 6), of which 13% reported sexual satisfaction (k = 3) (Ghorbani et al., 2022; Liu et al., 2024; Nitu et al., 2023; Zhang et al., 2016) measured through the Female Sexual Function Index (FSFI), International Index of Erectile Function-5 (IIEF-5), Sexual Interaction Inventory (SII), Dyadic Sexual Communication Scale (DSCS), and Index of Sexual Satisfaction (ISS), 8.7% reported marital intimacy (k = 2) (Kulshreshtha et al., 2023; Nitu et al., 2023) using the Marital Intimacy Questionnaire (MIQ) and Intimate Partner Violence (IPV), and one reported infertility history (k = 1) (Gül & Kirka, 2021) measured with Medical Infertility History (MIF).

Fourteen cross-sectional studies have been conducted to assess depressive and anxiety symptoms in individuals with RPL across diverse countries (L. Gao et al., 2020; R. Gao et al., 2022; Gül & Kirka, 2021; He et al., 2019; Hedegaard et al., 2021; Kolte et al., 2015; Kukulskien & Žemaitiené, 2022; Kulshreshtha et al., 2023; Slot et al., 2022; Tavoli et al., 2018; Van et al., 2023; T. Wang et al., 202; Y. Wang et al., 2021; Zhang et al., 2016).

Overall, women with RPL consistently showed significantly higher rates of depression and anxiety compared to women without RPL. In the Asian continent, specifically in China, L. Gao et al. (2020) reported anxiety in 45% and depressive symptoms in 37% of participants with RPL (p < .05). Perceived social support was inversely associated with anxiety (adjusted OR = 2.89, p < .001) and depression (adjusted OR = 2.56, p < .001) (L. Gao et al., 2020). A subsequent study confirmed the high prevalence of depression in women with RPL compared to those without (45% vs. 30%, p < .01) (R. Gao et al., 2022). Moreover, He et al. (2019) observed significantly higher levels of anxiety and depression among women with RPL and 1PL compared to women without RPL (p = .006), and low educational level was revealed as an independent risk factor (adjusted OR = 1.75-1.80, p < .01-.03) (He et al., 2019).

Studies with larger samples corroborated these findings. Y. Wang et al. (2021) found a higher prevalence of depression (49%) and anxiety (29%) in women with RPL (p < .001), and a strong association between increased levels of depression and anxiety with an increased risk of RPL (adjusted OR = 4.23-5.95, p = .001-.05) (Y. Wang et al., 2021). Likewise, T. Wang et al. (2023) reported state anxiety in 60.3%, trait anxiety in 51.7%, and depression in 33.9% of participants, with ≥ 4 RPL, low income, and recent loss ( < 6 months) being the main predictive factors (adjusted OR = 1.98-5.52, p = .01) (T. Wang et al., 2023). Moreover, Zhang et al. (2016) revealed that men with partners experiencing RPL also experienced significantly higher rates of depression (26.3%), anxiety (39.1%), and reduced sexual satisfaction (p = .001) (Zhang et al., 2016).

Similarly, we find converge results in other Eastern countries. In India, Kulshreshtha et al. (2023) reported higher rates of depression (27%) and generalized anxiety (15%) in women with RPL, along with elevated pregnancy-related anxiety (p = .01). RPL were identified as an independent risk factor for depressive symptoms (adjusted OR = 26.45, p = .01) (Kulshreshtha et al., 2023). Tavoli et al. (2018) found that Iranian women with RPL scored significantly higher on anxiety (p < .0001) and depression (p < .0001) symptoms and lower on emotional functioning (p = .0001-.05) (Tavoli et al., 2018), while, in Turkey, Gül and Kirka (2021) found increased dream anxiety in women with RPL compared to 1PL and non-RPL groups (p = .001) (Gül & Kirka, 2021).

Likewise, similar results have also been found in Western countries, such as Denmark, Lithuania and USA. For example, in Danish population, Hedegaard et al. (2021) observed that women with RPL showed a significantly higher prevalence of depressive symptoms compared to men (p = .001), and a strong correlation between depressive symptoms and number of RPL was reported among affected couples (adjusted RR = 3.74, p = .001) (Hedegaard et al., 2021). In line with these findings, Kolte et al. (2015) found that individuals with RPL exhibited significantly higher levels of depression (8.6% vs. 2.2%) than those without RPL (p = .01), with symptom severity closely associated with both outcomes in the RPL group (adjusted OR = 5.53, p = .01) (Kolte et al., 2015). Moreover, Slot et al. (2022) demonstrated that primary RPL (32.1%) was associated with a greater psychological burden than secondary RPL (18.4%), including sadness, guilt, and low self-esteem (p = .007) (Slot et al., 2022). Similar findings were also observed in Nigerian women (Eleje et al., 2024).

In Lithuanian population, Kukulskien and Žemaitiené (2022) indicate that 59.1% of women with RPL are at increased risk of postnatal depression, with 48.9% classified as being at high risk, while 44.7% were considered to be at increased risk for post-traumatic stress symptoms (adjusted OR = 2.45-2.45, p = .001-.01) (Kukulskien & Žemaitiené, 2022). Moreover, Van et al. (2023) observed poorer sleep quality in women with RPL, which was significantly associated with depressive symptoms among US individuals (adjusted OR = 2.03, p < .05) (Van et al., 2023).

Several observational studies have examined psychological stress in women affected by RPL, revealing consistent evidence of elevated stress levels and their associations with sociodemographic, emotional, and reproductive factors (Hedegaard et al., 2021; Issakhanova et al., 2023; Kolte et al., 2015; Kuhlmann et al., 2023; Kukulskien and Žemaitiené, 2022; Kulshreshtha et al., 2023; Nitu et al., 2023; Slot et al., 2022; Van et al., 2023).

On the European continent, specifically in Denmark, Hedegaard et al. (2021) found significantly higher emotional stress in women with RPL compared to their male partners (p = .02-.01). The number of RPL was positively associated with perceived stress (adjusted RR = 1.24; p = .02), while men showed a protective effect (adjusted RR = 0.67, p = .02) (Hedegaard et al., 2021). Another Danish study by Kolte et al. (2015) reported elevated stress levels in women with RPL (41.2%) compared to controls (23.2%), and the number of losses was positively correlated with stress severity (adjusted OR = 1.59, p = .01) (Kolte et al., 2015). Similarly, in a prospective cohort study, Slot et al. (2022) observed that women with primary RPL reported higher levels of stress (32.7%) compared to those with secondary RPL (19.9%), feelings of helplessness, guilt, and emotional exhaustion were more common in the primary RPL group (p = .01) (Slot et al., 2022).

In Lithuania, Kukulskien and Žemaitiené (2022) found that 44.7% of women with RPL were at increased risk of post-traumatic stress symptoms, significantly associated with the number of RPL (OR = 2.45, p = .001) (Kukulskien & Žemaitien, 2022). Likewise, Nitu et al. (2023) observed significantly higher stress levels (61% vs. 47%) and maladaptive coping strategies (43% vs. 27%) in Romanian women with RPL. These included a greater tendency toward emotion-focused coping, lower self-esteem, and greater relationship dissatisfaction compared to individuals without RPL (p = .01). In addition, stress was related to sexual dissatisfaction and lower marital intimacy (p = .001) (Nitu et al., 2023). Similar findings were also observed in Nigerian women (Eleje et al., 2024).

The psychological burden of RPL extends to partners. In Germany, Kuhlmann et al. (2023) reported that women (13.7%) showed significantly higher levels of post-traumatic stress compared to men (3.9% (p = .01) (Kuhlmann et al., 2023). Stress in women was closely related to limited social support and relationship strain, which in turn contributed to partner distress and sexual dissatisfaction. Moreover, in USA, Van et al. (2023) reported that no direct association was established between sleep and stress (Van et al., 2023).

Similar findings have been observed in some countries on the Asian continent. Kulshreshtha et al. (2023) identified significantly higher levels of perceived stress (3%) in Indian women with RPL compared to people without RPL (0.8%), and RPL was found to be an independent predictor of psychological distress (adjusted OR = 26.45, p = .01) (Kulshreshtha et al., 2023). Likewise, Issakhanova et al. (2023) found that women with RPL in Kazakhstan had significantly higher stress scores compared to women without RPL (p < .001). The number of RPL was strongly associated with stress symptoms (adjusted OR = 3.67, p < .001) (Issakhanova et al., 2023).

Several cross-sectional and case-control studies have reported consistent deficiencies in psychosocial areas such as psychopathology, sleep quality, perceived social support, and quality of life (Abbaspoor et al., 2016; Adib-Rad et al., 2019; L. Gao et al., 2020; Gül & Kirka, 2021; Kuhlmann et al., 2023; Nitu et al., 2023; Tavoli et al., 2018; Van et al., 2023; Voss et al., 2020; T. Wang et al., 2023).

Psychopathology has been observed to be more prevalent and severe in populations with RPL. The great majority of evidence comes from Eastern countries, specifically Iran and China. In Iran population, Abbaspoor et al. (2016) reported significantly higher scores on multiple domains of the SCL-25 in women with RPL, including anxiety, depression, phobia, and obsession-compulsion, compared to women without RPL (p = .01-.0001). The number of previous RPL was positively associated with psychopathological symptoms (adjusted OR = 4.22, p = .001) (Abbaspoor et al., 2016). Similarly, Adib-Rad et al. (2019) found higher overall psychopathology and intolerance to uncertainty among women with RPL (p < .0001), with psychopathology being the strongest predictor of RPL recurrence (adjusted OR = 6.05, p ≤ .0001) (Adib-Rad et al., 2019). In Chinese individuals, Wang et al. (2023) provided comprehensive evidence that women with RPL experience high levels of depression, anxiety, and general psychopathology (p < .05-.0001). Mean scores were significantly higher in women with RPL compared to women with 1PL and without RPL in areas such as hostility, phobic anxiety, paranoid ideas, and obsessive-compulsive symptoms (p = .001). Depression was closely related to low family income, recent losses, and lack of sleep (adjusted OR = 1.98-5.52, p = .001) (T. Wang et al., 2023).

Moreover, although to a lesser extent similar results have been found in Western populations. Nitu et al. (2023) found that women with RPL showed greater emotional stress (p = .02-.001). Emotion-focused coping was more common among women with RPL (p = .01), along with lower self-esteem and poorer psychological profiles (impairment relational and interpersonal) (p = .01-.001) (Nitu et al., 2023).

Sleep quality was also revealed to be a major concern in Eastern countries, specifically Turkey and China. Gül and Kirka (2021) and T. Wang et al. (2023) found significantly worse insomnia and dream-related anxiety among women with RPL compared to the 1PL and non-RPL groups in Turkish population (p = .001) (Gül & Kirka, 2021) and only compared with non-RPL group (p = .01) (T. Wang et al., 2023). Likewise, in the USA, Van et al. (2023) reported significantly worse sleep in participants with RPL, which correlated with depressive symptoms (adjusted OR = 2.03, p = .04) (Van et al., 2023).

Overall quality of life and well-being were also compromised in Eastern population. In Iran, Abbaspoor et al. (2016) and Tavoli et al. (2018) reported lower scores on multiple domains of the quality of life among participants with RPL, including physical functioning, vitality, emotional role, and mental health (p = .01-.001). These deficiencies were strongly related to increased psychological distress (adjusted OR = 4.22, p = .001) (Abbaspoor et al., 2016; Tavoli et al., 2018). In China, L. Gao et al. (2020) assessed perceived social support and found that women with RPL reported moderate levels of support (p = .01). Notably, low perceived support was associated with an increased risk of anxiety (OR = 2.8, p = .001) and depression (OR = 2.56, p = .001), suggesting a protective role for social relationships (Gao et al., 2020).

It should be noted that the perspectives of the couple were addressed by Kuhlmann et al. (2023) and Voss et al. (2020) in Western population, who demonstrated that women with RPL had significantly higher levels of anxiety, depression, and post-traumatic symptoms compared to men (p = .01-.001) (Kuhlmann et al., 2023). Limited social support, maladaptive coping styles, and psychosocial risk were strongly associated with greater distress in both partners (r = .47-.62, p = .05-.001) (Voss et al., 2020).

Sexual health is increasingly recognized as a fundamental dimension of well-being that is affected by RPL. Several studies reported alterations in various domains of sexual function, satisfaction, and intimacy in women with RPL and their partners (Ghorbani et al., 2022; Gül & Kirka, 2021; Kulshreshtha et al., 2023; Liu et al., 2024; Nitu et al., 2023; Zhang et al., 2016).

Most of the findings are found in Eastern countries. In Iran, Ghorbani et al. (2022) found that women with RPL had reduced sexual function, showing significantly lower scores in desire, arousal, lubrication, orgasm, satisfaction and pain (p < .001), indicating a marked reduction in sexual well-being after RPL (Ghorbani et al., 2022). Likewise, Zhang et al. (2016) provided a gender perspective by examining the male partners of women with RPL in Chinese population. The study found a significantly higher prevalence of sexual dysfunction among men in the RPL group, with lower sexual satisfaction (p = .02) and higher rates of erectile dysfunction (p = .01). Psychological symptoms, such as depression and anxiety, were strongly associated with sexual dissatisfaction and performance problems, indicating a bidirectional relationship between emotional distress and sexual health (OR = 1.47-1.86, p < .001) (Zhang et al., 2016). Similar findings were also observed in women with RPL (Liu et al., 2024). Although not directly focusing on sexual function itself, Gül and Kirka (2021) observed higher levels of dream anxiety and insomnia among women with RPL in Turkey (p = .30), factors that have been linked to reduced sexual desire and relational tension (p = .001). These sleep-related symptoms may indirectly affect sexual health and intimacy (Gül and Kirka, 2021).

Similar findings have been found in Western countries, although to a lesser extent. Specifically, in Romania, Nitu et al. (2023) highlighted that women with RPL reported significantly greater sexual dissatisfaction (p < .001) compared to women without RPL (p < .001), as well as decreased marital intimacy and reduced quality of sexual communication (p = .02-.008). These findings suggest that RPL affects not only the physiological aspects of sexuality, but also the relational and emotional components, especially in the context of marital dynamics (Nitu et al., 2023).

Finally, Kulshreshtha et al. (2023) and Nitu et al. (2023) emphasized the interconnection between emotional distress and sexual health in women with RPL. Higher levels of depression, anxiety, and perceived stress in women with RPL were accompanied by lower satisfaction in intimate relationships, poor sexual communication, and less emotional connection with their partners (p = .03-.001). Emotion-focused coping styles and low self-esteem were also more prevalent, which could contribute to the observed decline in sexual quality of life.

The assessment of risk of bias and methodological quality was performed using the NOS (see Table 2). Overall, the included case-control and cohort studies had a low risk of bias and good methodological quality. The majority (k = 18) obtained scores ranging from 7 to 9, indicating low risk of bias in terms of selection, comparability and exposure or outcome assessment. However, some studies had specific methodological limitations, such as lack of clear information on exposure ascertainment, or potentially insufficient or undescribed follow-up, which could compromise the internal validity of the findings (k = 5) (He et al., 2019; Kukulskien & Žemaitien, 2022; Kulshreshtha et al., 2023; Nitu et al., 2023; Van et al., 2023). Despite these limitations, overall compliance with key quality criteria was observed, especially in studies such as those by Eleje et al. (2024), Hedegaard et al. (2021), Issakhanova et al. (2023), Kolte et al., (2015), Kuhlmann et al. (2023), Voss et al. (2020), Y. Wang et al. (2021), T. Wang et al. (2023), and Zhang et al. (2016) (k = 9), which achieved the maximum score, reflecting robust methodological execution. Overall, the included studies present an acceptable methodological level to support their conclusions.

The assessment of the quality of information included using the STROBE tool showed the most articles had a good quality of information. Specifically, 91.3% (k = 21) of the studies had a score above 75%, indicating a good quality of information. The remaining 8.7% (k = 2) (He et al., 2019; Kulshreshtha et al., 2023) of articles showed “fair” quality of information, with scores between 50% and 75%, suggesting some concerns regarding to quality of information. No studies with low quality of information (score below 50%) were identified (see Table 3).

This systematic review has several limitations that affect the interpretation and generalization of the findings. First, the predominance of cross-sectional studies prevents the establishment of causal relationships between RPL and the psychological effects observed. Second, the frequent use of self-reports as a source of data introduces recall and social desirability biases. Third, there is marked methodological heterogeneity among the studies, both in the scales used to measure psychopathology and quality of life and in the definitions of RPL, making it difficult to compare research. Fourth, the diverse sociodemographic characteristics of the samples, coupled with the low representation of vulnerable populations (adolescent women, women living in poverty, ethnic minorities, or women with multiple pregnancies), limit the generalizability of the results to other contexts. Fifth, there are gaps in the inclusion of important covariates such as the number of living children, the time elapsed since the loss, or the presence of previous psychological symptoms. Sixth, many studies did not differentiate between types of loss (miscarriage, stillbirth, neonatal death), which prevents an assessment of their specific impact. Furthermore, the limited availability of data on stress, on couples (including same-sex couples), and on specific experiences such as RPL restricts a comprehensive understanding of the phenomenon. These weaknesses reflect the need for more longitudinal, representative, and culturally diverse research that addresses the emotional impact of RPL in greater depth. Lastly, due to the predominance of cross-sectional designs, causal inferences cannot be made, and the results should be interpreted as associations.

Despite its limitations, this review also has significant strengths that reinforce its validity and clinical utility. First, it highlights the exclusive use of validated psychological instruments and a systematic assessment of the quality of the included studies, which ensures greater reliability of the results. Second, relevant patterns were identified in levels of anxiety, depression, and stress, suggesting that psychopathology may serve as a risk indicator in this population. Third, it emphasizes the need to offer continuous psychological care throughout pregnancy and after RPL, and to adapt the health system to respond empathetically and appropriately to the emotional needs of these women and their partners. Fourth, it highlights the importance of effective communication between professionals, individualized follow-up, easy access to mental health services, and explicit recognition of experiences of loss. Finally, the report highlights the urgent need to develop specific psychosocial interventions aimed at reducing emotional distress during subsequent pregnancies, with the aim of improving the well-being of mothers, their partners, and future children.

Future research should focus on examining the differential impact of different types of RPL on mental health during subsequent pregnancies, considering understudied variables such as attachment, post-traumatic stress disorder, number of living children, and the interval between the loss and the current pregnancy. It is essential to broaden the representativeness of samples, including women at higher risk of loss, same-sex couples, and different cultural and socioeconomic contexts. Likewise, definitions and methodologies need to be standardized to improve comparability between studies and analyze how levels of anxiety, depression, and stress vary in the different trimesters of pregnancy after loss. Another crucial line of research is the development and evaluation of specific psychosocial interventions, as well as the implementation of clinical care that ensures continuity, effective communication, and mental health monitoring in these cases. Finally, it is important to explore the long-term consequences for the mother, the couple, and the family, including the impact on attachment and emotional well-being, and to control for previous baseline variables to avoid bias in the results.