INTRODUCTION
When women become mothers, they are under tremendous pressure during the perinatal period, leading to anxiety and stress due to prevailing role models and social expectations1. The perinatal period is considered a critical phase in a woman’s life, and is not without difficulties for mental health2. In particular, the postpartum period can be emotionally challenging for mothers, with high levels of anxiety, stress, and depression.
Anxiety manifests itself through agitation, palpitations, nausea, control problems, irrational thoughts, or social avoidance3. Anxiety can begin to be experienced by a mother during pregnancy and up to the first year postpartum and is referred to as postpartum anxiety3.
Although postpartum anxiety is an underdiagnosed problem1, it can be present in up to 42% of women after giving birth4-6. This figure varies depending on severity and other factors, but globally, it is estimated that 1 in 4 women will experience postpartum anxiety, which can last up to a year after childbirth4.
Although the causes associated with the appearance of anxiety are varied, the literature has described socio-economic factors7, low social support8, obstetric variables such as the experience of childbirth and the type of birth, and even previous mental health disorders of the woman herself 9,10. Cohabitation with the partner and the couples’ relationship has also been described as a variable associated with the onset of postpartum anxiety11.
Anxiety, to whatever degree, has been associated with an increased likelihood of developing adverse outcomes in both mother and newborn12,13. Thus, in the postpartum period, anxiety can influence the attachment between the baby and the mother14. It can also affect the social and occupational level in the late postpartum period15.
Women who suffer from anxiety after childbirth may develop other disorders during the postpartum stage, such as postpartum depression7,12,16, maternal stress11 or suicidal ideation, which is the main predictor of perinatal suicide17,18. It has been estimated that anxiety-associated comorbidity costs an estimated £8.1 billion per birth cohort per year in countries such as the UK19.
Considering the magnitude and impact of postpartum anxiety on maternal and neonatal health, together with the few studies that address this pathology (most studies focus on postpartum depression), as well as the recommendation to carry out studies on postpartum anxiety20, the present study aims to study the factors associated with the presence of this postpartum anxiety, as well as to determine the prevalence of this disorder.
METHODS
Design and participant selection
A cross-sectional study was conducted on women who had given birth in Spain during the last half of 2021 and the first half of 2022 who met the following inclusion criteria: having given birth less than 18 months ago and not having suffered a neonatal loss. Women under 18 years of age and those who did not understand Spanish (language barrier) were excluded.
In an attempt to estimate the sample size to obtain valid estimations, the maximum modeling principle21 was followed. This requires 10 events (women with anxiety) per each included variable. Considering that the prevalence in this reference population and sociodemographic context ranges 20–25%5, it would be necessary to recruit a sample of 800 women (200 with anxiety) for an initial model of 20 variables.
The questionnaire was distributed by collaborating health workers in the clinical setting, including during postpartum and postnatal consultations, as well as midwifery consultations in medical centers (including hospital or health center as well as midwife-led clinics). This allowed more women to be recruited (via purpose sampling method) including during the subsequent check-ups where the mother attended with the newborn. Once women were selected, they were given the information about the study and lately the choice to participate and sign the informed consent form. Mechanisms for resolving queries were established through a WhatsApp group among collaborating professionals in order to provide homogeneous answers to any questions that might arise.
Ethical considerations
The present study was approved by the Research Ethics Committee of the Province of Jaén (DCVA-21/2012-N-21). All women participating in the study signed the informed consent form.
Information source and study variables
The data were collected using a self-developed, previously piloted questionnaire, which was distributed in different hospitals and health centers. This questionnaire contained open and closed questions, with a language understandable to all educational levels. It included sociodemographic variables such as age, income level, lifestyles such as alcohol or tobacco consumption, obstetric, family, and personal history of the pregnancy itself, and also variables related to the newborn. To assess the presence of anxiety, the Generalized Anxiety Disorder Screener (GAD-7) scale was used in its validated version in a population similar to that of our study20. The GAD-7 is a self-administered 7-item scale with four response options in ascending order from 0 to 4 points (never, several days, half of the days, almost every day), with a score ≥5 indicating the presence of anxiety and ≥10 indicating moderate or severe anxiety20.
To measure mother-child bond and attachment, the "Maternal - Child Bond and Attachment" ("VAMF", for its initials in Spanish) tool was used. The VAMF tool is a self-administered 29-item scale that measures mother-child bonding and attachment, and is designed and validated for application in the postpartum period and up to 18 months of age of the infant. The Edinburgh Postpartum Depression Scale (EPDS)23 was used to determine the risk of postpartum depression (PPD). The Woman Abuse Screening Tool (WAST) was used to screen for Intimate Partner Violence (IVP)24. All these instruments have been validated in a population similar to that of the present study.
Statistical analysis
For sociodemographic and clinical data, absolute (n) and relative frequencies (%) were used to describe qualitative variables, and mean and standard deviation (SD) were used to describe quantitative variables. A bivariate and multivariate analysis was then performed between anxiety risk and moderate-severe anxiety risk with possible associated factors. In the first case, anxiety risk was considered as ‘No anxiety’ with scores <5 on the GAD-7 and ‘Anxiety’ with scores ≥5. In the case of moderate-severe anxiety risk, the comparison was between ‘Low or no anxiety levels’ with scores <10 and ‘Moderate-severe anxiety’ with scores ≥10.
Binary logistic regression was used to estimate odds ratios (ORs) and adjusted odds ratios (AORs) with their respective 95% confidence intervals. When multivariate analysis was performed, the backward stepwise procedure was used, and all variables from the bivariate analysis were included whether or not statistical significance was observed. For this analysis, the dependent variable GAD-7 score was dichotomized, while the independent variables were entered into the model in their original form (age, the VAMF and EPDS scales continuously and without transformations or categorizations).
Finally, the predictive ability of the model was estimated using the area under the receiver operating characteristic (ROC) curve (AUC). In order to assess the prediction in qualitative terms, the Swets criterion25 was used with values: 0.5–0.6 (bad), 0.6–0.7 (poor), 0.7–0.8 (satisfactory), 0.8–0.9 (good), and 0.9–1.0 (excellent). The statistical program SPSS 29.0 was used for data analysis.
RESULTS
A cross-sectional study was conducted with 820 postpartum women in Spain. The mean age was 34.30 years (SD=4.06), 58.7% (481) were primiparous and 41.2% (338) were multiparous; 9.0% (47) of the women had an unplanned pregnancy, while 13.5% (111) needed fertility treatment. A total of 24.9% (204) had experienced some mental issues during their lifetime. In terms of experience during labor, 11.5% (94) had a bad experience, and 70.4% (577) defined it as good or very good. The treatment received by the professional healthcare team was defined as bad or very bad by 3.4% (28) women, while 88.9% (729) described it as good or very good; 33.8% (277) of the women were screened as high-risk from suffering IPV. The mean score of EPDS was 7.44 (SD=4.70). Related to anxiety, 52.7% (423) of the women did not suffer any kind of anxiety, whereas 36.1% (296) suffered from mild anxiety, 8.5% (70) moderate, and 2.7% (22) severe anxiety. The mean score of anxiety in the whole sample was 4.94 (SD=3.83). The rest of the data can be seen in Table 1.
Table 1
Sociodemographic and clinical characteristics of the study sample, Spain, 2021–2022 (N=820)
The relation between the different scales was studied, finding a statistically significant relation between the mean score of EPDS (p≤0.001), VAMF bonding (p≤0.001), and VAMF attachment (p=0.002), and the presence of any kind of anxiety. This can be seen in Table 2.
Table 2
Distribution of scores between EPDS, VAMF bonding, VAMF attachment, and the presence of anxiety, Spain, 2021–2022 (N=820)
Subsequently, bivariate and multivariate analyses were performed to determine which factors were associated with any kind of anxiety on the GAD-7 questionnaire compared to no anxiety. Factors that were associated with a higher GAD-7 score were a high EPDS score (AOR=1.68; 95% CI: 1.55–1.81), smoking habit (AOR=1.97; 95% CI: 1.01–3.82), had suffered from any mental issues during lifetime (AOR=1.77; 95% CI: 1.13–2.79), baby having any problem currently (AOR=2.70; 95% CI: 1.34–5.47), or WAST screen positive as high-risk (AOR=1.53; 95% CI: 1.01–2.31). Protective factors associated with anxiety included the VAMF bonding score (AOR=0.90; 95% CI: 0.85–0.96) and a monthly income level between 1000–1999 € (AOR=0.55; 95% CI: 0.31–0.95) appeared. This can be seen in Table 3. The predictive capability for anxiety risk presented a AUC-ROC of 0.90 (95% CI: 0.88– 0.92), with an exceptional capability to classify subjects according to the Swets criterion. The ROC curve can be seen in Figure 1.
Table 3
Bivariable and multivariate analysis of the factors associated with anxiety compared to no anxiety, on the GAD-7 questionnaire, Spain, 2021–2022 (N=820)
[i] The backward stepwise procedure was used, and all variables from the bivariate analysis were included whether or not statistical significance was observed. For this analysis, the dependent variable GAD-7 was dichotomized, while the independent variables were entered into the model in their original form (age, the VAMF and EPDS scales continuously and without transformations or categorizations). ® reference categories.
When we compared no anxiety or mild anxiety and the presence of moderate or severe anxiety by performing bivariate and multivariate analyses to determine which factors were associated with these categories, a high EPDS score (AOR=1.35; 95% CI: 1.26–1.44) appeared as a risk factor. However, VAMF bonding score (AOR=0.92; 95% CI: 0.85–0.98), a monthly income level >2000 € (AOR=0.35; 95% CI: 0.15–0.80), and feeling well treated by healthcare professionals (AOR=0.21; 95% CI: 0.07–0.70) emerged as protective factors. The rest of the results of these analyses can be seen in Table 4. The predictive capability for moderate-severe anxiety risk had a AUC-ROC of 0.90 (95% CI: 0.88–0.93), with an exceptional capability to classify subjects according to the Swets criterion. The ROC curve can be seen in Figure 2.
Table 4
Bivariable and multivariate analysis of the factors associated with no or mild anxiety and the presence of moderate or severe anxiety, on the GAD-7 questionnaire, Spain, 2021–2022 (N=820)
[i] The backward stepwise procedure was used, and all variables from the bivariate analysis were included whether or not statistical significance was observed. For this analysis, the dependent variable GAD-7 was dichotomized, while the independent variables were entered into the model in their original form (age, the VAMF and EPDS scales continuously and without transformations or categorizations). ® Reference categories.
DISCUSSION
The study found that nearly 50% of women experienced some level of anxiety postpartum, with over 30% reported mild anxiety, 8% moderate, and nearly 3% severe anxiety. Factors increasing the likelihood of anxiety included higher EPDS scores, smoking, a history of mental health issues, positive WAST results, and having a baby with current problems. In contrast, protective factors against anxiety included strong bonding (VAMF), feeling well-treated by healthcare professionals, and having a moderate to high monthly income, particularly for moderate or severe anxiety.
The prevalence of anxiety detected in our study is slightly higher than that found by other studies, where the prevalence does not exceed 42%4-6. Fawcett et al.5 found in their systematic review with meta-analysis that 2 in 10 women suffer from postpartum anxiety. Likewise, Dennis et al.26, who differentiated between symptoms and the presence of an anxiety diagnosis, also found a lower prevalence than in our population and coincides with those of other authors4. This may be due to the tools used. The GAD-7, which has been used in this study, represents a tool that is easy to administer to establish a rapid screening for anxiety in perinatal populations27.
Women with higher EPDS scores suffer more frequently from postpartum anxiety, which is in line with different studies7,12,16. Sit et al.12 found in their study with 628 women, that at least 1 in 2 women has depression and also anxiety as a second diagnosis. Similarly, 41% of women with anxiety had depressive disorders as a second diagnosis. Both pathologies can coexist in such a way that the presence of one of them favors the development of the other28-30. Anxiety remains one of the most important risk factors for postpartum depression, highlighting the need for effective screening for its possible presence at multiple points in time31,32.
Smoking causes mood swings that can be conducive to the onset of anxiety33. Munafo et al.34 found an association between depressive symptoms and smoking, with a large proportion of women facing both depression and anxiety, and these two conditions are often interrelated28-30.
Mothers whose babies had a problem were more likely to suffer from anxiety, which is in line with other studies35,36. Support from healthcare professionals is crucial, especially if newborns are admitted to hospital. Indeed, the proper treatment by health professionals involved in postpartum care emerged as a protective factor against anxiety. Zhou et al.37 found in their systematic review with meta-analysis where they identified a total of 11 studies and a sample of 2424 women from six different countries, that this is crucial for positive effects on maternal mental health. Women’s experiences would improve if they were given the opportunity to establish a trusting relationship with health professionals38.
Poor partner relationships and even abuse have been shown to be a variable recurrently associated with the onset of anxiety, postpartum depression, and even perinatal suicide11,17 which coincides with our findings. Women consider the emotional and practical support of their partners to be fundamental to mitigating anxiety38 although there are researchers who found opposite results, with the role of the partner being of little importance and not being associated with the presence of anxiety39.
If mothers had experienced a mental health problem during their lifetime, they were more likely to suffer from anxiety, which is in line with other authors40.
The mother–baby bond acts as a protective factor for the development of anxiety during the postpartum period, something that has been found by other authors41. Figueiredo and Costa42 found that anxiety is associated with poorer bonding, which may produce strong negative emotions toward the baby and less emotional involvement with the baby. This can have implications for the cognitive and physiological development of the newborn at an estimated cost of almost £6 billion43.
Several authors found that socioeconomic status plays an important role in increasing the risk of anxiety in women who do not have the resources to meet their financial needs11,44,45. This is in line with our findings that having a moderate/high monthly income protects against the possibility of postpartum anxiety.
Strengths and limitations
The cross-sectional nature of this study limits the ability to draw causal inferences between the identified risk factors and postpartum anxiety. However, this limitation also presents an opportunity for further research in this area. While many of the assessment tools employed were self-reported questionnaires, the presence of a trained midwife or healthcare professional during the completion of these questionnaires helped to minimize potential bias. Consideration should be given to the temporal context of data collection and its potential impact on mental health outcomes in future studies. Enhancing the diversity of samples, including participants from various regions or countries, could improve the reliability of the findings. The questions were designed to be comprehensible across varying educational levels, thereby reducing the likelihood of classification bias. Furthermore, the sample was representative of the target population, with the mean age of participants aligning with the national mean maternal age in the country where the research was conducted. Although recall bias was a consideration, it is unlikely to have significantly affected the results, as the information solicited pertained to recent and salient experiences that would be difficult for participants to forget. One of the primary strengths of the study lies in the use of tools that have been specifically validated and adapted for the population studied, thereby enhancing the relevance and applicability of the findings.
CONCLUSIONS
The present study shows that a high percentage of women experience anxiety, with a notable prevalence of mild anxiety, and a significant number facing moderate and severe levels. Factors such as a history of mental health problems, smoking, and baby-related complications increase the risk of developing anxiety, while strong bonding, adequate care from health professionals, and a stable economic situation seem to offer greater protection against this condition. Knowledge of risk factors will help health professionals providing care to postpartum women to recognize warning signs, enabling early detection and care.
