Zhang XY, Shi Y, Lu YT, Zhong YJ, Wu JX, Wang XQ. Constructing a predictive model for postpartum anxiety in patients with preeclampsia based on multidimensional indicators and its application. World J Psychiatry 2026; 16(3): 113273 [DOI: 10.5498/wjp.v16.i3.113273]
Corresponding Author of This Article
Xiao-Qing Wang, MD, Attending Doctor, Department of Gynaecology and Obstetrics, Suzhou Ninth People’s Hospital (Affiliated with Soochow University), No. 2666 Ludang Road, Taihu New Town, Wujiang District, Suzhou 215200, Jiangsu Province, China. wang_xqing@126.com
Research Domain of This Article
Psychology
Article-Type of This Article
Observational Study
Open-Access Policy of This Article
This article is an open-access article which was selected by an in-house editor and fully peer-reviewed by external reviewers. It is distributed in accordance with the Creative Commons Attribution Non Commercial (CC BY-NC 4.0) license, which permits others to distribute, remix, adapt, build upon this work non-commercially, and license their derivative works on different terms, provided the original work is properly cited and the use is non-commercial. See: http://creativecommons.org/licenses/by-nc/4.0/
Mar 19, 2026 (publication date) through Feb 28, 2026
Times Cited of This Article
Times Cited (0)
Journal Information of This Article
Publication Name
World Journal of Psychiatry
ISSN
2220-3206
Publisher of This Article
Baishideng Publishing Group Inc, 7041 Koll Center Parkway, Suite 160, Pleasanton, CA 94566, USA
Share the Article
Zhang XY, Shi Y, Lu YT, Zhong YJ, Wu JX, Wang XQ. Constructing a predictive model for postpartum anxiety in patients with preeclampsia based on multidimensional indicators and its application. World J Psychiatry 2026; 16(3): 113273 [DOI: 10.5498/wjp.v16.i3.113273]
Author contributions: Zhang XY and Wang XQ researched and wrote the manuscript, and conducted the analysis; Shi Y provided guidance for the research; Lu YT, Zhong YJ, and Wu JX contributed to conceiving the research and analyzing data; and all authors reviewed and approved the final manuscript.
Supported by 2023 Academy-Level Research Start-Up Fund Project, No. YK202313.
Institutional review board statement: This study has been approved by the Ethics Committee of Suzhou Ninth People’s Hospital (Affiliated with Soochow University), No. KY2023-015-01.
Informed consent statement: Patients were not required to give informed consent to the study because the analysis used anonymous clinical data that were obtained after each patient agreed to treatment by written consent.
Conflict-of-interest statement: All the authors report no relevant conflicts of interest for this article.
STROBE statement: The authors have read the STROBE Statement-checklist of items, and the manuscript was prepared and revised according to the STROBE Statement-checklist of items.
Data sharing statement: No additional data are available.
Corresponding author: Xiao-Qing Wang, MD, Attending Doctor, Department of Gynaecology and Obstetrics, Suzhou Ninth People’s Hospital (Affiliated with Soochow University), No. 2666 Ludang Road, Taihu New Town, Wujiang District, Suzhou 215200, Jiangsu Province, China. wang_xqing@126.com
Received: October 17, 2025 Revised: November 20, 2025 Accepted: December 16, 2025 Published online: March 19, 2026 Processing time: 134 Days and 0.3 Hours
Abstract
BACKGROUND
Preeclampsia (PE) substantially increases the risk of postpartum anxiety, yet limited research has examined how disease onset and clinical features, such as blood pressure control and body mass index (BMI) changes during pregnancy, affect this risk.
AIM
To develop and apply a predictive model for postpartum anxiety disorder in patients with PE based on multidimensional indicators.
METHODS
A cross-sectional study was conducted among 196 patients with PE admitted to the Department of Obstetrics, Ninth People’s Hospital of Suzhou (Affiliated with Soochow University), from June 2019 to June 2024. According to the self-rating anxiety scale at six weeks postpartum, participants were divided into anxiety and no-anxiety groups. Two data sets were analyzed, and multivariate logistic regression was performed to identify risk and protective factors. Regression coefficients and constants were used to construct the predictive model. Model performance was evaluated using the receiver operating characteristic curve and area under the curve, along with a goodness-of-fit test. The model was then validated with clinical data.
RESULTS
Of the 196 patients with PE evaluated using the self-rating anxiety scale at six weeks postpartum, 51 (26.02%) patients showed anxiety symptoms. Significant group differences (P < 0.05) were observed for blood pressure control, BMI increase, hematocrit (Hct), family relationships, and psychological resilience. Logistic regression indicated that, poor blood pressure control, greater BMI increase, elevated Hct levels, and strained family relationships during pregnancy were risk factors for postpartum anxiety in patients with PE (P < 0.05), whereas higher psychological resilience was a protective factor (P < 0.05). The prediction model was defined as: Logit (P) = 0.684 × pregnancy blood pressure control + 0.805 × pregnancy BMI increase + 0.756 × Hct + 1.063 × family relationship - 1.105 × psychological resilience score - 5.487. The model’s area under the curve (0.908) exceeded that of individual indicators: Blood pressure control (0.794), BMI increase (0.814), Hct (0.808), family relationships (0.840), and psychological resilience (0.833). The goodness-of-fit test showed no overfitting (χ2 = 1.904, P = 0.725). Clinical validation demonstrated sensitivity of 85.71%, specificity of 87.72%, and accuracy of 87.18%.
CONCLUSION
Postpartum anxiety risk in patients with PE is associated with poor blood pressure control, excessive BMI gain, elevated Hct index, and poor family relationships, while strong psychological resilience serve as a protective factor. The developed prediction model effectively supports clinical assessment and targeted management of postpartum anxiety in patients with PE.
Core Tip: Preeclampsia, as an idiopathic disease during pregnancy, is not conducive to the physical and mental health of pregnant women and the growth and development of the fetus, and will increase the risk of postpartum anxiety in pregnant women. This study identified the influencing factors of postpartum anxiety in patients with preeclampsia and constructed a predictive model, offering guidance for clinical risk assessment and preventive management.
Citation: Zhang XY, Shi Y, Lu YT, Zhong YJ, Wu JX, Wang XQ. Constructing a predictive model for postpartum anxiety in patients with preeclampsia based on multidimensional indicators and its application. World J Psychiatry 2026; 16(3): 113273
Anxiety disorder is a mental disorder characterized by excessive worry, tension, unease, and fear[1]. Patients often present with autonomic symptoms and behavioral manifestations such as restlessness. When anxiety disorder develops after childbirth, it can directly impair maternal role adaptation and negatively affect the mother’s physical and mental health, the mother-infant relationship, infant feeding, and child growth and development. In severe cases, it may progress to postpartum depression. Studies report that the incidence of postpartum anxiety disorder can reach 20%[2]. Preeclampsia (PE), an idiopathic disorder of pregnancy[3], significantly increases the risk of postpartum anxiety disorder[4], though its specific mechanism remains unclear. Overweight and obesity are key risk factors for PE[5], while elevated blood pressure is its primary clinical feature[6]. These factors are closely linked to the postpartum mental health of patients with PE[7]. However, existing research largely focuses on marital relationships, postpartum estrogen fluctuations, and family-related discrimination[8], often overlooking the disease etiology and clinical characteristics of PE and their effects on postpartum mental health. Therefore, this study integrates multiple dimensions, including blood pressure control during pregnancy, gestational weight change, psychological resilience, and family relationships to analyze the mechanism underlying postpartum anxiety in patients with PE. Based on these factors, a multidimensional prediction model for postpartum anxiety in patients with PE was developed to support obstetricians in assessing postpartum anxiety risk in patients with PE and implementing appropriate management strategies.
MATERIALS AND METHODS
Research object
A cross-sectional survey was conducted among 196 patients with PE treated at the Department of Obstetrics, Suzhou Ninth People’s Hospital (Affiliated with Soochow University) between June 2019 and June 2024, all of whom completed delivery. According to the self-rating anxiety scale (SAS) administered at six weeks postpartum, participants were divided into anxiety and nonanxiety groups. Data from both groups were analyzed to establish a predictive model.
Inclusion criteria: (1) Meeting the diagnostic criteria for PE in the 9th edition of Obstetrics and Gynecology[9]: Systolic blood pressure (SBP) ≥ 140 mm of mercury (1 mm of mercury = 0.133 kilopascal) and/or diastolic blood pressure (DBP) ≥ 90 mm of mercury after 20 weeks of gestation, accompanied by proteinuria or, in its absence, complications such as thrombocytopenia, liver or kidney dysfunction, pulmonary edema, or new-onset central nervous system abnormalities, or visual impairment; (2) Completion of all prenatal checkups at the study hospital; (3) Age ≥ 18 years old; (4) Natural conception, singleton pregnancy, and gestational age ≥ 37 weeks at delivery; and (5) No prior cognitive impairment or mental illness, with normal communication ability.
Exclusion criteria: (1) Preexisting anxiety or depression during pregnancy; (2) Severe adverse pregnancy outcomes, such as intrauterine fetal death, neonatal asphyxia, malformations, or serious maternal complications (e.g., postpartum hemorrhage, infection); (3) Concurrent malignant tumors; (4) First-degree relatives with mental illness; (5) Severe hepatic dysfunction (Child-Pugh grade C); (6) Severe renal dysfunction (glomerular filtration rate < 30 mL/minute or creatinine > 500 μmol/L); and (7) Ongoing medical disputes with the hospital.
Research methods
General information collection: A self-designed information form was used to collect patients’ demographic and clinical data, including age, pre-pregnancy body mass index (BMI), monthly family income, educational level, fertility history, hematologic parameters at full term (37 weeks) [hematocrit (Hct), hemoglobin, platelet count, and white blood cell count], delivery information (gestational age and mode of delivery), neonatal sex, birth weight, and feeding method.
Blood pressure control during pregnancy: Following diagnosis of PE, antihypertensive therapy was initiated. According to the guidelines for the management of gestational hypertension[10], the target blood pressure for pregnant women without organ dysfunction is an SBP of 130-155 mm of mercury and DBP of 80-105 mm of mercury. For pregnant women with concurrent organ dysfunction, SBP should be maintained at 130-139 mm of mercury and DBP 80-89 mm of mercury. Blood pressure meeting these criteria was defined as well controlled; otherwise, control was considered poor.
Changes in BMI during pregnancy: BMI was calculated as weight (kg)/height (m2), and change during pregnancy was defined as: ΔBMI = predelivery BMI - pre-pregnancy BMI[11].
Family relationships: Family relationships were assessed using the Chinese version of the Family Adaptability and Cohesion Scale[12], which evaluates two dimensions: Intimacy and adaptability. Intimacy reflects emotional closeness between the patient and family members, whereas adaptability reflects the patient’s ability to adjust to situational or developmental changes. The scale contains 30 items scored on a 5-point Likert scale, yielding a total of 150 points. Higher scores indicate better family relationship; > 70 points indicate harmonious, and ≤ 70 points indicate non-harmonious relationships.
Psychological resilience: The Chinese version of the Connor-Davidson Resilience Scale[13] was used to evaluate the psychological resilience status of the PE patients, consisting of three dimensions: Resilience (13 items), strength (8 items), and optimism (4 items). Each item is rated on a 5-point Likert scale, with a score range of 25-125 points; higher scores indicate stronger psychological resilience.
Diagnosis of anxiety disorder: Postpartum anxiety was assessed at six weeks after delivery using the SAS[14]. The scale includes 20 items (15 positively scored, 5 negatively scored, each rated on a 4-point Likert scale. The gross score multiplied by 1.25 yields the standard score; a score > 50 indicates the presence of anxiety disorder.
Statistical analysis
Data were analyzed using SPSS 25.0. Measurement data were tested for normality using the Shapiro-Wilk test and expressed as mean ± SD. Independent t-tests were applied for group comparisons. Count data were expressed as n (%) and compared using χ2 test. Multivariate logistic regression was used to explore the risk and protective factors for postpartum anxiety in patients with PE, and a logistic regression formula was derived from regression coefficients (β) and constants. A nomogram was generated using R 4.0.3 for intuitive visualization. The area under the receiver operating characteristic (ROC) and area under the curve was used to evaluate predictive efficiency; the goodness-of-fit test was used to evaluate overfitting. The model was subsequently validated using real data. P value < 0.05 indicated that the difference was statistically significant.
RESULTS
Basic information of patients with PE and occurrence of postpartum anxiety disorder
Among the 196 patients with PE, the mean age was 29.09 ± 2.78 years and the mean pre-pregnancy BMI was 23.38 ± 2.78 kg/m2. Regarding education, 97 patients (49.49%) had a high school education or below, while 99 (50.51%) had a college degree or higher. Monthly household income was < 8000 RMB in 85 (43.37%) and ≥ 8000 RMB in 111 (56.63%) participants. There were 40 primiparous (20.41%) and 156 multiparous (79.59%) women. Blood pressure during pregnancy was well controlled in 133 (67.86%) and poorly controlled in 63 (32.14%) patients. The mean gestational BMI increase was 4.97 ± 1.38 kg/m2, and the mean gestational age at delivery was 38.05 ± 0.32 weeks. Delivery methods included 122 cesarean sections (62.24%) and 74 vaginal deliveries (37.76%). Neonatal sex distribution was 114 males (58.16%) and 82 females (41.84%), with a mean birth weight of 3.21 ± 0.39 kg. Feeding methods included 34 artificial (17.35%), 93 mixed (47.45%), and 69 exclusive breastfeeding (35.20%) cases. Family relationships were harmonious in 108 (55.10%) and inharmonious in 88 (44.90%) patients. The mean maternal psychological resilience score was 81.75 ± 7.85 points. At the 6-week postpartum follow-up, SAS assessments identified 51 cases of anxiety (anxiety group), yielding an incidence of 26.02% (51/196); 145 had no-anxiety symptoms (no-anxiety group).
Univariate analysis of postpartum anxiety in patients with PE
Significant differences (P < 0.05) were observed between the anxiety and no-anxiety groups in blood pressure control during pregnancy, BMI increase, Hct, family relationships, and psychological resilience scores (Table 1).
Table 1 Univariate analysis of postpartum anxiety disorder in preeclampsia patients, n (%)/mean ± SD.
Multivariate logistic regression analysis of postpartum anxiety in patients with PE
Taking postpartum anxiety (0 = no; 1 = yes) as the dependent variable, variables with statistical significance in the univariate analysis (maternal blood pressure control during pregnancy, BMI increase, Hct, family relationship, and psychological resilience score) were entered as independent variables (Table 2). Multivariate logistic regression revealed that poor blood pressure control, greater BMI increase, elevated Hct index, and poor family relationships were risk factors for postpartum anxiety in PE patients (P < 0.05), whereas psychological resilience was a protective factor (P < 0.05) (Table 3).
Construction of a risk prediction model for postpartum anxiety in patients with PE
Based on multivariate analysis, the risk prediction model for postpartum anxiety in patients with PE was established as follows: Logit (P) = 0.684 × gestational blood pressure control (0 = good; 1 = poor) + 0.805 × gestational BMI increase (actual value) + 0.756 × Hct (actual value) + 1.063 × family relationships (0 = harmonious; 1 = discordant) - 1.105 × psychological resilience score (actual value) - 5.487. A nomogram visualizing this logistic regression model was created using R software, converting the total score of each indicator into a predicted probability (Figure 1). The area under the curve value under the ROC curve for this model was 0.908, exceeding that for gestational blood pressure control (0.794), gestational BMI increase (0.814), Hct (0.808), family relationship (0.840), and psychological resilience score (0.833) (Figure 2; Table 4). The goodness-of-fit test (χ2 = 1.904, P = 0.725) showed no significant deviation between predicted and actual values, indicating the absence of overfitting.
Figure 2 Receiver operating characteristic curve analysis of the model and each index.
A: Receiver operating characteristic (ROC) curve analysis of blood pressure control during pregnancy; B: ROC curve analysis of body mass index increases during pregnancy; C: ROC curve analysis of hematocrit index; D: ROC curve analysis of family relationship; E: ROC curve analysis of psychological resilience score; F: ROC curve analysis of each index combination prediction model. AUC: Area under the curve; CI: Confidence interval.
Table 4 Comparison of area under the curve values between the model and each index.
Application effect of prediction model in clinical practice
The predictive model was applied to 78 patients with PE who delivered at Suzhou Ninth People’s Hospital (Suzhou Ninth Affiliated Hospital of Soochow University) between August 2024 and June 2025. The model’s sensitivity was 85.71% (18/21), specificity was 87.72% (50/57), and overall accuracy was 87.18% (68/78) (Table 5).
Table 5 Predicted and actual values of the prediction model.
The clinical manifestations of PE are diverse, often involving multiple organ systems and reducing blood flow velocity, which can affect the liver, kidneys, and brain. Microvascular leakage may cause edema and proteinuria, and impaired uterine perfusion directly endangers fetal health. These complications can increase psychological stress and the risk of postpartum anxiety disorder[15], although the underlying mechanisms remain unclear. Therefore, understanding these mechanisms is essential for developing interventions to reduce postpartum anxiety in patients with PE.
Abnormal blood pressure elevation is the hallmark of PE. After diagnosis, persistently high arterial pressure reflects poor blood pressure control and may cause physical discomfort and fear concerning maternal and infant safety. Alers et al[16] found that poor blood pressure control during pregnancy contribute to postpartum cognitive decline, increasing susceptibility to anxiety and depression. Consistent with this, the present study identified poor blood pressure control as a risk factor for postpartum anxiety in patients with PE. The likely mechanism is that poor blood pressure control during pregnancy in patients with PE induces cerebral small-artery spasm, damages the blood-brain barrier’s endothelial function, and causes symptoms such as headache and visual impairment. These distressing experiences may persist as postpartum fears of permanent neurological damage, contributing to postpartum anxiety[17]. Additionally, mild or transient cerebral hemodynamic changes and endothelial injury caused by poor blood pressure control during pregnancy may subtly affect emotion-regulating regions, such as the amygdala and prefrontal cortex, heightening vulnerability to postpartum anxiety disorders. BMI is a standard measure of adiposity, and overweight or obesity is a key contributor to PE[18]. Women with higher pre-pregnancy BMI often experience excessive gestational weight gain[19], leading to postpartum weight retention. Bazzazian et al[20] reported that postpartum weight retention increases maternal anxiety, although they did not examine the role of BMI increase during pregnancy. Hill et al[21] found that anxiety, low self-esteem, and unsatisfactory body image are associated with excessive weight gain during pregnancy. Similarly, this study found that excessive BMI increase during pregnancy is a risk factor for postpartum anxiety in patients with PE. The mechanism may involve abnormal lipid metabolism and chronic low-grade inflammation, which disrupt neurotransmitter function and the hypothalamic-pituitary-adrenal axis, increasing susceptibility to anxiety. Moreover, substantial gestational weight gain during pregnancy can negatively affect body image, causing shame, loss of control, and self-consciousness[22], which have lasting psychological effects[23]. It may also increase upper respiratory tract resistance and the risk of obstructive sleep apnea, reducing sleep quality and impeding physical and neurological recovery, thus heightening irritability and emotional instability that contribute to postpartum anxiety disorder. Biological factors may further underlie postpartum anxiety in patients with PE. Ramiro-Cortijo et al[24] found that Hct index was significantly higher in patients with PE than in healthy pregnant women and correlated positively with disease severity. Similarly, Roomruangwong et al[25] reported that Hct index in the late pregnancy (at 37 weeks of gestation) could accurately predict anxiety scores 4-6 weeks postpartum, aligning with the present findings. Elevated Hct increases blood viscosity and alters hemorheology; combined with oxidative stress and endothelial injury associated with polycythemia, these changes promote chronic inflammation and establish a biological and physiological basis for psychological disorders such as anxiety and depression, thereby increasing the risk of postpartum anxiety in patients with PE.
The emotional connection between postpartum women and family members, particularly mother-in-law and marital relationships, plays a critical role in family dynamics[26]. In China, the 30-42 days postpartum recovery period is often overseen by in-laws. However, differing understandings of traditional culture, lifestyle habits, and childcare concepts between generations may provoke emotional fluctuations, dissatisfaction, and even conflict, imposing invisible psychological pressure on new mothers[27]. When family relationships are disharmonious, the resulting lack of family support can significantly increase maternal psychological burden and lead to postpartum anxiety[28]. This study identified family disharmony as a risk factor for postpartum anxiety in patients with PE, corroborating previous findings. Psychological resilience refers to an individual’s capacity to adapt to adversity or trauma. Sójta et al[29] found that maternal psychological resilience predicts postpartum anxiety and depression, while Akkuş and Akkuş[30] demonstrated a negative correlation between resilience and anxiety, depression, and stress. These findings suggests that postpartum women with greater psychological resilience, manifested in stress management, emotional regulation, positive cognition, help-seeking, and self-efficacy, are far less likely to develop anxiety than those with low psychological resilience[31]. Therefore, psychological resilience serves as a protective factor against postpartum negative emotions. The possible reason might be that high psychological resilience function as a “buffer”, enabling pregnant women with PE to better cope with the physical and psychological trauma of the disease, the demands of childcare, and the uncertainty about the future[32]. Consequently, assessing psychological resilience before and after delivery is crucial in PE management. Early identification of women with low psychological resilience allows targeted psychological support and timely intervention, mitigating the psychological impact of PE and preventing postpartum anxiety.
Postpartum anxiety in PE patients is influenced by multiple factors. A multi-indicator predictive model, as a quantitative risk assessment tool, can aid clinicians in estimating the likelihood of anxiety occurrence. In this study, several significant predictors were combined to develop such a model. The resulting area under the curve (0.908) exceeded that of individual indicators: Blood pressure control during pregnancy (0.794), BMI increase during pregnancy (0.814), Hct (0.808), family relationship (0.840), and psychological resilience score (0.833). The goodness-of-fit test showed no overfitting, confirming the model’s stability and clinical value. Verification further demonstrated a high overall prediction accuracy of 87.18%, indicating that the multi-indicator model offers superior predictive performance and provides a practical tool for identifying high-risk patients.
This study has several limitations: (1) It is a single-center study with limited representativeness, which constrains the generalizability of the model; multicenter studies with larger sample size are needed for validation; (2) The variable “blood pressure control during pregnancy” was defined using standard management guidelines for hypertensive disorders in pregnancy, but without detailing specific antihypertensive types or dosages, which may have influenced results. Future research should refine this analysis to include the types and dosages of antihypertensive medications for each patient; (3) Due to the low incidence of PE, validation was conducted with a small cohort; larger prospective studies are needed to enhance model reliability; and (4) Although the SAS scale is a widely used self-assessment tool for anxiety, reliance on a single scale may misclassify transient stress responses as anxiety symptoms. Future work should integrate clinical interviews to improve diagnostic precision.
CONCLUSION
In summary, poor blood pressure control during pregnancy, excessive BMI increase during pregnancy, elevated Hct levels, strained family relationships, and low psychological resilience scores are closely associated with postpartum anxiety in patients with PE. Early detection of abnormalities in these indicators should prompt vigilance for postpartum anxiety. A predictive model based on these factors can effectively predict postpartum anxiety risk and guide timely interventions to reduce its occurrence.
Brown MA, Magee LA, Kenny LC, Karumanchi SA, McCarthy FP, Saito S, Hall DR, Warren CE, Adoyi G, Ishaku S; International Society for the Study of Hypertension in Pregnancy (ISSHP). Hypertensive Disorders of Pregnancy: ISSHP Classification, Diagnosis, and Management Recommendations for International Practice.Hypertension. 2018;72:24-43.
[RCA] [PubMed] [DOI] [Full Text][Cited by in Crossref: 724][Cited by in RCA: 1412][Article Influence: 201.7][Reference Citation Analysis (0)]
Provenance and peer review: Unsolicited article; Externally peer reviewed.
Peer-review model: Single blind
Specialty type: Psychiatry
Country of origin: China
Peer-review report’s classification
Scientific Quality: Grade B, Grade C
Novelty: Grade B, Grade C
Creativity or Innovation: Grade B, Grade B
Scientific Significance: Grade C, Grade C
Open Access: This article is an open-access article that was selected by an in-house editor and fully peer-reviewed by external reviewers. It is distributed in accordance with the Creative Commons Attribution NonCommercial (CC BY-NC 4.0) license, which permits others to distribute, remix, adapt, build upon this work non-commercially, and license their derivative works on different terms, provided the original work is properly cited and the use is non-commercial. See: https://creativecommons.org/Licenses/by-nc/4.0/
P-Reviewer: Duplenne L, PhD, France; Zahn R, PhD, France S-Editor: Jiang HX L-Editor: A P-Editor: Zhang YL
