Development and validation of a marital anxiety scale for married patients with breast cancer of childbearing age Authors: Yiou Fu, Yuqiao Xiao, Mingxia Yao, Yan Zhu, Ke Li, Shiqing Liu, Jinhua Li Published: 2026 ABSTRACT AbstractObjectiveThis study aimed to develop and develop and preliminarily validate a scale for assessing marital anxiety among patients with breast cancer.MethodsGuided by Social Ecological Theory, we employed an exploratory sequential design within a mixed-methods research framework. The initial draft of the scale was developed through a systematic literature review, in-depth qualitative interviews, and expert consultations. A convenience sample of 509 patients with breast cancer was recruited for this study. Participants were randomly divided into two independent sub-samples for exploratory factor analysis (EFA) and confirmatory factor analysis (CFA).ResultsA total of 509 valid questionnaires were collected. The final scale comprised three first-level dimensions, nine second-level dimensions, and 25 items. The Cronbach’s α coefficient for the entire scale was 0.865, with test-retest reliability of 0.902 and split-half reliability of 0.780. The item-level content validity index (I-CVI) for individual items ranged from 0.88 to 1.00; scale-level content validity index (S-CVI)/Ave = 0.987, and S-CVI/Universal Agreement (UA) = 0.81. Convergent validity was confirmed by average variance extracted values exceeding 0.5 and composite reliability values exceeding 0.7 across all dimensions. Several CFA fit indices met acceptable criteria; however, RMSEA and GFI suggested marginal fit, indicating that further validation is needed.ConclusionsThe scale demonstrated preliminary reliability and validity after preliminary verification and can provide a reference tool for assessing marital anxiety among married patients with breast cancer who are of childbearing age. Nevertheless, follow-up studies need to verify its structural stability and applicability to a wider population.Keywords: Patients with breast cancer, Married women, Marital anxiety, Scale development, Reliability, Validity Introduction According to the 2022 global cancer burden data released by the International Agency for Research on Cancer of the World Health Organization, breast cancer remains the most common malignant tumor among women worldwide in terms of both incidence and mortality, endangering women’s health.1 Although the relative survival rate of breast cancer has improved,2 the patient population has become significantly younger.3 Multiple research reports targeting different populations have shown that the proportion of married individuals among breast cancer patients ranges from 55% to 79.5%.4,5 Breast cancer treatment-related issues, including breast loss, sexual dysfunction, and fertility risks, may directly affect marital relationships,6,7 triggering marital anxiety among patients. Marital anxiety refers to the conflicting psychological states of individuals regarding marital stability, marital competence, fertility, and child-rearing when they perceive potential threats to their marriage because of social, personal, or family factors.8 Studies show that patients with breast cancer may become overly sensitive because of concerns about marital stability, undermining the quality of spousal communication.9,10 Long-term marital anxiety can also stimulate the body to continuously release stress hormones, resulting in endocrine diseases.11 Furthermore, a 10.3-year follow-up study of young Chinese patients with breast cancer revealed that poor marital quality nearly tripled the risk of recurrence and death,12 and unmarried patients had a 19.0% higher risk of death compared with married patients.13 Current research on marital anxiety mainly explores healthy married adult women using assessment tools such as the Relationship Assessment Scale (RAS),14 the Marital Adjustment Test (MAT), and the Marital Quality Questionnaire. However, most of these instruments conflate marital anxiety among patients with breast cancer with general cancer-related anxiety, leaving the precise clinical significance of marital anxiety among patients with breast cancer unclear. Therefore, to assist health care providers in accurately identifying marital anxiety among patients with breast cancer and respond to the mental health promotion goals of the World Health Organization’s WHO World Mental Health Report 2022,15 this study aims to develop and validate a Marital Anxiety Scale for patients with breast cancer from the patients' perspective, grounded in three levels of Social Ecological Theory: microsystem, mesosystem, and macrosystem.16 Ultimately, the study aims to enhance public awareness and improve the capacity to maintain marital health, effectively reducing marital anxiety among this patient population. Methods Study design We employed an exploratory mixed methods design. The initial draft of the scale was developed through a literature review and qualitative interviews, followed by a cross-sectional survey to measure and validate marital anxiety among Chinese patients with breast cancer. Stage 1: scale development Determining the initial scale Social Ecological Theory explains the complex interactions between individuals and their environments, encompassing three levels: microsystem, mesosystem, and macrosystem.16 It emphasizes that individuals are embedded within a series of interdependent environmental systems that interact with them and influence their development.16 Numerous domestic and international studies have employed Social Ecological Theory to investigate issues such as the aging status of empty nesters, substance use disorders among adolescents, and childhood obesity prevention.17, 18, 19 These studies effectively demonstrate that the social environment influences individual behaviors and that individual development interacts with and is affected by subsystems (microsystem, mesosystem, and macrosystem). Regarding marital anxiety among patients with breast cancer, core sources of anxiety at the individual level within the microsystem include changes in body image, shifts in self-worth perception, and emotional fluctuations caused by the disease. The mesosystem refers to spousal interactions and dyadic coping, encompassing family responsibilities, sexual experiences, and emotional support from husbands.20 The macrosystem encompasses the economic burden of treatment, reproductive pressures from elders or relatives, and societal perceptions. Based on this framework, the marital anxiety scale for patients with breast cancer is structured around three first-level dimensions: self-perception (microsystem), couples’ response (mesosystem), and family and social pressure (macrosystem) (Fig. 1). Theoretical model of ecosystem theory and marital anxiety in breast cancer patients. To systematically summarize the research status of marital anxiety among patients with breast cancer and provide a solid theoretical and empirical foundation for the scale’s dimensional construction and item design, we conducted a systematic literature search and screening adhering to the PRISMA (Preferred Reporting Items for Systematic Reviews and Meta-Analyses) guidelines.21 The search strategy is detailed in Supplementary File 1. The article inclusion criteria were original research, clinical guidelines, or systematic reviews focusing on patients with breast cancer and examining the association between marital status and anxiety. The exclusion criteria were conference abstracts, dissertations, case reports, duplicate publications, low-quality studies, and studies lacking full-text access or key data. After full-text evaluation, nine studies were ultimately included for subsequent analysis, excluding low-quality and duplicate research. Among the 9 included studies, 2 were clinical guidelines, 1 was a systematic review, and 6 were clinical studies. Although no guidelines or expert consensus statements directly targeting breast cancer-related marital anxiety were identified during the literature search phase, all included studies explored core research themes related to marital anxiety in patients with breast cancer. Marital anxiety was consistently identified as a result of the combined effects of self-perception, couple dual coping, and environmental stress. Based on this, three first-level dimensions were established for the scale: self-perception, couples' response, and family and social pressure. Subsequently, we conducted in-depth interviews and data analysis with 15 patients with breast cancer, and performed data analysis on the interview content. The inclusion criteria were pathologically diagnosed with breast cancer, married women aged 20–49 years, awareness of their diagnosis, and voluntary participation in the study. The exclusion criteria were coexisting malignant tumors or severe physical illnesses that would prevent participation, previous history of mental illness, or cognitive impairment. Through analysis, 3 main themes and 12 sub-themes were identified. Combining these findings with the literature, an initial scale framework comprising 3 first-level dimensions, 8 second-level dimensions, and 28 initial items was developed. Detailed data can be found in the previously published qualitative research results.9 Content validity There is no strict limit to the number of experts required for the Delphi method. It is generally accepted that selecting 15 to 30 experts can yield relatively satisfactory results.22 In this study, 16 experts were selected through purposive sampling, based on principles of complementary professional expertise and reasonable knowledge structure, as well as research objectives, expert authority, and representativeness. To ensure expert representativeness and comprehensive opinions, the panel included three psychology experts, six oncology specialists, five breast cancer specialists, and two nursing management experts. The inclusion criteria were an intermediate or higher professional title, at least 10 years of relevant field experience, and interest in and willingness to participate in the study. The content validity of the scale items was assessed using the expert inquiry method, with content validity calculated separately for first-level dimensions, second-level variables, and initial items. Item importance and relevance were rated on a five-point Likert scale (5 = Very Important to 1 = Very Unimportant) and a four-point Likert scale (4 = Strongly Related to 1 = Weakly Related). The scale items were revised, supplemented, and improved based on expert feedback. Face validity Thirty eligible married patients of childbearing age with breast cancer evaluated the initial item pool for clarity,22 appropriateness, and difficulty. Based on their feedback, the items were refined for language clarity and cultural relevance, resulting in a preliminary test version of the Marital Anxiety Scale for Patients with Breast Cancer. Stage 2: scale validation Recruitment Using convenience sampling, this study was conducted from May to October 2025 at Hunan Cancer Hospital, a tertiary referral cancer hospital in Hunan Province, China, which hosts a dedicated breast cancer center with four specialized wards. The hospital is a regional leader in breast cancer diagnosis and treatment, ensuring access to a large and representative patient population. The inclusion criteria for the participants were pathologically diagnosed with breast cancer, married women aged 20–49 years, aware of their diagnosis, and voluntary participation in the study. The exclusion criteria were coexisting malignant tumors or severe physical illnesses that would prevent participation, previous history of mental illness, or cognitive impairment. The sample size was determined based on exploratory factor analysis (EFA) and confirmatory factor analysis (CFA) standards. For EFA, following the “5–10 times the number of items” rule of thumb,23 a priori statistical power analysis24 was further used to determine the minimum sample size. Assuming a statistical power of 80%, α = 0.05, and a minimum factor loading of 0.45, the required sample size for EFA was calculated as 190. Accounting for a 10% invalid response rate, the target sample size for EFA (n1) was set to 210. For CFA, the sample size was required to be ≥ 200 and larger than the EFA sample,25 therefore, the target sample size for CFA (n2) was set at 220. In total, we initially planned to recruit 430 married patients of childbearing age with breast cancer. Data were collected in two stages. In the first stage, 440 questionnaires were collected, and 10 invalid samples were eliminated, obtaining 430 valid questionnaires (effective rate: 97.7%). The preliminary reliability and validity analyses based on these 430 samples showed that some psychometric indicators did not meet the ideal standards. To further ensure the reliability and stability of the findings, supplementary data collection was conducted from March to April 2026. In the second stage, 300 questionnaires were distributed; after excluding 14 low-quality responses, 286 valid questionnaires remained. The quality-screening criteria included: (1) an excessively short completion time (< 300 s), (2) a clear response pattern, such as selecting the same option for all items, and (3) missing values for key variables exceeding 5%. After merging the first- and second-stage samples, 207 additional low-quality cases were removed, yielding 509 final valid cases. These cases were then randomly divided into two independent and non-overlapping sub-samples for EFA (n1 = 247) and CFA (n2 = 262). The sample sizes of both sub-samples exceeded the minimum requirements determined by a priori power analysis and met the requirements for factor analysis (Fig. 2). Flowchart of questionnaire distribution, retrieval, and sample selection. EFA, exploratory factor analysis; CFA, confirmatory factor analysis. Research tools A self-designed general information questionnaire was used to collect demographic data (age, education level, marital duration, household income, number of children) and stress-related information (sources of stress). Data collection and quality control To ensure data authenticity and reliability, the research members underwent standardized training. Data were collected and entered using a combination of paper surveys and Questionnaire Star, a data-entry software developed by Chinese researchers. Study participants were selected following the inclusion and exclusion criteria, after obtaining permission from the site administrator. Paper surveys were distributed individually. Before distribution, the purpose and significance of the survey were explained to the patients, who were assured of the anonymity and confidentiality of their responses before signing an informed consent form. Whenever possible, participants completed the scale independently. The completed scales were examined and verified immediately after submission. If any items were missing or abnormal, patients were asked to supplement or correct them until the scale was confirmed error-free. The data were double-checked within 24 h and exported to Excel for further analysis. Data analysis SPSS 27.0 and IBM SPSS AMOS 28.0 were used for statistical analyses. Item analysis included critical ratio (CR) testing and Pearson correlation analysis to evaluate item discriminability. The mean and standard deviation were calculated for each item. Reliability was assessed using Cronbach’s α coefficient (internal consistency), test-retest reliability (intraclass correlation coefficient for scores at 2-week intervals), and split-half reliability (correlation between odd- and even-numbered items). Content validity was evaluated using the item-level content validity index (I-CVI) and scale-level content validity index (S-CVI). Kendall’s coefficient of concordance was used to assess consistency among expert ratings. Construct validity was examined using EFA and CFA. The Kaiser-Meyer-Olkin (KMO) test and Bartlett’s test of sphericity were used to confirm the suitability of the data for EFA. For EFA (Sample 1), principal factor analysis method was performed, and factors with eigenvalues greater than 1 were retained. For CFA (Sample 2), model fit was evaluated using multiple indices: chi-square/degrees of freedom (χ2/df), root mean square error of approximation (RMSEA), goodness-of-fit index (GFI), comparative fit index (CFI), normed fit index (NFI), and incremental fit index (IFI). Convergent validity was assessed using AVE and CR. The AVE square root method was employed to examine discriminant validity. Results Stage 1 scale development A two-round Delphi expert consultation was employed to refine the scale. In the first round, expert consensus was high, with an average coefficient of variation (CV) of 0.11 across all items. Subsequent revisions were guided by quantitative consensus metrics and qualitative feedback from experts, followed by iterative group discussions. The research team adjusted the scale by removing four items and adding four new items. The second round of Delphi consultation maintained high consensus, with the average CV remaining at 0.11. This round further optimized the scale by removing three items and adding one. Concurrently, the research team invited experts to collaboratively refine individual vocabulary terms for greater clarity and alignment with Chinese linguistic conventions. Lastly, based on the core content of the initial scale (including 3 first-level dimensions, 8 second-level dimensions, and 28 items), the revised scale comprised 3 first-level dimensions (self-perception, couples’ response, and family and social pressure), 9 second-level dimensions, and 26 items. A pilot survey was conducted by distributing 30 questionnaires, all of which were retrieved, resulting in a 100% recovery rate. All 30 questionnaires were valid, yielding a 100% valid recovery rate. No item had a missing response rate of 20% or higher. All items were clearly articulated, and patients demonstrated a consistent understanding of the questions, with no ambiguous feedback reported. Interviews confirmed that no key dimensions of anxiety were omitted, eliminating the need to add items. The average time for patients to complete the scale was 7.2 ± 1.3 minutes, with no evidence of response fatigue or reluctance. Overall, the administration process was straightforward and practical. Stage 2: scale validation Sample characteristics The total sample size was N = 509 (n1 = 247, n2 = 262). All patients with breast cancer in this sample were married (Table 1) (Note: Treatment modalities and major sources of stress were multiple-choice questions; therefore, percentages may sum to more than 100%). Table 1Demographic and clinical characteristics of participants (N = 509).Table 1VariablesEFACFAn1 = 247%n2 = 262%Age, years ≤ 358634.89235.1 36-4411144.911845.0 ≥ 455020.25219.8Family annual income (CNY) < 20,0005522.36324.1 20,000-50,0006827.59235.1 > 50,00012450.210740.8Medical insurance Yes22791.924292.4 No208.1207.6Family structure Nuclear family9940.110439.7 Stem family3112.63413.0 Couple-only family8534.49034.4 Special family3213.03413.0Education level Junior high school and below9739.310640.5 High school or technical secondary school7630.88733.2 Junior College/University Bachelor5421.94216.0 Master’s degree or above208.12710.3Marital status First marriage24097.225195.8 Remarried72.8114.2Marital duration, years ≤ 56024.33111.8 6-159839.710038.2 > 158936.013150.0Couples living conditions Living together20482.616161.5 Separate rooms3313.44216.0 Separated104.05922.5Number of children 02911.793.4 110743.38934.0 ≥ 211144.916462.6Duration of illness, months < 616566.810439.7 6-123915.84316.4 > 124317.411543.9Treatment methods Surgery17872.118972.1 Chemotherapy10743.319674.8 Radiotherapy5421.95621.4 Endocrine therapy4417.89937.8 Immunotherapy20.84316.4 Targeted therapy2911.75822.1Pressure source Economic pressure20482.620578.2 Child support12550.610138.5 Fertility pressure4016.23714.1 Appearance change19277.79937.8 Disease prognosis16466.417667.2 Therapeutic effect15964.418871.8 Worry about disease recurrence19076.917968.3 None3815.45320.2EFA, exploratory factor analysis; CFA, confirmatory factor analysis. Demographic and clinical characteristics of participants (N = 509). EFA, exploratory factor analysis; CFA, confirmatory factor analysis. Item analysis Based on the total scale scores of 247 participants, the sample was divided into high (top 27%) and low marital anxiety subgroups (bottom 27%).26 The critical scores for the high and low groups were 80 and 71, respectively. The CR value for each item was calculated by comparing the mean scores between the two subgroups. A CR value greater than 3 (P < 0.05) is generally considered indicative of good discrimination, warranting retention of the item. In this study, the original Item 16 had a CR value of 2.462 (P < 0.05), which did not meet this criterion; therefore, it was deleted.27 Subsequently, all entries were renumbered to 25 entries. Pearson correlation analysis was further conducted to examine the relationship between each item and the total scale score. The correlation coefficients between the items and the total score ranged from 0.442 to 0.656 (P < 0.001), all exceeding 0.4, indicating a strong association between each item and the total scale score. After item analysis, the original Item 16 was deleted because of insufficient discrimination, resulting in the final 25-item MAS-PBC with three first-level dimensions and nine second-level dimensions. Cronbach’s α coefficient Cronbach’s α coefficient, ranging from 0 to 1 and positively correlated with reliability (with 0.7 considered the acceptable standard),28 was used to evaluate the internal consistency of the scale. The Cronbach’s α coefficient for the total scale was 0.865. Regarding the three first-level dimensions, the coefficients were 0.944 (self-perception), 0.917 (couples' response), and 0.945 (family and social pressure). For the nine second-level dimensions, Cronbach’s α coefficients ranged from 0.815 to 0.913. Retest reliability The same measurement method was used to re-survey 30 participants, with an interval of at least two weeks between the two assessments. The correlation coefficient was used to evaluate test-retest reliability; a coefficient greater than 0.7 indicates good temporal stability and consistency of the scale. The results demonstrated that the test-retest reliability of the total scale was 0.902, while the reliability of each dimension ranged from 0.898 to 0.947, indicating strong temporal stability. Split-half reliability Split-half reliability reflects the internal consistency of scores obtained from two equivalent halves of a scale, with a coefficient greater than 0.7 indicating acceptable internal consistency. Split-half reliability analysis was conducted for the 25 items of the scale (originally 26 items, with the original Item 16 removed after item analysis), three first-level dimensions, and nine second-level dimensions. All items were divided based on odd–even order. The results showed that the split-half reliability of the total scale was 0.78. For the three first-level and nine second-level dimensions, the split-half reliability ranged from 0.924 to 0.959 and 0.893 to 0.921, respectively, indicating good internal consistency for the total scale and its dimensions. Content validity The I-CVI for individual items ranged from 0.88 to 1.00, S-CVI/Ave = 0.987, and S-CVI/Universal Agreement (UA) = 0.81, indicating good content validity. Construct validity Construct validity evaluates the internal structure of an instrument and its alignment with theoretical concepts. In this study, construct validity was assessed using EFA and CFA. Exploratory factor analysis A total of 247 participants were included in the EFA. The results indicated a Kaiser-Meyer-Olkin (KMO) measure of sampling adequacy of 0.901 and a statistically significant Bartlett’s test of sphericity (P < 0.001), confirming the suitability of the data for factor analysis.29 Common factors were extracted using the principal factor analysis method, combined with the varimax orthogonal rotation method. Using eigenvalues greater than 1 as the criterion, three common factors were ultimately extracted, accounting for a cumulative variance contribution of 69.313%. The factor loadings of the 25 items ranged from 0.600 to 0.886, all exceeding the threshold of 0.4, with no substantial cross-loadings (Table 2). Table 2Results of exploratory factor analysis.Table 2Common factorSelf-perceptionFamily and Social PressureCouples' Response1. I often feel like a burden to my family and lose confidence in the future.0.8492. I worry about the treatment outcomes and prognosis of my disease.0.7333. I worry about my husband’s loyalty to our marriage, such as whether he might engage in extramarital relationships.0.8714. I am very concerned about how others perceive my appearance (e.g., breast loss, scarring, hair loss).0.8095. Due to disease-related physical changes (e.g., breast loss, scarring), I often feel inferior.0.8036. I want to reshape my body image through breast reconstruction surgery or by altering my appearance.∗0.7527. I worry about being unable to have children or about not being able to take good care of them.0.7218. I feel very guilty that I cannot fulfill my duties as a wife.0.8179. I worry that my illness will affect the equal relationship between my husband and me.0.82110. My husband’s attitude toward me is just as good as it was before.∗0.83511. When I am sad, my husband is indifferent toward me.0.60012. My husband accepts the changes in my body and comforts me.∗0.85913. My husband actively discusses treatment options with me.∗0.74714. I am very satisfied with the adjustments made to the division of household chores.∗0.80115. I am very satisfied with the division of responsibilities between my husband and me concerning the care of our parents and children.∗0.82016. My husband intentionally avoids being intimate with me because of my illness.0.71517. I have sexual needs, but my experience is negatively affected due to treatment side effects, such as vaginal dryness.0.71118. The cost of my medical treatment has placed a financial strain on my family.0.88619. My illness has forced me to postpone my original family plans, such as buying a house and traveling.0.86120. I worry that my elders will have negative opinions about my marriage.0.84521. I worry about the health of my family members.0.82722. The care and concern I receive from my relatives and friends is sufficient.∗0.83023. When I have financial problems, someone (e.g., relatives, friends, or social organizations) will help me.∗0.86524. After becoming ill, I am reluctant to participate in social activities.0.84225. Due to long-term treatment, I worry about losing my job or my social identity.0.784Items marked with “∗” are reverse-scoring questions. Results of exploratory factor analysis. Items marked with “∗” are reverse-scoring questions. Confirmatory factor analysis The three dimensions of the scale developed in this study corresponded to the three factors extracted in the preliminary EFA. The CFA data were obtained from 262 cases collected from both the initial and supplemental samples. The results showed that the model fit based on the three first-level dimensions of the original scale was generally acceptable; however, some indicators did not meet the ideal standard. The model fit indices included χ2 = 899.458, df = 265, χ2/df = 3.394 (< 5), CFI = 0.922 (> 0.85), NFI = 0.893 (> 0.8), and IFI = 0.922 (> 0.8). However, RMSEA = 0.096 and GFI = 0.753 indicate marginal fit.30 Therefore, this three-factor structure should be considered preliminarily supported, but further validation is needed in larger samples31 (Fig. 3). Confirmatory factor analysis model. The values on the double-headed arrows indicate standardized latent factor correlations. Convergent validity Convergent validity refers to the degree to which multiple observed variables converge on the same construct when measuring the same characteristic using different methods. Two methods are mainly used to assess convergent validity: standard factor loadings > 0.7 are considered ideal, and > 0.5 are acceptable; and judgment based on AVE and CR, where AVE > 0.5, and CR > 0.7 indicate high convergent validity.32 We used AVE and CR to assess convergent validity (Table 3). Table 3Convergent validity (AVE and CR) of each dimension.Table 3DimensionnConvergent ValidityAVECRSelf-perception90.7560.965Couples'response80.7610.962Family and social pressure80.7400.958AVE, Average variance extracted; CR, Composite reliability. Convergent validity (AVE and CR) of each dimension. AVE, Average variance extracted; CR, Composite reliability. Discriminant validity The AVE square root method was used to test discriminant validity. Discriminant validity was established if the square root of the AVE of a dimension was greater than the correlation coefficients between that dimension and all other dimensions.32 The results showed that the square root of the AVE of all dimensions was greater than the inter-dimensional correlation coefficients, meeting the discriminant validity criteria. This indicates that the three first-level dimensions have clear construct boundaries with no overlap, and can independently measure different aspects of marital anxiety (Table 4). Table 4Discriminant validity (correlation coefficients and AVE square roots).Table 4DimensionAVE Square RootCorrelation Coefficient with Self-PerceptionCorrelation Coefficient with Couples’ responseCorrelation Coefficient with Family and Social PressureSelf-perception0.869–0.1380.706Couples’ response0.8720.138–0.220Family and social pressure0.8600.7060.220–Note: Table 4 presents observed dimension score correlations, whereas Fig. 3 presents standardized latent factor correlations from the confirmatory factor analysis. AVE, Average variance extracted. Discriminant validity (correlation coefficients and AVE square roots). Note: Table 4 presents observed dimension score correlations, whereas Fig. 3 presents standardized latent factor correlations from the confirmatory factor analysis. AVE, Average variance extracted. Discussion Main findings This study scientifically develops a high-quality scale by integrating a structured theoretical framework, a systematic literature review, in-depth semi-structured interviews, and expert consultations. First, guided by Social Ecological Theory and informed by the results of the literature search and semi-structured interviews, an initial scale item pool containing three first-level dimensions, nine second-level dimensions, and 28 initial items was constructed. Second, experts from different fields were invited to conduct two rounds of Delphi consultations to systematically evaluate the dimensions and items. The results showed that 16 experts provided 135 specific and valuable suggestions for modification, laying a solid foundation for scale optimization. The scale’s validity was further established through psychometric testing. Thirty patients were selected for a pilot study to evaluate the readability and acceptability of the scale items. Based on feedback from the pilot study participants, the wording of some items was optimized and adjusted (e.g., clarifying ambiguous expressions, revising awkward sentences) to better align with patients' language habits and comprehension levels. The final scale comprised three first-level dimensions (Self-Perception, Couples' Response, and Family and Social Pressure), nine second-level dimensions (Emotional Perception, Body Perception, Role Perception, Emotional Support, Family Responsibility, Sexual Experience, Economic Pressure, Family Pressure, and Social Pressure), and 25 items. The final 25-item MAS-PBC is provided in Supplementary File 2. Reliability was assessed through internal consistency, test-retest reliability, and split-half reliability. The results showed that the total scale Cronbach’s α coefficient was 0.865; the Cronbach’s α coefficients for the three first-level dimensions were 0.944, 0.917, and 0.945, respectively; the split-half reliability was 0.78, and the test-retest reliability was 0.902. All values were well above the acceptable standard of 0.7, indicating good internal consistency and temporal stability. Furthermore, item analysis showed that after deleting the original Item 16 (CR = 2.462 < 3) because of insufficient discrimination, the corrected item–total correlations for the remaining items ranged from 0.442 to 0.656 (P < 0.001), all reaching statistical significance, indicating good item discrimination. Validity was evaluated using both content validity and construct validity. The I-CVI ranged from 0.88 to 1.00, S-CVI/Ave = 0.987, and S-CVI/UA = 0.81, indicating good content validity. EFA and CFA were used to assess construct validity. The EFA results showed a KMO value of 0.901 and a statistically significant Bartlett’s test of sphericity (P < 0.001). Three common factors were extracted, with a cumulative variance contribution rate of 69.313%. The CFA results showed acceptable fit for several incremental indices (CFI = 0.922, NFI = 0.893, IFI = 0.922), while RMSEA (0.096) and GFI (0.753) suggested only a marginal fit. Therefore, the three-factor structure should be interpreted as preliminarily supported and should be further verified in larger, independent, multicenter samples. Existing marital relationship assessment tools (e.g., RAS and MAT) were mostly developed based on general populations and do not account for the specific psychological state of patients with breast cancer. On the one hand, influenced by traditional Confucianism, Chinese women often express their emotions implicitly and reservedly. Even when experiencing marital anxiety, they may hide their feelings because they are reluctant to talk about them or worry about burdening their families. Additionally, the breast cancer burden remains high in China, with a proportion of younger patients exceeding that in Europe and the United States. These patients are often in critical stages of their marital roles (e.g., fulfilling duties as wives and mothers). They may be unwilling to bring up issues because of embarrassment, may not receive spousal support because of poor communication,33 and misunderstandings caused by differences in doctor-patient perceptions can affect intervention outcomes. On the other hand, changes in body image triggered by breast-cancer-specific surgical treatments (e.g., mastectomy) may weaken patients' sense of self-worth in the marital relationship, further exacerbating anxiety and depression. Therefore, existing generic assessment tools do not adequately incorporate the specific disease characteristics of breast cancer and the unique manifestations of marital anxiety, making it difficult to accurately identify patients’ anxiety within the marital relationship, and further limiting the development of targeted interventions. Implications for nursing practice and research Preliminary evidence from this single-center sample suggests that the scale has potential adaptability and clinical value in assessing marital anxiety among married patients with breast cancer, especially those of childbearing age and those experiencing greater economic pressure. We systematically identified the main sources of marital anxiety among patients at three levels: individual, spousal, and social support. This helps clinical staff provide personalized intervention plans based on dimensional scores. For example, when high scores are found for items related to body perception within the self-perception dimension, targeted body image reconstruction guidance34 can be provided to help patients manage the negative impacts of body image changes and take positive steps to improve their condition. Within the “Couples' Response” dimension, the scores directly reflect the strengths and weaknesses of couple interaction. Targeted interventions, such as couple communication skills training and intimacy repair,35 can be implemented to help patients adjust their role coping as a couple and improve their engagement with treatment and marital information. The “Family and Social Pressure” dimension focuses on stressor sources unique to Chinese families, such as economic burdens and children’s education, aligning with real-life scenarios. Economic burden is a significant trigger of family relationship tension for patients with cancer. For patients with breast cancer, long-term treatment costs and lost working hours further exacerbate marital anxiety. Furthermore, the influence of family values and societal expectations on patients' marital anxiety within the Chinese cultural context is unique and differs from the focus on individual autonomy found in Western studies on marital anxiety.16 Existing research on marital anxiety mostly examines the individual and family levels.14 Therefore, this dimension improves the framework of factors influencing marital anxiety by incorporating macro factors such as social support and social identity. In clinical application, combining scores from this dimension with those from other dimensions forms an integrated care plan that addresses both internal and external factors. This scale-based personalized intervention can compensate for the shortcomings of general tools that overlook the particularities of diseases and the Chinese cultural backgrounds of patients with cancer. Based on the current research sample, the scale preliminarily demonstrates that it can evaluate marital anxiety among married patients with breast cancer who are of childbearing age, and may help identify the root causes of anxiety through the scores of each dimension, providing a reference for medical staff to formulate targeted interventions. However, its general applicability to different populations requires further research. Limitations While we successfully developed and validated a scale to measure marital anxiety among married patients with breast cancer of childbearing age, effectively assessing their levels of marital anxiety, there are still some limitations in the study’s design and execution. First, the sample lacks representativeness and geographic coverage; all cases were concentrated in a single Chinese hospital, and a single-center sample cannot adequately capture variations in marital anxiety characteristics stemming from regional cultural differences. Second, the inclusion criteria were limited to “married patients with breast cancer who were of childbearing age.” In the sample, 80.4% reported economic pressure, 66.8% reported concerns about disease prognosis, and 54.6% had an annual household income of CNY 50,000 or less. Although this homogeneity helps control for confounding variables, it also limits the generalizability of the scale to patients with different economic levels, marital statuses, and disease stages. Third, the results of the CFA showed that some fit indicators did not meet the ideal standards (GFI = 0.753, RMSEA = 0.096). Although the incremental fit indices, such as CFI and IFI, meet the standards, the above results suggest that the current three-factor model is only at an acceptable level of data fitting, rather than excellent. Therefore, the current sample data do not accurately reflect the status of marital anxiety among married patients with breast cancer who were of childbearing age across different hospital tiers and regions nationwide. Future research should validate the cross-population applicability of the scale using a broader sample, including patients with various income levels, marital statuses, and treatment stages, across multi-center and cross-regional sites. Conclusions Based on a single-center sample, this study preliminarily explores a scale to measure marital anxiety among married patients with breast cancer of childbearing age. The results showed that it demonstrated a certain degree of reliability and validity, providing a reference for assessing marital anxiety in this specific population. The scale comprised three first-level dimensions, nine second-level dimensions, and 25 items. Importantly, the applicability of the scale among patients with breast cancer from different social and cultural backgrounds, economic levels, and marital statuses needs to be further verified because the sample source is limited and concentrated at a specific economic level. CRediT authorship contribution statement Yiou Fu: Conceptualization, Methodology, Investigation, Writing – Original Draft; Writing – Review & Editing. Yuqiao Xiao: Conceptualization, Methodology, Investigation, Writing Original Draft, Writing Review & Editing. Mingxia Yao: Validation, Investigation, Data Curation. Yan Zhu: Validation, Investigation, Data Curation. Ke Li: Validation, Investigation, Data Curation. Shiqing Liu: Validation, Investigation, Data Curation. Jinhua Li: Conceptualization, Resources, Supervision, Project administration, Funding acquisition. All authors have read and approved the final manuscript. Ethics statement This study was approved by the Ethics Committee of Hunan Cancer Hospital (Approval No. KY2025587) and was conducted in accordance with the 1964 Helsinki Declaration and its later amendments or comparable ethical standards. All participants provided written informed consent. Data availability statement The data that support the findings of this study are available from the corresponding author, JL, upon reasonable request. Declaration of generative AI and AI-assisted technologies in the writing process No AI tools/services were used during the preparation of this work. Funding This study was supported by the High-Level Talent Support Program of Hunan Cancer Hospital (Grant No. 20250731-1018); and the Hunan Provincial Natural Science Foundation of China (Grant No. 2025JJ80854). The funders had no role in considering the study design or in the collection, analysis, interpretation of data, writing of the report, or decision to submit the article for publication. Declaration of competing interest The authors declare no conflict of interest. Footnotes Supplementary data to this article can be found online at https://doi.org/10.1016/j.apjon.2026.100996. Appendix A. Supplementary materials The following are the Supplementary data to this article: Multimedia component 1 Multimedia component 2 References Associated Data Supplementary Materials Multimedia component 1 Multimedia component 2 Data Availability Statement The data that support the findings of this study are available from the corresponding author, JL, upon reasonable request.