Developing the parenting concerns scale for cancer survivors and assessment of its validity and reliability in China

Abstract

Background

Parenting concerns impose a substantial psychological burden on cancer survivors, who must cope not only with concerns related to disease treatment but also with worries about the healthy development of their minor children. However, the early identification and assessment of this complex psychological burden remain insufficiently addressed owing to the lack of a valid and reliable assessment instrument. Therefore, this study aims to provide a scientifically valid and reliable assessment tool for evaluating parenting concerns among cancer survivors.

Methods

1. An initial item pool was constructed through a literature review, Delphi expert consultation, and patient cognitive interviews to develop the preliminary pre-test version of the scale.

2. A total of 165 cancer survivors with minor children were recruited to conduct the initial item screening of the pre-test version using the critical ratio method, item–total correlation analysis, and exploratory factor analysis, thereby establishing the formal test version of the scale. Subsequently, an additional 200 eligible cancer survivors were recruited to evaluate the scale's reliability and validity.

Results

1. The item pool of the preliminary scale was revised through expert consultation and patient cognitive interviews and ultimately consisted of 3 dimensions and 25 items. The Kendall’s W coefficients for the two rounds of expert consultation were 0.217 and 0.232, respectively, indicating good consistency among expert opinions.

2. The Kaiser-Meyer-Olkin (KMO) value for the first exploratory factor analysis was 0.890, and Bartlett’s test of sphericity yielded χ²=2370.625(P < 0.001). The cumulative variance explained was 57.427%. Therefore, the scale retained 3 dimensions and 21 items. In the second exploratory factor analysis, all item factor loadings exceeded 0.50, and no cross-loadings were observed. The cumulative variance explained reached 61.629%.

3. Regarding the reliability and validity results, the item-level content validity index (I-CVI) and scale-level content validity index (S-CVI) were both 1.00. Confirmatory factor analysis demonstrated acceptable structural validity, with χ²/df=1.940, RMSEA=0.069, and CFI, TLI, and IFI values of 0.939, 0.929, and 0.940, respectively. The Cronbach’s α coefficient of the scale was 0.906, the split-half reliability was 0.704, and the test–retest reliability was 0.821.

Conclusion

The scale was developed in accordance with scientific principles and demonstrated good reliability and validity, making it an optimized tool for assessing parenting concerns among cancer survivors.

Supplementary Information

The online version contains supplementary material available at https://doi.org/10.1186/s40359-026-04790-1.

Introduction

The term “cancer survivor” refers to an individual from the moment of cancer diagnosis until the end of life, regardless of whether they have received treatment or achieved treatment success [1]. Advances in early cancer screening and treatment have led to a growing and increasingly younger population of cancer survivors [2, 3], many of whom are raising children while undergoing anticancer treatment [4, 5]. Research has shown that approximately 20% of patients coping with cancer also bear the responsibility of caring for minor children [6]. Parenting responsibilities not only influence treatment-related decision-making but also exacerbate psychological distress. Compared with cancer survivors without child-rearing responsibilities, those with parenting responsibilities report significantly higher levels of anxiety and depression [7]. This form of distress, commonly referred to as parenting concerns, encompasses worries about the potential impact of cancer on children as well as concerns about one’s ability to provide adequate care [8]. Parenting concerns place a substantial psychological burden on cancer survivors, who must simultaneously cope with disease management and fears regarding their children’s well-being. Such concerns are associated with increased anxiety, depression, guilt, and persistent distress that may extend for several years following diagnosis [9]. Moreover, parenting concerns may adversely affect children, contributing to emotional difficulties, including anxiety and depression. Given the genetic risk, potential for recurrence, and financial burden associated with cancer, parenting concerns are particularly prominent among cancer survivors with child-rearing responsibilities and warrant careful attention [10]. From the perspectives of medical humanities and patient-centered care, addressing parenting concerns during cancer treatment has important clinical and social implications for improving support for survivors and their families.

Parenting concerns represent a distinctive form of psychological distress among cancer survivors with child-rearing responsibilities. These concerns typically manifest as difficulties in disclosing the illness to children, concerns about children’s health and future development, unmet needs for emotional support, and increased financial burden [11, 12]. Such concerns may arise from uncertainty regarding survivors’ own health, the potential long-term impact of cancer on their children, and the challenge of balancing family responsibilities with the physical and psychological demands of cancer treatment [13]. The severity and manifestation of parenting concerns are shaped by interrelated factors. Disease-related factors, including advanced cancer stage, comorbidities, and poor prognosis, are associated with increased parenting-related psychological distress [6, 14]. Family-related factors also play a crucial role: effective communication and strong family support may alleviate distress and improve psychological well-being among cancer survivors [15, 16]. In contrast, increased financial burden is closely associated with elevated levels of anxiety and worry, whereas better family economic conditions are associated with improved mental health outcomes [17]. Demographic and psychological factors further contribute to variations in parenting concerns. Mothers tend to report higher levels of parenting concerns than fathers [18], and concerns are generally more pronounced among survivors with younger children [6, 19]. Additionally, negative cognitive emotion regulation strategies and more severe depressive symptoms are positively associated with increased parenting concerns [20]. Given the multifaceted and context-specific nature of parenting concerns, comprehensive assessment and the development of targeted and individualized interventions are essential, particularly within the Chinese sociocultural context.

However, few instruments are available in China for evaluating parenting concerns among cancer survivors, and the existing measures have several limitations. The most widely used tool, the Parenting Concerns Questionnaire (PCQ), was developed by Muriel in 2012. It comprises 15 items across three dimensions: practical impact of illness on children, emotional impact of illness on children, and concerns about the co-parent.It has demonstrated good internal consistency (Cronbach’s α = 0.83) [21]. Although a Chinese version of the PCQ was translated and preliminarily validated in 2021 (Cronbach’s α = 0.850), its applicability remains uncertain due to a validation sample largely limited to highly educated women. Moreover, the PCQ was developed in a Western cultural context, whereas parenting practices in China are strongly influenced by Confucian values and the concept of guan, which emphasizes parental responsibility, behavioral control, and academic expectations in ways that differ from Western parenting norms [22]. Cross-cultural evidence indicates systematic differences between Chinese and Western parents in parenting practices and educational priorities [23–25], and Chinese cancer survivors often express concerns regarding their children’s academic achievement, social development, and fulfillment of family responsibilities—domains not adequately captured by the PCQ [26–28]. Therefore, there is a clear need to develop a culturally appropriate and comprehensive assessment instrument to evaluate parenting concerns among Chinese cancer survivors and to facilitate the provision of targeted supportive care.

This study was guided by Family Systems Theory and the Social Ecological Model, which together provided the theoretical foundation for developing the Parenting Concerns Scale for Cancer Survivors. The Family Systems Theory conceptualizes the family as an interdependent system in which changes in one member or subsystem may affect the functioning of the entire family [29]. Within this framework, family relationships are conceptualized as multiple interconnected subsystems, including the marital and parent–child subsystems, and disruptions in one subsystem may produce cascading effects on others subsystems [30]. From this perspective, parenting concerns among cancer survivors extend beyond concerns about caregiving capacity to encompass concerns related to children’s development, family roles, and emotional dynamics within the family system. Persistent parenting concerns may contribute to chronic anxiety, which may subsequently impair family functioning, communication, and emotional well-being. Accordingly, this theoretical perspective informed the identification of domains reflecting intrafamilial relationships and emotional dynamics during scale development. The Social Ecological Model provided a complementary framework by emphasizing that individual experiences are shaped by multiple interacting levels, including the individual, family, community, and broader sociocultural context [31]. From a Social Ecological Model perspective, parenting concerns among cancer survivors are influenced not only by individual health status and family support, but also by interactions with healthcare systems, community resources, and sociocultural expectations. This model underscores the multidimensional nature of parenting concerns and highlights the importance of considering both proximal and contextual influences. In this study, the Social Ecological Model guided the inclusion of items capturing not only individual and family-related concerns, but also concerns related to social support, educational environments, and broader contextual factors among cancer survivors in China. These theoretical frameworks informed the initial item generation and dimensional structure of the scale, thereby providing a foundation for the subsequent scale development and validation procedures.

Method

This study developed a preliminary version of the Parenting Concerns Scale for Cancer Survivors through a literature review, two rounds of Delphi expert consultations, and cognitive interviews. Items were screened using item analysis and item–total correlation analysis, resulting in a scientifically robust final version of the scale. The finalized scale was administered to 200 cancer survivors with minor children, and its psychometric properties were evaluated in terms of construct validity, content validity, internal consistency, split-half reliability, and test–retest reliability.

Conducting a literature search and constructing the item pool

The research team systematically searched databases, including PubMed, Web of Science, Embase, Cochrane Library, Scopus, CNKI, Wanfang Medical Online, and VIP Chinese Journal Database, with the search period spanning from the establishment of each database to August 2022. After iterative preliminary searches, the final search terms was determined as: (“cancer” OR “tumor” OR” neoplasm “OR” cancer survivor) AND (“raising children” OR “nurture” OR “parenting concerns” OR “parent” OR “father” OR “mother” OR “mom” OR “child” OR “parent-child” OR “mother-child” OR “father-child”) AND (“qualitative research” OR “interview” OR “feeling” OR “experience”) NOT (“childhood cancer” OR “pediatric oncology” OR " children with cancer”). A total of 14,839 articles were initially identified. After duplicate removal, two stages of screening, and quality assessment, 19 articles related to parenting concerns among cancer survivors were finally included. Based on the Family Systems Theory, the Social Ecological Model, and literature analysis, the research team conducted discussions and preliminarily developed an initial item pool for the Parenting Concerns Scale for Cancer Survivors, consisting of 34 items.

Delphi expert consultation

Following a literature review and group discussions, an initial item pool for the scale was developed, and subsequently, a Delphi expert consultation questionnaire was constructed. In this study, each round of the Delphi consultation was conducted over 2–4 week intervals. Experts were contacted in advance via WeChat (a commonly used communication app in China, similar to iMessage) or email to explain the study objectives and obtain informed consent to participate in the Delphi consultation. After each round, expert ratings and qualitative feedback were summarized and analyzed, and items were added, removed, or revised according to predefined criteria. The revised questionnaire was then redistributed for the next round. The Delphi process was terminated once expert consensus was achieved; otherwise, additional rounds were conducted until consensus was reached [32].

Cognitive interviewing

Cognitive interviewing enables researchers to assess whether respondents correctly understand questionnaire items and plays a critical role in scale testing, development, evaluation, and refinement processes. It helps determine the consistency of item interpretation within the target population and reduces avoidable measurement error [33]. Prior to the interviews, participants were informed of the study purpose, procedures, potential benefits, and risks, and provided written informed consent. Participants completed the scale based on their own understanding, and the scale was subsequently revised based on their feedback. Interview rounds continued until data saturation was achieved [34]. The interview guide included five questions: (1) whether the items were understandable and contained any difficult wording; (2) whether any relevant parenting concerns were missing; (3) how response options were interpreted and selected; (4) whether any options were difficult to choose; and (5) whether any items required modification.

Participants

The inclusion criteria were as follows: (1) participants diagnosed with cancer; (2) having at least one child younger than 18 years; (3) voluntary participation in the study and provision of written informed consent. The exclusion criteria were as follows: (1) minor patients; (2) cognitive impairment, psychiatric disorders, or non-cooperation. The withdrawal criteria: (1) patients unable to continue participation due to physical condition during the study; (2) withdrawal for any other reason.

Data collection

Participants

Participants were recruited from a tertiary hospital in Zhejiang Province using a convenience sampling method. For the item analysis stage, According to the COSMIN criteria, the sample size for scale development and validation should be 5–10 times the number of items [35]. In addition, a 20% attrition rate was taken into consideration [36]. Data collection was conducted in two stages. During the item analysis stage, 170 questionnaires were distributed, of which 170 were collected, and 165 were deemed valid, resulting in an effective response rate of 97.59%. During the reliability and validity testing stage, a total of 210 questionnaires were distributed, and 206 were returned. After excluding 6 invalid questionnaires because of patterned responses, 200 valid questionnaires were included in the final analysis, resulting in an effective response rate of 95.24%.

Instruments

1. General information questionnaire: designed by the research team, including variables such as gender, age, education level, marital status, income level, cancer stage, and child-related information.

2. Parenting Concerns Scale for Cancer Survivors: developed by the research team and preliminarily developed through two rounds of expert consultation and cognitive interviews.

Data analysis

Item analysis was performed using the critical ratio (CR) method and item–total correlation analysis. Participants were divided into high-score (top 27%) and low-score (bottom 27%) groups, and independent-samples t tests were conducted for each item. Items with non-significant group differences, t values < 3.00, or item–total correlation coefficients < 0.30 (P > 0.05) were considered for deletion. Exploratory factor analysis (EFA) was conducted to examine the factor structure of the scale [37]. Item retention criteria included factor loadings ≥ 0.50, cross-loading differences ≥ 0.15, and a minimum of three items per factor [38, 39]. The validity and reliability of the final scale were evaluated using a validation sample of cancer survivors with minor children. Validity was assessed using content validity and construct validity. Content validity was evaluated using the content validity index (CVI), including item-level CVI (I-CVI) and scale-level CVI (S-CVI) [40]. Construct validity was examined using confirmatory factor analysis (CFA), with model fit evaluated using χ²/df, RMSEA, CFI, TLI, and IFI [41]. The recommended threshold values are presented in Table 1. Reliability was assessed using internal consistency, split-half reliability, and test–retest reliability. Cronbach’s α coefficients, split-half reliability coefficients, and test-retest reliability coefficients ≥ 0.70 were considered acceptable [42].

Table 1.

Model Fit Reference Criteria

Index	Acceptable Values
Chi-square to degrees of freedom ratio (χ²/df)	χ²/df < 2 indicates good model fit; 3–5 is acceptable
Root Mean Square Error of Approximation (RMSEA)	< 0.05 indicates good model fit; 0.05–0.08 indicates acceptable fit; < 0.10 is acceptable
Incremental Fit Index (IFI)	> 0.90 indicates good fit
Tucker–Lewis Index (TLI)	> 0.90 indicates good fit
Comparative Fit Index (CFI)	> 0.90 indicates good fit

Result

Results of the literature review

Through an extensive review of domestic and international literature and repeated discussions, a pool of scale items was developed, comprising three dimensions: the impact of my illness on my child (12 items), decline in my ability to care for my child (8 items), and unmet support needs (14 items), resulting in a total of 34 items across three dimensions.

Results of the Delphi expert consultation

General characteristics of the experts

A total of 20 experts were invited to participate in the Delphi expert consultation, including 18 experts from 7 hospitals (two tertiary cancer specialty hospitals, one tertiary pediatric specialty hospital, and four tertiary general hospitals) and two experts from medical universities. Among them, nine were engaged in clinical nursing, one in clinical medicine, seven in nursing management, two in nursing education, and one in research institutes. The general characteristics of the experts are presented in Table 2.

Table 2.

Basic Information Table For Correspondence Experts (N = 20)

Items		Number of people(n)	Composition ratio (%)
Age(years)	30–39	1	5
	40–49	9	45
	50years~	10	50
Professional work experience (in years)	10–19	4	20
	20–29	4	20
	30~	12	60
Education	Doctoral students	3	15
	Master’s degree	7	35
	Undergraduate	10	50
Title	Senior	11	55
	Associate senior	7	35
	Intermediate	2	10
Professional field	Clinical care	9	45
	Clinical medicine	1	5
	Care management	7	35
	Nursing education	2	10
	Social psychology	1	5
Position	Clinical doctor	1	5
	Clinical nurse	16	80
	Nursing teacher	2	10
	Researcher	1	5

Expert response rate and authority coefficient

Two rounds of Delphi consultation were conducted in this study, with a 100% valid response rate in both rounds, including a high level of expert engagement [43]. The expert authority coefficient (Cr) reflects the experts’ level of familiarity with the research field and influences the reliability of the consultation results. Cr is determined by the experts’ familiarity with the consultation domain (Cs) and the basis of their judgments regarding the consultation content (Ca). The calculation formula is Cr = (Ca + Cs)/2. Generally, a Cr value ≥ 0.70 indicates adequate expert authority and acceptable consultation results [44]. In the first round, the coefficients were as follows: Cr = 0.838, Ca = 0.895, and Cs = 0.780; in the second round, Cr = 0.865, Ca = 0.900, and Cs = 0.830. The expert authority coefficient (Cr) exceeded 0.700 in both rounds, suggesting that the experts possessed a high level of authority in the relevant fields and that the results were reliable [45]. The Kendall’s W coefficients for two rounds were 0.217 and 0.232(P<0.001), with the second-round coefficient being higher than that of the first round, indicating greater consistency among expert opinions and improved consensus on the items [46]. The relevant data are presented in Table 3.

Table 3.

Expert Coordination Coefficients (Kendall’s W) Across Delphi Rounds

	The first round			The second round
Overall index	Kendall’s W	χ2	p	Kendall’s W	χ2	p
	0.217	143.417	P<0.001	0.235	116.482	P<0.001

Modification status of the items

Regarding the first round, five items were deleted, six items were merged, five items were added, and 11 items were revised. All indices in the second round met the predefined criteria, and expert opinions gradually converged. After two rounds of Delphi expert consultation, the final scale structure comprised three dimensions and 24 items. Details of the item revisions are provided in Supplementary Material 1.

Results of the cognitive interviews

In the cognitive interview phase, participants were recruited using convenience sampling; 10 cancer survivors were recruited. Following the cognitive interviews, the pretest version of the Parenting Concerns Scale for Cancer Survivors was finalized, consisting of three dimensions and 25 items. The detailed structure of the scale is presented in Supplementary Material 2.

Results of the scale item analysis

General characteristics of the participants

During the exploratory factor analysis stage, 165 valid questionnaires were included in the analysis. The general characteristics of the participants are presented in Table 4.

Table 4.

The general information table of research objects(n = 165)

Category	N (%)	Category	N (%)
Sex		Education level
Mela	46(27.9%)	Primary school or below	18(10.9%)
Female	119(72.1%)	Junior high school	50(30.3%)
Age		Senior high school	23(13.9%)
20–30	10(6.1%)	Junior college	27(16.4%)
31–40	78(47.3%)	Bachelor’s degree	42(25.5%)
>40	77(46.7%)	Master’s degree or above	5(3.0%)
Marital status		Caregiver
Married	148(89.7%)	Spouse	126(76.4%)
Single	0(0%)	Other	31(18.8%)
Divorced	17(10.3%)	None	8(4.8%)
Widowed	0(0%)	Cancer type
The number of minor children		Breast cancer	55(33.3%)
1	56(33.9%)	Cervical cancer	14(8.5%)
2	104(63%)	Lung cancer	8(4.8%)
3	4(2.4%)	Thyroid cancer	21(12.7%)
>3	1(0.6%)	Nasopharyngeal carcinoma	31(18.8%)
Type of medical insurance		Lymphoma	2(1.2%)
Urban–rural residents’ medical insurance	60(36.4%)	Liver cancer	6(3.6%)
Employee medical insurance	93(56.4%)	Colorectal cancer	10(6.1%)
Self-paying	12(7.3%)	Gastric cancer	14(8.5%)
Employment status		Kidney cancer	4(2.4%)
Employed	91(55.2%)	Comorbidity
Unemployed	74(44.8%)	Yes	25(15.2%)
Monthly household income		No	140(84.8%)
<2000	14(8.5%)	Tumor origin
2000–5000	37(22.4%)	Primary	124(75.2%)
5001–8000	40(24.2)	Secondary	41(24.8%)
8001–10,000	28(17.0%)	Cancer stage
>10,000	46(27.9%)	I	26(15.8%)
		II	62(37.6%)
		III	34(20.6%)
		IV	43(26.1%)

Results of the scale item analysis

Critical ratio results

The results showed that the critical ratios for all items in the scale were greater than 3.00, indicating that no items needed to be deleted [47]. And the results demonstrated that all item–total correlation coefficients exceeded 0.30; therefore, all items were retained [48]. Details are presented in Table 5.

Table 5.

The statistical measures of the critical ratio and correlation coefficient for each item(n = 165)

Item	t	p	r	Item	t	p	r
1	9.288	<0.001	0.636**	14	9.608	<0.001	0.622**
2	11.018	<0.001	0.667**	15	7.521	<0.001	0.524**
3	9.146	<0.001	0.669**	16	11.437	<0.001	0.617**
4	9.089	<0.001	0.601**	17	5.152	<0.001	0.468**
5	9.616	<0.001	0.641**	18	7.891	<0.001	0.579**
6	9.008	<0.001	0.603**	19	7.166	<0.001	0.483**
7	9.859	<0.001	0.662**	20	6.853	<0.001	0.529**
8	9.274	<0.001	0.624**	21	8.656	<0.001	0.564**
9	11.606	<0.001	0.688**	22	3.727	<0.001	0.344**
10	7.646	<0.001	0.500**	23	6.953	<0.001	0.519**
11	12.411	<0.001	0.677**	24	5.443	<0.001	0.481**
12	10.900	<0.001	0.655**	25	6.070	<0.001	0.531**
13	10.065	<0.001	0.594**

The * indicates the level of statistical significance, ** means P < 0.01

Results of exploratory factor analysis

After item analysis, 25 items were retained for exploratory factor analysis (EFA). The results showed a KMO value of 0.890 and a Bartlett’s test of sphericity of χ² = 2370.625, P<0.001, indicating that the data were suitable for EFA [47]. Three factors were extracted, with a cumulative variance contribution rate of 57.427%. Item 9 showed cross-loadings on Factor 1 (0.538) and Factor 2 (0.509) and was deleted. Items 8,19 and 22 had factor loadings < 0.50 across all factors and were deleted. In total, four items were removed, leaving three factors with 21 items. Then, a second EFA was conducted on the remaining 21 items. The results showed that all items had factor loadings > 0.50, with no cross-loadings were observed(Table 6). The cumulative variance contribution rate of the three factors was 61.629%, indicating that the scale demonstrated good construct validity [38].

Table 6.

Rotated component matrix from the second factor analysis

Term	Factor1	Factor2	Factor3
Factor 1: Concerns about the Child’s Development
I am concerned that my child may not be able to accept the fact that I have cancer.	0.777	0.137	0.158
I am concerned that my illness may cast a shadow over my child’s development.	0.849	0.098	0.159
I am concerned that my child may worry excessively about my health condition.	0.717	0.165	0.233
I am concerned that my negative emotions may affect my child.	0.594	0.115	0.266
I am concerned that my illness may cause my child to feel inferior.	0.802	0.129	0.122
I am concerned that my child may experience discrimination.	0.734	0.156	0.112
I am concerned that my child’s social interactions may be affected.	0.807	0.121	0.166
I am concerned that my illness may be passed on to my child.	0.550	-0.022	0.266
Factor 2: Concerns about One’s Caregiving Ability
I am concerned that I may be unable to take care of my child’s daily activities.	0.258	0.805	0.084
I am concerned that my physical condition may prevent me from engaging in outdoor activities with my child.	0.204	0.790	0.131
I am concerned that I may miss important moments in my child’s development.	0.129	0.797	0.066
I am concerned that I may be unable to meet my child’s emotional needs.	0.206	0.736	0.123
I am concerned that I may not spend enough time accompanying my child.	0.132	0.750	-0.020
I am concerned that I may be unable to fulfill my original family roles.	0.223	0.685	0.125
I am concerned that I may be unable to provide sufficient financial support for my child.	0.130	0.047	0.690
I am concerned that my parental role may influence my treatment decisions.	0.346	-0.021	0.733
Factor 3: Concerns about the Support System
I do not know how to answer my child’s questions about cancer.	0.209	0.038	0.727
I am concerned that I may not receive adequate guidance on parenting from healthcare professionals.	0.068	0.272	0.708
I am concerned that my child may not receive support from immediate family members.	0.094	0.121	0.753
I am concerned that my spouse may not provide sufficient daily care for the child.	-0.064	0.189	0.795
I am concerned that my spouse may not provide adequate emotional support for the child.	-0.030	0.222	0.812

Reliability and validity testing of the scale

The general characteristics of participants

In the confirmatory factor analysis (CFA) stage, 200 valid questionnaires were included in the final analysis. The general characteristics of the participants are presented in Table 7.

Table 7.

The general information table of research objects(n = 200)

Category	N (%)	Category	N (%)
Sex		Education level
Mela	67(33.5%)	Primary school or below	30(15.0%)
Female	133(66.5%)	Junior high school	52(26.0%)
Age		Senior high school	33(16.5%)
20–30	13(6. 5%)	Junior college	38(19.0%)
31–40	102(51.0%)	Bachelor’s degree	43(21.5%)
>40	85(42.5%)	Master’s degree or above	4(2.0%)
Marital status		Caregiver
Married	183(91.5%)	Spouse	143(71.5%)
Single	1(0.5%)	Other	53(26.5%)
Divorced	16(8.0%)	None	4(2.0%)
Widowed	0(0%)	Cancer type
The number of minor children		Breast cancer	46(23.0%)
1	74(37.0%)	Cervical cancer	21(10.5%)
2	115(57.5%)	Lung cancer	19(9.5%)
3	10(5.0%)	Thyroid cancer	37(18.5%)
>3	1(0.5%)	Nasopharyngeal carcinoma	9(4.5%)
The age of minor children(years)		Lymphoma	11(5.5%)
<6	56(28.0%)	Liver cancer	13(6.5%)
6–12	92(46.0%)	Colorectal cancer	24(12.0%)
13–18	52(26.0%)	Gastric cancer	11(5.5%)
Type of medical insurance		Kidney cancer	9(4.5%)
Urban–rural residents’ medical insurance	82(41.0%)	Comorbidity
Employee medical insurance	111(55.5%)	Yes	32(16.0%)
Self-paying	7(3.5%)	No	168(84.0%)
Employment status		Tumor origin
Employed	124(62.0%)	Primary	156(78.0%)
Unemployed	76(38.0%)	Secondary	44(22.0%)
Monthly household income		Cancer stage
<2000	17(8.5%)	I	49(24.5%)
2000–5000	47(23.5%)	II	51(25.5%)
5001–8000	40(20.0%)	III	57(28.5%)
8001–10,000	20(10.0%)	IV	43(21.5%)
>10,000	76(38.0%)

Validity and reliability analysis results

Content validity was excellent. All 21 items were rated 3 or 4 by experts, resulting in both the item-level content validity index (I-CVI) and the scale-level content validity index (S-CVI) being 1.00 [49]. Construct validity was assessed using confirmatory factor analysis (CFA) based on data from 200 participants. The initial model demonstrated suboptimal fit indices (χ²/df = 2.655, RMSEA = 0.091, CFI = 0.890, TLI = 0.875, IFI = 0.891). After model modification guided by modification indices, the revised model showed good fit (χ²/df = 1.940, RMSEA = 0.069, CFI = 0.939, TLI = 0.929, IFI = 0.940) [42, 50]. All standardized factor loadings met the recommended criteria. The average variance extracted (AVE) for each factor exceeded 0.50, and composite reliability (CR) values were all > 0.80, indicating good convergent validity. Inter-factor correlations ranged from 0.417 to 0.619, and the square roots of the AVE for each factor exceeded the corresponding inter-factor correlations, supporting satisfactory discriminant validity [51]. Reliability analysis demonstrated good internal consistency and stability. The Cronbach’s α coefficient for the overall scale was 0.906, with subscale coefficients ranging from 0.832 to 0.895. The split-half reliability coefficient was 0.704 for the total scale and ranged from 0.754 to 0.835 across subscales. Test–retest reliability assessed over a 2–4-week interval showed a coefficient of 0.821 for the overall scale, with subscale coefficients ranging from 0.791 to 0.842, indicating good temporal stability [42].

Discussion

Currently, the psychological burden of parenting concerns among cancer survivors is substantial and indicates a clear need for supportive care. This study aimed to develop and validate a Parenting Concerns Scale for Cancer Survivors to effectively assess parenting concerns among Chinese cancer survivors. Compared with the Parenting Concerns Questionnaire(PCQ)developed by Muriel [21], the present scale demonstrates notable differences in structure, psychometric properties, and cultural sensitivity. The PCQ comprises three dimensions: practical impact of illness on the child, emotional impact of illness on the child, and concerns about the co-parent, consisting of 15 items. Although the structure of the PCQ captures parenting concerns, it tends to emphasize observable domains and lacks clearly defined and well-differentiated constructs. Guided by Family Systems Theory and the Social Ecological Model, the present scale conceptualizes parenting concerns as a multidimensional construct influenced by individual, family, and societal factors. It yields a theoretically more coherent three-factor structure, encompassing concerns about children’s development outcomes, concerns about caregiving capacity, and concerns about the support system. From a psychometric perspective, this factor structure demonstrates superior construct validity, with higher internal consistency across dimensions and more clearly distinguishable constructs, thereby enhancing interpretability and measurement precision.

More importantly, this scale incorporates culturally specific elements highly relevant to the Chinese context. In China, education has traditionallybeen highly valued, with a long-standing belief that education can change one’s destiny. Compared with parents in many Western countries, Chinese parents tend to place greater emphasis on their children’s academic achievement and educational attainment [52]. Consequently, cancer survivors are particularly concerned that their illness may negatively affect their children’s academic progress. Influenced by long-standing Confucian cultural traditions, Chinese families particularly value family ties and blood relations, and parents demonstrate a stronger sense of responsibility for raising their minor children [53]. In addition, the indirect communication style commonly observed in Chinese parents, together with a tendency to avoid discussing death, makes it difficult for them to communicate about their illness to their children, thereby exacerbating parenting concerns [54]. However, the PCQ doesn’t adequately capture these domains, a limitation that is particularly critical in the Chinese context, where strong educational expectations and family role obligations constitute major sources of parenting concerns [55]. The present scale incorporates these culturally grounded elements into its items, providing a more comprehensive instrument that is better aligned with the Chinese context. In addition, the inclusion of concerns regarding insufficient information and professional support highlights existing gaps in family-centered care within the Chinese healthcare system.

In this study, both the item-level and scale-level content validity indices reached 1.00, indicating a high level of agreement among the experts [56]. Such a perfect content validity index may be attributable to the fact that most experts in this study were from East China; a high degree of academic consensus or homogeneity in perspectives may have reduced the independence of evaluations, resulting in limited variability in ratings. In future research, it is recommended to include experts from diverse disciplines, geographic regions, and clinical backgrounds, or employ additional approaches, such as the modified kappa statistic or cross-validation in independent samples, to further evaluate the content validity of the scale. The initial model in this study didn’t fully meet the recommended fit thresholds; therefore, modifications were made based on modification indices(MI). However, it should be acknowledged that MI-based modifications are data-driven, post hoc adjustments, which may lead to overfitting to the current sample and thereby compromising the model’s external validity [47]. To mitigate the risk of overfitting, strict theoretical constraints were applied when implementing MI-based modifications. Residual correlations or cross-loadings suggested by modification indices were adopted only when supported by clear theoretical or methodological justification. All modifications were documented during the model respecification phase, and purely data-driven exploratory modifications were avoided. Nevertheless, the modified model structure requires cross-validation in independent samples. Future studies should employ multi-group confirmatory factor analysis(MG-CFA) or exploratory structural equation modeling(ESEM) to examine the cross-sample stability of the factor structure.

Limitations and future directions

Although this study makes important contributions to the development and validation of the Parenting Concerns Scale for Cancer Survivors, several limitations should be acknowledged. First, regarding sample selection, this study employed a single-center design and convenience sampling, which may introduce selection bias and limit the external validity and generalizability of the findings. As all participants were recruited from a single medical institution, the sample may not fully represent cancer survivors across different regions and socioeconomic backgrounds within China. Second, about data sources, this study relied on participants’ self-reports, which may be subject to social desirability and recall biases, thereby potentially compromising the accuracy of the measurements. Third, this scale was developed within a specific cultural context, and its applicability beyond China or in other cultural settings remains unclear. Therefore, future studies are warranted to conduct cross-cultural adaptation and validation to evaluate the scale’s validity across diverse cultural populations. In addition, this study employed a cross-sectional design, which precludes assessment of the scale’s temporal stability and responsiveness to change. In conclusion, future research should prioritize improving sample diversity, improving the objectivity of data sources, cross-cultural applicability, and the use of longitudinal designs, in order to further enhance the scientific value and generalizability of the scale.

Conclusion

This study developed and validated a Parenting Concerns Scale for Cancer Survivors, comprising three dimensions: concern about children’s growth (8 items), concerns about self-care ability (8 items), and concerns about the support system (5 items), consisting of a total of 21 items. The scale demonstrated a clear and theoretically grounded factor structure, along with satisfactory reliability and validity, supporting its robustness as a psychometric instrument. The scale incorporates culturally relevant items, particularly those reflecting educational expectations and family support dynamics within the Chinese context. It can be used to identify cancer survivors experiencing significant parenting concerns and to facilitate targeted, family-centered communication, thereby supporting the implementation of tailored psychosocial and family-centered interventions.

Authors’ contributions

Study conception and design: M.Z.; Data collection and analysis: M.Z.、L.Y.、 J.Z.; Drafting of the article: M.Z.、L.Y.; Critical revision of the article: Z.J.

Funding

This study was supported by the 2024 Zhejiang Provincial Medical and Health Science and Technology Program (General Project) (2024KY837).

Data availability

The datasets used and/or analyzed during the current study are available from the corresponding author on reasonable request.

Declarations

Ethics approval and consent to participate

This study involving human participants was reviewed and approved by the Ethics Committee of Zhejiang Cancer Hospital (Approval No.: IRB-2022-444, dated 7 August 2022). Participants were informed about the study before giving their consent and completing the questionnaire. The article uses correct referencing for all cited materials, and the Declaration of Helsinki was followed at all stages of this research.

Consent for publication

Not applicable.

Competing interests

The authors declare no competing interests.