Translation & psychometrics of the Parenting Concerns Questionnaire (PCQ) in Iranian woman with breast cancer.

Introduction
Breast cancer is the most common cancer and the cause of most cancer-related deaths among women [1]. According to a report by the Global Cancer Observatory (GCO) in 2022, breast cancer is the second most common cancer among new cancer cases [2]. Reports indicate that new cases of breast cancer in Iran account for 11.3% of the cancer cases, and 25.5% of the cancers in the female population, hence, ranked first in this regard [3].
The average age of breast cancer diagnosis has decreased during recent years, and many of the diagnosed women had at least one child under the age of 18, indicating their maternal role [4]. Cancer diagnosis and related treatments cause critical changes in family life and daily routines [5]. When a mother is diagnosed with cancer, changes in her appearance, energy level, mood, and memory affect her daily activities and the quality of her care for her children [6]. They must adjust the balance between the responsibility of parenthood and managing their symptoms [7]. In addition, They worry about how to deal with children’s emotional reactions, fear of loss and inability to see their children grow [8] Even cancer survivors who completed their treatment may experience such challenges because cancer patients face uncertainty in their lives [9]. A study in 2021 demonstrated that Iranian women with breast cancer experience physical limitations and psychological tension due to their disease, which poses a threat to their parental role [10]. In other words, concern about parental role and the effect of disease on their children is another unresolved psychosocial need reported by the patients [11]. The parental concern may be more frequent in patients with advanced stages and metastatic cancers [12]. Various studies revealed that parental concerns were positively associated with the psychosocial tensions of the cancer patients, with adverse outcomes for parents, including anxiety, depression, and reduced quality of life [13].
Considering that psycho-oncology guidelines emphasize caring for the entire family system, assessing the parental concerns using a valid tool is the first step to identifying the unmet needs of cancer patients [14]. Accordingly, several tools, such as quality-of-life research instruments, may include items related to the general family concerns and responsibilities, but they will not exclusively cover the parental concern [15]. Also, another questionnaire, such as the Parenting Stress Index, deals more with the child’s inherent issues than specific stressors such as parental illness. To address this gap, Muriel et al. (2012) were the first to develop a tool in USA after qualitative and quantitative studies that specifically assessed cancer patients’ parenting concerns. This questionnaire is the only existing tool to measure the severity of parenting concerns in cancer.
The Parenting Concerns Questionnaire (PCQ) includes 15 items and three primary identified dimensions: “Practical impact of illness on child”, “Emotional impact of illness on child” and “Concerns about coparent.” Each item is scored from one to five on a Likert scale, and generally, higher scores indicate higher levels of parental concern [16]. The psychometric evaluation of the questionnaire was conducted in several European and Asian countries, such as Portugal, Germany, Japan, Sweden, and Poland, which confirmed the validity and reliability of the questionnaire [14, 17].
Parental concern is a missing link in palliative care in Iran and has not been sufficiently addressed by healthcare workers of oncology care centers in Iran [18]. Although in Iranian culture, parenthood is an important aspect of life that plays a pivotal role, recognizing and prioritizing these concerns is a crucial step toward psycho-oncologic and in addressing this gap in the literature.
In Iran, no previous study has quantitatively evaluated parental concerns in women with breast cancer, and qualitative studies have been conducted so far. Therefore, one of the obstacles for its evaluation is the lack of translated version of PCQ and targeted Persian instruments. Accordingly, considering the high frequency of breast cancer in Iran and the socio-cultural background of the Iranian society regarding the value of the maternal role, more dependence of children on mothers and more responsibility for taking care of children, it is necessary to assess the validity of this questionnaire in Iranian society, similar to those examinations of the cancer patients in various countries, cultures, and populations. Hence, the current study aimed to translate and evaluate the psychometric properties of the Parenting Concerns Questionnaire in Iranian women with breast cancer.

Methods

Study design
The present methodological study was conducted in 2024, aiming to translate and evaluate the psychometric properties of the PCQ among the women population with breast cancer who referred to a hospital affiliated with Shahid Beheshti University of Medical Sciences, Tehran, Iran.

Study participants
The participants of the current study included women with breast cancer who were referred to one of the hospitals affiliated with the Oncology Center of Shahid Beheshti University of Medical Sciences. A total of 368 questionnaires were distributed in this study, 346 of which were completed by the patients, indicating a 94% response rate. Accordingly, 176 cancer patients participated in the exploratory factor analysis, and 170 patients participated in the confirmatory factor analysis (Fig. 1). The samples were obtained using the convenience method for both phases. The inclusion criteria included the individual’s willingness to participate in the study, adults who were at least 18 years old, having a minimum of one child under 18, pathologically confirmed breast cancer diagnosis, and being able to read and write in Persian. The exclusion criteria also included not returning the questionnaire and partial completion of the questionnaire.

Data collection
After obtaining permission from center’s authorities, the researcher explained the purpose of the study to the patients who had been referred to the center for treatment and identified those eligible to participate in the study. Written consent was obtained from the patients, and they received the questionnaires before starting their chemotherapy or radiotherapy sessions. The participants completed the questionnaires and returned them to the researcher on the same day.

Study instruments

Socio‑demographic information
The demographic and clinical information form included age, marital status, education level, occupation, financial condition, number of children under 18, type of treatment, and duration of the disease.
Muriel et al. designed the Parenting Concerns Questionnaire in 2012 in the USA. The questionnaire includes 15 items and three primary identified dimensions “Practical impact of illness on child” (5 items) “Emotional impact of illness on child” (5 items) and “Concerns about coparent” (5 items). Each item is scored from one to five (1 = not at all concerned, 2 = a little bit concerned, 3 = somewhat concerned, 4 = very concerned, and 5 = extremely concerned) based on a Likert scale. The scale scores range from one to five. Generally, higher scores indicate higher parental concerns. The Cronbach’s alpha for the internal consistency of the original version of the scale was α = 0.83 [16].
European Organization for Research and Treatment of Cancer QoL Questionnaire (EORTC QLQ-C30). This questionnaire has 30 items in three dimensions (Functional Scales, Symptom Scales, and Global QoL). Out of these 30 items, 28 are scored on a 4-point Likert and 2 are scored on a 7-point Likert scale. The total score of each dimension ranges from 0 to 100. For Functional Scales and Global QoL, a higher score indicates higher functionality and better health. But a higher score for ​​Symptom Scales means the patient is suffering from more symptoms. In general, scores of 75 and above in this questionnaire are interoperated as favorable, 50–75 as moderate, and below 50 as unfavorable. In Iran, Safaei and Dehkordi confirmed the validity and reliability of the questionnaire, as Cronbach’s alpha of most of the dimensions was higher than 0.7, and the construct validity, calculated as convergent validity, was higher than 0.4 for every dimension [19].
Zigmond and Snaith (1983) designed the Hospital Anxiety and Depression Scale, containing seven questions about the symptoms of anxiety and seven questions about the signs of depression. The questionnaire is scored on a four-point scale (0 to 3), ranging from 0 to 21, and higher scores have higher clinical importance, indicating higher levels of anxiety or depression [20]. The Persian version of this scale has acceptable validity and reliability. Cronbach’s alpha coefficient has been found to be 0.78 for the anxiety subscale and 0.86 for the depression subscale [21].

Translation process of the scale
The translation process of the questionnaire was based on the recommendations of the World Health Organization [22], including translating the tool to the target language by two Persian-speaking translators, merging and matching the translations by specialized groups, backward translation to the primary language by two independent native translators, conducting pretest and cognitive revision with a target group, preparing the final version, and documentation of the process. In the present study, the researchers first asked for psychometric permission from Muriel, the designer, by email. Then, two individuals fluent in English independently translated the questionnaire into Persian. In the next step, the researchers and the translators examined the differences in the translation and revised the inappropriate phrases to achieve a single version. In the backward translation,‌ two native translators translated the final version of the translation from Persian to English, while they were not aware of the original version. After another review by the research team, the questionnaire draft was designed and sent to the tool’s designer for confirmation. Then, the psychometric process of the translated questionnaire was conducted.

Psychometric assessment of the scale
The psychometric properties of the questionnaire, including face validity, content validity, construct validity (exploratory factor analysis, confirmatory factor analysis and convergent validity), and reliability in the form of internal consistency, were evaluated. The assessments were conducted using the Consensus-based Standards for the selection of health measurement instruments (COSMIN) checklist [23].

Face validity
Face validity refers to the extent to which a test is seemingly designed to measure. In other words, the users’ opinions are essential to the development of the tools. In the present study, the researcher used cognitive interviews to assess the qualitative face validity after finalizing the translation process [24]. Ten of the eligible patients were interviewed in person and asked to comment on the comprehensibility of the items (the level of difficulty), the relevance of the items to the concepts, and cases of incomprehensible phrases. This process led to slight changes in items 6, 9, 14, and 15. The revised version was sent to the tool designer again and confirmed.
To assess the quantitative face validity, the same 10 patients were asked to explain their understanding of each item on a five-point Likert scale (1 = not at all understandable to 5 = completely understandable) to measure the face validity of the questionnaire. The formula ‘Impact Score = Frequency × comprehensiveness was used to calculate the impact score, in which the frequency indicates the number of individuals who assigned 4 or 5 points to each item [25]. Items with an impact score equal to 1.5 or higher were retained, while the rest were removed.

Content validity
In this study, the content validity of the final translated version was quantitatively and qualitatively assessed. We presented the questionnaire to field experts, including four nursing professors, two oncologists, two experienced individuals in tool development, and two oncology nurses, to qualitatively examine the content validity, and they were asked to express their opinions about the grammar, phrasing, and comprehensibility of the items.
The Content Validity Index (CVI) was calculated to evaluate the content validity. A version of the questionnaire was sent to the same 10 experts, again, and they were asked to score each item on a four-point Likert scale (1 = not relevant to, 4 = completely relevant) based on its clarity and relationship with the questionnaire. In the following step, the CVI was calculated for each item using the following formula:

CVI: content validity index,
n: number of experts who scored an item 3 and 4 points,
N: total number of experts who responded to the subject [25].
The total score of each item’s CVI was divided by the total number of items to calculate the total questionnaire CVI. CVIs higher than 0.8 were considered acceptable [26].

The ceiling and floor effects
If more than 15% of the respondents received the lowest or highest score, floor and ceiling effects were present, indicating the insufficiency of the content validity [27]. The floor and ceiling effect was calculated among the 346 individuals participating in the study.

Construct validity
The current study used exploratory and confirmatory factor analyses and convergent validity assessment to determine the construct validity.

Exploratory factor analysis
EFA is used to discover the underlying structure of a relatively large number of variables. The minimum required sample for the exploratory factor analysis was 3 to 10 participants per item [28]. The sampling adequacy was evaluated using the KMO and Bartlett’s sphericity tests. KMO values higher than 0.5 were acceptable [29]. The Maximum Likelihood (ML) extraction method was employed, and an oblique Promax rotation was used to allow for correlated factors. Factor retention was determined based on eigenvalues greater than 1 (Kaiser’s criterion), with the final factor solution selected when these indicators converged. Items with factorial load lower than 0.4 were excluded from the questionnaire. The EFA was performed on data collected from 176 patients with cancer who met the inclusion criteria.

Confirmatory Factor Analysis (CFA)
The factors extracted in the EFA were assessed with CFA. A total of 170 participants, other than those who entered the EFA phase, participated in the confirmatory factor analysis. Data analysis was conducted using Amos version 24, and several fit indices were used to measure the model’s fitness. The indices used in this study include RMSEA (Root Mean Square Error of Approximation), AGFI (Adjusted Goodness of Fit Index), GFI (Goodness of Fit Index), NFI (Normed Fit Index), RFI (Fit Index Relative), IFI (Index Tucker-Lewis), TLI (Fit Index Relative), CFI (Comparative of Fit Index), CMIN/DF (Minimum Discrepancy Function by Degrees of Freedom divided).

Convergent validity
The current study examined the relationship between the PCQ score, hospital anxiety and depression (HADS), and quality of life in 100 cancer patients who entered the exploratory factor analysis phase. The presented hypothesis stated that the average parenting concerns score has a positive association with anxiety and depression score, and it is negatively related to the quality of life.

Reliability
The internal consistency (Cronbach’s alpha) was calculated to assess the reliability of the parenting concerns questionnaire. The scores of 100 patients were examined to calculate the Cronbach’s alpha, and coefficients higher than 0.7 were considered acceptable [30].

Data analysis
Data analysis was conducted using SPSS version 25 and Amos software. Frequency distribution tables, central indices (mean and median), and dispersion indices (standard deviation) were used to describe the data. Statistical tests (including Pearson’s test) were used to examine the convergent validity, and χ2 was used to assess the model fit in construct validity. Cronbach’s alpha was also used to ensure the internal consistency. P-values lower than 0.05 were considered statistically significant.

Results

Socio‑demographic information
A total of 346 patients with breast cancer participated in this study, among whom 312 (90%) were married and 34 (10%) were divorced. Most patients were university graduates (182), and most of them were housewives (294). The majority of the participants had one child under the age of 18 (225) (65%). Table 1 presents other demographic information about the participants.

Face validity
Ten breast cancer patients cooperated to assess the face validity of the tool. Regarding the qualitative validity, patients described the tool’s items as simple, clear, and related to the study goal. However, considering the existence of words such as death and mortality, the patients had a consensus that these words deliver a negative meaning to their minds, and asked to change them. Accordingly, after contacting the tool designer through email and receiving permission, these items were slightly changed without changing the meaning of the phrases. In items 6, 9, 14, and 15, the word ‘death’ was replaced with the phrase ‘progression of the disease.’ Since the item’s impact score was higher than 1.5 in the quantitative phase, all the items were retained and none were removed.

Content validity
In the qualitative phase of the content validity assessment, all the experts deemed every item understandable and necessary. Therefore, the tool didn’t undergo any change in this phase. In the quantitative phase, CVI was calculated for each item, which ranged between 0.8 and 1. Finally, S-CVI scale equal to 0.9 was obtained.

The ceiling and floor effects
The lowest and highest scores were 15 and 73, respectively, with 2 participants obtaining the minimum score and 2 participants obtaining the maximum score. The ceiling and floor effects were therefore 0.6%, indicating acceptable ceiling and floor effects of the tool.

Exploratory factor analysis
KMO = 0.846, and Bartlett’s sphericity was statistically significant, demonstrating the adequacy of the sample size (χ2 = 1564.67; df = 105, P < 0.001). The factorial load of all items was over 0.5 and none of the items was removed in this phase. However, the result of exploratory factor analysis revealed the three dimensions, similar to the original version of the questionnaire. In the original version, the dimension ‘Practical impact of illness on child’ included items 1, 3, 5, 8, and 10, the dimension ‘Emotional impact of illness on child’ included items 2, 4, 7, 9, 11, and the dimension ‘Concerns about coparent’ included items 6, 12, 13, 14, and 15. In the present study, the item number 1 was allocated to the emotional dimension and items 4 and 6 were assigned to the ‘dimension practical impact of cancer on children.’ The percentage of the total variance explained was 66.270. Also, the factorial load of the items varied from 0.605 to 0.795 (Table 2) (Fig. 2).

Confirmatory factor analysis
Model modifications were performed based on covariance modification indices (MI) to improve the overall model fit, specifically by allowing theoretically justifiable correlated error terms. The results demonstrated that the modified model achieved acceptable fit. According to commonly recommended cut-off criteria (e.g., RMSEA < 0.08, CFI > 0.90, TLI > 0.90, GFI > 0.90), the fit indices indicated that the measurement model provided a satisfactory representation of the data. These results are summarized in Table 3 and confirm the construct validity of the factor structure identified in the EFA. The chi-squared test showed a suitable (p < 0.001, 173.003) and goodness of fit index (GFI) (0.88). Other indicators measured in this model were as follows: RMSEA = 0.08, CFI = 0.92, NFI = 0.87 and IFI = 0.92. Measured indices in this model confirmed the fitness. The following Table 3 presents the results of other indices (Fig. 3).

Convergent validity
The following Table 4 presents the results of the correlation tests between the mean parenting concerns score and the total anxiety and depression score, in addition to the quality of life score. Findings revealed that parental concerns had a positive relationship with anxiety and depression (r = 0.490, r = 0.627, P < 0.001), while showing a negative relationship with the quality of life (r=-0.477, P < 0.001).

Reliability
The internal consistencies were calculated as α = 0.89 and α = 0.77–0.89, for the questionnaire and its dimensions, respectively, which proved the acceptability of the internal consistency Table 5.

Discussion
Focusing on the family system, and specifically knowing the concerns of the parents with cancer, is critical to psycho-oncologic care. The present research aimed to translate and conduct the psychometric assessment of the parenting concerns questionnaire in Iranian women with breast cancer. Our findings revealed that PCQ is the only valid tool to measure the level of concern of parents with cancer. The results of the current study showed that the Persian version of this questionnaire has an acceptable validity and reliability, similar to the original version. Accordingly, so far, this questionnaire has been psychometrically analyzed in various countries, including Japan, Portugal, Sweden, Poland, Germany, and among different cancer patient populations during the treatment, those that recovered, and those with metastatic cancers [14, 17, 31–34].
The three-factorial structure of the questionnaire was confirmed in this research, similar to the original, Japanese, and Portuguese versions [17, 33]. A study by Inhestern et al. in Germany in 2015 showed that the three-factorial structure of the original version of the questionnaire was also confirmed in the population of cancer survivors in Germany [14]. However, Świgoń et al. reported that the Polish version of the questionnaire did not fully support the three-factor structure of the questionnaire, which may be related to the sample size, as factorial confirmation is sensitive to sample size. Sample size of 145 parents during cancer treatment participated in the study [34]. Park et al. reported in 2017 that the three-factorial structure of the questionnaire was not entirely supported by the confirmatory factor analysis in female patients with metastatic cancer, and showed a lower proportionality in that population compared with other cancer populations. In other words, the parental concerns of those in advanced cancer and near-death cases can be different from other cancer populations, and this questionnaire cannot cover the entire parental concerns of this population. Eventually, Park et al. matched the PCQ with the parental concerns of parents involved with advanced cancer stages and developed the PCQ.AD version [31].
The assessment of parenting concerns for parents with advanced cancer may be different from the concerns of parents with earlier stage or curable illness. Concerns about the impact of progressive illness and death may be more pertinent for this population than for other parents with cancer.
The results of the exploratory factor analysis of the Persian version showed that the dimensions of the tool had changed in three items, which, considering the cultural differences, seems reasonable. On the practical dimension, the item ‘If I die, there’s no one to take good care of my children,” although placed under the spouse concerns, yet, in since in the Persian culture looking after children is not merely on the parents’ shoulders and parents’ relatives also play supportive roles during the critical periods, such as diseases. In other words, in the Iranian context, family plays a central role in emotional and practical support during illness. The extended family often becomes deeply involved in caregiving and decision-making, which may alleviate parental concerns [35]. As this statement in Persian was related to receiving support, from the participants’ perspective, this could be considered as help from everybody other than their spouse [36]. Another item placed under this dimension was, ‘My children may require professional mental care.’ This statement was placed under the emotional dimension in the original version; however, since this item refers to seeking help, it could be considered related to the practical dimension from the participants’ point of view.
The item ‘My mood, concerns, or emotions affect my children’ was placed under the emotional dimension. Since this item refers to the emotional cases regarding the children, women with breast cancer interpreted this item under the dimension’ concerns about the emotional impact of cancer on children.’
Regarding the face validity of the tool, some of the mentioned items had slight changes in the results. The items included phrases related to death and mortality, and since cancer in Persian culture is still related with stigma and refractory disease, and talking about death and dying is associated with taboo in Iranian society, and won’t happen conveniently, after obtaining permission from the tool’s designer, these phrases were changed into disease progression and words close to the meaning of death [37, 38]. On the other hand, strong social expectations for mothers to remain emotionally resilient and prioritize their children’s well-being, even in the face of severe illness, may contribute to heightened feelings of guilt or anxiety [39].
The results of the convergent validity assessment also showed that the original version of PCQ has a significant relationship with the tools intended to measure the psychosocial tensions and quality of life. In this research, the Persian version of the questionnaire was positively correlated with the anxiety and depression questionnaire, and showed a negative correlation with quality of life. Accordingly, the results of other studies were consistent with the present research. For instance, a study on cancer survivors demonstrated a positive relationship between the general parental concerns score and its dimensions with HADS, and a negative association with the mental and physical health dimension and the quality of life [14]. Among the population of parents with metastatic cancer, there was a positive association between the parents’ concerns and anxiety and depression, and a negative relationship with the quality of life related to FACT-G [31]. Yanai et al. showed that the Japanese version of PCQ had a positive relationship with HADS and a negative association with quality of life [33].
Besides the mentioned cases of validity, the results of this study confirmed the acceptable reliability of the Persian version of the questionnaire, showed the acceptable Cronbach’s alpha, and the three dimensions of the questionnaire. Other studies on this subject were consistent with the present research. The values of alpha in different cancer populations, including patients under treatment and survivors, patients in advanced stages of the disease, and metastatic patients, were acceptable [14, 31].

Limitations
This study had some limitations. The participants were selected by convenience sampling. Moreover, the breast cancer patients in this study were selected from one of the hospitals of the capital (Tehran), the samples may not represent the entire population of the patient with breast cancer throughout Iran. In other words, Since Iran is a country with different cultures, generalizability of findings needs to be considered carefully. Another limitation some clinical information of participants was missing, such as the cancer stage, Considering that parenting challenges and concerns may change in different cancer stages, it is necessary to examine this questionnaire in other cancer populations, such as those in advanced stages of the disease, and among cancer survivors. Furthermore, considering that the nature of concerns and parents’ responsibilities may be different, it is recommended to investigate this questionnaire among the population of male cancer patients.

Conclusion
The results of this study demonstrated that the Persian version of PCQ can be a useful screening tool in psycho-oncologic care and to determine the effectiveness of interventions aiming to reduce the concerns of parents with cancer, and helping these patients to effectively fulfill this role, in addition to be a significant factor in supportive and palliative care.
PCQ can be considered as a useful screening instrument for the subjective needs of patients with cancer and might be helpful to individualize psychosocial care. Moreover, the PCQ can be used as an outcome measure in prospective studies on interventions for families affected by parental cancer.