Abstract
Parents of child with cancer often perceive their child as vulnerable and are at risk of overprotecting their child because of the challenges associated with the illness. We examined the relationships of perceived child vulnerability, parental overprotection, and parenting stress among Korean parents of a child with cancer. A sample of 163 parents of a child with cancer completed questionnaires. Perceived child vulnerability was associated with parenting stress. Parental overprotection was associated with parental distress and difficult child, but not dysfunctional parent–child interactions. This study emphasizes the need to consider parenting capacity variables in the context of childhood cancer.
Parenting a child with cancer may pose additional stressors related to the child’s challenges in performing normal self-care activities and lack of understanding of the nature of the illness (An et al., 2011; Hung et al., 2004). Family systems theories posit that chronic illness in childhood affects the entire family, disrupting normal family life (Patterson & Garwick, 1994). Due to increased demands of care, the role for parents of children with cancer is intensified. In addition to the conventional parental role, these parents often take on new roles as a trainer, coach, teacher, and therapist to support the child with cancer (Norberg & Steneby, 2009). Therefore, parental distress combined with illness-related stressors may affect parenting practices (Long et al., 2014).
Parents of children with cancer experience high levels of parenting stress (Bennett et al., 2013; Rabineau et al., 2008; Rodriguez et al., 2012; Sulkers et al., 2015; Vrijmoet-Wiersma et al., 2010), higher than among parents of children with physical disabilities (Hung et al., 2004), those with other pediatric conditions (Guilfoyle et al., 2012), those with psychological problems (Feizi et al., 2014), and healthy children (An et al., 2011). The negative consequences of long-term parenting stress can seriously affect outcomes for both the child (Colletti et al., 2008; Fedele et al., 2011; Hile et al., 2014; van der Geest et al., 2014; Wolfe-Christensen et al., 2010) and parents (Litzelman et al., 2011). As a construct, parenting stress reflects the level of stress present in the parent–child relationship, including general emotional distress and dissatisfaction in the parenting role, dysfunctional parent–child interactions, and difficult child behavior or characteristics (Abidin, 1990). Because more children are being diagnosed with cancer (American Cancer Society, 2020) and childhood cancer can be a chronic illness that requires treatment for years (Young et al., 2002), further examination of factors related to parenting stress is needed to help parents reduce parenting stress and improve adjustment for both the parents and child.
Despite evidence of the high prevalence of parenting stress among parents of children with cancer, it is unclear to what extent this stress is related to variables such as parental perceptions of child vulnerability and parental overprotection. Referred to as parenting capacity variables (National Institutes of Health, 2006), these three constructs have been examined in the context of parents of children with a chronic illness (Hullmann et al., 2010). Knowledge about the relationship between parenting stress and these variables would help identify parents at risk of parenting stress and inform interventions to help such families.
Although the concepts of perceived child vulnerability and parental overprotection have been often linked and used interchangeably, they are distinctively different (Thomasgard & Metz, 1997; Tluczek et al., 2019). Perceptions of child vulnerability, first used by Green and Solnit (1964) to describe parental anxiety about their child’s health, refer to parental beliefs or attitudes toward their child as more susceptible to illness or injury (Hullmann et al., 2010). In contrast, parental overprotection is defined as parental behaviors considered excessive given the child’s developmental stage (Thomasgard et al., 1995). Both are reactions to a child’s illness, but the first is cognitive and the latter behavioral, and they influence each other.
Parenting stress can disrupt a parent’s ability to meet the child’s needs (van der Geest et al., 2014) and may be related to how parents perceive their children, especially when the child is unhealthy. Previous research found an association between parenting stress and parental perception of child vulnerability among parents of children with cancer (Lemos et al., 2020; Vrijmoet-Wiersma et al., 2010), chronic pain (Connelly et al., 2012), and other disabilities (Driscoll et al., 2018). Children with a chronic health condition are perceived as more vulnerable by their parents compared to healthy children (Houtzager et al., 2015). Parents who are less educated tend to perceive their children as more vulnerable (Anthony et al., 2003; Sharkey et al., 2019). Younger children (Houtzager et al., 2015; Sharkey et al., 2019) and children with greater treatment risk are perceived as more vulnerable (Sharkey et al., 2019).
Parents of children with cancer are more likely to perceive their child as fragile due to lack of physical strength and motor function, even years after the completion of treatment (Staba Hogan et al., 2018), and this increases the risk of overprotecting their child (Norberg & Steneby, 2009). Older child age and advanced maternal education were associated with less parental overprotection among parents of childhood cancer survivors (Sharkey et al., 2019). Parental overprotection of a child with cancer has been linked to many negative psychosocial outcomes for children, including depression (Stein et al., 2000), anxiety (Spada et al., 2012), and decreased autonomy (Holmbeck et al., 2002). However, when compared with parents of healthy children, evidence suggests no differences in overprotective parenting practices in parents of children with cancer from either the parent’s perspective (Davies et al., 1991) or child’s perspective (Tillery et al., 2014). Furthermore, when compared with parents of children with other serious illnesses such as asthma, diabetes, and cystic fibrosis, parents of children with cancer did not show any differences in overprotective behaviors (Hullmann et al., 2010). Tillery et al. (2014) suggested that child distress, rather than the child’s illness, is more strongly associated with parental overprotection.
The family adjustment and adaptation response theoretical model (Patterson, 1988) is often useful in understanding the reciprocal relationships between a child’s chronic illness and functioning of the family system. According to this model, when a family faces a stressful life event such as a chronic illness, it goes through cycles of adjustment, crisis, and adaptation in an attempt to achieve homeostasis.
Method
Study Design
Drawing on the family adjustment and adaptation response theoretical model, this study examined the relationships of three parenting capacity variables—parenting stress, parental perceptions of child vulnerability, and parental overprotection—in a sample of Korean parents of a child with cancer. We hypothesized that both parental perceptions of child vulnerability and parental overprotection would be significantly related to parenting stress, beyond the effects of parent and child characteristics. We used a cross-sectional survey design with a sample of parents of child with cancer to obtain information on these variables.
Participants
Korean parents of child with cancer who met the following criteria were recruited between May and November 2015 from multiple organizations and support groups related to childhood cancer in Korea. Participants were included if they were (a) parents of a child diagnosed with cancer; (b) their child’s cancer was diagnosed before age 19; and (c) it had been less than 5 years since their child’s diagnosis. Generally, people with cancer who survive 5 years or more after diagnosis are considered to have completed cancer treatment without recurrence, a benchmark for cancer survivorship. Thus, parents of child with cancer within 5 years of diagnosis are most likely to be more affected by the child’s cancer. Parents whose child died were excluded from the study. Of the 173 eligible parents, 10 were excluded because they did not meet the inclusion criteria (eight due to more than 5 years elapsing since diagnosis and two due to no cancer diagnosis). Thus, 163 parents (117 mothers and 46 fathers) were included in the analyses.
Procedures
Participants were recruited through multiple strategies to ensure effective access to eligible participants residing in metropolitan and remote or rural areas. First, the research team visited branches of two major childhood cancer organizations in the community (i.e., Korea Childhood Leukemia Foundation and Korean Association for Children with Leukemia and Cancer). Also, the team recruited participants at childhood cancer-related events, such as parent education or family camps throughout Korea. Second, we contacted the leaders of support groups for parents of a child with cancer in hospitals. We sent a brief letter about the study to their group members to recruit participants. Finally, we posted recruitment flyers with contact information on the websites of the two aforementioned childhood cancer organizations and self-help groups for parents of a child with cancer. When eligible participants approached our research team, we explained the purpose and goals of the study, possible risks and benefits, voluntary nature of the study, confidentiality, and incentives for participation. Written informed consent was obtained from each participant according to the regulations of the ethics committee of the university with which the principal investigators are affiliated.
Questionnaire packages with return envelopes were delivered to potential participants via mail or in person according to their preference. After participants completed the 20-min questionnaire, they received a bookstore gift voucher worth about 5,000 won (equivalent to $4).
Measures
Parenting Stress
Participants’ level of parenting stress was assessed using the Korean version of the Parenting Stress Index-Short Form (K-PSI-SF; Chung et al., 2008), which was validated for Koreans based on the Parenting Stress Index (PSI; Abidin, 1995). The measure features 36 items consisting of three subscales: parental distress (PD), parent–child dysfunctional interaction (P-CDI), and difficult child (DC). PD includes items such as “I feel trapped by my responsibilities as a parent”; P-CDI includes items such as “My child is not able to do as much as I expected”; and DC includes items such as “I feel that my child is very moody and easily upset.” Each item is scored on a 5-point Likert scale ranging from 1 (strongly disagree) to 5 (strongly agree). Each subscale is summed to obtain an overall PSI total score, ranging from 36 to 180. Higher scores indicate high levels of parenting stress. According to the manual (Abidin, 1995), a PSI-SF total score in or exceeding the 90th percentile indicates clinically significant levels of parenting stress.
Several studies (Bennett et al., 2013) have examined parenting stress associated with having a child with cancer using the PSI (Abidin, 1995) or PSI-SF (Abidin, 1990). The PSI-SF has good internal consistency (Abidin, 1990; Solis & Abidin, 1991; Yeh et al., 2001) and has been used with parents of childhood cancer survivors in both the United States (Bennett et al., 2013; Kazak & Barakat, 1997; Patel et al., 2013) and Korea (Yoo et al., 2013). Previous studies found different levels of parenting stress with the three subscales (Hung et al., 2004), indicating the importance of identifying the source of parenting stress in families with a chronic illness. The K-PSI-SF has been validated to measure three factors (Chung et al., 2008) and thus, we used the three subscales of the K-PSI-SF. Similar to other research using the PSI-SF (Hung et al., 2004; Patel et al., 2013), we used raw scores for analyses, and Cronbach’s alpha ranged from .856 to .930.
Parental Perceptions of Child Vulnerability
Perceptions of child vulnerability among parents of a child with cancer were assessed using the Child Vulnerability Scale (CVS; Forsyth et al., 1996). The scale was translated into Korean by the bilingual and bicultural authors (first three authors), who have PhDs in social work and extensive knowledge and research experiences in childhood cancer or parenting. This scale features eight items. Each item is scored on a 4-point Likert scale ranging from 0 (definitely disagree) to 3 (definitely agree). The CVS has adequate internal reliability and high test–retest reliability (Forsyth et al., 1996; Thomasgard & Metz, 1996). A score equal to or greater than 10 is defined as clinically high vulnerability (Forsyth et al., 1996). Cronbach’s alpha in the present study sample was .799.
Parental Overprotection
Parental overprotection toward a child with cancer was assessed using the Parent Protection Scale (PPS; Thomasgard et al., 1995), which measures parenting behaviors related to child autonomy, individuation, and separation. This measurement consists of 25 items with four subscales (supervision, separation problems, dependence, and control). Items include “I decide when my child eats,”“I encourage me my child depend on me,” or “I protect my child from criticism.” Each item is scored on a 4-point scale (0 = never, 1 = sometimes, 2 = most of the time, 3 = always). Seven items are reverse coded, with a higher score reflecting greater levels of protective behaviors toward the child. A total score of one standard deviation above the sample mean is used as a clinical cutoff for elevated parental overprotection (Thomasgard & Metz, 1997) and thus, a cutoff score equal to or greater than 41.35 was utilized for the current sample. The PPS has acceptable internal reliability and high test–retest reliability, and validity for the total score was confirmed with a nonchronically ill pediatric sample (Thomasgard et al., 1995; Thomasgard & Metz, 1999). A recent study suggested that the PPS may be a unidimensional instrument in the context of a pediatric population with chronic illness (Fedele et al., 2010). Thus, we used the total score as a continuous variable, ranging from 0 to 75. Cronbach’s alpha in the present study sample was .845.
Control Variables
Sociodemographics for participating parents were controlled, including age, gender, educational level, and household monthly income. Marital status and residence area were not used as a control variable for hypothesis testing due to low variance. Employment status was also used for descriptive purposes but not as a control variable because of its high correlation with gender. Parents were asked about the child’s demographic and medical data, such as age, gender, diagnosis, age at diagnosis, time since diagnosis, treatment completed, and relapse or second cancer. Time since diagnosis and treatment status are potential correlates of parenting stress (Vrijmoet-Wiersma et al., 2010) and thus, treatment status (i.e., completed or not) was used as a control variable. Time since diagnosis was not used as a control variable because all participants’ children were within 5 years since diagnosis in our study.
Statistical Analyses
Descriptive analyses were conducted to understand the levels of parenting stress, perceived child vulnerability, and parental overprotection. Additionally, bivariate correlational analyses were conducted to examine multicollinearity among study variables. To examine the main research questions, three multiple regression analyses were conducted to assess how parenting stress may be related to perceived child vulnerability and parental overprotection among parents of a child with cancer, controlling for parents’ characteristics (age, gender, education level, and household monthly income) and child’s characteristics (age, gender, diagnosis, treatment completion, and relapse). For the current study sample, 98.1% of the participants were in relationships, and employment status was strongly correlated with gender. Thus, marital status, residence area, and employment status were not included as covariates. In the three multivariate models, the dependent variable (parenting stress) was measured via three K-PSI-SF subscale scores: PD, P-CDI, and DC. Several items of PSI were reverse coded, and the sum of 12 items under each subscale were used to calculate the three K-PSI-SF subscale scores. Finally, K-PSI-SF total scores were calculated by summing the three subscale scores. For the independent variables, PPS and CVS total scores were calculated as the sum of the item scores as a continuous variable. All data analyses were conducted using the SPSS 23.0 statistical package.
Results
Characteristics of the Study Participants
Table 1 shows the demographic characteristics of parents and their child with cancer. Approximately 71.8% were mothers and 28.2% were fathers. The mean age of the participants was 40.4 years (SD = 4.87), and their age ranged from 32 to 49 years. Most participating parents (n = 160, 98.1%) were in a relationship (either married or living with a partner), and their highest education (n = 122, 74.8%) was a 2 year college or 4 year university degree or more. More than half of the participants (n = 89, 54.6%) were not in the workforce (either stay-at-home parent or unemployed), and about 16.6% (n = 27) reported a monthly income of less than $2,000. All participants were living with and parenting their child who had cancer.
Participants’ and Their Child Characteristics (N = 163).
Regarding the child’s gender, there were more boys (n = 94; 57.7%) than girls (n = 69, 42.3%). Children’s mean age was 8.7 years (SD = 4.28), ranging from 2 to 19 years. The most common diagnosis were hematological cancers (n = 120, 73.6%) including acute lymphoblastic leukemia and lymphoma, followed by other solid or soft tissue tumors (n = 28, 17.2%), including osteosarcoma, hepatoblastoma, germ cell tumor, and brain or central nervous system tumors (n = 15, 9.2%). Their mean age at diagnosis was 6.10 years (SD = 4.34), ranging from birth to 18 years old. The average length of illness from the first diagnosis to the time of the study was 2.55 years (SD = 1.45). Approximately 51.2% (n = 83) of the children had completed treatment for cancer and 8% (n = 13) experienced relapse or a second cancer.
Parenting Stress, Child Vulnerability, and Parental Overprotection
Table 2 presents descriptive statistics for the K-PSI-SF, CVS, and PPS as reported by the parents of a child with cancer. The mean K-PSI-SF total score was 86 (SD = 19.19) with a range of 43 to 139, and approximately 15.7% (n = 25) of the parents experienced clinically significant levels of parenting stress (>90th percentile). The mean score on the CVS was 11.03 (SD = 4.20), and 68.8% (n = 110) scored 10 or greater, indicating they perceived their child as highly vulnerable. The mean score on the PPS was 31.86 (SD = 9.49), and 14.4 % (n = 23) of the parents reported parental overprotective behaviors toward their child with cancer.
Descriptive Statistics of Parenting Stress Index Short Form, Child Vulnerability Scale, and Parent Protection Scale.
Note. Clinical cutoffs were ≥41.35 for parental overprotection, ≥10 for perceived child vulnerability, and ≥90 percentile for parenting stress.
Correlations Between Study Variables
As shown in Table 3, the three subscales of the K-PSI-SF were found to be intercorrelated. PD was positively correlated with P-CDI (r = .524, p < .01) and DC (r = .513, p < .01); P-CDI was positively correlated with DC (r = .677, p < .01). Perceived child vulnerability was correlated with the three subscales of the K-PSI-SF (r = .415, p < .01; r = .247, p < .01; r = .287, p < .01, for PD, P-CDI, and DC, respectively). Parental overprotection was correlated with PD (r = .511, p < .01) and DC (r = .303, p < .01). A moderate correlation existed between perceived child vulnerability and parental overprotection (r = .369, p < .01).
Correlation Analysis of Three Subscales of Parenting Stress Index Scores with Child Vulnerability and Parental Overprotection (N = 158).
p < .01.
Regression Analysis
Table 4 shows the results of multiple regression analyses. The first model accounted for 32.2% of the variance in PD scores, F(1, 143) = 7.648, p < 001. Higher levels of perceived child vulnerability (β = .245, p < .01) and parental overprotection (β = .410, p < .001) were associated with higher PD scores. The second model accounted for 7.6% of the variance in P-CDI scores, F(1, 143) = 2.158, p < .05. Higher levels of perceived child vulnerability (β = .219, p < .05) were associated with higher P-CDI scores, but not parental overprotection. The third model accounted for 10.6% of the variance in DC scores, F(1, 143) = 2.668, p < .01. Higher levels of perceived child vulnerability (β = .227, p < .01) and parental overprotection (β = .178, p < .05) were associated with higher DC scores. In the three models, only having a diagnosis of brain or central nervous system tumors was associated with lower P-CDI scores. Other parent and child characteristics were not significantly associated with K-PSI-SF total or subscale scores.
Regression Analyses of Factors Associated with PSI Scores and Three Subscales.
Note. PD: Parental Distress; P-CDI: Parent-Child Dysfunctional Interaction; DC: Difficult Child.
0 = female, 1 = male.
0 = high school or 2-year college graduate, 1 = university graduate or graduate.
0 = $2,000,000, 1 = over 2,000,000.
0 = girl, 1 = boy.
0 = other tumors (brain, central nervous system, or other solid or soft tissue tumors), 1 = hematological cancers.
p < .05. **p < .01. ***p < .001.
Discussion
This study examined three parenting capacity variables—parenting stress, parental perceptions of child vulnerability, and parental overprotection—among Korean parents of a child with cancer. Specifically, this study sought to investigate whether parental perceptions of child vulnerability and parental overprotection were related to parenting stress after controlling for parent and child characteristics.
We found that significant numbers of parents experienced clinical levels of parenting stress (at or exceeding the 90th percentile, as recommended by Abidin, 1995). Several statistics on parenting stress provide a wider context to understand our sample’s levels of parenting stress. For example, in the United Kingdom, 51% of caregivers of a child with brain tumors reported clinical levels of parenting stress (Bennett et al., 2013), whereas in the United States, 19.3% of parents of a child with cancer reported clinical levels (Hullmann et al., 2010). The average parenting stress in our study was slightly lower that that reported by Bennett et al. (2013), but higher than that of parents whose child survived cancer in the United States (Patel et al., 2013). Caution is necessary when comparing these statistics because the conceptualization of parenting stress and timing of assessment differed in these studies. The study finding that almost 16% of our sample experienced clinical levels of parenting stress needs research and clinical attention. Korean parents of a child with cancer reported feeling overwhelmed by the dual role of caretaking and assisting their child with treatment (M. A. Kim et al., 2019) and lacked systematic support and thus, had unmet needs related to parenting (M. A. Kim et al., 2016).
In our study, many participants perceived their child as highly vulnerable. Previous research (Staba Hogan et al., 2018) found that a significant proportion of parents of childhood cancer survivors still perceive their child as vulnerable years after cancer therapy completion, independent of current health status. It is important to recognize that Korean parents’ attitudes toward a child with chronic illness may differ from those of Western parents. In Korea, when a child experiences a chronic illness, their family members accept it as “fate” and “punishment” for wrongdoings in a previous life (E. S. Park et al., 2009). These Korean parents often see their child as “a cross to bear”—someone who is weak and unable to have a normal childhood due to their illness. Therefore, the high percentage of parents who perceive their child as vulnerable in our sample should be interpreted in relation to their Korean cultural background. Furthermore, high perceptions of child vulnerability need attention because they have been linked to inadequate emotional adjustment (Colletti et al., 2008) and inadequate quality of life (Vance et al., 2001) among children with cancer. We also found a significant association between parenting stress and parental perceptions of child vulnerability, which is consistent with other study findings regarding parents of a child with cancer (Lemos et al., 2020; Vrijmoet-Wiersma et al., 2010) and other disabilities (Driscoll et al., 2018). As Patterson and Garwick (1994) explained using family systems theories, parental roles and obligations are reshaped with additional burden after a child’s cancer diagnosis. If parents perceive that their child, whose vulnerabilities have been intensified by cancer, is struggling with behavioral or emotional challenges, this may lead to even more parenting stress. Parental responsibility for and protection of an ill child have been identified as a major source of distress for mothers of a child with cancer (Young et al., 2002). Furthermore, previous research showed that parents of children with a chronic illness, such as chronic arthritis, diabetes, or asthma, who perceive their child as highly vulnerable have higher levels of anxiety and uncertainty (Anthony et al., 2003, 2011; Lopez et al., 2008; Ryan et al., 2011). Parents who perceive that their ill child is not adjusting well may think that the child might experience adverse consequences due to cancer, such as negative sequelae and even death, which increases the risk of anxiety (Wolfe-Christensen et al., 2014) and psychological symptoms (Mullins et al., 2004) among parents, which may lead to increased parenting stress.
In traditional Korean culture, deeply influenced by Confucianism, the mother’s role is often emphasized as caring for the child (M. Park & Chesla, 2007). Despite changes in parental role allocations in modern Korean society, traditional gender roles are still valued and mothers are expected to take the main responsibility of caring for children, whereas fathers take on the role of provider (H. O. Kim & Hoppe-Graff, 2001). Therefore, when a child has an illness, the mother is often blamed. Given that almost 72% of the parents in our sample were mothers, this finding may indicate that parents who perceive their child as vulnerable feel highly stressed in rearing the child with illness due to the cultural belief that they should be blamed for not having done a proper job as a parent. Only a few participants reported having overprotective behaviors toward their child with cancer. This is consistent with Sharkey et al. (2019), who also found only a small subset of parents reported overprotection in the clinical range. Previous research showed no differences in overprotective parenting practices between parents of healthy children and those of children with cancer (Tillery et al., 2014) or other types of illness (Hullmann et al., 2010). However, because parental overprotection has been linked to child negative outcomes, such as low health-related quality of life in children with cancer (Hullmann et al., 2010), this finding warrants more attention.
In this study, parental overprotection was positively associated with the PD and DC subscales. Previous research has found increased overprotection to be related to child anxiety (Howard et al., 2017; Spada et al., 2012) and both internalizing and externalizing behaviors in children with spina bifida (Holmbeck et al., 2002). As such, it is possible that parents show overprotective behaviors when experiencing stress related to having a difficult child and being unable to manage the child. Previous research also found evidence that parents experiencing higher levels of parenting stress report higher levels of both emotional and behavioral problems in their children (Renk et al., 2006). Furthermore, parental overprotection has been shown to affect parental anxiety symptoms (Clarke et al., 2013), which may lead to stress relating to the parental role. Such explanations are speculative, and future research would benefit from investigating possible mechanisms.
Contrary to expectations, parental overprotection was found to be unrelated to the P-CDI subscale. Parental responses to protect the child’s physical and emotional well-being may be seen as an act of caring and thus, not necessarily damaging interactions with the child. Although few in number, some studies have documented associations between overprotection and positive outcomes in children, such as parental support for the child’s affect (Gagnon et al., 2020), whereas others found no association (Colletti et al., 2008). As such, it may be that parents in our study were more involved in supportive and caring behaviors when the child was feeling anxious or depressed, and such parenting practices might have produced positive effects on the child’s affect management (Gagnon et al., 2020). Thus, it is plausible that parental overprotection is not necessarily related to parent–child dysfunctional interaction and stress resulting from the interaction. Ungar (2009) suggested that the problem may be the inappropriateness of excessive concerns in low-risk environments rather than overprotective behavior. Further studies on motivations for parental overprotection are needed to verify the relationship between parental overprotection and parenting stress.
Unlike previous research that found associations of parenting stress with sociodemographic variables, such as income (Mullins et al., 2011), or child’s medical variables, such as treatment status (Vrijmoet-Wiersma et al., 2010), no control variables in our study were associated with parenting stress, except having a diagnosis of brain or central nervous system tumors, which was associated with lower P-CDI scores. These findings are contrary to a previous study that showed the rate of parent–child dysfunctional interaction was significantly higher in families with a child with a brain tumor than in families with healthy controls (Radcliffe et al., 1996). Families with a child with brain, central nervous system, or other tumors have less interaction due to functional limitations. Future studies oversampling this group and examining the impact of cancer type on parenting stress would be beneficial.
Consistent with our hypothesis, perceived child vulnerability and parental overprotection were found to be important variables to explain parenting stress. These findings could help provide psychosocial oncology professionals with support while developing appropriate parenting-based interventions that assess parental beliefs about child vulnerability. Particularly, greater levels of parental education were associated with less parental child vulnerability (Anthony et al., 2003) and parental overprotection (Sharkey et al., 2019) and thus, parental education that reduces perceptions of child vulnerability is needed (Anthony et al., 2003). Parents’ perceptions of child vulnerability and overprotection can result from illness uncertainty, which is influenced by the child’s health status and parents’ perceived social support (Lin et al., 2010). Parents of children with cancer tend to have health anxiety due to feelings of helplessness and lower acceptance (Bilani et al., 2019). Thus, interventions that improve parents’ understanding of their child’s illness and health status can help them develop appropriate expectations for their child. Interventions that increase parents’ perceived social support would also alleviate anxiety and uncertainty, which ultimately would alleviate perceptions of child vulnerability and overprotection behaviors.
These two parenting variables ultimately affect later child adjustment (e.g., social competence; Fedele et al., 2011; Shin & Shim, 2011), child distress (Tillery et al., 2014), and other health-related quality of life factors. Parental overprotection and perceived child vulnerability may decline (Fedele et al., 2011; Vrijmoet-Wiersma et al., 2010) or increase (Driscoll et al., 2018) over time, depending on the child’s health status and adjustments. Thus, future studies could examine how parental variables change throughout survivorship trajectories, using a longitudinal study design.
Limitations
Several limitations of the current study must be noted. First, the sample included both parents of a child currently receiving treatment and those whose child already completed treatment. Although we controlled for treatment stage, types and patterns of parenting stress may differ. Thus, future studies should examine how parenting stress may manifest differently depending on the treatment stage. Second, the participants were recruited through support groups. Therefore, the parents in our sample may have had more distress or concerns about parenting compared to those not involved in support groups. Third, parental overprotection and perceptions of child vulnerability may differ from the perceptions of the child. Thus, a parent–child dyad approach to the assessment of their perceptions could be beneficial. Fourth, our study could not confirm causality because of its cross-sectional nature. Thus, the directions of observed associations are not clear. Finally, the current study did not differentiate whether parenting stress resulted from general or illness-specific issues. Parenting stress could involve managing daily demands and specific responsibilities related to assistance for cancer treatment. Thus, specific measures could improve our understanding of parenting stress and inform interventions for targeted populations.
Conclusions
In conclusion, parents of a child with cancer who witnessed the tough treatment process that their child went through may perceive their child as vulnerable and wish to protect them from other risks. Thus, parents of children with cancer tend to perceive their child as highly vulnerable and show overprotective behaviors toward their child, but this adversely affects their psychological health by increasing parenting stress. We suggest that systematic support and psychoeducation is needed for parents of a child with cancer, ensuring they receive information related to their child’s vulnerability and encouraging developmentally appropriate independence.
Footnotes
Declaration of Conflicting Interests
The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Funding
The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This research was supported by funding from the Korean Association for Children with Leukemia and Cancer.
Ethical Approval
Ethical approval was obtained from the institutional review board of the Myongji university (MJU-2015-04-001-02)
Data Availability Statement
The participants of this study did not give written consent for their data to be shared publicly, so due to the sensitive nature of the research supporting data is not available.
