Family Communication and Psychological Health in Children and Adolescents Following a Parent’s Death From Cancer

Abstract

Parentally bereaved children and adolescents are at risk of developing psychological health problems. Evidence for a correlation between communication and broad measures of psychological health exists in other populations. The aim of this study was to examine associations between family communication and specific aspects of psychological health for children and adolescents following a parent’s death from cancer using parent-proxy and adolescent self-reports. Parent-proxy reports for children and adolescents, and adolescent self-reports for Parent–Adolescent Communication, Strengths and Difficulties Questionnaire, and Prolonged Grief-13 child were analyzed using descriptive statistics and Spearman’s correlation. Parents rated communication as moderate in quality and reported good psychological health for children and adolescents. Adolescent self-reports indicated low-quality communication with their parent and poor psychological health. Significant associations between Parent–Adolescent Communication subscales and Strengths and Difficulties Questionnaire subscales were found for each group. Prolonged grief was associated with emotional problems but not communication for all three groups.

Keywords

childhood bereavement family communication internalizing problems externalizing problems prolonged grief

Introduction

Children and adolescents are at risk of developing increased internalizing problems such as depression, anxiety, somatic complaints, emotional problems, or difficulty forming relationships with peers; externalizing problems such as aggression, delinquency hyperactivity, and conduct disorder; as well as prolonged grief following the death of a parent (Ayers et al., 2014; Dowdney, 2000; Spuij, Dekovic, & Boelen, 2015; Stikkelbroek, Bodden, Reitz, Vollebergh, & van Baar, 2016). For example, Stikkelbroek et al. (2016) found that adolescents were significantly more likely to develop or have increased internalizing problems within 2 years of a family member’s death compared with nonbereaved controls while accounting for preloss internalizing problems. This was not found for externalizing problems. However, the study included all types of family bereavement and was not specifically focused on parental bereavement which is often considered more traumatic for the child as they are permanently losing a primary attachment figure (Dowdney, 2000; Howell, Shapiro, Layne, & Kaplow, 2015).

Prolonged grief disorder, a type of complicated grief (Cohen & Mannarino, 2004; Spuij, van Londen-Huiberts, & Boelen, 2013), is characterized as a sense of disbelief that the death happened, feelings of anger, guilt or bitterness, prolonged intense yearning, intense painful emotions, rumination, catastrophizing, and avoidance of reminders of the loss (Shear, 2012; Shear & Shair, 2005; Spuij et al., 2015). Prolonged grief disorder in children and adolescents has been associated with increased suicidal ideation and impairments in health and quality of life (Spuij et al., 2013).

Clinical case studies indicate that child and adolescent grief reactions differ based on the child’s age and development, but research has not yet established specific guidelines regarding differential disturbance in bereaved children and adolescents (Dowdney, 2000). Knowledge of stages of child and adolescent development along with clinical case studies may provide a foundation for understanding children’s varied responses to loss (Dowdney, 2000; Green & Connolly, 2009). It is important to keep in mind that children’s cognitive understanding or interpretation of death may not match their level of development or maturity in other areas and may differ based on life experiences associated with death and loss; therefore, both age and development must be considered.

Generally, children’s grief may appear sporadic and short lived, but their limited language skills may be masking the actual duration and intensity of their grief. Children’s grief reactions differ from adult grief reactions in that children may have a limited capacity to tolerate emotional pain; children may be more sensitive to being different or perceived as different from their peers; and some children are unable to understand the implications of death such as irreversibility, universality, and inevitability (Green & Connolly, 2009).

Children under the age of 5 years know that death exists but struggle to understand that it is final. They often ask questions about when the dead person will return and may believe that their family and friends cannot die. They need continuous reassurance about the well-being of the people closest to them. Young children typically react strongly to significant loss, even if they cannot express it in words. Children aged 5 to 10 years begin to grasp the finality and universality of death (Pettle & Britten, 1995). Disturbances in mood such as jealousy, clinginess, aggression, anxiety, irritability, impatience, and even dysphoria are common (Dowdney, 2000). Adolescents have developed advanced cognitive skills which cause them to be more critical of information and explanations given by adults and may begin to develop their own beliefs and ideas about death that contradict with their parents/caregivers (Pettle & Britten, 1995).

A known protective factor for children and adolescent’s psychological health following the death of a parent is parent–child communication (Howell et al., 2016; Shapiro, Howell, & Kaplow, 2014). Communication is the primary process by which children and adolescents receive information and socioemotional support (Houck, Rodrigue, & Lobato, 2007). Furthermore, aspects of family communication such as exchange of positives, listening, and expression skills are important aspects of positive parenting which can reduce mental health problems for children independent of the impact of negative life events (Haine, Wolchik, Sandler, Millsap, & Ayers, 2006). The child’s age and developmental level help to determine how parents and children communicate with each other. Furthermore, communication within the family may be affected by new challenges which arise following a parent’s death and the family’s ability to cope with the loss (Weber, Alvariza, Kreicbergs, & Sveen, 2019).

Evidence for a correlation between communication and psychological health exists in several populations other than bereaved children and adolescents, but these studies often rely on broad measures of psychological health or examine only one specific psychological disorder (Heidari, Mortezaee, Masomi, & Raji, 2016; Hollmann, Gorges, & Wild, 2016; Houck et al., 2007; Wilson, Chernichky, Wilkum, & Owlett, 2014).

Even though communication is known to be a protective factor following the death of a parent, there is still a need to better understand which aspects of psychological health are associated with communication in order to develop psychosocial interventions targeting the most relevant problems for parentally bereaved children and adolescents. Furthermore, to our knowledge, there are no studies examining the relation between family communication and psychological health in children and adolescents following a parent’s death from cancer which use both parent reports and adolescent self-reports. The aim of this study was therefore to use both parent and adolescent reports to examine the relationship between family communication and specific aspects of psychological health of children and adolescents including emotional problems, conduct problems, prosocial behavior, hyperactivity, relationship problems, and prolonged grief following a parent’s death from cancer.

Methods

Participants

Individuals who died of cancer between 2013 and 2015 and were aged between 25 and 65 years at the time of death were identified by the Swedish National Causes of Death Register. The deceased were linked to offspring by the Multi-Generational Register at Statistics Sweden to identify if he or she was a parent of a child or children aged between 1 and 18 years at the time of death. If the deceased had been living in Stockholm County with a partner, the family was eligible for the study. The participants are the surviving partner and children. Participants must reside in Stockholm County during data collection and speak and understand written and spoken Swedish. Specific inclusion criteria from the National Board of Health and Welfare included: The deceased parent and surviving parent must have lived at the same address and children and adolescents must be registered at the same address as their parent at the time of the study. This effectively excludes families with separated or divorced parents and adolescents who have moved from the family home.

In total, 214 eligible families were identified and 49 families consented to participation. Parents (n = 39), with a mean age of 48.90 years (SD = 6.07), filled in parent report questionnaires for a total of 55 children, where 25 were girls and 30 were boys. Children’s mean age was 12.78 years (SD = 4.42). Fourteen of the participating parents filled in questionnaires for more than one child. Adolescent self-reports were collected from 23 adolescents aged 12 to 20 years, where 15 were girls and 8 were boys, with a mean age of 16.2 years (SD = 1.88) some adolescents participated even though their parent did not.

According to parent-proxy reports, none of the children or adolescents had sought professional help for psychological health before their parent became ill. Adolescent self-reports indicated that one adolescent had sought professional help for psychological health prior to their parent’s illness. Mean time since loss for parent reports was 2.78 years (SD = 0.78) and 2.9 years (SD = 0.91) for adolescent self-reports.

Procedure

The study was approved by the Regional Ethics Committee 2016/1192–31/1. Special consideration was taken with regard to the ethical aspects of this study because children and adolescents were involved and the harm a project inflicts must be balanced against the scientific knowledge it produces. Surviving parents identified as potential participants were sent information about the study by Statistics Sweden. The information letter included brief information about the study, a link to the questionnaire website where they could sign up for the study, and contact information for the research group. If families chose to participate, parents went online to the website, provided their contact information, indicated how many children they had in each age-group, and marked that they gave informed consent for their own and their children’s participation. Adolescents aged 15 years and older were able to give consent for participation even if their parent did not, and all children were required to assent to participation in accordance with Swedish law. If families preferred to participate using a paper questionnaire, they could request paper questionnaires from the research team. As the informational letters were sent out by Statistics Sweden, the research group did not have access to the families’ contact information and were therefore unable to contact potential participants unless they signed up for the study.

A link to the relevant questionnaires based on children’s age was then sent to parent’s e-mail address. Parents were asked to fill in a parent report questionnaire for each of their children regardless of the child’s age. Adolescents aged 12 years or older received a separate adolescent questionnaire once the surviving parent and the adolescent gave consent. If the participants did not fill in the questionnaire within 2 weeks, a reminder e-mail was sent.

Measurements

Demographic variables

Demographic questions were included in the parent and adolescent questionnaires and included participants’ age and gender: Time since the parent’s death, and the surviving parent’s employment and marital status were included in the parent version.

Parent and adolescent communication

The Parent and Adolescent Communication (PAC) Scale consists of 20 items, comprised of a parent form which was included in the parent questionnaire and an adolescent form included in the adolescent questionnaire. It measures family communication on two subscales: Open Family Communication and Problems in Family Communication with each subscale consisting of 10 items. Open communication reflects responses related to free expression and understanding, whereas problem communication measures hesitancy to disclose concerns and negative interaction patterns. Higher scores on the Open Communication subscale indicate higher levels of open communication. The scores for the Problem Communication subscale are reversed so higher scores indicate less problem communication. The two subscales are added together to get a total score for parent–adolescent communication with higher scores indicating a better quality of communication (Barnes & Olson, 1985, 2003). Scores below 70 indicate low communication and is used as the cutoff score in this study. The PAC has been validated using factor analysis (Barnes & Olson, 1985). The PAC was forward translated from English to Swedish by two independent researchers and back-translated to English by a native English speaker. The questions were then validated face to face with parents, children, and adolescents who participated in an earlier study, and families in the research group’s professional network. Internal consistency for the parent reports for children in this study was found to be high, α = .76 for the Open Communication subscale, α =.73 for the Problem Communication subscale.

For the parent reports for adolescents, internal consistency was high with α = .71 for the Open Communication subscale and α = .71 for the Problem Communication subscale.

For the adolescent self-reports, α = .89 for the Open Communication subscale and α = .84 for the Problem Communication subscale. While this scale was developed using age-appropriate language so that it could be completed by adolescents aged 12 years and older, we have chosen to use parent-proxy reports for children under the age of 12 years, as the only known reason for the age limit for the scale is children’s reading level. This decision made it possible to assess how parents communicate with younger children and adolescents.

Prolonged grief

The Prolonged Grief Disorder-13 Child (PG-13 Child) was completed by parents and adolescents to assess prolonged grief. The PG-13 Child is based on the PG-13 for adults (Pohlkamp, Kreicbergs, Prigerson, & Sveen, 2018; Prigerson et al., 2009) and is comprised of 13 items including 2 on duration and impairment which are answered with yes or no and 11 items assessing cognitive, behavioral, and emotional symptoms related to grief. Items 1, 2, 4, and 5 are rated on a 5-point scale measuring frequency with answers ranging from not at all to several times a day. Items 6 to 12 measure intensity of symptoms from not at all to overwhelming. The PG-13 total score is used in this study which is a continuous measure calculated by summing the symptom items, with scores ranging from 11 to 55. The Swedish version of PG-13 Child and the parent report version have previously been translated by the National Centre of Disaster Psychiatry at Uppsala University in 2017. Internal consistency for the PG-13 parent report for children was high with α = .88, and for parent reports regarding adolescents, α = .92. For the adolescent self-report, α = .82.

Strengths and difficulties

Parents completed the parent report and adolescents completed the self-report of the Strengths and Difficulties Questionnaire (SDQ). The SDQ is comprised of 25 questions which are divided into five subscales: Emotional Symptoms, Conduct Problems, Hyperactivity, Peer Relationship Problems, and Prosocial Behavior. These subscales can be combined to produce a score for internalizing problems, externalizing problems, and total difficulties. Each item is scored on a scale of 0 to 2 with 0 indicating not true, 1 indicating somewhat true, and 2 indicating certainly true. Questions 7, 11, 14, 21, and 25 are reverse scored. Higher scores indicate more problems. Cutoff scores for each subscale were defined based on a population-based study in the United Kingdom such that 80% of children’s scores were categorized as normal, 10% as borderline, and 10% as abnormal. Validity of the SDQ’s five-factor structure has been supported in large-scale surveys (Goodman, 2001; Malmberg, Rydell, & Smedje, 2003; Smedje, Broman, Hetta, & Von Knorring, 1999). Internal consistency was acceptable for the parent-proxy reports regarding children for total problem score α = .69 and parent reports regarding adolescents was high with α = .80. Internal consistency was acceptable for the adolescent self-reports with α = .69.

Analysis

Descriptive statistics was used to describe demographic variables, family communication, and psychological health. Parent reports were divided into two age groups, children aged 4 to 11 years (n = 22) and adolescents (n = 33). Spearman correlation coefficients were calculated, along with p values, using cluster robust covariance estimates, taking into account that some parents answered for several children in the case of siblings. The functions rank, lm, cluster-vcov, and coeftest, from packages base, stats, multiwayvcov, and lmtest, in R version 3.5.0 were used. For adolescent self-reports, Spearman correlation coefficients were calculated, along with p values in SPSS version 22. Spearman correlation was chosen due to the small sample size.

Results

Communication and Psychological Health Following a Parent’s Death From Cancer

The PAC total score for parent reports was 75.36 (SD = 9.36) for children aged 4 to 11 years and 78.4 (SD = 9.89) for adolescents which indicates a moderate level of communication with their child (Table 1). Ten parents (eight mothers and two fathers) rated their communication with their children (four daughters and seven sons) as being low or very low. Of these, six were children and five were adolescents. One of these parents rated their communication with both of their adolescents as low and one parent rated low communication with only one of his or her three adolescents.

Table 1.

Parent Reports of Psychological Health and Communication.

	Parent report for children Age 4–11(n = 22)			Parent report for adolescentsAge 12–20(n = 33)
	Mean	SD	Number above/below cutoff (% of sample)	Mean	SD	Number above/below cutoff (% of sample)
PAC total	75.36	9.36	6 (27%)	78.4	9.89	5 (15%)
SDQ Total Problem Score	8.59	5.21	2 (9%)	9.09	6.83	5 (15%)
PG13 total	13.9	3.71	NA	18.41	8.08	NA

Note. PAC = Parent–Adolescent Communication Scale; SDQ = Strengths and Difficulties Questionnaire; PG-13-child = PG-13 Prolonged Grief Disorder-13-child.

Parent ratings of child and adolescent psychological health indicated that most children were within the normal range, with a mean SDQ total difficulties scores of 8.59 (SD = 5.21) for children aged 4 to 11 years and 9.09 (SD= 6.83) for adolescents (Table 1). Seven parents rated their child as being in the abnormal range for total difficulties, four girls and three boys, of which two were children and five were adolescents. According to the parents, children’s symptoms of prolonged grief following a parent’s death from cancer were low, as PG-13 mean total score was 13.9 (SD = 3.71) for children aged 4 to 11 years and 18.41 (SD= 8.08) for adolescents (Table 1).

The adolescents (n = 23) reported low quality of communication with the surviving parent with PAC total score of 63.43 (SD = 6.50). One adolescent who rated communication with his or her parent above the cutoff for low communication had a parent who rated their communication as low (Table 2).

Table 2.

Adolescent Reports of Psychological Health and Communication.

	Mean	SD	Number above/below cutoff (% of sample)
PAC total	63.43	6.50	20 (87%)
SDQ Total Problem Score	29.21	7.40	22 (95%)
PG13 total (n = 52)	23.86	7.36	NA

Note. PAC = Parent–Adolescent Communication Scale; SDQ = Strengths and Difficulties Questionnaire; PG-13-child = PG-13 Prolonged Grief Disorder-13-child.

Adolescents reported SDQ Total difficulties mean score of 29.21 (SD = 7.40) indicating that 22 adolescents scored in the abnormal range for total difficulties and one scored in the borderline range. These scores imply that all of adolescents in our sample may be experiencing some type of clinically significant psychological distress. The adolescent mean scores for PG-13 were 23.86 (SD = 7.36; Table 2).

Correlations Between Communication Psychological Health Following a Parent’s Death From Cancer

According to the parent reports, better communication (PAC total) between the parent and child was associated with fewer psychological health problems in children aged 4 to 11 years (n = 22) following a parent’s death from cancer. More open communication was associated with fewer conduct problems and less hyperactivity. Furthermore, increased problem communication was associated with increased conduct and emotional problems. There were no significant associations between communication and, peer problems, prosocial behavior or symptoms of prolonged grief (Table 3).

Table 3.

Parent Reports: Spearman Correlation Coefficient (p Value) Between Communication and Psychological Health.

	Parent report for childrenAge 4–11(n = 22)			Parent report for adolescentsAge 12–20(n = 33)
	PAC Total	PAC Open	PAC Problem	PAC Total	PAC Open	PAC Problem
SDQ Total Problem Score	−0.38 (0.09)	−0.28 (0.20)	−0.38 (0.07)	−0.35 (0.04)	−0.22 (0.19)	−0.34 (0.05)
SDQ Conduct	0.13 (0.55)	0.15 (0.46)	0.19 (0.42)	0.45 (0.00)	0.41 (0.00)	0.35 (0.014)
SDQ Hyper-activity	−0.41 (0.04)	−0.27 (0.23)	−0.40 (0.04)	−0.30 (0.05)	−0.16 (0.33)	−0.33 (0.03)
SDQ Peer	−0.56 (0.00)	−0.55 (0.00)	−0.37 (0.03)	−0.43 (0.00)	−0.22 (0.24)	−0.49 (0.00)
SDQ Emotional	−0.34 (0.11)	−0.43 (0.03)	−0.22 (0.35)	−0.17 (0.41)	−0.10 (0.57)	−0.19 (0.35)
SDQ Prosocial	0.01 (0.94)	0.17 (0.41)	−0.17 (0.42)	−0.16 (0.32)	−0.08 (0.66)	−0.09 (0.58)
PG13-child total	−0.07 (0.71)	0.04 (0.84)	−0.21 (0.29)	−0.23 (0.30)	−0.25 (0.22)	−0.19 (0.39)

Note. PAC = Parent–Adolescent Communication Scale; SDQ = Strengths and Difficulties Questionnaire; PG-13-child = PG-13 Prolonged Grief Disorder-13-child.

p < .05.

Similarly, parent reports for adolescents (n = 33) indicated that better communication (PAC total) between the parent and child was associated with fewer psychological health problems in the children following a parent’s death from cancer. However, for this age-group, open communication was associated with more prosocial behavior and not associated with conduct problems. Similarly, less problem communication was associated with more prosocial behavior. Furthermore, increased problem communication was associated with increased conduct and emotional problems. There were no significant associations between communication and hyperactivity, peer problems, or symptoms of prolonged grief (Table 3).

Adolescents’ self-reported psychological health problems were related less open communication with the parent. There were low to moderate correlations between communication and prosocial behavior but these were not statistically significant. Adolescents self-reported conduct problems were associated with lower quality of parent–adolescent communication (PAC total, PAC open, and PAC problem). Unlike in the parent report, there was no association between adolescents’ self-reported emotional problems and quality of communication (Table 4).

Table 4.

Adolescent Self-Reports: Spearman Correlation Coefficient (p Value) Between Communication and Psychological Health.

	PAC Total	PAC Open	PAC Problem
SDQ: Total Problem Score	−0.32 (0.12)	−0.47 (0.02)	−0.22 (0.31)
SDQ: Conduct	−0.52 (<0.01)	−0.50 (0.01)	−0.50 (0.01)
SDQ: Hyperactivity	−0.02 (0.91)	−0.12 (0.57)	0.00 (0.98)
SDQ: Peer	−0.17 (0.43)	0.31 (0.15)	−0.08 (0.70)
SDQ: Emotional	−0.16 (0.44)	−0.34 (0.10)	−0.3 (0.88)
SDQ: Prosocial	0.39 (0.06)	0.40 (0.05)	0.35 (0.09)
PG13-child total (n = 52)	0.04 (0.80)	−0.09 (0.68)	0.12 (0.56)

Note. PAC = Parent–Adolescent Communication Scale; SDQ = Strengths and Difficulties Questionnaire; PG-13-child = PG-13 Prolonged Grief Disorder-13-child.

p < .05.

Correlations Between Strength and Difficulties and Prolonged Grief Following a Parent’s Death From Cancer

For children aged 4 to 11 years, only the Emotional Problems subscale was significantly positively correlated with symptoms of prolonged grief. There were significant correlations between parent-reported prolonged grief symptoms for adolescents and the SDQ total difficulties score as well as the subscale emotional problems (Table 5). Adolescents’ self-reported symptoms of prolonged grief were associated with emotional and conduct problems (Table 5). These scores indicate that children and adolescents with more emotional problems may also be experiencing more symptoms of prolonged grief or that the symptoms of prolonged grief overlap with symptoms related to emotional problems.

Table 5.

Spearman Correlation Coefficient (p Value) Between Psychological Health (SDQ) and Prolonged Grief (PG-13).

	PG-13Parent report for children age 4–11(n = 22)	PG-13Parent report for adolescents age 12–20(n = 33)	PG-13Adolescent self-reports(n = 23)
SDQ: Total problems	0.31 (0.16)	0.44 (<0.001)	0.39 (0.06)
SDQ: Prosocial	−0.15 (0.54)	−0.07 (0.59)	0.17 (0.42)
SDQ: Emotional	0.54 (0.00)	0.60 (<0.001)	0.48 (0.02)
SDQ: Conduct	0.21 (0.38)	0.06 (0.77)	0.48 (0.02)
SDQ: Hyperactivity	0.16 (0.46)	0.15 (0.41)	−0.01 (0.94)
SDQ: Peer	0.02 (0.90)	0.22 (0.17)	0.25 (0.24)

Note. PAC = Parent–Adolescent Communication Scale; SDQ = Strengths and Difficulties Questionnaire; PG-13-child = PG-13 Prolonged Grief Disorder-13-child.

p < .05.

Discussion

This is the first study, to our knowledge, examining the association between family communication and specific aspects of psychological health in children and adolescents following a parent’s death from cancer, using both parent and adolescent reports. Our results showed that many adolescents reported poor psychological health and low or very low quality of parent–child communication. Parents rated their children and adolescents as being in the normal range for psychological health problems and reported moderate communication with both age groups. A significant correlation between communication and conduct problems was found for all three groups. Communication was not associated with symptoms of prolonged grief for any group, but prolonged grief was associated with emotional problems in all three groups.

Parent reports in our study indicated that between 3% and 27% of children and adolescents may have a clinically significant level of psychological symptoms which is in line with statistics cited in previous research (Cipriano & Cipriano, 2017; Howell et al., 2016; Melhem, Porta, Shamseddeen, Walker Payne, & Brent, 2011; Worden & Silverman, 1996). In this study, adolescent self-reports indicated that this percentage was much higher, with 95% of adolescents reporting scores above the cutoff for SDQ total difficulties. This indicates that the adolescents who participated in our study may be experiencing poor overall psychological health which is in contrast to the findings of Heidari et al. (2016) that only 8% of adolescent self-reports indicated poor psychological health and 48% indicated moderate psychological health. One reason for this may be that cross-cultural studies have shown that Scandinavian parents often report fewer symptoms and Scandinavian adolescents often report more symptoms, than participants from other countries (Broberg et al., 2001; Heyerdahl, Kvernmo, & Wichstrøm, 2004; Rescorla et al., 2007; Van Roy, Groholt, Heyerdahl, & Clench-Aas, 2010). It is also possible that the adolescents exhibiting the most symptoms are those who chose to respond to the questionnaire.

According to parent and adolescent reports, better quality communication was associated with fewer psychological health problems in children and adolescents following a parent’s death from cancer. Specifically, more open communication was associated with better overall psychological health and more problem communication was associated with reduced overall psychological health. This was similar to findings of Houck et al. (2007) that better quality of parent–adolescent communication when a parent was chronically ill was associated with fewer adolescent reports of posttraumatic stress disorder. Howell et al. (2015) found that adolescents with clinically significant psychological health problems were more likely to avoid, suppress, or hide their feelings, in other words not communicate with their parent, whereas adolescents who used expressive coping strategies and were willing to process and openly express difficult emotions displayed more adaptive functioning following their parent’s death.

The significant positive association between parent reports of open communication and prosocial behavior in adolescents and moderate negative association between adolescent self-reports of problem communication and prosocial behavior indicates that communication may be important for increasing prosocial behavior in adolescents but not children under the age of 12 years. Young (2012) points out that cognitive and psychosocial development occur simultaneously. In other words, as a child’s cognitive skills are developing, so are their social and affective skills, although not necessarily at the same rate. The children in our study, aged 4 to 11 years, are likely able to use inductive reasoning or focus on a multifaceted problem, whereas adolescents are not only capable of these tasks but can also actively choose to engage in them or not. The issue is the adolescent’s disposition or desire to use a skill or execute a task (Kuhn, 2008). This may help to explain why communication was associated with prosocial behavior in adolescents as parents may be able to use communication to convince or encourage adolescents to engage in prosocial behavior thereby changing their disposition. This may not work the same way with younger children who are less skilled at inductive reasoning and problem-solving and less aware of their conscious decisions to engage in or refrain from specific tasks.

Conduct problems were associated with communication for all three groups. Parent reports for children aged 4 to 11 as well as adolescent self-reports showed that increased open communication was associated with fewer conduct problems and decreased open communication is associated with more conduct problems. Openness in communication is often associated with more sharing of information which may reduce the child’s uncertainty and anxiety (Houck et al., 2007). Furthermore, increased problem communication was associated with more conduct problems according to parent reports for both age groups and adolescent self-reports. Houck et al. (2007) found that more problem communication was associated with higher levels of posttraumatic stress and anxiety and that increased conflict may lead to adolescents developing feelings of vulnerability, anxiety, and isolation. For younger children, these conduct problems may appear as age-appropriate behavioral and emotional problems (Dowdney, 2000) such as temper tantrums, irritability, and impatience, whereas adolescents are more likely to exhibit aggression and delinquency (Ayers et al., 2014), concentration and behavioral problems (Bergman, Axberg, & Hanson, 2017; McClatchey & Vonk, 2005), and risk-taking behaviors such as substance abuse, criminal behavior, promiscuity, and reckless driving (Ellis, Dowrick, & Lloyd-Williams, 2013; McClatchey & Vonk, 2005).

Higher levels of problem communication were also associated with more emotional problems according to parent reports for both age groups but not for adolescent self-reports. As the adolescents’ ability to reflect on their own thinking is still developing, and the content and meaning of what the adolescent is thinking about affect the type of thinking that occurs (Kuhn, 2008), they may not be able to accurately separate their behavioral responses into the categories of conduct and emotion. A behavior such as throwing or kicking something out of frustration or yelling at a parent may be attributed by the adolescent to conduct problems as these are behavioral reactions but the parent may interpret it as an emotional response. Adolescents may not be able to identify and reflect on the emotion that is responsible for the behavioral response and therefore do not answer questions about their worry or fears accurately. Worden (1996) states that children who show more aggression, anger, or acting out behavior during the first 2 years following their parent’s death were more fearful or anxious regarding their surviving parent’s health and safety, were less able or likely to speak about their deceased parent, and had a lower sense of self-efficacy.

The correlation between the scores on PG-13 and SDQ indicates that symptoms of prolonged grief may overlap with symptoms related to emotional or conduct problems. Melhem et al. (2011) found that children with prolonged grief show greater functional impairment within the first 4 years following the death and that the combination of prolonged grief in the child and surviving parent was a reliable predictor of the child developing depression up to 3 years following their parent’s death. Similarly, lifetime comorbidity between prolonged grief and mood or anxiety disorders is a common occurrence in adults (Shear et al., 2011) and this could be assumed to be true for children and adolescents as well.

Strengths and Limitations

This study had several strengths and limitations. A strength is including the use of both parent and adolescent reports. Special consideration was taken with regard to the ethical aspects of this study because children and adolescents were involved and the harm a project inflicts must be balanced against the scientific knowledge it produces.

A limitation of this study is the small sample size. While our results had a high level of significance, they may not be generalizable to all children and adolescents who have lost a parent to cancer. Specific guidelines from The National Board of Health and Welfare and Statistics Sweden did affect our ability to recruit participants as families with separated or divorced parents and adolescents who had moved out of the family home were excluded. The rational given for these criteria was that there may be a possibility that the child or adolescent may not have had contact with their deceased parent and therefore not been aware of the deceased parents’ illness or death and that learning such information would possibly be traumatic or harmful to the children.

The fact that the research group could not contact participants during the recruitment process may have also affected the response rate for this study as the participants had to actively contact the research group. The requirement of active participation on the part of participants during the recruitment process may have also led to a more biased sample which included only those families already possessing adequate coping skills and possibly higher levels of psychological health and communication.

A limitation of the study is that we do not have information on the family members history of psychological health prior to the loss; hence, we do not know if they already had psychological problems. Another limitation is a lack of information regarding additional stressors, such as other losses and financial problems in the family, which may affect the families’ psychological health.

Furthermore, most families in our study seemed to have a high degree of functionality with parents who were educated and working full time. Despite this, our results indicate a high level of psychological health problems. It could be that nonresponders may be less educated, with less resources and coping skills and may have even higher levels of psychological health problems and poor-quality communication.

A methodological consideration for this study was the use of PAC for parents’ communication with children under the age of 12 years. The scale was developed to be readable by children as young as 12 years, but information regarding age restrictions for the use of the parent report was not found. The research group decided that, since the age limit seemed to be due to issues of comprehension and readability for the child or adolescent and not the parent, the parent report should be used for both age groups in order to have a consistent measurement for communication for all three groups.

Conclusions and Clinical Implications

An association between family communication and children’s psychological health was found. Our results indicate that focusing on communication between the surviving parent and child may be a useful target of clinical intervention. It can be hypothesized that teaching families to reduce problem communication and increase open communication could lead to increased prosocial behavior and decreased conduct problems in children and adolescents following a parent’s death from cancer. Similarly, interventions targeting specific behaviors such as reinforcing positive behaviors and teaching new skills or coping strategies to replace negative behaviors may in turn lead to improved communication such as a reduction in conflict or arguments. The association between symptoms of prolonged grief and conduct and emotional problems helps to illuminate the many ways in which children and adolescents may express their grief following a parent’s death from cancer which provides useful information for parents, teachers, and clinicians working to meet the psychological and emotional needs of these children and adolescents.

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 work was supported by The Kamprad Family Foundation (Grant No. 20150044), Gålö Foundation, The Erling-Persson Family Foundation, and Ersta Sköndal Bräcke University College.

Author Biographies

Megan Weber, MSc, is a PhD student at Ersta Sköndal University College in Stockholm, Sweden. She is studying the psychological health of children and adolescents after their parent has died of cancer. She is also designing and testing a supportive family intervention for parentally bereaved children and adolescents. Her background is psychology and she has previously worked as a school counselor.

Anette Alvariza, RN, is an associate professor with a PhD degree in caring sciences at the Department of Health Care Sciences, Palliative Research Centre, Ersta Sköndal Bräcke University College in Stockholm, Sweden. She is also a care development manager at a specialist palliative care unit. Her research focuses mainly interventions aiming to support families in palliative care. She has contributed in development and evaluation of interventions that proved to be effective with increased preparedness and competence resulting in safe care for patients with enhanced communication between family members and patients.

Ulrika Kreicbergs, RN, PhD, holds a professorship in palliative care for children and youth at Ersta Sköndal University College and is also associated to Karolinska Institutet, Stockholm, Sweden. Her clinical work in pediatric oncology led to her research on children and families affected by cancer. Her work has been acknowledged worldwide and Kreicbergs has continued her research in the field of palliative care. Her work is focused on factors in the health care that can be modified or avoided to reduce suffering and thereby enhance quality of life for the child and their families.

Josefin Sveen is associate professor in medical psychology at the Department of Neuroscience, Psychiatry, Uppsala University and Department of Health Care Sciences, Palliative Research Centre, Ersta Sköndal Bräcke University College in Sweden. Her research focus on psychosocial aspects of trauma and loss, including posttraumatic stress and prolonged grief. In her research she has evaluated psychosocial interventions for children and adults, who have experienced a trauma and/or loss of a family member.

References

Ayers

T. S.

Wolchik

S. A.

Sandler

I. N.

Twohey

J. L.

Weyer

J. L.

Padgett-Jones

, . . . Kriege

(2014). The family bereavement program: Description of a theory-based prevention program for parentally-bereaved children and adolescents. OMEGA—Journal of Death and Dying, 68(4), 293–314.

Barnes

H. L.

Olson

D. H.

(1985). Parent-adolescent communication and the circumplex model. Child Development, 56(2), 438–447.

Barnes

H. L.

Olson

D. H.

(2003). Parent-Adolescent Communication Scale. Retrieved from https://lifeinnovations.com

Bergman, A. S., Axberg

Hanson

(2017). When a parent dies–a systematic review of the effects of support programs for parentally bereaved children and their caregivers. BMC Palliative Care, 16(1), 39.

Broberg

A. G.

Ekeroth

Gustafsson

P. A.

Hansson

Hägglöf

Ivarsson

Larsson

(2001). Self-reported competencies and problems among Swedish adolescents: A normative study of the YSR. European Child & Adolescent Psychiatry, 10(3), 186–193.

Cipriano

D. J.

Cipriano

M. R.

(2017). Factors underlying the relationship between parent and child grief. Omega—Journal of Death and Dying. 80(1), 120–136. doi:10.1177/0030222817726935

Cohen

J. A.

Mannarino

A. P.

(2004). Treatment of childhood traumatic grief. Journal of Clinical Child and Adolescent Psychology, 33(4), 819–831.

Dowdney

(2000). Childhood bereavement following parental death. Journal of Child Psychology and Psychiatry, 41(7), 819–830.

Ellis

Dowrick

Lloyd-Williams

(2013). The long-term impact of early parental death: Lessons from a narrative study. Journal of the Royal Society of Medicine, 106(2), 57–67.

10.

Goodman

(2001). Psychometric properties of the Strengths and Difficulties Questionnaire. Journal of the American Academy of Child and Adolescent Psychiatry, 40(11), 1337–1345.

11.

Green

E. J.

Connolly

M. E.

(2009). Jungian family sandplay with bereaved children: Implications for play therapists. International Journal of Play Therapy, 18(2), 84–98.

12.

Haine

R. A.

Wolchik

S. A.

Sandler

I. N.

Millsap

R. E.

Ayers

T. S.

(2006). Positive parenting as a protective resource for parentally bereaved children. Death Studies, 30(1), 1–28.

13.

Heidari

Mortezaee

Masomi

Raji

A. R.

(2016). The relationship between family communication patterns and mental health in adolescents [Special issue]. International Journal of Humanities and Cultural Studies, 416–427.

14.

Heyerdahl

Kvernmo

Wichstrøm

(2004). Self-reported behavioural/emotional problems in Norwegian adolescents from multiethnic areas. European Child & Adolescent Psychiatry, 13(2), 64–72.

15.

Hollmann

Gorges

Wild

(2016). Motivational antecedents and consequences of the mother–adolescent communication. Journal of Child and Family Studies, 25(3), 767–780.

16.

Houck

C. D.

Rodrigue

J. R.

Lobato

(2007). Parent-adolescent communication and psychological symptoms among adolescents with chronically ill parents. Journal of Pediatric Psychology, 32(5), 596–604.

17.

Howell

K. H.

Barrett-Becker

Burnside

Wamser-Nanney

Layne

Kaplow

(2016). Children facing parental cancer versus parental death: The buffering effects of positive parenting and emotional expression. Journal of Child & Family Studies, 25(1), 152–164.

18.

Howell

K. H.

Shapiro

D. N.

Layne

C. M.

Kaplow

J. B.

(2015). Individual and psychosocial mechanisms of adaptive functioning in parentally bereaved children. Death Studies, 39(5), 296–306.

19.

Kuhn

(2008). Formal operations from a twenty-first century perspective. Human Development, 51(1), 48–55.

20.

Malmberg

Rydell

A-M.

Smedje

(2003). Validity of the Swedish version of the Strengths and Difficulties Questionnaire (SDQ-Swe). Nordic Journal of Psychiatry, 57(5), 357–363.

21.

McClatchey

R. S.

Vonk

M. E.

(2005). An exploratory study of post-traumatic stress disorder symptoms among bereaved children. Omega—Journal of Death and Dying, 51(4), 285–300.

22.

Melhem

N. M.

Porta

Shamseddeen

Walker Payne

Brent

D. A.

(2011). Grief in children and adolescents bereaved by sudden parental death. Archives of General Psychiatry, 68(9), 911–919.

23.

Pettle

Britten

(1995). Talking with children about death and dying. Child: Care, Health and Development, 21(6), 395–404.

24.

Pohlkamp

Kreicbergs

Prigerson

H. G.

Sveen

(2018). Psychometric Properties of the Prolonged Grief Disorder-13 (PG-13) in bereaved Swedish parents. Psychiatry Research, 267, 560–565.

25.

Prigerson

H. G.

Horowitz

M. J.

Jacobs

S. C.

Parkes

C. M.

Aslan

Goodkin

, . . . Neimeyer

R. A.

(2009). Prolonged grief disorder: Psychometric validation of criteria proposed for DSM-V and ICD-11. PLoS Med, 6(8), e1000121.

26.

Rescorla

Achenbach

T. M.

Ivanova

M. Y.

Dumenci

Almqvist

Bilenberg

, . . . Döpfner

(2007). Epidemiological comparisons of problems and positive qualities reported by adolescents in 24 countries. Journal of Consulting and Clinical Psychology, 75(2), 351.

27.

Shapiro

D. N.

Howell

K. H.

Kaplow

J. B.

(2014). Associations among mother–child communication quality, childhood maladaptive grief, and depressive symptoms. Death Studies, 38(3), 172–178.

28.

Shear

(2012). Grief and mourning gone awry: Pathway and course of complicated grief. Dialogues in Clinical Neuroscience, 14(2), 119.

29.

Shear

Shair

(2005). Attachment, loss, and complicated grief. The Journal of the International Society for Developmental Psychobiology, 47(3), 253–267.

30.

Shear

Simon

Wall

Zisook

Neimeyer

Duan

, … Keshaviah

(2011). Complicated grief and related bereavement issues for DSM-5. Depression and Anxiety, 28(2), 103–117.

31.

Smedje

Broman

J.-E.

Hetta

Von Knorring

A.-L.

(1999). Psychometric properties of a Swedish version of the “Strengths and Difficulties Questionnaire. European Child & Adolescent Psychiatry, 8(2), 63–70.

32.

Spuij

Dekovic

Boelen

P. A.

(2015). An open trial of ‘Grief‐Help’: A cognitive–behavioural treatment for prolonged grief in children and adolescents. Clinical Psychology & Psychotherapy, 22(2), 185–192.

33.

Spuij

van Londen-Huiberts

Boelen

P. A.

(2013). Cognitive-behavioral therapy for prolonged grief in children: Feasibility and multiple baseline study. Cognitive and Behavioral Practice, 20(3), 349–361.

34.

Stikkelbroek

Bodden

D. H. M.

Reitz

Vollebergh

W. A. M.

van Baar

A. L.

(2016). Mental health of adolescents before and after the death of a parent or sibling. European Child & Adolescent Psychiatry, 25(1), 49–59.

35.

Van Roy

Groholt

Heyerdahl

Clench-Aas

(2010). Understanding discrepancies in parent-child reporting of emotional and behavioural problems: Effects of relational and socio-demographic factors. BMC Psychiatry, 10, 56.

36.

Weber

Alvariza

Kreicbergs

Sveen

(2019). Communication in Families with Minor Children Following the Loss of a Parent to Cancer. European Journal of Oncology Nursing, 39, 41–46.

37.

Wilson

S. R.

Chernichky

S. M.

Wilkum

Owlett

J. S.

(2014). Do family communication patterns buffer children from difficulties associated with a parent's military deployment? Examining deployed and at-home parents’ perspectives. Journal of Family Communication, 14(1), 32–52.

38.

Worden , J. W. (1996). Children and grief: When a parent dies: ERIC. Guilford Press.

39.

Worden, J. W., & Silverman

P. R.

(1996). Parental death and the adjustment of school-age children. Omega—Journal of Death and Dying, 33(2), 91–102.

40.

Young

(2012). A unitary Neo-Piagetian/Neo-Eriksonian model of development: Fundamental assumptions and meta-issues. New Ideas in Psychology, 30(2), 241–249.