Abstract
The death of a parent in a child’s life is a significant risk factor for later mental and physical health problems. While much has been written about the surviving parent’s functioning and its effects on their bereaved children, little work has been done to look into factors underlying this effect such as how the parent copes. The present study recruited 38 parent–child dyads from a community-based grief support center. Parent and child, independently, completed various measures of emotional functioning, including grief symptoms and coping such as social support and locus of control. The results indicated that parental coping did have an impact on children’s grief symptoms. This represents a unique view of adaptation in bereaved children: Parental coping strategies can have an impact on the child, independent of the child’s coping strategies. By focusing on parent coping, we have highlighted another possible pathway through which parental functioning affects children’s grief.
Parent and Child Grief
Although most children who experience the death of a parent will achieve a healthy adjustment, such a loss does become a risk factor for psychopathology in adulthood (Brown, Sandler, Tein, Liu, & Haine, 2007; Draper & Hancock, 2011; Pitman, Osborn, King, & Erlangsen, 2014; Wilcox et al., 2010). Early loss of a parent also puts individuals at risk for negative physical health outcomes (Luecken, 2008) including cortisol dysregulation (Biank & Wener-Lin, 2011) and later-life mortality (Smith, Hanson, Norton, Hollingshaus, & Mineau, 2014). A consensus is forming that around 20% of parentally bereaved children will experience depression or other mental health problems beginning 1 to 2 years after the death (Cerel, Fristad, Verducci, Weller, & Weller, 2006; Grenklo et al., 2013; McClatchy, Vonk, & Palardy, 2009; Melhem, Porta, Shamseddeen, Payn, & Brent, 2011; Worden & Silverman, 1996). Because it has been estimated that 3% to 4% of children in Western societies experience the death of a parent (Guldin et al., 2015), this is a significant public health concern. The wide range of possibilities in adaptation to the death of a parent suggests the importance of underlying factors which can influence the outcome of this major life event (Haine et al., 2006; Luecken & Roubino, 2012). In other words, there is a push to search beyond main effects and to identify mechanisms and moderators of outcomes (Haine, Ayers, Sandler, Wolchik, & Weyer, 2003; Luecken, 2008; Stroebe, Schut, & Stroebe, 2007; Williams & Aber, 2016). Potential interacting variables include the psychological functioning of the surviving parent, other within-parent factors, within-child factors, and certain environmental stressors. As occurred with the early stress and coping literature, this area of research has evolved into a search for resiliency factors (Ahern, Ark, & Byers, 2008; Lin, Sandler, Ayers, Wolchik, & Luecken, 2004).
The psychological functioning of the surviving parent, now the primary caretaker of the child in most cases, has emerged as a significant predictor of negative outcomes in children who have lost a parent (Lin et al., 2004). In fact, parental depression is a major risk factor for psychological disorders in children and adolescents in general (Luecken, 2008). This effect seems to take place via several key aspects of parenting following the death of a parent (Haine, Ayers, Sandler, & Wolchik, 2008). Primary among these is parenting effectiveness which can be reduced if the surviving parent is struggling emotionally (Christ & Christ, 2006; Hagan et al., 2012; Nickerson, Bryant, Aderka, & Hinton, 2013; Wolchik et al., 2008). They may be less consistent, structured, and organized. Surviving caregivers have been found to use less consistent discipline following the death of a parent, and this is related to children’s mental health (Brown et al., 2007).
The surviving parent’s warmth and caregiving, the latter being in relation to the parent’s reaction to the child’s grief can also suffer (Hung & Rabin, 2009). Hagan et al. (2012) term this combination of warmth, acceptance, and effective discipline “effective parenting” and cite much evidence for its being a protective factor for children encountering adversity. In particular, effective parenting following parental death is related to lower levels of mental health problems in children and adolescents (Hagan et al., 2012, 2006). Haine et al. (2006) identified a similar construct which they termed “positive parenting” and identified it as being comprised of parental warmth and consistent discipline. These researchers found that positive parenting helped improve outcomes for parentally bereaved children regardless of their gender (Haine et al., 2006). Bereaved parents who are struggling themselves may be less aware of depressive and other grief-related symptoms in their children (Hung & Rabin, 2009; Saldinger, Cain, Porterfield, & Lohnes, 2004; Werner-Lin & Biank, 2012).
Finally, surviving parents play an important role in facilitating a connection with the deceased through rehearsing memories, providing mementos, establishing rituals, and maintaining contact with the deceased’s family and friends (Saldinger, Cain, et al., 2004). Saler and Skolnick (1992) also found that the surviving parent’s facilitation of mourning rituals predicted better outcomes in adulthood.
Attention has been called to within-parent factors as potential determinants of child outcomes following the death of a parent. These can include various aspects of the parent’s personality as well as the prior relationship and modes of interacting with the child (Saldinger, Porterfield, & Cain, 2004). Among factors hypothesized to be important are the surviving parent’s social support, ability to express their emotions, and other aspects of coping (Haine et al., 2008; Hung & Rabin, 2009 ; Silverman, Baker, Cait, & Boerner., 2003; Werner-Lin & Biank, 2012).
There are within-child factors which seem to predict poor outcomes in parentally bereaved children as well. These include external locus of control (Christ & Christ, 2006; Haine et al., 2003; Haine et al., 2008), which bereaved children are more likely to develop (Brown et al., 2007; Haine et al., 2003). Closely related is a child’s lack of understanding of why events occur or “unknown control beliefs” (Lin et al., 2004). Fear of abandonment (i.e., fear of the loss of the surviving parent) is a common psychological reaction in children who have experienced the loss of a parent (Wolchik et al., 2008). Support seeking from adults, including the surviving parent, has been found to predict good adjustment following the death of a parent (Wolchik et al., 2008). Age at time of death (developmental level), child gender, and self-esteem (Haine et al., 2003, 2008) are other within-child factors which have been studied in reference to potential intervening variables following the death of a parent. Findings on gender, for example, have suggested that girls have more internalizing problems while boys have greater externalizing problems following the death of a parent (Haine et al., 2008). Others have not found gender to be a potent factor in a predictive model (Cerel et al., 2006). Positive or effective parenting can influence these factors including self-esteem, ability to seek social support and to enact coping strategies (Biank & Werner-Lin, 2011; Haine et al., 2006).
Apart from parenting other environmental factors have been found to impact child outcomes of parental death. Financial strain, geographical relocation, loss of contact with extended family, loss of routines, and positive events have all been found to be related to poor outcomes in children who have experienced the death of a parent (Brown et al., 2007).
Surviving Parent’s Grief
There is a growing body of evidence pointing to the importance of the surviving parent’s emotional functioning in predicting children’s outcomes following the death of a parent. Wolchik, Ma, Tein, Sandler, and Ayers (2008) hypothesized that negative effects on children’s sense of control and coping efficacy occur through the surviving parent’s emotional state. They proposed that these effects occur through less effective parenting as seen in less supportive behavior and less time to spend with their children. Families who had recently experienced the death of a parent were recruited from the community. A subset of these were assessed at the initial contact and again at 3 months and 11 months. The children’s life stressors were measured as well as the caregiver–child relationship quality (as reported by both child and parent). The researchers also videotaped parent and child dyads discussing two different family issues. In addition, the children’s fear of abandonment, self-esteem, coping efficacy, and grief were measured. In the cross-sectional analysis, relationship quality had no direct effect on grief but was significantly negatively related to fear of abandonment and positively related to coping efficacy. The longitudinal analyses allowed for the construction of a mediational model. This revealed that Time 1 stressors had an effect on fear of abandonment which in turn had a marginal effect on intrusive grief thoughts. Without the hypothesized mediators, Time 1 relationship quality had a significant negative effect on intrusive grief thoughts. The authors felt that their results supported prior research which demonstrated the importance of parenting in reducing mental health problems of bereaved children.
Saldinger, Porterfield, et al. (2004) studied the parenting capacities of 41 bereaved spouses who had school-aged children. Parents were interviewed and the responses were coded on the Child-Centered Parenting Coding System. Parents also filled out a self-report symptom inventory as well as one on their children. Children completed two self-report symptom inventories. The Child-Centered Parenting Coding System includes the following categories: facilitating continued attachment to the deceased parent, communicating information about illness and death, communicating about feelings, maintaining a stable environment, obtaining support for children, awareness/responsiveness to a child’s loss-related needs, exposure to the dying or deceased parent, participation in funeral or memorial services and meaning making. As predicted, the child-centered parenting was associated with less symptoms in the child. All measures of the child’s functioning were in the predicted direction, while only the parent-completed child inventory was statistically significant.
Positive parenting defined as consisting of caregiver warmth and consistent discipline was studied by Haine et al. (2006) in a sample of 313 recently bereaved children. Parents completed a series of questionnaires to assess positive parenting and children completed self-report symptom inventories. Parent–child interactions were videotaped and coded. The results indicated positive parenting was associated with less mental health problems in children regardless of gender of child or who was reporting on the child’s health problems. The authors proposed several mechanisms through which positive parenting may affect child outcomes following the death of a parent: through the satisfaction of the child’s needs for social connections, esteem, and control; a reduction of negative reinforcement and harsh punishment; promoting competencies such as social skills and problem-solving; and by encouraging active coping processes.
Lin et al. (2004) studied 179 bereaved children and their surviving caretaker. Measures included various aspects of caregiver functioning such as mental health problems, discipline, and warmth and children’s self-report regarding emotional expression, control beliefs, self-esteem, coping efficacy, and appraisal of threat. Environmental stressors were assessed as well as children’s mental health as measured by caregiver report (parent and teacher) and child self-report. Resilient children were found to have parents who exhibited higher levels of warmth and consistent discipline and who had lower mental health problems of their own.
Research Question
Although much has been said about the surviving parent’s functioning, particularly their parenting and its effects on their bereaved children, little work has been done to look into factors underlying this effect such as how the parent copes (Werner-Lin & Biank, 2012). In fact, few studies have looked into how multiple variables from each domain (stress, environment, and within person) combine to predict poor outcomes in bereaved children (Lin et al., 2004). The present study includes data on the surviving caretaker’s social support, locus of control, and engagement in participation in a grief program. Considering these aspects of the parent’s coping will enable an exploration of these as possible moderators of the parent’s emotional functioning and subsequent effect on their grieving children. Thus, the research question that has evolved is this: Will the parents’ coping act as a moderator between parental distress and child distress? Meaning, does the parents’ coping affect the strength of the relationship between the parents’ distress and the children’s distress (i.e., distress buffer or distress exacerbator)?
Following Haine et al.’s (2003) study, the moderated effect is investigated by entering parental distress and grief symptoms along with the coping measures as the first step in a multiple regression with the corresponding interaction term entered as the second step. The interaction terms can be summed up as Parental Distress/Grief × Parental Locus of Control, Parental Distress/Grief × Parental Social Support, and Parental Distress/Grief × Parental Engagement in the grief support group. The dependent variable is the child’s distress and grief symptoms.
Hypotheses
Children whose parents report an internal locus of control will report less emotional distress such as sad mood and worry and lower levels of grief symptoms such as difficulty accepting the death or anger related to the death. Children whose parents report more social support will be experiencing less distress and grief symptoms. Finally, children whose parents are more engaged in the grief center programming will report less distress and grief symptoms.
Method
Recently-bereaved families who were participating in a grief program in Milwaukee, WI, were recruited to participate in the present study. The grief center is Kyle’s Korner and its program is based on the Dougy Center model headquartered in Portland, OR. This model, which is used in over 100 children’s bereavement centers across the country, incorporates parental support along with the children’s programming. A typical session involves children’s grief programming occurring in one part of the center while parents or caregivers are discussing the impact of the death on the family and learning how to help their children cope. Kyle’s Korner has paid professionals overseeing programming, but the direct implementation of programming is done by trained community volunteers. Participation is free of charge.
In preparation for subject recruitment and administration of the study, the grief center’s staff were given a presentation by the researchers as to the consent/assent process as well as the nature of the measure to be used. Graduate students assisted with this process and were present on the evenings of data collection to ensure appropriate consent procedures as well as privacy measures. These data were originally collected for the purposes of an outcome study.
Participants were recruited during regular programming at the grief center. They were given a brief overview of the questionnaire that they were being asked to fill out as well as the consent/assent process. All potential subjects were assured that their participation in the study was voluntary and had no impact on their continued inclusion in the grief center’s programs. Parents were asked to consent for themselves and for their children, while children were asked to assent separately. If parents did not consent for their children, those children were not asked to participate. Individuals who did consent to participate were immediately given the questionnaire and asked to complete it at that time. Parents completed the Parent Self-Report and Parent Report on Child Functioning versions, while children completed the Child Self-Report version (discussed later). The graduate students remained in the room to answer questions and to collect the completed questionnaires.
Measures
Subjects completed the Dougy Center Evaluation (Coleman, DeCristofaro, Schuurman, Salazar, & Noh, 2012) which was originally designed for program evaluation and development. Permission was received from the authors for its use in the present study. Consultation was received from the first author. This instrument incorporated items from well-known and empirically validated measures such as the Texas Revised Inventory of Grief (Faschingbauer, Zisook, & DeVaul, 1987), Hogan Grief Reaction Checklist (Hogan, Greenfield, & Schmidt, 2001), and the Jimmerson Youth Common Grief Reaction Checklist (Jimerson, 2002). Demographic data as well as circumstances surrounding the death were also obtained. Response options were on a 5-point Likert-type scale indicating agreement or disagreement with the item. The full version of the Dougy Center Evaluation was used in the present study. There were several versions of the questionnaire: parent self-report (37 items), parent’s report on his or her child’s functioning (45 items), and child self-report. The child self-report appeared in two versions, one for youth ages 12 and older (56 items) and one for ages 6 to 11 (46 items) with developmentally appropriate wording and response options. For younger children (ages 6–9), the items were either read to them, or reading comprehension was checked informally. Parents filled out their self-report version and the version on his or her perceptions of the child’s functioning. All versions of the questionnaire took between 15 and 20 minutes to complete.
Several subscales were determined based on the origin of the items. The subscales are not demarcated on the questionnaire but instead appear as a continuous list. General Distress incorporates items regarding mood symptoms such as sad mood, tearfulness, energy level, ruminating, worry, and irritability. These items came from the Depression Self Rating Scale (Birleson, Hudson, Buchanan, & Wolff, 1987) and the Depression, Anxiety and Stress Scales, 2nd Edition (Lovibond & Lovibond, 1995). Higher scores indicate greater general distress symptoms. Grief symptoms included items which indicate emotional struggles with the death in particular (e.g., anger, risk-taking, denial) and items which suggest the possibility for growth from the experience (e.g., more caring, better coping). Sources for this subscale were the Texas Revised Inventory of Grief (Faschingbauer et al., 1987), the Jimmerson Youth Common Grief Reaction Checklist (Jimerson, 2002), and the Hogan Grief Reaction Checklist (Hogan et al., 2001). Higher scores indicated a greater number of grief-related symptoms (items measuring growth were reverse scored). The Locus of Control subscale includes seven items from the Nowicki–Strickland Children’s Locus of Control Scale (Nowicki & Strickland, 1973). Higher scores indicate greater external locus of control. The Social Support subscale is made up of five items measuring the subjects’ felt accessibility to family and friends on whom they can count and communicate with. A higher value equates with greater social support. The Engagement subscale consists of eight items which measure one’s satisfaction with and felt benefit from participation in the children’s grief program at Kyle’s Korner. Higher scores indicate greater satisfaction with and engagement at the center. These subscales appeared on all versions of the questionnaire. All subscales were scored by summing the responses on the Likert-type scales. Raw scores were used for statistical analyses rather than standardized scores.
Results
Demographics
The participating parent or caretaker in the study was predominantly female (84%) and White (84%). There was one African American parent, two Asian American parents, and three Latino parents who participated in the study. These caretakers ranged in age from 29 to 71 with a mean age of 43. All were educated at the high school (or equivalency) level at least, with 29% having college degrees and another 34% having graduate degrees. They had an average income of US$40,000 to $62,500 to support an average family size of three to four people. Furthermore, these caretakers were largely the biological parent of a child participating in the study (71%), while 23% were stepparents and just two were grandparents. The great majority of the children involved in the study were grieving a parent who had died (84%) and the majority of those had lost a father (71%). Four children had lost a sibling and two others had lost another relative or person close to them.
A total of 38 children (17 boys and 21 girls) who participated in the study ranging in age from 4 to 18 with an average age of 11. The ethnicity of the children was slightly different from the caretakers with 8 African American children, 2 Asian American children, 26 Caucasian children, and 2 Latino children. Some children were very new to the grief center (seven having only attended three or less times), whereas the average number of times attended was 15. While some had just started at Kyle’s Korner, six had been there over a year and the average length of attendance was 7 months. The death had occurred between 1 month and 3.5 years prior to the study with the average time since the loss being 14 months. Over half of the deaths (55%) were due to an illness, while seven were described as sudden and four were the result of suicide. All but two of the children lived with the surviving parent and the others lived with a grandparent. Only two of the children had experienced a previous trauma, and the average number of other losses (e.g., other deaths, geographic moves, divorce) was 1.5.
Statistics
Six of the children in the present study experienced the death of someone other than a parent as noted earlier. To determine the necessity of partitioning the data set based on this, differences on child self-reported grief symptoms were analyzed between those who had lost a parent and all others. Independent samples t tests were performed, and there were no significant differences on children’s general level of distress, t(36) = .28, p = .78, or on specific grief symptoms, t(36) = −.73, p = .47). Therefore, all further analyses were conducted without partitioning the set by who died. Likewise, time since the death did not significantly predict children’s general distress (z = −1.00, p = .32) or grief symptoms (z = −.44, p = .66) in a linear regression analysis, and so that variable is also excluded from subsequent analyses.
To test the hypothesis that the effect of parental emotional functioning (as measured by the subscales General Distress and Grief Symptoms) on children’s emotional well-being (measured by child self-reported General Distress and Grief Symptoms) is moderated by certain aspects of the surviving parent’s coping, linear regression analyses were performed. Using Statistical Analysis System (SAS), a linear model was used with the assumption that the response variables will follow normal distribution. The shared variance (in some cases) of multiple child subjects being reported on by a single parent was accounted for within this model. This dependency among children nested within a family resulted in generalized estimating equations. Furthermore, all continuous predictors were centered to simplify the interpretation of the intercept and interaction terms. To begin with, parental general distress, grief symptoms, locus of control, and social support were loaded into the equation along with child age and sex, followed by the interaction variables. These interaction variables were computed as parental General Distress × parental Locus of Control, parental General Distress × parental Social Support, parental Grief Symptoms × parental Locus of Control, and parental Grief Symptoms × parental Social Support. These were regressed onto child self-reported General Distress and child self-reported Grief Symptoms. Initial analyses also included parental Engagement at the grief center. In this regression model, parental engagement did not predict child self-reported distress (z = .29, p = .77) or grief (z = −.09, p = .93). Therefore, this variable was left out of subsequent analyses.
Variables Regressed on Child Self-Reported Distress.
Wald Global Test of Moderation for Parental Grief and Distress on Child Grief.
Variables Regressed on Child Self-Reported Grief.
Note. This study was approved by the Medical College of Wisconsin/Froedtert Hospital Institutional Review Board (PRO ID PRO00024820) on June 8, 2015. Statistical consultation provided by Aniko Szabo, PhD, and Yayun Xu, Division of Biostatistics, Medical College of Wisconsin.
Using this model, two moderation effects were found (see Table 3). The first is parental social support, which was moderated by the impact of parental general distress on the child’s self-reported grief (z = −3.13, p = .001). The effect of parental social support depends on the level of parental distress. When parental distress is high, the parent’s social support has little impact on children’s grief-specific symptoms. When parental distress is relatively low, however a counter-intuitive finding appears: High social support is associated with greater children’s grief symptoms. In other words, when parents are not in distress but report having plenty of people in their lives who care about them and on whom they can count, their children are struggling more with their grief.
The second interaction effect found was regarding parental locus of control and it’s moderation by parental distress on the child’s grief (z = −2.18, p = .02). Again, at high levels of parental distress, parental locus of control seems to have little impact on children’s grief symptoms. At lower levels of parental distress however, parental internal locus of control is associated with lower levels of children’s grief symptoms and parental external locus of control is associated with high children’s grief symptoms.
To summarize, as parental distress increases, the impact of parents’ coping (in this case social support and locus of control) on children’s grief decreases. Thus, these coping variables appear active only when parental distress is relatively low. Another way to interpret the data is to consider the converse of the interaction effect, that is, the effect of the level of the coping variables on parental distress. Taking into account the fact that there was no main effect for parental distress on child’s grief symptoms, we can say that with average levels of social support and a balanced level of locus of control (not too internal and not too external), parental distress does not affect the child’s grief substantially.
Discussion
The much-documented effect of parental distress on a child’s grief following the death of a loved one has focused mainly on various aspects of diminished parenting effectiveness (Haine et al., 2008). This has been thought to take place via reduced structure and organization in the household, changes in disciplinary practices, and diminished caregiver warmth. Various aspects of these have been studied previously (Haine et al., 2006; Lin et al., 2004; Saldinger, Cain, et al., 2004; Wolchik et al., 2008). Such research has provided not only mechanisms for a well-documented phenomenon but can also be used to guide intervention.
The search for moderators in this equation is also useful in that it clarifies which children and families are at risk for poor outcomes and can therefore guide preventive interventions. We focused on certain aspects of parental coping as potential moderators of this effect and found that the parents’ social support and their locus of control did indeed act in just such a way. The first effect was not as predicted indicating that the more social support the parent had, the more grief symptoms their child experienced especially when parents’ level of distress was relatively low. The second effect was that having an internal locus of control served to buffer the effects of the parents’ distress (again, especially when at low levels) to produce fewer grief symptoms in their children. Parents’ coping does appear to interact with their emotional state to predict their child’s grief. In other words, parental coping can be a stress modifier of the parent’s functioning and how it impacts the child.
The counterintuitive finding that higher social support among parents was associated with higher grief symptoms in their children (when parental distress was relatively low) bears some interpretation. High levels of social support equated with feelings of having friends and relatives who care about one and who one can count on for help. Perhaps children experience their parents’ orientation toward these support persons as a threat to their own access to that parent. At such a time of loss, it is not uncommon for children to fear losing their remaining parent and thus can become preoccupied with that parent’s safety and whereabouts. They may even become possessive and protective of their remaining parent. It may also be that parental social support is manifested by time away from their children and thus has a negative impact on the child’s grief.
The fact that there were no main effects of parental grief or general distress on the child’s grief and distress does seem to mirror the overall finding for a wide range of possible outcomes in children who have lost a loved one. On the other hand, such main effects have been found previously and in fact often (Lin et al., 2004; Wolchik et al., 2008). It may be that this group is already selected for lessened negative outcomes by virtue of the fact that the sample was obtained from a children’s grief support center. This is also a highly educated group of surviving parents whose income level is certainly above the poverty level. Thus, perhaps the resources that the parents offered to their children had already reduced the impact of parental functioning on the children’s grieving process.
There were also stark differences between the constructs being measured by the Grief Symptoms subscales and the General Distress subscales. Parental functioning had no effect (either main effect or interaction) on the child’s general distress but did on the child’s grief symptoms. Children may feel what they are going to feel, in terms of distress or depression, but they may look to the adults around them to navigate through grief. This also highlights how grief and depression are two separate processes and experiences, a widely held belief in the grieving literature.
Also, parental grief did not have an impact on the children’s grieving while parental general distress did. This again highlights the unique experiential process of grief versus depression but may also point out that children may find their parents’ grief validation of their own. It is after all, in many families, a shared journey. Whereas if a parent is showing signs of depression, this could be more troubling for a child and signal danger or risk to a young mind. Parents after all have other practical reasons to be experiencing distress apart from normal grief after the death of a loved one, particularly a spouse. There are increased financial hardships and other responsibilities which now fall on the surviving parent, and these stressors can accumulate to increase distress or depression.
Finally, it was somewhat surprising that elapsed time since the death did not have an effect on children’s symptoms of grief of general distress. This sample had a wide range of time span between the occurrence of the death and the point in time that the subjects participated in our study (between 1 month and 3.5 years). This again may be a reflection of selection bias seen in a sample drawn from a grief center. In other words, these families had sought support and some were continuing to seek support even over3 years later, likely reflecting their own self-knowledge that they were still in a very active stage of grieving.
The present study is a unique view of the stress and coping literature in bereaved children. We focused on the coping of the parent and its impact on the child, rather than the relationship of the child’s coping to the child’s outcomes. By doing this, we have highlighted another possible pathway through which parental functioning affects children’s grief outcomes. So, in addition to the much-called for and well-supported focus on parenting skills to support grieving children (see Haine et al., 2008), the present study suggests that attention be paid to supporting the parents’ coping resources, so that they in turn can support their grieving children.
Shortcomings of the present study include a low number of subjects and a largely homogenous sample which renders results less generalizable. Further affecting the generalizability of these findings was the fact that subjects were drawn from a children’s grief center which introduces a selection bias toward improved coping and resources available to these families. Although the Distress and Grief subscales of the questionnaire were well defined, the other subscales such as Social Support, Locus of Control, and Engagement had relatively few items in each, perhaps reducing the validity of their stated objective. In fact, validity of the measure of these constructs may have been enhanced by using the original scales, rather than abbreviated versions which appeared on the Dougy Center Evaluation. The fact that the present design was cross sectional in nature means that causal inferences are more difficult to make. Such a design also does not allow for studying possible cofounding factors such as passage of time or certain stabilizing events that may have happened in the parents’ lives such as adjusting to single parenthood.
Future research should focus on community-residing populations rather than what amounted to a clinic-referred sample in the present study. Repeated measures over time would enhance the power of conclusions about the causality of parental functioning on children’s grief. Adding observations of or interviews with families would help clarify findings such as the unexpected way that parental social support impacted children’s grief. Such experiential measures would also help explain what is happening in families in which parental distress is very high.
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) received no financial support for the research, authorship, and/or publication of this article.