Facilitating Mourning Following Parental Death

September 1, 2001
Volume 18, Issue 9

Do children's reactions to losing a parent to cancer differ according to the child's age and mental status? What treatment options are most efficacious for each age group?

According to the 1990 U.S. census, approximately 2 million children and adolescents under the age of 18 (over 2%) have experienced the death of a parent. In contrast to the emphasis of earlier psychoanalytic interpretations on children's inability to mourn parent death (Freud, 1917; Wolfenstein, 1966), there is now general agreement that children as young as 3 years old do indeed grieve, that their mourning is different from that of adults, and that supporting children's bereavement may help to prevent later adverse psychological consequences (Bowlby, 1980; Furman, 1983; Weller et al., 1991).But systematic descriptions of how children's responses vary by their developmental status have been limited. This has led to confusion and uncertainty about how professionals, parents and other supporting adults can understand children's grief, distinguish normative from more pathological responses, and mediate the more stressful aspects of the experience.Healing Children's Grief: Surviving a Parent's Death from Cancer (Christ, 2000) reports the findings from a qualitative analysis of a parent guidance intervention study that began about six months before and ended 14 months after the death of one parent from cancer. The findings presented in this book emphasize the changing manifestations of mourning in 157 3- to 17-year-old children from 88 families.

There were several unique features of the parent guidance intervention research we conducted from 1988 to 1994 at Memorial Sloan-Kettering Cancer Center. First, rather than beginning after the death as did previous studies, families were identified in which the ill parent had an estimated survival time of three to six months. This permitted exploration of anticipatory responses of both children and adults. Secondly, only families where the parent had cancer were studied. Sudden or traumatic deaths such as homicide or suicide that may elicit traumatic stress responses, in addition to grief reactions, were not included. Finally, data were collected for both a quantitative and qualitative arm of the study, which permitted a more in-depth understanding of the phenomena. An expanded account of the qualitative methodology is available at http://www.childrensgrief.com .

Quantitative Arm

The first aim of the quantitative arm was to determine whether depression, anxiety and self-esteem scores of bereaved children were different from those of a non-bereaved community sample before and after the death of a parent. The non-bereaved control sample was selected because the parents had similar education and family incomes to those of the study sample and they were matched with the study children on age, gender and number of children in the family. The scores of children in the study sample were significantly worse (by about half the standard deviation) during the pre-death evaluation but were no longer different from those of the community sample at eight and 14 months after the death (Siegel et al., 1996a, 1996b).

A second aim for the quantitative analyses was to determine whether children whose parents participated in the more intensive parent-guidance intervention had better scores on three standardized measures after the intervention than those that participated in an alternative intervention. Although the scores of children in the experimental group were better than those in the control group, the differences did not reach significance except for greater improvement in trait anxiety of children in the experimental group. An assessment of the children's rating of their parent's parenting skills and communication by a measure developed for this study (Siegel et al., 1996a, 1996b) showed that the children who participated in the experimental intervention rated their parents as more skillful and more communicative than those who participated in the alternative telephone monitoring one.

Qualitative Arm

The qualitative arm of this study had a broader aim: to deepen our understanding of the bereavement process in children and adolescents, to understand the role of development in the children's mourning, to understand how positive and negative mediators and moderators impacted on bereaved children, and to develop a baseline that would permit longer term prospective follow-up. Of the 137 families who participated in the parent-guidance intervention, 88 were selected who completed most of the six pre- and post-death therapeutic interviews and the pre- and post-death assessments so that extensive therapy notes and audiotaped therapeutic interviews were available (Christ, 2000).

Developmental Grouping

The sample included 51 surviving mothers and 37 fathers and their 3- to 17-year-old children (78 sons and 79 daughters). Since an arbitrary age category is not a good indicator of development (Rutter, 1989), grouping children by several developmental attributes rather than arbitrary age groups has been recommended, but it remains a challenge to implement. For this analysis, three developmental areas (cognitive, psychodynamic and ecological) were found to be most pertinent in this particular sample. Grouping children by these developmental attributes yielded five age groups with the most developmental similarities.

Developmental Variations

The 3- to 5-year-old children, even when well-prepared by parents during the terminal illness, did not understand that death was permanent and irreversible. It was often three to six months after the death before they understood the parent would never return. Cognitively, they were pre-operational (Ginsberg and Opper, 1997); emotionally, they were intolerant of separations from primary caregivers but adjusted well to surrogates; and ecologically, the family microsystem was paramount (Bronfenbrenner, 1989). Their mourning was nonspecific and included temporary irritability, toileting regression, sleeping with the surviving parent, somatic symptoms, nightmares and increased separation anxiety. They wanted an intact family--one 5-year-old girl asked her mother to get two daddies in case one died. Anxiety was apparent during parental separation, especially during the terminal illness period. Parents found that play bereavement groups helped improve communication and decrease symptoms.

The 6- to 8-year-old children were in transition. Cognitively, they were late pre-operational with some early concrete operational attributes. They struggled most with misunderstandings of the situation based on magical thinking and their focus on themselves as the cause when bad things happened. Emotionally, they tolerated separations better, but needed clear preparation for emergency separations. Although the school microsystem became important, self-esteem remained predominantly dependent on parental approbation. When informed of their parent's death, they immediately understood its finality. Their mourning included joyful reminiscences. They displayed pictures of the parent, liked and wore deceased parental belongings, and many spoke of wanting to die to be with the deceased parent but, when evaluated by a psychiatrist, had no suicidal intent.

Most felt comforted by "speaking" with the dead parent, whom they located in a place (usually heaven), and most were convinced the parent watched and cared for them from there. Crying and sadness were sporadic--they quickly returned to play and activities. Play bereavement groups or contacts with peers with similar loss experiences could be helpful, as were reminiscences and rituals like going to the cemetery, observing Father's/Mother's day and others. It was harder for fathers to understand and be supportive of children in these first two groups because of the children's extensive cognitive, emotional and caregiving needs.

The 9- to 11-year-old children were in developmental equilibrium. They were solidly concrete operational. Their self-esteem was enhanced by parental and other adult approbation and by their own experience of competence. They left home to be with peers rather than to distance themselves from family, as did the next two groups. Ecologically, school, home, and peer and group activities (sports) were important. Intellectualization was an important defense, and they were able to compartmentalize their feelings. Although intolerant of strong emotions, their mourning included a strong need for intellectual understanding. In school, they wrote reports about cancer, its treatments and even about euthanasia. They often looked for ways to be helpful with caretaking, benefitting from active mastery.

It was vitally important that parents provide clear, detailed, concrete information and balance the children's intolerance of strong emotion with encouragement of appropriate reminiscences. More structured rituals seemed to help children in this age group overcome their fears of speaking informally about the parent who died. Like their younger peers, they mourned the parent's caretaking role, but more so their teaching and supportive attributes. Both mothers and fathers found boys and girls in this group easier to understand and support than younger and older children.

The 12- to 14-year old group was characterized by marked egocentrism and emotional and physiological disequilibrium. Inconsistent fluctuations in emotional responses and needs, such as dependence-independence, were common and, for parents, confusing. Formal operational attributes began to emerge. Emotionally, boys began to distance from parents (Blos, 1962); girls also began to change their relationship, especially with mothers, and started to enjoy talking about relationships, sharing shopping ideas and so on (Jordan et al., 1991). Both boys and girls seemed to need to be in control of their withdrawal, shown by assertions of independence and fury when their dependent needs were not met. Paradoxically, they also reacted strongly to the emotional and physical withdrawal of both parents during the terminal illness period. Boys and girls were intolerant of strong emotions, but they were also avoidant of information about the illness and death.

The peer microsystem not only included peers to do things with, as did the younger group, but also close friendships. Unlike the 15- to 17-year-old group, sharing personal grief with peers was very unusual. Grief was very private: "I cry in my room, alone" was their mantra. Surprisingly, they generally felt free to talk about dreams after the parent died, describing a strong sense of their continuing presence. Many said they spoke with the dead parent and generally felt a connection with them. Obvious identification with the parent who died, e.g., wearing their clothes, becoming interested in their vocations and avocations was most frequent in this age group. A marked contrast grief expression occurred with the few 12- to 14-year-old adolescents whose parents were recent immigrants. These adolescents more openly shared their grief with parents and interviewers suggesting the important role of culture in shaping grief responses.

Setting appropriate limits and providing formal family reminiscence experiences was important for this early adolescent group, as it was for the 9- to 11-year-old children. These rituals also helped them express feelings in formal ways that they avoided in more informal discussions.

The 15- to 17-year-old group showed more consistent pervasive formal operational attributes. Their anticipatory grief during the terminal illness included clear recognition that their family's future would be changed. Many experienced the full weight of adult grief shortly after the death, although its duration was generally shorter than their parents'. Where parents were concerned that 9- to 11-year-old and 12- to 14-year-old children were not grieving, parents of the 15- to 17-year-old children were more likely to be concerned about the intensity of the grief. The anhedonia was pervasive, including sleeping problems, waking early and inability to concentrate in school, even in sports.

Emotionally, their peer involvement was more intense, and it often included sharing grief experiences with others their age who had lost parents to death or divorce. They were more accepting of family responsibility, including that of caring for younger siblings. They mourned more complex attributes of the parent than did younger peers and thought in more specific ways about what the parent who died expected of them and how they might fulfill their dreams. As with the 12- to 14-year-olds, the challenge to the parent was setting appropriate limits and staying out of or getting out of spiraling conflicts. An additional task was to encourage confidence in independent functioning.

Factors in Bereavement

What was impressive was that 83.5% of the children had returned to previous levels of functioning as measured by a qualitative assessment of psychological evaluations. This compares with 12% who had high depression scores both before and after the death (Siegel et al., 1996a, 1996b). As described in the book Healing Children's Grief (Christ, 2000), most mothers and fathers succeeded in their struggle to balance their own powerful grief and their children's emotional needs. Most of these parents made excellent use of help offered by the intervention, but also outside that domain. They appropriately enlisted school personnel, family and neighbors. Surviving mothers were generally more psychologically attuned to their children, especially younger children, while fathers were more comfortable setting limits and allowing adolescents their independence.

A few more surprising findings that might account for the high proportion of youngsters who reconstituted well included the following: fathers fell into two groups. There were a number who could not or would not be supportive and understanding of their children, especially the younger ones. But there was a sizable group who learned these skills through different methods. Five of the 37 fathers actually did not work for three to six months after the death of their wives while they learned to understand and look after their children. A few elected to use their financial assets to hire competent caretakers and worked out realistic, though limited, times for interacting with their children. One father recognized his poor relationship with his young daughter and worked at improving it during his wife's terminal illness. When last seen, his daughter gave him high marks for improved communication. A few remarried fairly quickly, but selected a mate who was fond of the children and whom the children liked. Of interest was that these men were often still deeply mourning the death of their wives; their children and the new spouse were aware of this and responded positively. Finally, there were a few parents who dealt with psychosis or alcoholism in their background successfully by acknowledging their problem and returning for maintenance treatment. They sought clarity for those needs of their children that they were likely to be unable to meet, and they looked for people who would be able to substitute for them.

The intent of this study was to learn about children's grief experiences in situations with limited additional stressors: i.e., families with two parents, adequate resources and opportunities for preparation for parental loss. In future research, these results can be compared with children who confront multiple stresses with fewer resources. The results of this study are also based on evaluations eight and 14 months after the death. A follow-up study is planned that will explore the longer-term outcomes.




Blos P (1962), On Adolescence. New York: Macmillan.


Bowlby J (1980), Attachment and Loss. New York: Basic Books.


Bronfenbrenner U (1989), Ecological systems theory. In: Six Theories of Child Development: Annals of Child Development, Vesta A, ed. Greenwich, Conn.: JAI Press.


Christ G (2000), Healing Children's Grief: Surviving a Parent's Death From Cancer. New York: Oxford University Press.


Freud S (1917), Mourning and Melancholia. In: The Complete Works of Sigmund Freud, vol. 14, (1957) Strachey J et al., eds. London: Hogarth Press.


Furman E (1983), Studies in childhood bereavement. Can J Psychiatry 28(4):241-247.


Ginsburg H, Opper S (1997), Piaget's Theory of Intellectual Development, 3rd ed. Englewood Cliffs, N.J.: Prentice Hall.


Jordan J, Kaplan A, Miller J et al. (1991), Women's Growth in Connection: Writings from the Stone Center. New York: Guilford Press.


Rutter M (1989), Age as an ambiguous variable in developmental research. Int J Behavioral Dev 12(1):1-34.


Siegel K, Karus D, Raveis V (1996a), Adjustment of children facing the death of a parent due to cancer. J Am Acad Child Adolesc Psychiatry 35(4):442-450.


Siegel K, Raveis V, Karus D (1996b), Patterns of communication with children when a parent has cancer. In: Cancer and the Family, Cooper C, Baider L, Kaplan De-Nour A, eds. New York: John Wiley & Sons, pp109, 128.


Weller RA, Weller EB, Fristad MA, Bowes JM (1991), Depression in recently bereaved prepubertal children. Am J Psychiatry 148(11)1536-1540.


Wolfenstein M (1966), How is mourning possible? Psychoanal Study Child 21:93-123.