SEARCH

SEARCH BY CITATION

Keywords:

  • family;
  • parent;
  • children;
  • adolescent;
  • family function;
  • internalizing problems;
  • externalizing problems

Abstract

  1. Top of page
  2. Abstract
  3. MATERIALS AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. Conflict of Interest Disclosures
  7. References

BACKGROUND:

This study aimed to evaluate prevalence and risk factors for emotional and behavioral problems in dependent children of cancer patients using a multinational research design.

METHODS:

The sample comprised 350 ill parents, 250 healthy partners, and 352 children. Parents assessed the child's psychological functioning using the Child Behavior Checklist, parental depression using the Beck Depression Inventory, family functioning using the General Functioning subscale of the Family Assessment Device, quality of life using short-form questionnaire, and adolescents (N = 168) self-reported psychological functioning using the Youth Self Report.

RESULTS:

Children and adolescents, in particular latency-aged boys and adolescent girls, were of higher risk of psychosocial problems than norms. There was a higher risk of problems when the father was ill than when the mother was ill, but it remains unclear whether this difference was due to the different diagnoses of fathers and mothers, gender or other factors. The best predictor of internalizing problems in children and adolescents was parental depression, and the best predictor of externalizing problems in children and adolescents was family dysfunction.

CONCLUSIONS:

The results indicate the need for a family-oriented approach to psychological support of cancer patients. Cancer 2009. © 2009 American Cancer Society.

It is estimated that about 24% of adult cancer patients have dependent children.1 Having a parent with cancer may be a significant stressor to children.2-4

Between-group studies, where children and adolescents of parents with cancer have been compared with control groups or national norms, have produced mixed results. Some studies have shown that children and adolescents function in a similar way to, or even better than, other children.5,6 Others have found that children and adolescents, in particular female adolescents and latency-aged boys, show more emotional problems than normative groups.7-10 In a recent review of the psychosocial impact of early stage parental cancer on children and adolescents,11 it is concluded that children and adolescents do not experience serious psychological difficulties compared with reference groups, but that they are at a slightly increased risk for internalizing problems, and that adolescent daughters seem to be the most negatively affected group. The studies were, however, often characterized by using nonrepresentative comparison groups, low statistical power, an under-representation of fathers with cancer, and a lack of data from multiple informants (eg, patients, partners, and children).

Studies using a within-group design, focusing on risk factors for developing psychosocial problems, have explored a range of factors. Poor prognosis and longer disease duration of the parent have, in some studies, been associated with higher levels of distress in the child.5,12 Others have found that objective disease characteristics did not predict distress in the child, but a stronger predictor was the child's subjective perception of illness severity.13,14 A recent study with a relatively large sample size (N = 136 families) found a higher prevalence of problems in the children if the father was ill.10 In 3 studies on children of mothers with breast cancer, distress of the ill mother was positively associated with emotional and behavioral problems in the child.9,13,15 In Osborn's review,11 it was concluded, that there is little evidence of an association between disease/treatment variables and child outcomes, but there is suggestive evidence for associations between maternal depression and family factors and child outcomes. More recent data also suggest high stress levels in male (33%) and female (45%) adolescents and psychological morbidity were linked to poor family cohesion as well as maternal depression.16 There remains a need to further examine the impact of paternal cancer on children.

Given gaps in the existing literature, this study aimed to evaluate prevalence and risk factors for emotional and behavioral problems in dependent children of cancer patients using a multinational research design. The identification of “at-risk” groups will clarify the planning and targeting of family care services within oncology.

The goals of this study were as follows: In a multinational context, we 1) examined the rates of emotional and behavioral functioning of latency-aged and adolescent children of parents diagnosed with cancer by comparing them to normative groups; and 2) assessed risk factors for emotional and behavioral problems in children (factors include age and gender of child, gender of the ill parent, cancer diagnosis, illness duration, parental depression, family functioning, and ill parent's health-related quality of life).

MATERIALS AND METHODS

  1. Top of page
  2. Abstract
  3. MATERIALS AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. Conflict of Interest Disclosures
  7. References

Participants

Dependent children and their parents were consecutively recruited between May 2002 and April 2004 from hospitals in 6 European countries: Hamburg (Germany), Basel (Switzerland), London and Sutton (United Kingdom), Turku (Finland), Vienna (Austria), and Aarhus (Denmark). The inclusion criteria for patients were as follows: 1) cancer diagnosed at least 2 months before the start of the study and 2) at least 1 biological child aged 4-17 years living or having regular contact with the patient.

The total sample comprised 352 families, 350 ill parents, 250 healthy partners, and 352 children (who were reported on by at least 1 parent). Only 1 child selected at random per family was included in the study.

Seventy-two percent of the families comprised both biological parents. Socioeconomic status was categorized using the highest educational level of the parents. Eighty-7 families (25%) belonged to the lowest economic status category, 129 (36%) to the middle category, and 132 (37%) to the highest category. Data were missing for 6 families. Patients were diagnosed with various types of cancer, the majority (62%) with breast cancer.

Demographic and medical characteristics are summarized in Table 1

Method

Parent participants and adolescents (aged 11-17 years) completed self-report questionnaires. Ethical review boards of all hospitals approved the study. There were different recruitment procedures between countries, eg, the questionnaires were mailed, handed over to the participants, or completed in an interview. Limited information about nonresponders was available across all centers.

Instruments

Child emotional and behavioral functioning

Parents completed the Child Behavior Checklist (CBCL) to assess child mental health and behavior.17 Adolescents completed the self-report version of the CBCL, the Youth Self-Report (YSR), designed for children between ages 11 and 18 years.18 The CBCL comprised 120 problem items and the YSR 102 problem items. This study used the internalizing, externalizing, and total problem scale of the CBCL/YSR. The internalizing scale reflects the internalizing spectrums of psychopathology of children and consists of the syndrome subscales of withdrawal, somatic complaints, and anxiety/depression. The externalizing scale represents socially unacceptable behavior and comprises the CBCL/YSR syndrome subscales of delinquent and aggressive behavior. The total problem scale represents the total score derived from the sum of all items. In addition to the internalizing and externalizing spectrum of symptomatology, it includes the subscales thought problems and attention problems. Higher scores denote more problems. Except for the UK and Finland, national normative data of the CBCL and YSR were available,19-22 with separate norms for latency-aged children (4-10 years), adolescents (11-18 years), and boys and girls. For the UK data, the US norms and clinical cutoffs as reported by Achenbach,17,18 were used. For the Finnish data, Danish norms and clinical cutoffs, as reported by Bilenberg,22 were used.

According to national norms, raw values were transformed into standardized T-scores. To discriminate between relevant problems (need for diagnostic, counseling, or treatment) and nonrelevant problems, we used the criteria T ≥ 60, proposed by Achenbach, to define C values (ie, clinically significant cases). It is estimated that approximately 10% of the children in the general population are rated as having clinically significant behavioral and emotional problems. If one includes subclinical or so-called borderline scores of abnormalities as cases, then 17% of the general population are expected to be rated as having behavioral or emotional problems.17

Depression in ill and healthy parents was assessed by the Beck's Depression Inventory (2nd ed.; BDI),23 a 21-item, self-report questionnaire measuring symptoms experienced during the last 2 weeks. The response format consists of 4 rank-ordered sentences with corresponding scores ranging from 0 to 3. Both ill and healthy parents completed the BDI. Internal consistency was satisfactory. Alpha coefficients for the mother and the father were .87 and .85, respectively.

Ill parents' health-related quality of life (HRQL) was measured by the short-form questionnaire (SF-8) of the Medical Outcomes Health Survey,24 giving an HRQL profile of 8 scores summarized into physical component (PCS) and mental component (MCS). No clinical cutoffs are available. A score of 50 is taken to be average, with higher scores indicating greater frequency or severity of symptoms. Acceptable reliability and content validity has been reported.24 Only the ill parents completed the SF-8 and only the physical component summary was used in the analysis. Internal consistency reliability was satisfactory. Alpha coefficients for mothers and fathers were .92 and .89, respectively.

Family functioning was measured by the General Functioning subscale of the Family Assessment Device (FAD-GF),25 which assesses overall healthy functioning or dysfunction of intrafamilial relationships. The scale was derived by summing items that sampled the 6 domains included in the McMaster Model of Family Functioning: problem solving, communication, roles, affective responsiveness, affective involvement, and behavioral control. Higher scores indicate greater family dysfunction. Both parents completed the FAD-GF. Internal consistency was satisfactory with alpha coefficients of .90 and .88, respectively, for the mother and the father. The translations were either previously available or produced for the study according to scientific standards.

Analysis

The data were analyzed in 3 steps. In step 1, descriptive statistics were performed for each of the measures provided by each of the participants (ill parent, healthy parent, and adolescents). For the CBCL and YSR, standardized T-scores were used. Differences between countries were analyzed with chi-square statistics or analyses of variance (ANOVA's) with posthoc comparisons controlling for multiple comparisons (Tukey HSD test). One-sample t tests or chi-square statistics were used to compare CBCL and YSR results with norm values.

In step 2, analyses of variance (ANOVA) were conducted to test for differences in T-scores on the CBCL and YSR scores as a function of child age, gender, and the gender of the ill parent, or as a function of the interaction of these 3 variables. T-tests were conducted to analyze whether differences in CBCL and YSR scores were found between children of breast cancer and nonbreast cancer patients.

In step 3, a series of multivariate hierarchical regression analyses with CBCL internalizing, externalizing, and total scores, provided by ill and healthy parent and adolescent YSR internalizing, externalizing, and total scores as dependent variables, were performed. In each regression model, countries coded as dummies were entered first to control for the associations of these variables and the dependent variable, followed by diagnosis (coded as 1 = breast cancer, 2 = all other diagnoses), disease duration, SF-8 physical functioning score of the ill parent, and BDI and FAD general functioning score of the ill or healthy parent.

RESULTS

  1. Top of page
  2. Abstract
  3. MATERIALS AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. Conflict of Interest Disclosures
  7. References

Demographic and Clinical Data

Sixteen percent of families were German, 15% Austrian, 8% Danish, 20% Finnish, 4% Swiss, and 37% English.

Seventy-nine percent of patients were female, and 21% male. There was no significant difference between disease duration of ill mothers and fathers. In total, 56% of the children were girls and 46% were below 12 years (see Table 1).

Table 1. Summaries of the Demographic Data of Parents and Children, Means of the Clinical Data of Parents, and Diagnosis
 No.AgeYears Since DiagnosisBDIFAD-GFSF-8DiagnosisMalesFemales
  1. BDI indicates Beck Depression Inventory; FAD-GF, Family Assessment Device – General Functioning; SF-8, short-form questionnaire of the Medical Outcome Survey; SD, standard deviation.

Families352     Breast cancer 211
Male patients7344.5, SD=7.42.0, SD=2.813.5, SD=9.020.5, SD=5.836.8, SD=9.9Malignant neoplasm of female genital organs 6
Female patients27744.0, SD=5.61.8, SD=2.613.2, SD=8.621.9, SD=6.443.2, SD=9.9Malign lymphoma and leukaemia2019
Male partners18245.6, SD=7.0 7.6, SD=6.922.1, SD=5.8 Malignant neoplasms of brain159
Female partners6844.3, SD=5.6 12.2, SD=9.121.1, SD=6.8 Malignant neoplasms of digestive organs98
Boys15511.1, SD=3.7    Malignant neoplasms of respiratory and intrathoracic organs84
Girls19712.1, SD=3.8    Malignant neoplasms of urinary tract34
       Melanoma35
       Other malignant neoplasms114
       Benign neoplasms4 

Differences Between Countries

The following significant differences between countries were found: gender distribution of the patients; disease duration; SF-8 PCS scores of ill parents; mean CBCL internalizing, externalizing, and total scores; CBCL C internalizing; and CBCL C total scores reported by ill parents; and mean self-reported YSR internalizing scores (data not shown). The first author can provide information about results for each country.

Child Emotional and Behavioral Problems

Parent reports for latency-aged children (Table 2)

Ill parents reported, relative to gender-specific norms, significantly more internalizing, externalizing, and total problems in their sons. Ill parents rated a significantly higher proportion of sons as having abnormal scores in internalizing (31.9%) and total problems (30.6%), as compared with 17%, respectively, in a normative population. Healthy parents reported significantly more internalizing and total problems in their sons, significantly more internalizing, externalizing, and total problems in their daughters, and rated their sons as having a significantly higher proportion of abnormal externalizing (32.7%) and total problems (27.3%). Significant gender effects were found for healthy parents CBCL reports. Healthy mothers reported significantly more total problems in their sons (M = 59.5 ± 15.8) and daughters (M = 56.4 ± 8.1) than healthy fathers (sons: M = 51.6 ± 10.5, P < .04; daughters: M = 50.4 ± 12.4, P < .02). 342:

Table 2. Child Behavior Checklist T* and C Values Evaluated by Both Ill and Healthy Parents, and T-Test and Chi-Square Tests for Differences Between Parental Scores for Latency-Aged Children and Adolescents and Those From the Norm Group
 Ill ParentsHealthy Parents
 SonsDaughtersSonsDaughters
 CBCL TCBCL CCBCL TCBCL CCBCL TCBCL CCBCL TCBCL C
  • CBCL indicates Child Behavior Checklist; M, mean; SD, standard deviation.

  • *

    T values are compared with 50 (norm).

  • C values are compared with 17%.

  • P < .05.

  • §

    P < .01.

  • P < .001.

ChildrenMSD MSD MSD MSD 
Internalizing54.311.5§31.9§51.610.525.753.611.423.652.89.526.4
Externalizing53.411.623.651.79.519.752.911.732.7§53.18.325.9
Total problems54.712.2§30.6§51.39.821.453.712.527.352.58.322.6
Adolescents            
 Internalizing53.911.1§34.853.110.7§23.153.39.621.958.111.344.8
 Externalizing50.511.418.850.09.918.749.110.612.952.411.026.9
 Total problems51.912.424.651.410.519.651.411.424.255.710.537.3
Parent reports for adolescents (Table 2)

Compared with norms, ill parents reported significantly more internalizing problems in their adolescent sons and daughters, and they rated their adolescent sons as having a significantly higher proportion of abnormal internalizing problems (34.8%). Ill fathers reported significantly more total problems (M = 56.2 ± 10.3) and internalizing problems (M = 57.6 ± 10.7) in their adolescent daughters than did ill mothers (total problems: M = 50.4 ± 10.4, P < .03; internalizing problems: M = 52.1 ± 10.6, P < .04).

Healthy parents reported significantly more internalizing and total problems in their adolescent daughters, and rated their adolescent daughters as having a significantly higher proportion of abnormal internalizing (44.8%), externalizing (26.9%), and total problems (37.3%). Healthy mothers reported significantly more internalizing problems in their adolescent daughters (M = 62.3 ± 12.3) than did healthy fathers (M = 56.1 ± 10.4, P < .04).

Adolescent self-reports (Table 3)

No significant differences were found between adolescent self-reports and normative values.

Table 3. Youth Self Report T and C Values Evaluated by the Adolescents*
 Adolescent Boys (n=61)Adolescent Girls (n=107)
 YSR TYSR CYSR TYSR C
 MSD MSD 
  • YSR indicates Youth Self Report; M, mean; SD, standard deviation.

  • *

    T values are compared with 50 (norm), and C values are compared with 17%.

Internalizing51.511.41851.511.023.6
Externalizing49.410.31850.49.817.8
Total problems50.29.819.751.210.221.5

Predictors of Child Emotional and Behavioral Problems

Age and gender differences

Healthy parents reported that adolescent daughters had significantly more internalizing problems than latency-aged daughters (t = 2.7, P = .009). With respect to the report of the healthy parents, significant effects were found in CBCL scores depending on the gender of the ill parent: more internalizing (F = 6.6, P = .011), externalizing (F = 6.0, P = .015), and total problems (F = 11.8, P = .001) were reported when the father was ill than when the mother was ill. Adolescent daughters whose father was ill (N = 23) reported significantly more internalizing (M = 56.9, SD = 8.9; t = 3.71, P = .001) and total (M = 55.3, SD = 8.9; t = 2.86, P = .001) YSR scores than the norm value of 50. Significant higher internalizing and total YSR scores were found among adolescent daughter whose fathers were ill than whose mothers were ill. No other significant differences were found.

Diagnosis

Sixty-two percent of ill parents were women with breast cancer (Table 1). When comparing CBCL scores, reported by breast cancer patients with those of all other patients, the total problem scores reported by breast cancer patients (M: 50.9 ± 11.6; N = 207 women), were significantly lower than those of all other patients (M = 56.7 ± 11.1; N = 65 male and 59 female patients, P < .0001).

Adolescents of breast cancer patients had lower YSR total problem scores (M = 49.4 ± 10.3; N = 101) than adolescents of all remaining patients (M = 53.5 ± 9.0, P < .02; N = 67). Also, the internalizing and externalizing CBCL scores reported by breast cancer patients of their latency-aged and adolescent children and the adolescent self-reported YSR scores were significantly lower in breast cancer families than the scores of all other families (data not shown). Duration of parental illness was not significantly different between the 2 groups. When analyzing the nonbreast cancer patient group separately, no significant differences were found between children of ill fathers and ill mothers for any CBCL and YSR scores.

Hierarchical regression analysis

To test which variables best predicted child problems, 12 hierarchical regression analyses with child emotional and behavioral problems (CBCL internalizing, externalizing, and total scores of ill and healthy parents and YSR internalizing, externalizing, and total scores of adolescents) as dependent variables were performed. Because differences between countries with respect to several of the dependent variables were found in the previous analyses, countries coded as dummies were entered simultaneously in the first step, diagnosis, illness duration, and SF-8 physical sum score of ill parents, and BDI and FAD of the ill and healthy parents were entered simultaneously in the second step.

CBCL scores

As shown in Table 4, country variables did not account for a large amount of variance in the dependent measures, ranging from 4% to 7%. At step 2, the independent variables added significant incremental variance in the prediction of all parent reported child problems, ranging from 14 % for internalizing scores reported by ill parents to 21 % for internalizing scores reported by healthy parents. The self-reported depression scores of both ill and healthy parents and the family functioning scores of ill parents significantly predicted all parent-reported child problems. The family functioning scores of healthy parents significantly predicted externalizing and total CBCL scores reported by healthy parents. No other significant predictors were found.

Table 4. Hierarchical Multiple Regression Analysis Predicting Child Behavior Checklist Scores of Ill and Healthy Parents
 CBCL
 Ill ParentHealthy Parent
 InternalizingExternalizingTotalInternalizingExternalizingTotal
  • CBLC indicates Child Behavior Checklist; 1, breast cancer patients; 2, non–breast cancer patients. BDI, Beck Depression Inventory; FAD-GF, Family Assessment Device – General Functioning; SF-8, short-form questionnaire of the Medical Outcome Survey.

  • *

    P < .01.

  • P < .05.

  • P < .001.

1. Country      
 R2.06*.05.06.04.04.07
2. Predictor variables      
 R2 change.14.15.16.21.14.17
 Diagnosis (B).08.11.08.07.09.09
 Duration of illness (B).06.08.07.07.06.08
 SF-8 ill parent (B)−.02.04−.01−.17.01−.07
 BDI ill parent (B).30.16*.25   
 BDI healthy parent (B)   .28.20*.27
 FAD ill parent (B).12.25.21   
 FAD healthy parent (B)   .12.21*.15
YSR scores

Country did not make significant contributions to the prediction of adolescent reported problems (Table 5). At step 2, the independent variables added significant incremental variance in the prediction of YSR externalizing and total scores. The depression scores of ill parents significantly predicted all YSR scores, and the family functioning scores of the ill parent significantly predicted YSR externalizing scores. The scores of healthy parents did not significantly predict any of the YSR scores (data not shown).

Table 5. Hierarchical Multiple Regression Analysis Predicting Youth Self Report Scores of Adolescents
 YSR
 InternalizingExternalizingTotal problems
  • YSR indicates Youth Self Report; 1,† breast cancer patients; 2, non–breast cancer patients; BDI, Beck Depression Inventory; FAD-GF, Family Assessment Device – General Functioning; SF-8, short-form questionnaire of the Medical Outcome Survey.

  • *

    P < .001.

  • P < .01.

  • P < .05.

1. Country   
 R2.07.05.07
2. Predictor variables   
 R2 change.07.12*.11
 Diagnosis (B).06.14.09
 Duration of illness (B).03.05.02
 SF-8 ill parent (B)−.01.08.03
 BDI ill parent (B).22.21.26
 FAD ill parent (B).06.20.15

DISCUSSION

  1. Top of page
  2. Abstract
  3. MATERIALS AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. Conflict of Interest Disclosures
  7. References

The first goal of this study was to examine the emotional and behavioral functioning of children and adolescents of parents diagnosed with cancer by comparing them to a normative population. That UK and Finland national normative data of the CBCL and YSR were not available is clearly a limitation of the study. However, there is a close resemblance between available Scandinavian norms,22 as well as between Danish norms and an older small scale Finnish study,26 which suggests that Finnish CBCL norms possibly are comparable to Danish norms. US norms are higher than existing European norms,22 which indicates, that the UK CBCL scores in this study may underestimate rather than overestimate the prevalence of problems.

The results show that both ill and healthy parents reported significantly more problems and a greater percentage of relevant problems in their latency-aged sons than were reported for the normative group. Healthy (but not ill) parents reported significantly more problems in their latency-aged daughters.

Our results concerning the ill parents' evaluation of the behavioral and emotional functioning of latency-aged sons are consistent with those of Visser et al,10 who used the same measures as in the present study. Our findings concerning the healthy parents' evaluation of their sons and latency-aged daughters, however, are not in line with previous findings. In the Visser study, healthy parents reported their sons to be in the normal range, and, in other studies, the functioning of latency-aged girls have been reported to be similar to that of a normative group.5,10,13 The present study includes a greater proportion of healthy mothers than previous studies that mostly have a research bias toward ill mothers. In general, mothers tend to report more problems in their children than do fathers,27,28 and a possible explanation is linked to differences in parenting contact of mothers and fathers, making mothers a more reliable source of information. Previous studies may, therefore, have underestimated the number of problems in latency-aged children as reported by healthy mothers.

Adolescent daughters in our study had, compared with norms, higher scores on internalizing problems reported by both ill and healthy parents, and they had higher scores on total problems and a greater percentage of relevant problems reported by healthy parents. That adolescent daughters of parents with cancer have a higher incidence of problems has been found by several previous studies,5,10,12 and it has been hypothesized that adolescent girls compared with boys may feel a greater responsibility for caretaking in the family and more responsible for the well-being of their ill parents.5 Concerning the adolescents self-report, we found that self-reported problems of adolescent sons and daughters were not statistically different from problems reported by normative groups.

The second goal of the study was to investigate risk factors of emotional and behavioral problems in the children. According to the healthy parents, adolescent daughters experienced more internalizing problems than latency-aged daughters, and children as well as adolescents experienced more problems when the father was ill than when the mother was ill. Also, adolescent daughters whose fathers were ill reported significantly more internalizing and total problems than the norm group and daughters whose mother was ill. Our results are not in concordance with most previous studies that have shown adolescent daughters of ill mothers to be the most distressed.5,12,29 These studies had a relatively small number of participants and used the CBCL subscales for anxiety/depression and aggression. The internalizing and externalizing scales, which are used in the present study includes a wider range of problems, and the number of ill fathers is (N = 73) is greater than in previous studies. Visser et al10 used the same measures and found, similarly, that children and adolescents had more problems when the father was ill than when the mother was ill. The results may be interpreted in relation to our finding that both the CBCL scores reported by breast cancer patients and the YSR scores of adolescents of breast cancer patients (who were, of course, only mothers) were significantly lower than the CBCL and YSR scores reported by patients and adolescents of patients with other diagnoses. The group of patients with diagnoses other than breast cancer comprised an almost equal number of fathers and mothers, and, in this group, no differences were found between problems reported by fathers and mothers and between reports of adolescents of father and mothers. It is, thus, not clear whether the particular vulnerability of children and adolescents of ill fathers were due to gender of the ill parent or type of cancer. In some studies, poorer patient prognosis, worse functional impairment, disease recurrence, and more intensive treatment has been associated with negative outcomes among children and adolescents.11 It might be that the patients with diagnosis other than breast cancer may have more of the above characteristics than the patients with breast cancer; however, because the information was unavailable, this remains speculative.

Parental depression and family dysfunction were the most important predictors of emotional and behavioral problems in children for the reports of both ill and healthy parents, even when controlling for illness related variables. For the reports of the adolescents themselves, depression reported by the ill parents predicted internalizing, externalizing, and total problems scores, and family dysfunction reported by the ill parents predicted externalizing problems, after controlling for illness related variables. In several previous studies, a link between parental depression and child outcomes has been found.6,16,30 In these studies, however, except from the study by Edwards et al,16 the same pattern has not been found using adolescent self-report of psychosocial functioning, That an association between parental depression and child outcomes was found for the reports of parents as well as the adolescents themselves indicates that parental depression in our sample influenced child functioning and that the evaluation of problems in the children by the parents was not just an artifact of parent's depression causing a tendency in the parent to report more child problems. Adolescents, however, seemed to be more vulnerable in relation to depression in the ill parent than in the healthy parent. Associations between various aspects of family functioning and child problems have also been confirmed by several previous studies.31-33 That poorer family functioning in the present study, in particular, was related to behavioral/externalizing problems and parental depression, in particular, to emotional/internalizing problems is an interesting finding, which warrants further examination.

In conclusion, children and adolescents, in particular latency-aged boys and adolescent girls, were at higher risk of psychosocial problems than norms. There was a higher risk of problems when the father was ill than when the mother was ill, but it remains unclear whether this difference was due to the different diagnoses of fathers and mothers, gender or other factors. The best predictor of internalizing problems in children and adolescent was parental depression, and the best predictor of externalizing problems in children and adolescents was family dysfunction.

This study has several strengths: a large sample in a multinational context, a relatively large number of ill fathers, and data from multiple informants. There are however also several limitations. The sample is heterogeneous with respect to disease duration and the duration of children's exposure to the parental cancer. Several medical variables are lacking (eg, cancer prognosis, stage, course, and treatment). No self-report data on behavioral and emotional problems of latency-aged children are available. Because the study was cross-sectional, it is not known whether parental depression and dysfunctional family functioning caused the child problems or vice versa.

Clinical Implications

The clinical implications of the findings need to be carefully considered given that some unavailable variables may contribute to our findings.

However, there appears to be enough evidence to suggest that depression in ill parents needs to be targeted. Support for depressed parents will provide wider benefits to families by ensuring that their children are not adversely affected. Also, interventions might need to focus on aspects of family functioning such as problem solving, communication, roles, affective responsiveness, affective involvement, and behavioral control.

The development of appropriate interventions is a challenge for the future within oncology; clinicians need to look beyond the parent patient to the wider context and the needs of the whole family to safeguard the future development of patients' children.

Conflict of Interest Disclosures

  1. Top of page
  2. Abstract
  3. MATERIALS AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. Conflict of Interest Disclosures
  7. References

The authors made no financial disclosures.

References

  1. Top of page
  2. Abstract
  3. MATERIALS AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. Conflict of Interest Disclosures
  7. References
  • 1
    Rauch PK, Muriel AC, Cassem NH. Parents With Cancer: who's looking after the children? J Clin Oncol. 2003; 21( 9 suppl): 117s-21s.
  • 2
    Worsham NL, Compas BE, Sydney EY. Children's coping with parental illness. In: Sandler W&. Handbook of children's coping: Linking theory and intervention. New York: Plenum Press; 1997: 195-214.
  • 3
    Christ GH. Healing Children's Grief. New York: Oxford University Press; 2000.
  • 4
    Romer G, Barkmann C, Schulte-Markwort M, Thomalla GA, Riedesser P. Children of somatically ill parents: A methodological review. Clin Child Psychol Psychiatry. 2002; 7: 17-38.
  • 5
    Welch AS, Wadsworth ME, Compas BE. Adjustment of children and adolescents to parental cancer-Parents' and children's perspectives. Cancer. 1996; 77: 1409-1418.
  • 6
    Hoke LA. Psychosocial adjustment in children of mothers with breast cancer. Psychooncology. 2001; 10: 361-369.
  • 7
    Siegel K, Mesagno FP, Karus D, Christ G, Banks K, Moynihan R. Psychosocial adjustment of children with a terminally ill parent. J Am Acad Child Adolesc Psychiatry. 1992; 31: 327-333.
  • 8
    Armsden GC, Lewis FM. Behavioral adjustment and self-esteem of school-age children of women with breast cancer. Oncol Nurs Forum. 1994; 21: 39-45.
  • 9
    Heiney SP, Bryant LH, Walker S, Parrish RS, Provenzano FJ, Kelly KE. Impact of parental anxiety on child emotional adjustment when a parent has cancer. Oncol Nurs Forum. 1997; 24: 655-661.
  • 10
    Visser A, Huizinga GA, Hoekstra HJ, et al. Emotional and behavioural functioning of children of a parent diagnosed with cancer: a cross-informant perspective. Psychooncology. 2005; 14: 746-758.
  • 11
    Osborn T. The psychosocial impact of parental cancer on children and adolescents: a systematic review. Psychooncology. 2007; 16: 101-126.
  • 12
    Compas BE, Worsham NL, Eppingjordan JE, et al. When mom or dad has cancer-markers of psychological distress in cancer-patients, spouses, and children. Health Psychol. 1994; 13: 507-515.
  • 13
    Howes MJ, Hoke L, Winterbottom M, Delafield D. Psychosocial effects of breast-cancer on the patients children. J Psychosoc Oncol. 1994; 12: 1-21.
  • 14
    Compas BE, Worsham NL, Ey S, Howell DC. When mom or dad has cancer: II. coping, cognitive appraisals, and psychological distress in children of cancer patients. Health Psychol. 1996; 15: 167-175.
  • 15
    Watson M, James-Roberts I, Ashley S, et al. Factors associated with emotional and behavioural problems among school age children of breast cancer patients. Br J Cancer. 2006; 94: 43-50.
  • 16
    Edwards L, Watson M, James-Roberts I, et al. Adolescent's stress responses and psychological functioning when a parent has early breast cancer. Psychooncology. 2008; 17: 1039-1047.
  • 17
    Ascenbach TM. Manual for the child behavior checklist/4-18 and 1991 profile. Burlington: University of Vermont, Department of Psychiatry. 1991.
  • 18
    Ascenbach TM. Manual for the Youth Self-Report and 1991 Profiles. Burlington: University of Vermont, 2004.
  • 19
    Dopfner M, Schmeck K, Lehmkuhl G. Handbuch: Fragebogen fur Jugendliche, Forschungsergebnisse zur deutchen Fassung des Youth Self-Report (YSR) der Child Behaviour Checklist (CBCL). 1st ed. Koln: KJFD, Arbeitsgruppe Kinder Jugend-und Familiendiagnostic; 1994.
  • 20
    Dopfner M, Schmeck K, Berner W. Elternfragebogen uber das Verhalten vin Kindern und Jugendlichen, Forschungsergebnisse zur deutchen Fassung der Child Behaviour Checklist (CBCL). 1st ed. Koln: KJDF, Arbeitsgruppe Kinder Jugend-und Familiendiagnostic; 1994.
  • 21
    Steinhausen HC, Metzke CW, Kannenberg R. Elternfragebogen uber das Verhalten von Kindern und Jugenlichen. Zurich: Psychiatrische Universitats-Poliklinik fur Kind; 1996.
  • 22
    Bilenberg N. The Child Behavior Checklist (CBCL) and related material: standardization and validation in Danish population based and clinically based samples. Acta Psychiatr Scandsuppl. 1999; 398: 2-52.
  • 23
    Beck AT, Steer RA, Brown.G.K. Manual: Beck Depression Inventory. 2nd ed. San Antonio: The Psychological Corporation; 1996.
  • 24
    Ware.J.E., Kosinski M, Dewey JE, Gandak B. How to score and interpret Single-Item Healt Status Measures: A Manual for Users of the SF-8 Health Survey. London, RI: Quality Metric Incorporated; 2001.
  • 25
    Epstein NB, Baldwin LM, Bishop DS. The McMaster family assessment device. Journal of Marital and Family Therapy. 1983; 9: 171-180.
  • 26
    Larsson B, Frisk M. Social competence and emotional/behaviour problems in 6-16 year-old Swedish school children. Eur Child Adolesc Psychiatry. 1999; 8: 24-33.
  • 27
    Duhig AM, Renk K, Epstein MK, Phares V. Interparental agreement on internalizing, externalizing, and total behavior problems: A meta-analysis. Clinical Psychology: Science and Practice. 2000; 7: 435-453.
  • 28
    Luoma I, Koivisto AM, Tamminen T. Fathers' and mothers' perceptions of their child and maternal depressive symptoms. Nord J Psychiatry. 2004; 58: 205-211.
  • 29
    Grant KE, Compas BE. Stress and anxious-depressed symptoms among adolescents-searching for mechanisms of risk. J Consult Clin Psychol. 1995; 63: 1015-1021.
  • 30
    Lewis FM, Darby EL. Adolescent adjustment and maternal breast cancer: A test of the “faucet hypothesis”. J Psychosoc Oncol. 2003; 21: 81-104.
  • 31
    Harris CA, Zakowski SG. Comparisons of distress in adolescents of cancer patients and controls. Psychooncology. 2003; 12: 173-182.
  • 32
    Huizinga GA, Visser A, van der Graaf WT, Hoekstra HJ, Hoekstra-Weebers JE. The quality of communication between parents and adolescent children in the case of parental cancer. Ann Oncol. 2005; 16: 1956-1961.
  • 33
    Schmitt F, Santalahti P, Saarelainen S, Savonlahti E, Romer G, Piha J. Cancer families with children: factors associated with family functioning-a comparative study in Finland. Psychooncology. 2008; 17: 363-372.