Age differences in perceived social support by paediatric haematopoietic progenitor cell transplant patients: a longitudinal study

Original Article

doi:10.1111/j.1365-2214.2007.00785.x

Age differences in perceived social support by paediatric haematopoietic progenitor cell transplant patients: a longitudinal study M. Barrera, G. S. Andrews, D. Burnes and E. Atenafu Department of Psychology, Program in Population Health Sciences, The Hospital for Sick Children, Toronto, ON, Canada Accepted for publication 17 April 2007

Abstract

Keywords adolescents, cancer, children, paediatric haematopoietic progenitor cell transplantation, social support Correspondence: Maru Barrera, PhD, Department of Psychology, The Hospital for Sick Children, 555 University Avenue, Toronto, ON, Canada, M5G1X8 E-mail: maru.barrera@ sickkids.ca

Objectives To describe longitudinally different sources of perceived social support by children and adolescents who undergo haematopoietic progenitor cell transplant (HPCT). Methods Thirty-six (20 males, 16 females) survivors of paediatric HPCT, aged 8–18 years (Mean = 11.73), were assessed pre-HPCT and 6 months, 1 year and 2 years post-HPCT. Survivors were compared with siblings (n = 22) during the last assessment. The Social Support Scale for Children (SSSC), a self-report measure, provided scores on perceived social support from parents, teachers, friends and classmates. Results Throughout the 2 years post-HPCT, perceived social support from all sources was generally higher for survivors than for population norms. Over time, both children and adolescents reported higher social support from parents than the normative values. Over time, children perceived higher support from teachers than did adolescents, whereas adolescents’ perceived social support from classmates increased but children’s decreased. Survivors and siblings did not differ significantly in their perceived social support 2 years post-BMT. Conclusions This exploratory study suggests that perceived parental support is equally important for children and adolescents but classmate and teacher social support varies with age. These developmental differences have important clinical implications for the care of these patients.

Introduction Paediatric haematopoietic progenitor cell transplant (HPCT) is a highly risky procedure, which incurs a high morbidity rate. For some children, however, this treatment offers hope for a remission from rare childhood diseases such as cancer or lifethreatening blood and metabolic disorders. The uncertainty of the treatment outcome (Patenaude & Kupst 2005) and adverse acute and long-term medical effects, including complications of the liver and gut and even death (Dreyer et al. 2002), may place survivors at a high risk for psychosocial adjustment and poor quality of life (Andrykowski 1994; Phipps et al. 1995, 1999,

© 2007 The Authors Journal compilation © 2007 Blackwell Publishing Ltd

2000, 2002; Phipps & Barclay 1996; Parsons et al. 1999; Barrera et al. 2000). The examination of factors such as social support, which may facilitate psychosocial adjustment after childhood HPCT, has only received minimal attention. This longitudinal study aims to examine perceived social support in this population. Specifically, it describes social support from different sources as perceived by children and adolescents prior to and 6 months, 1 years and 2 years post-HPCT. Social support is the individual’s perception of positive regard from others (Barrera 1986). This may include the extent to which people care about them, the way they are liked, the degree to which they are asked to join in and how they are

19

20

M. Barrera et al.

valued by others (Harter 1985). Social support also refers to the help resulting from relationships with others, which may lead to feelings of being loved, being part of a group, reassurance of self-worth and reliable alliance with others (Weiss 1974). In children newly diagnosed with cancer, high social support was associated with low anxiety and low depressive symptoms (Varni et al. 1994; Varni & Katz 1997). Yet, there are also reports indicating that the level of social support for adolescents with cancer is lower than that of healthy adolescents (Nichols 1995). It is critical to examine the perception of social support from various sources in children and adolescents who are seriously ill and undergo aggressive medical treatment, such as HPCT, for a number of reasons. First, support from some sources (parents, peers) may be more salient at certain developmental stages (adolescence vs. early childhood) than others; and second, many physical and social restrictions related to treatment may change the availability of some sources of social support for the ill child or adolescent, which in turn may have an impact on psychosocial well-being. In one study approximately 50% of adolescents with cancer identified parents as the most important people in their lives, while only 18% considered teachers as critical (Stern et al. 1993). In another study, adolescents with cancer identified mothers and friends as their key sources of emotional and social support (Ritchie 2001). For 10–16 years olds treated for cancer, the following people were described as sources of support: at least one parent, siblings, other relatives, friends and professionals (Nichols 1995). Although the amount of social support children and adolescents with cancer receive from their social network may change over time (Manne & Miller 1998), longitudinal changes have rarely been investigated. A related factor influencing perception of social support is age (Manne & Miller 1998). In one study, children and adolescents who were treated for cancer perceived similar support from friends, but children perceived more support from their mothers than adolescents did (Nichols 1995). Thus, whether children and adolescents who undergo major risky treatment procedures for cancer or other life-threatening illness differ in their perception of different sources of social support is still an open question. This longitudinal study investigated children’s and adolescents’ perceived social support from parents, teachers, friends and classmates at pre-HPCT (T1) and 6 months (T2), 1 year (T3) and 2 years (T4) post-HPCT using a standardized selfreport questionnaire that has been used successfully with children with cancer (Varni & Katz 1997). Survivors’ perceptions of social support were compared with population norms at each assessment time and with their siblings at T4. Given adolescents’ developmental need for autonomy from authority figures

(parents and teachers) and for acceptance in their peer group, we hypothesized that children will perceive greater social support from parents and teachers than adolescents. Adolescents, on the other hand, will report more social support from friends and classmates than children. We also hypothesized that compared with normative values and with their siblings, perceived social support will be higher in these survivors.

Method Participants Potential participants for this investigation were 53 children over the age of 8 years who participated in the second cohort (2001) of a longitudinal study of paediatric HPCT. Of those 53 children (21 females, 32 males), 36 participated at T1 (within 2 weeks pre-HPCT). Non-participation was due to illness or scheduling conflicts. At T2 (6 months post-HPCT), 30 children participated; six children had died. At T3 (1 year post-HPCT) 19 completed the measure, four children had died and seven were followed up at other centres and did not return the questionnaires. Finally, at T4 (2 years post-HPCT) five of the seven families who did not return the questionnaire at T3 were able to complete the questionnaire at the last assessment, making the total sample 24. Children’s age ranged from 8 to 18 years, with a mean age of 11.73 years (SD = 3.5 years). By 1 year post-HPCT most children had returned to school. Twenty-seven families had other children who met the inclusion criteria (absence of cancer or chronic illnesses, close in age to the target child, and able to speak and write English at age level). Of the 27 families 22 (seven females, 15 males) participated. Their mean age was 12.05 years (SD = 3.34 years). Siblings were within two and a half years, older and younger, than survivors with survivors being slightly older. Ninety percent of the participants came from a two-parent family. This reflects a similar distribution of the majority of families in Canada (Statistics Canada 2007) which consist of two-parent families. The mother was the primary caregiver in 100% of the cases and the majority of mothers (72%) described their ethnicity as European Canadian. Mean time since diagnosis was 20.49 months (SD = 30.06), the median was 10.60 months, ranging from 1 to 162 months. The participating children were diagnosed with acute lymphoblastic leukaemia (ALL; 31.4%), other leukaemia types (i.e. acute myelogenous leukaemia, chronic myelogenous leukaemia; 27.5%), neuroblastoma (5.9%), solid tumours (Hodgkin’s disease, non-Hodgkin’s disease, Ewing’s sarcoma; 21.6%), and haematologic disorders (Fanconi’s anaemia, thalassaemia; 13.7%). Seventy-five percent

© 2007 The Authors Journal compilation © 2007 Blackwell Publishing Ltd, Child: care, health and development, 34, 1, 19–24

Social support of children treated with haematopoietic transplantation

of the children were scheduled to have an allogeneic transplant (63% related, 37% unrelated) and 25% an autologous transplant. Children received fractionated radiation treatments of either 300 cGy or less (53.3%), as part of their conditioning for the HPCT, or 1200 cGy or more (45.0%) in children who received prophylactic radiation of the central nervous system. Medical charts confirmed a diagnosis of GVHD (acute or chronic) for 35% of the sample, 16% had no symptoms of the disease and data were inconclusive or missing for 51% of children.

Recruitment, procedure and measures Recruitment and data collection procedures have been thoroughly described elsewhere (Barrera et al. 2000, 2006). Briefly, after the hospital’s Research Ethics Board approved the study, informed parental consent and child’s consent (when appropriate) or assent were obtained from the participants by the project co-ordinator prior to HPCT. Demographic information was provided by the mother and medical information was abstracted from medical charts. At each time period, survivors completed the Social Support Scale for Children (SSSC) (Harter 1985), a 24-item self-report measure normed with elementary and middle school age children from primarily Caucasian, middle class families. The scale is divided into four subscales (six items each) to assess the degree to which children perceive parents, teachers, close friends and classmates as caring and understanding. The parent support subscale assesses children’s perception of how parents like them, care about their feelings and treat them like a person. Perceived teacher support assesses the extent children perceive their teachers as caring and encouraging them to do their best. Close friends refer to whether the child has someone to tell problems to, whereas classmate support is assessed by how well children feel liked and accepted (i.e. not made fun of) by classmates (Harter 1985). Validity was tested by children’s perceptions of their social competence and general self-worth [r = 0.28–0.49, Harter (1985)]. Internal consistency ranges from 0.72 to 0.88. Although our sample consisted of very sick children with diverse cancers or haematological diseases rather than healthy children, their demographic variables were similar to those of the normed sample. This measure has also been successfully used with children newly diagnosed with cancer (Varni & Katz 1997).

Statistical analysis Mixed linear models analysis for the total social support and for each source of support was conducted with time and age

21

group as the factors. This model ensures that all data available at each assessment time are included in the analysis. We tested the assumption that missing data were missed at random by comparing the mean scores for parent, teacher, friend and classmate support at each assessment time considering patients missing one versus none, missing two versus none and missing three versus none using parametric (t-test) and non-parametric (Wilcoxon test) statistics. None of the comparisons was significantly different, supporting the assumption of having random missing data. Child’s age at the time of the transplant was stratified into two groups: 11 years or younger and 12 years or older. To further examine the age effect, mixed contrast linear analyses were conducted on each source of support adjusting for the time effects. Subsequently, because of the small sample, non-parametric Kruskal–Wallis tests were conducted to compare the age groups (with age as a categorical variable) at each time. Chi-square and one-sample t-tests were also conducted to compare participants with the norms at each time. In addition, paired t-test comparisons between survivors and the sibling control group were performed at T4. To control for multiple comparisons, for these t-test analyses, the alpha level was held at 0.01. For all the analyses, the statistical software sas version 9.1 (SAS Institute Inc. Cary, NC, USA) was used.

Results Age group differences in perceived social support from various sources over time Perceived social support was generally high and no significant changes over time were found from any source of support. There was an interaction between age group and time for Classmate social support [F (1,35) = 4.38, P = 0.043] (see Table 1). There was an increase in Classmates social support for adolescents from pre-HPCT to 2 years post-HPCT. Classmate support was significantly higher for adolescents than for children at 2 years post-HPCT (c2 [ d.f. = 1,24) = 4.55, P = 0.03]. In contrast, there was a decrease of Classmate social support for children for the same time period. There was also a main age group effect for Teacher social support, adjusted for time [F (1,63) = 4.18, P = 0.0452]. In general, children perceived significantly more social support from Teacher than did adolescents. This finding was significant at 1 year post-HPCT [c2 (d.f. = 1,19) = 4.51, P = 0.03). Parental support was perceived consistently high by both age groups across time. A similar pattern was observed with friends support scores.

© 2007 The Authors Journal compilation © 2007 Blackwell Publishing Ltd, Child: care, health and development, 34, 1, 19–24

22

M. Barrera et al.

Table 1. Means, standard deviation and effect size of children’s and adolescents’ perceived social support from various sources over time

Source Children† Parent Teacher Friend Classmate Adolescents‡ Parent Teacher Friend Classmate

Population

Pre-HPCT

Norms

SD

Mean

3.5 3.23 3.09 2.99

0.58 0.66 0.68 0.66

3.31 3.17 3.02 3.10

0.66 0.67 0.65 0.65

(n = 11) 3.67** 3.50* 3.53* 3.39 (n = 25) 3.71*** 3.32 3.69** 3.22

SD

ES

0.29 0.39 0.54 0.51

0.39 0.51 0.72 3.36

0.25 0.61 0.34 0.52

0.88 1.35

6 months Post-HPCT

1 year Post-HPCT

2 years Post-HPCT

Siblings

Mean

Mean

Mean

Mean

(n = 9) 3.76*** 3.75** 3.43 0.70 (n = 21) 3.76*** 3.45d 3.59* 3.38

SD

ES

0.21 0.31 0.74

0.66 1.07

0.34 0.55 0.61 0.69

0.90 0.46 0.90

(n = 6) 3.78*** 3.81** 3.70g 3.36 (n = 13) 3.77*** 3.26 3.54 3.23

SD

ES

0.17 0.31 0.54 0.66

0.75 1.20 1.00

0.37 0.54 0.53 0.73

0.89

(n = 10) 3.57 3.43 3.63** 3.15 (n = 14) 3.68** 3.26 3.62** 3.54***

SD 0.32 0.55 0.36 0.49 0.48 0.41 0.39 0.33

ES

1.04

0.65 1.15 0.90

(n = 9) 3.46 3.80** 3.54* 3.43 (n = 13) 3.53 2.98 3.51 3.5

SD

ES

0.56 0.28 0.47 0.63

1.21 0.78

0.63 0.94 0.69 0.53

Means are compared with the expected population norms. *P < 0.05, **P < 0.01, ***P < 0.001, d = 0.01, g = 0.05. †Children were 8–11 years old compared with combined normed means from grades 1–5. ‡Adolescents were 12–18 years old compared with combined normed means from grades 6–8. .To control for multiple comparisons, for these t-test analyses, the alpha level was held at 0.01. ES, effect size; HPCT, haematopoietic progenitor cell transplant.

Social support from each source compared with normative values Table 1 presents the means and standard deviations for each source of support for children and adolescents at each time and for sibling controls at 2 years post-HPCT. Effect size for significant results is also presented. An effect of 0.50 (medium size) or greater corresponds to a minimum perceptible difference and thus could be considered as a clinically important difference (Norman et al. 2003).

Survivors’ parent support The mean scores for the large majority of both children and adolescents were significantly higher than the normative mean at pre-HPCT [t (11) = 3.24, P < 0.01; t (25) = 9.81, P = 0.00 respectively], at 6 months post-HPCT [t (9) = 5.37, P < 0.0001; t (21) = 7.32, P < 0.0001 respectively] and 1 year post-HPCT [t (6) = 5.54, P < 0.0001; t (13) = 5.37, P < 0.0001 respectively]. At 2 years post-HPCT, perceived Parental support continued to be higher than the norm in both groups but reached significance only for the adolescent group [t (14) = 5.54, P = 0.004]. With one exception, the effect size for both children and adolescent ranged from medium to large.

derline at pre-HPCT [t (11) = 2.78, P = 0.02] but did not reach significance at 2 years post-HPCT. For adolescents, perceived Teacher support was significantly higher than the normative age group only at 6 months post-HPCT [t (21) = 2.84, P = 0.01]. The effect size for children scores was large except for the scores at pre-HPCT, which was medium. For adolescents, none reached clinical significance.

Survivors’ friend support Compared with population norms, Friend support for children was significantly higher at 2 years post-HPCT [t (10) = 4.36, P = 0.002], but only higher at borderline levels at pre-HPCT [t (11) = 2.37, P = 0.039] and 1 year post-HPCT [t (6) = 2.52, P = 0.05]. For the adolescent group, perceived Friend support was significantly higher than the norms at pre-HPCT [t (25) = 6.50, P < 0.0001], 6 months post-HPCT [t (21) = 2.57, P = 0.018] and 2 years post-HPCT (t (14) = 3.51, p = 0.004), but not at 1 year post-HPCT. Effect size ranged from medium to large.

Survivors’ classmate support Classmate support was found to be comparable with normative values for children at each time. Classmate support for adolescents was significantly higher than the normative value with a large effect size at 2 years post-HPCT [t (14) = 4.69, P < 0.0001].

Survivors’ teacher support Children’s perceived Teacher support was significantly higher than the normative value at 6 months post-HPCT [t (8) = 5.44, P = 0.001] and 1 year post-HPCT [t (6) = 5.15, P = 0.004], bor-

Siblings’ social support Compared with the population normative means, the mean scores for the children subgroup of siblings’ were significantly

© 2007 The Authors Journal compilation © 2007 Blackwell Publishing Ltd, Child: care, health and development, 34, 1, 19–24

Social support of children treated with haematopoietic transplantation

higher than the normative value, with a large effect size, on perceived Teachers’ social support [t (9) = 6.69, P < 0.0001], and at borderline level of significance for Friend support, but with a medium effect size [t (9) = 2.53, P = 0.04].

Comparison between survivors and siblings Comparisons with siblings’ mean scores of social support did not differ significantly for each source of support or for all sources of support combined.

Discussion This longitudinal study is the first to investigate how different sources of social support are perceived by children and adolescents who undergo HPCT and how these perceptions change over time. In general, these severely ill children and adolescents perceived higher social support than the normative values, with most effect size being medium to large, suggesting positive and clinically significant levels of social support for survivors. Children and adolescents both perceived high social support from parents over time, but differed in their perception of teacher and classmate social support. Children reported greater teacher social support than adolescents, particularly so 1 year postHPCT, whereas adolescents’ perceived social support from classmates significantly improved over time and it was significantly higher than for children 2 years post-HPCT. The unique pattern of high levels of perceived parental support is likely related to the illness experience: the challenges and demands of treatment and the related social isolation from their peers prior to and during the first year post-HPCT. This finding is consistent with previous reports of the critical role parents play in the life of adolescents treated for cancer. (Stern et al. 1993; Nichols 1995; Ritchie 2001). In contrast, the age differences on perceived social support from teachers and classmates partly support the developmental hypothesis and the unique adolescent’s milestone to thrive for autonomy from authority figures and belong to their peer group. In this study, adolescents reported less support from teachers (authority figures), but perceived more support from their classmates, their peer group. Typical of adolescent development (Offer et al. 1996), the improvement in adolescents’ perceived classmate support suggests that by 2 years post-HPCT adolescents are more likely to be reintegrated into their peer group than are younger children. Moreover, with younger children, in addition to relying more on authority figures, elementary or primary schools offer the same teacher in the year for all the subjects. This provides children the opportunity to build strong and

23

supportive relationships with one teacher, which is not the case with adolescents in high school. Regarding friends, HPCT survivors in general, perceived sustained high levels of social support from friends throughout the period of convalescence. This is a reassuring finding given that support from close friends is critical for social development (Keefe & Berndt 1996). In all, the differential perception of social support in children and adolescents who underwent HPCT supports Manne and Miller’s (1998) proposition that the sources of social support vary with age. The lack of differences in perceived social support between siblings and survivors 2 years post-HPCT was somewhat surprising given previous findings with siblings (Barrera et al. 2004) and did not support the hypothesis that survivors will perceive stronger social support from parents than will siblings. This finding may reflect a normalizing effect in the family after the child who underwent HPCT recovers and re-integrates into everyday family activities. Although one would suspect that perceived social support at T1 might have been different for patients and siblings because of the necessary extra attention the ill child needs during the most severe phases of the treatment, data for siblings were only available at 2 years post-HPCT. There are some limitations with this study. The small sample size is an inherent problem when studying severely ill children with rare medical conditions and high mortality and morbidity rates, as is the case with the sample of this study. Needless to say, longitudinal research with this population is difficult to conduct. A small sample prevented examination of the potential role of clinical factors (type of diagnosis and time since diagnosis), and familial and social factors in participants’ perception of social support. Given the current literature, perception of social support may not be influenced by clinical factors, but may be influenced by familial and other social factors. Future research with this population, however, will benefit from multisite participation to ensure larger samples and the examination of related factors. Small sample size also weakened the statistical power of the findings and made them exploratory in nature. Another potential limitation is the lack of a longitudinal comparison group, which is a great project for future studies. When comparing our mean scores for each source of support at pre-HPCT, 6 months and 1 year post-HPCT, to the mean scores of children tested shortly after a cancer diagnosis as well as 6 and 9 months later (Varni & Katz 1997) we found similar results. This consistency across the two studies supports the validity of our findings. Despite these limitations, the major contributions of this longitudinal study are the age differences in perceived social support by child and adolescent survivors of HPCT, and the

© 2007 The Authors Journal compilation © 2007 Blackwell Publishing Ltd, Child: care, health and development, 34, 1, 19–24

24

M. Barrera et al.

changes encountered over time. Age differences in perceived social support suggest that, like healthy peers, HPCT survivors undergo developmental social changes. This information must be kept in mind for the clinical care and understanding of the social networks of child and adolescent survivors of HPCT. It is vital to foster access to the various sources of social support, to ensure that survivors and families are able to resume normal daily lives.

References Andrykowski, M. A. (1994) Psychosocial factors in bone marrow transplantation: a review and recommendations for research. Bone Marrow Transplant, 13, 357–375. Barrera, M. (1986) Distinctions between social support concepts, measures, and models. American Journal of Community Psychology, 14, 413–445. Barrera, M., Boyd-Pringle, L. A., Sumbler, K. & Saunders, F. (2000) Quality of life and behavioral adjustment after pediatric bone marrow transplantation. Bone Marrow Transplantation, 26, 427–435. Barrera, M., Fleming, C. F. & Khan, F. S. (2004) The role of emotional social support in the psychological adjustment of siblings of children with cancer. Child: care, health and development, 30, 103–111. Barrera, M., Gee, C., Andrews, G. S., Armstrong, C. A. & Saunders, F. E. (2006) Health related quality of life of children and adolescents prior to hematopoietic progenitor cell transplantation: diagnosis and age effects. Pediatric Blood Cancer, 47, 320–326. Dreyer, Z. E., Blatt, J. & Bleyer, A. (2002) Late effects of childhood cancer and its treatment. In: Principles and Practice of Pediatric Oncology, 4th edn (eds P. A. Pizzo & D. G. Poplack), pp. 1431–1461. Lippincott Williams & Wilkins, Philadelphia, PA, USA. Harter, S. (1985) The Social Support Scale for Children and Adolescents 8/18. University of Denver, Department of Psychology, Denver, CO, USA. Keefe, K. & Berndt, T. J. (1996) Relations of friendship quality to self-esteem in early adolescence. Journal of Early Adolescence, 16, 110–129. Manne, S. & Miller, D. (1998) Social support, social conflict and adjustment among adolescents with cancer. Journal of Pediatric Psychology, 23, 121–130. Nichols, M. L. (1995) Social support and coping in young adolescents with cancer. Pediatric Nursing, 21, 235–240. Norman, G. R., Sloan, J. A. & Wyrwich, K. W. (2003) Interpretation of changes in health-related quality of life: the remarkable universality of half a standard deviation.[see comment]. Medical Care, 41, 582–592.

Offer, D., Schonert-Reichl, K. & Boxer, A. (1996) Normal adolescent development: empirical research findings. In: Child and Adolescent Psychiatry: a Comprehensive Textbook (ed. M. Lewis), pp. 278–290. Williams & Wilkins Co, Baltimore, MD, USA. Parsons, S. K., Barlow, S. E., Levy, S. L., Supran, S. E. & Kaplan, S. H. (1999) Health-related quality of life in pediatric bone marrow transplant survivors: according to whom? International Journal of Cancer – Suppl., 12, 46–51. Patenaude, A. F. & Kupst, M. J. (2005) Psychosocial functioning in pediatric cancer. Journal of Pediatric Psychology, 30, 9–27. Phipps, S. & Barclay, D. (1996) Psychosocial consequences of pediatric bone marrow transplantation. International Journal of Pediatric Hematology/Oncology, 3, 171–182. Phipps, S., Brenner, M., Heslop, H., Krance, R., Jayawardene, D. & Mulhern, R. (1995) Psychological effects of bone marrow transplantation on children and adolescents: preliminary report of a longitudinal study. Bone Marrow Transplantation, 15, 829–835. Phipps, S., Dunavant, M., Jayawardene, D. & Srivastiva, D. K. (1999) Assessment of health-related quality of life in acute in-patient settings: use of the BASES instrument in children undergoing bone marrow transplantation. International Journal of Cancer, 83, 18–24. Phipps, S., Dunavant, M., Srivastava, D. K., Bowman, L. & Mulhern, R. K. (2000) Cognitive and academic functioning in survivors of pediatric bone marrow transplantation. Journal of Clinical Oncology, 18, 1004–1011. Phipps, S., Dunavant, M., Garvie, P. A., Lensing, S. & Rai, S. N. (2002) Acute health-related quality of life in children undergoing stem cell transplant. I. Descriptive outcomes. Bone Marrow Transplantation, 29, 425–434. Ritchie, M. A. (2001) Sources of emotional support for adolescents with cancer. Journal of Pediatric Oncology Nursing, 18, 105–110. Statistics Canada. (2007) Number of children at home (8) and census family struture (9) for the census families in private households of Canada, provinces, territories. 2006 Census of Population. Statistics Canada, catalogue no 97-553-XCB2006007. Ottawa, Canada. Stern, M., Norman, S. L. & Zevon, M. A. (1993) Adolescents with cancer: self-image and perceived social support as indexes of adaptation. Journal of Adolescent Research, 8, 124–142. Varni, J. W. & Katz, E. R. (1997) Stress, social support and negative affectivity in children with newly diagnosed cancer: a prospective transactional analysis. Psycho-Oncology, 6, 267–278. Varni, J. W., Katz, E. R., Colegrove, R. & Dolgin, M. (1994) Perceived social support and adjustment of children with newly diagnosed cancer. Journal of Developmental and Behavioral Pediatrics, 15, 20–26. Weiss, R. S. (1974) The provision of social relationships. In: Doing Unto Others (ed. Z. Rubin), pp. 17–26. Prentice Hall, Englewood Cliffs, NJ, USA.

© 2007 The Authors Journal compilation © 2007 Blackwell Publishing Ltd, Child: care, health and development, 34, 1, 19–24