Pediatr Blood Cancer 2006;47:61–70 Leaving Home After Cancer in Childhood: A Measure of Social Independence in Early Adulthood Susanne Vinkel Koch, MD, PhD, 1,2 * Anne Mette Tranberg Kejs, MSc, 3 Gerda Engholm, MSc, 3 Henrik Møller, MSc, DMSc, 4 Christoffer Johansen, MD, PhD, DMSc, 2 and Kjeld Schmiegelow, MD, DMSc 1 INTRODUCTION In Denmark, 200 new cases of cancer are diagnosed annually among children and adolescents aged 0–19 years [1]. As the probability of cure has improved over the past few decades [2,3] investigations have been expanded to include late effects, such as impaired growth and reproduction, musculoskeletal abnormalities, cardiopulmonary dysfunc- tion [4,5], neurological [6,7], neuropsychological sequelae [8,9], and the risk for second malignancies [10]. As many as 30 – 40% of childhood cancer survivors have impairments [4,11–15] that may limit their quality of life. One way of estimating the long-term psychosocial effects of cancer and its treatment is to determine psychosocial and socio- economic achievements or life goals. Earlier studies have shown diverging results according to outcome, cancer type, study design, and method. Thus, educational problems was seen in subsets of cancer survivors [16–20], fewer survivors engaged in marriage [16,17,21], while employment rates recently was found to be normal [19,22]. Thus, most cancer survivors seem to live practically normal lives, but many studies agree that especially CNS tumor or other cancer survivors, who were cranially irradiated as part of their treatment, comprises an at-risk population [16–18,20,22]. From a life course perspective, ‘leaving the parental home’ precedes other significant social events of early adult life [21]. As a major developmental milestone it indicates the transition between child- and adulthood and separation of the adolescent from the parents, which eventually leads to mutual independence, one of the major developmental tasks of adolescence [23,24]. It has been suggested that childhood cancer survivors are less socially independent and that separation from their parents is delayed [21,25–29]. However, few studies have addressed the issue [17,25,27,28], showing that 11–52% fewer cancer survivors left home compared to healthy controls [17,27,28] or that they left home at an older age (N ¼ 16; 20.6 vs. 19.2 years) [25]. In a recent study 353 survivors left home as frequently as healthy controls [30]. These studies were based on relatively small, single center samples with varying participation rates, using different control groups, and univariate statistical methods. Thus, whether cancer survivors leave home less frequently or at older ages remains inconclusive. The aim of the present study was to investigate whether childhood cancer survivors from a large population-based cohort left the parental home with the same risk as a control cohort of persons sampled from the general Danish popu- lation, as a measure of social independence in early adult life, with the a priori hypothesis that cancer survivors had a reduced risk for leaving home. Our definition of indepen- dence was restricted to social independence, which enables cancer survivors to live apart from their parents, and not Background. Previous studies on psychosocial outcomes for childhood and adolescent cancer survivors have found diverse results concerning social independence. As a measure of social independence, we investigated whether cancer survivors displayed the same patterns of leaving home as population-based control group. Procedure. We identified 1,597 patients in the Danish Cancer Register, born in 1965–1980, in whom a primary cancer was diagnosed before they reached the age of 20 in the period 1965– 1995. The patients were compared with a random sample of the general population (n ¼ 43,905) frequency matched on sex and date of birth. By linking the two cohorts to registers in Statistics Denmark, we obtained socioeconomic data for the period 1980–1997. The relative risk for leaving home was estimated with discrete-time Cox regression models. Results. The risk for leaving home of survivors of hematological malignancies and solid tumors did not differ significantly from that of the control cohort. Adjustments for possible socioeconomic confounders did not change this pattern. In contrast, survivors of central nervous system (CNS) tumors had a significantly reduced risk for leaving home, which was most pronounced for men (relative risk, men: 0.66; 95% confidence interval, 0.55–0.80; women: 0.88, 95% confidence interval, 0.80–0.97). Conclusion. Overall, the psychosocial effects of cancer in childhood or adolescence and its treatment on the survivor and family did not appear to impede social independence in early adulthood, except for survivors of CNS tumors. Pediatr Blood Cancer 2006;47:61–70. ß 2006 Wiley-Liss, Inc. Key words: adolescence; childhood cancer; cohort studies; registries; social adjustment; socioeconomic factors ß 2006 Wiley-Liss, Inc. DOI 10.1002/pbc.20827 —————— 1 Section of Pediatric Hematology and Oncology, Pediatric Clinic II, Juliane Marie Center, University Hospital, H:S Rigshospitalet, Copenhagen, Denmark; 2 Department of Psychosocial Cancer Research, Institute of Cancer Epidemiology, Danish Cancer Society, Copenhagen, Denmark; 3 Department of Cancer Prevention & Documentation, Danish Cancer Society, Copenhagen, Denmark; 4 Thames Cancer Register, Guy’s, King’s and St. Thomas’ School of Medicine, King’s College London, London, United Kingdom Grant sponsor: Psychosocial Research Foundation; Danish Cancer Society; Grant number: 98 150 05; Grant sponsor: Danish Childhood Cancer Foundation. *Correspondence to: Susanne Vinkel Koch, Department of Psychosocial Cancer Research, Institute of Cancer Epidemiology, Danish Cancer Society, Strandboulevarden 49, DK-2100 Copenhagen, Denmark. E-mail: koch@cancer.dk Received 21 May 2004; Accepted 7 February 2006