Psychological Problems in Children with Cancer in the Initial Period of Treatment

Maria Gerali, PhD; Marina Servitzoglou, MD, PhD; Dimitra Paikopoulou, PhD; Helen Theodosopoulou, PhD; Michael Madianos, MD, PhD; Helen Vasilatou-Kosmidis, MD, PhD


Cancer Nurs. 2011;34(4):269-276. 

In This Article


Patients/Control Subjects

The study was conducted at the Oncology Unit of the Children's Hospital "P. & A. Kyriakou" in Athens, one of the largest pediatric oncology units in Greece.

Every child with newly diagnosed cancer and treated during a 30-month period was enrolled in the study. Overall, there were 132 children with cancer: 66 boys (50%) and 66 girls (50%). The ages of our enrolled participants ranged from 1 to 15 years (mean age, 7.1 [SD, 4.2] years). All children had both their parents alive, whereas 100 (75.75%) had siblings. Their diagnoses were as follows: leukemia (46.2%), lymphoma (12.9%), nephroblastoma (3.8%), neuroblastoma (9.1%), soft-tissue sarcoma (12.1%), bone tumor (7.6%), central nervous system tumor (2.3%), nasopharyngeal carcinoma (2.3%), and germ-cell carcinoma (3.7%). Among these children, 5 were excluded from the study because they were mentally disabled, relapsed, had a bone marrow transplant, or died during the initial 6 months of treatment. Also, 6 parents did not consent to completing the questionnaires (response rate, 95.5%).

A group of 100 children with no cancer, 60 boys (60%) and 40 girls (40%), was included as control group. Their ages ranged from 1 to 15 years (mean age, 4.3 [SD, 2.3] years). These children were selected from public nurseries and all grades of elementary and high schools in a medium socioeconomic area of Athens to match patients in age and background, provided they were physically and mentally healthy, they had not been previously admitted in hospital for any serious illness, and both parents and 1 teacher consented on completing the questionnaires. The mean age in the control population is higher than that of the participating children with cancer because a higher number of the former are from elementary and high schools and relatively few from nurseries. The socioeconomic background of children in the control group was equally matched with the participants with cancer. The demographic characteristics of the populations of patients and control subjects are described in Table 1A and Table 1B.

Study Design

Our study was prospective and comparative. Prior to accrual of eligible participants in the study, we obtained approval by the Ethics Committee of the Children's Hospital "P. & A. Kyriakou." Data were collected from the parents and teachers of children with cancer during a short interview by the principal investigator, after informed consent was given. The duration of the interview was less than 20 minutes, and the questionnaires were anonymous. The same parent (father or mother, usually the mother) completed the questionnaires for each child at the different time points.

The interviews with the parents were conducted at 3 time points: (a) 1 month after the diagnosis (T1); (b) 3 months after the diagnosis, in the middle of intensive treatment (T2); and (c) 6 months after the diagnosis, at the end of intensive treatment (T3). The interviews with the teachers were conducted only at T3, when they had developed a global opinion of their students and after parents had completed the questionnaires. Also, the parents and teachers of participants in the control group completed the questionnaires only once.

The Rutter instrument for parents and teachers was used to assess the psychological condition of children with cancer.[18–20] It is designed to evaluate the kind and frequency of psychological problems with respect to the attitude and behavior of a child in a specific place.

The term psychological problem refers to a constant problem in the behavior, personal feelings, or interpersonal relationships of a child that is severe enough to disable the child and disturb his/her family, school, and/or the community. The Rutter instrument for parents consists of 31 items. The first section consists of 8 problems (eg, complains of headaches, truants from school). The second section includes 5 questions, such as speech or eating difficulties. The third section consists of 18 descriptions of behavior against which the parent is asked to indicate whether and at what degree each description applies to the child. The total score for Rutter instrument represents the sum of the responses given to each item on a 3-category scale from 0 to 2, where 0 = "does not apply," 1 = "applies somewhat," and 2 = "certainly applies," with higher ratings indicating more severe adjustment problems. Possible ratings on the scale range from 0 to 62. There are no reverse-coded items. The Rutter instrument for teachers is a 26-item questionnaire designed to evaluate behavior at school. The items are rated on a scale of 0 to 2, with higher ratings indicating more severe problems. Possible ratings on the scale range from 0 to 52.

This specific version had already been validated for the Greek population, with Cronbach α scores of .911 for the parents and .658 for the teachers.[21] The present study was validated for consistency, reliability, and validity, and the Cronbach α scores of .818 for the parents and .609 for the teachers were similar to those of the previous study. To date, the Rutter questionnaire has rarely been used by parents and teachers of children with cancer in pediatric oncology research.[22]

Statistical Analysis

Statistical analysis was performed using SPSS software (SPSS Inc, Chicago, Illinois). Nonparametric control was used for most variables, because their distribution was not normal. Differences between the Rutter scores for patients and control subjects, according to the patients' and teachers' responses, were examined using Mann-Whitney U tests. The Friedman test was used for the comparison of the participants' total scores and subscores on Rutter questionnaire at 3 different time points, and the Kruskal-Wallis test and the Wilcoxon control for pairwise comparisons between different time points. The Spearman correlation was used for the calculation of correlations between score and age. Also, multivariate regression was used to evaluate the correlation of the participants' total scores on the Rutter questionnaire with the independent variables of age, sex, and type of cancer.


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.