Parents Perceptions of Their Children s Cancer-Related Symptoms During Treatment: A Prospective, Longitudinal Study

Transcription

1 Vol. 40 No. 5 November 2010 Journal of Pain and Symptom Management 661 Original Article Parents Perceptions of Their Children s Cancer-Related Symptoms During Treatment: A Prospective, Longitudinal Study Ulrika Pöder, PhD, Gustaf Ljungman, MD, PhD, and Louise von Essen, PhD Section of Caring Sciences (U.P.) and Section of Psychosocial Oncology and Supportive Care (U.P., L.v.E.), Department of Public Health and Caring Sciences, and Section of Paediatric Oncology (G.L.), Department of Women s and Children s Health, Uppsala University, Uppsala, Sweden Abstract Context. This article describes a comprehensive assessment of treatment-related symptoms in children aged 0e18 years undergoing cancer treatment from the perspective of their parents. Objectives. The aim was to describe which symptoms that, according to parents, cause most problems for children receiving cancer treatment and to explore whether there is a relationship between parents ratings of their children s symptoms and their own post-traumatic stress at one week (T1), two months (T2), and four months (T3) after a child s cancer diagnosis. Methods. In total, 214 parents (107 mothers and 107 fathers) of 115 children answered a modified version of the Memorial Symptom Assessment Scale 10e18 and the PTSD Checklist Civilian Version over the telephone at T1eT3. Results. According to parents, the following symptom areas cause the most problems for children undergoing cancer treatment: emotional distress, fatigue, nutrition, and pain. Pain is the most problematic area. The prevalence of most symptoms and the symptom burden decreases over time. Parents ratings of their children s symptom burden and their own emotional distress, and mothers and fathers ratings of their child s symptom burden, are associated. Parents of adolescents report a greater symptom burden for their child than the parents of the youngest children. Conclusion. The opinions of both the patient and the parent are important in pediatrics. The results of this study can be used to guide health care professionals within pediatric oncology in their discussions of cancer treatment s adverse effects with patients and families. Not only the expectations and potential interventions but also the sources of worry should be discussed. J Pain Symptom Manage 2010;40:661e670. Ó 2010 U.S. Cancer Pain Relief Committee. Published by Elsevier Inc. All rights reserved. Key Words Parents, childhood cancer, symptoms Address correspondence to: Ulrika Pöder, PhD, Department of Public Health and Caring Sciences, Uppsala Ó 2010 U.S. Cancer Pain Relief Committee Published by Elsevier Inc. All rights reserved. University, Box 564, Uppsala, Sweden. ulrika.poder@pubcare.uu.se Accepted for publication: February 17, /$ - see front matter doi: /j.jpainsymman

2 662 Pöder et al. Vol. 40 No. 5 November 2010 Introduction The prevalence and severity of children s cancer- and treatment-related symptoms have been investigated according to children s selfreports. 1e5 To the best of our knowledge, there is limited knowledge about parents perceptions about the symptom prevalence and symptom burden for their children. In one study, it was shown that the following symptoms, according to parents, are the most prevalent for children undergoing cancer treatment: mood swings (85%), fatigue (80%), and disappointment at missing activities with friends and peers (74%); 6 thesamesymptomswerealso most commonly identified as being significantly severe. Symptoms most often identified as being the most bothersome were as follows: disappointment at missing activities with friends and peers (50%); feeling worried about receiving treatment, procedures, or side effects (43%); dry mouth or lips (39%); and nausea (39%). The symptoms most commonly identified as being the most severe and bothersome were as follows: disappointment as missing activities with friends and peers (46%); feeling worried about receiving treatment, procedures, or side effects (40%); and painful, aching, or stiff bones, joints, or muscles (36%). 6 The findings indicate that, according to parents, the most common symptoms are not necessarily the most bothersome ones, and vice versa. Ratings of parents and children s psychosocial functions are often associated, 7e9 and it has been concluded that parents psychosocial functions influence their ratings of their children s psychosocial functions. 10,11 It also has been reported that perceiving one s child in distress may activate parents emotional distress, for example, post-traumatic stress symptoms (PTSS). 12e17 Other findings show that parents expectancies about a child s pain mediate between the child s expectancies about, and actual experience of, pain. 18 Previous findings from our group show that a subgroup of parents of children receiving cancer treatment (the sample in the present study), 33%, 28%, and 22%, fulfilled the symptom criteria method, indicating potential post-traumatic stress disorder (PTSD) at one week, two months, and four months after diagnosis, respectively. 19 As a step toward an increased understanding of the nature of the relationship between parents and children s psychosocial functions and children s symptom experiences, we set out to describe which symptoms, according to parents, cause the most problems for children receiving cancer treatment and to explore the relationship between parents ratings of their children s symptoms and their own PTSS. The following research questions were answered according to parents ratings at one week, two months, and four months after their child s cancer diagnosis: 1) Which are the prevalent symptoms and symptom burden for children receiving cancer treatment? and 2) Are children s symptom burdens during cancer treatment related to parents and child s characteristics, including parents ratings of their own PTSS? Methods Data were collected at the first three of seven assessments within an ongoing Swedish research project, with the overall aim to investigate the prevalence and development of PTSD among Swedish parents of children with cancer. Data presented in this report were collected at one week (T1), two months (T2), and four months (T3) after a child s cancer diagnosis. The assessment times are planned in relation to diagnosis (T1eT3) and last treatment (T4eT7). Diagnosis and last treatment are characterized by crisis, and regarding termination of treatment, many parents experience increased insecurity and anxiety. Sample Parents were included from 2002 to 2004, during 18 continuous months at each center, at four of the six Swedish pediatric oncology centers: Gothenburg, Linköping, Umeå, and Uppsala. The inclusion criteria were as follows: Swedish- and English-speaking parents (including stepparents) of children 0e18 years, diagnosed (#14 days ago) with cancer (no relapse), scheduled for chemotherapy and/or radiotherapy, and having access to a telephone. To be eligible at T2 and T3, the child should be on curative treatment. In total, 292 parents fulfilled the inclusion criteria at T1 and the eligibility criteria at T2eT3. At T1, 66 parents

3 Vol. 40 No. 5 November 2010 Parents Perceptions of Children s Cancer-Related Symptoms 663 declined participation, and at T2 and T3, 10 and 2 declined, respectively (27% of eligible parents at T1eT3). The sample consisted of 214 parents (50% mothers) of 115 children (47% girls), who participated in all assessments. Eighty-four children were represented by two parents (custodians), 24 by one, six by three, and one by four parents (Table 1). There was no difference with regard to parent or child age between eligible and noneligible parents. More parents of a child with a central nervous system (CNS) tumor (38%) than with other solid tumors (18%), lymphoma (8%), or leukemia (7%) (X 2 [3] ¼ 35.2, P < 0.01), were excluded. Most of these were excluded because, when it was possible to inform them about the study, more than 14 days had passed since diagnosis. There was no difference with regard to parent or child age or child diagnosis between parents who declined vs. accepted participation. There was no difference with regard to parent or child age or ratings of the child s symptom burden at T1 between those who participated at all assessments (the sample) vs. those who did not participate at all assessments. As more children with lymphoma had ended successful treatment at T3, fewer of their parents (67%) than parents of children with other solid tumors (80%), a CNS tumor (85%), or leukemia (91%) (X 2 [3] ¼ 13.6, P < 0.01) participated in all assessments. Instruments The Memorial Symptom Assessment Scale 10e 18. Parents answered a modified version of the Memorial Symptom Assessment Scale (MSAS) 10e18 1 over the telephone. The original MSAS was developed to be answered by adults with cancer. 20 The MSAS 10e18 1 is to be answered by children of approximately 10 years of age, consists of 30 items, and was translated into Swedish using a forward-backward procedure. 21 For this study, the questions were modified to be answered by parents according to their perceptions of their child s symptoms. In addition to the original 30 items, the instrument used in this study, in accordance with previous findings, 1 embraces questions about headache and hair loss. For each symptom, parents were asked to assess whether it had been prevalent during the last week, and if so, to rate it according to frequency, intensity, and distress. Answers were provided on Likert scales: frequency, from Table 1 Parent (n ¼ 214) and Child Characteristics at T1 a Characteristics n b % b Mothers/fathers 107/107 50/50 Age, years c < e $ Place of origin Nordic country Marital status c Married/cohabiting Single Education c #9 year compulsory Upper secondary University Work status before child s diagnosis c Full time/student 41/95 38/88 Part time 43/6 40/6 Not working d 23/6 22/6 Household income V c #32, ,300e43, ,100e53, $53, No answer 4 2 Parent of daughters/sons c 96/118 45/55 Age of the child, years c 0e e e e Sibling/s, yes c Diagnosis c Leukemia Lymphoma CNS tumor Other solid tumor Prognosis, % c 5-Year probability of survival >70% Treatment modalities at T1eT3 a Chemotherapy Chemotherapy and surgery Chemo- and radiotherapy 11 5 Chemo- and radiotherapy and surgery Days of hospitalization, in total at T1eT3 a M ¼ 53.2 SD ¼ 23.9, 7e107 days SD ¼ standard deviation. a T1 ¼ one week after the child s diagnosis (DI); T2 ¼ two months after DI; T3 ¼ four months after DI. b When there is a significant difference (P < 0.01) between mothers and fathers, figures for both genders are presented. c Statistical analyses have been performed to explore differences between perceived symptom burden and parent characteristics (for values, see text). d For example, retirement, long-term sick-leave because of doctor s order, unemployment, parent allowance.

4 664 Pöder et al. Vol. 40 No. 5 November 2010 almost never (1) to almost always (4); intensity, from slight (1) to very severe (4); and distress, from not at all (0) to very much (4). Nine symptoms, because of their nature, were not evaluated according to frequency. For the purpose of data analysis, symptoms reported as prevalent were categorized as frequent if rated as 3 (a lot) to 4, intense if rated as 3 (moderate) to 4, and distressing if rated as 3 (quite a bit) to 4. 1 For each symptom, the average, subsequently called symptom score, of the scores for frequency, intensity, and distress was calculated. 1 The instrument embraces three subscales: the Global Distress Index (GDI; Cronbach s alpha (a) at T1eT3 ¼ 0.77e0.78), the psychological symptoms subscale (PSYCH; a ¼ 0.73e0.77), and the physical symptoms subscale (PHYS; a ¼ 0.81e0.83). The total MSAS score (a ¼ 0.87e0.88) is the average of all symptom scores. 1 The symptom burden is illustrated by the mean number of prevalent symptoms; the total MSAS score; and the scores for GDI, PSYCH, and PHYS. For data analysis, symptoms reported as not present were given a value of 0 for frequency, intensity, and distress. 20 Most parents (96%e100%) provided responses about the prevalence as well as the frequency, intensity, and distress. After having answered the MSAS 10e18, parents were asked an open-ended question about whether their child, during the last week, had been bothered by any aspect/s of distress not included in the instrument. If so, the aspect/s was/were thereafter assessed according to frequency, intensity, and distress. The PTSD Checklist Civilian Version. Parents answered the PTSD Checklist Civilian Version (PCL-C) 22 over the telephone. This is a selfreport questionnaire consisting of 17 items keyed to a specific traumadin this study, the child s cancer. It was translated into Swedish using a forward-backward procedure. 21 The respondent is asked to report how much he has been bothered by each item during the last month (at T2 and T3) on a scale from not at all (1) to extremely (5). The total score ranges from 17 to 85 (at T2, T3: a ¼ 0.90, 0.92). Each item belongs to one of three subscales: re-experience (five items; a ¼ 0.77, 0.82), avoidance (seven items; a ¼ 0.76, 0.80), or hyperarousal (five items; a ¼ 0.83, 0.83). The English version of the PCL-C has shown convergent validity with the Impact of Event Scale and the Mississippi Scale 22,23 and satisfactory diagnostic effectiveness. 22,24 The PCL-C symptom criteria method, 22 that is, a score greater than or equal to 3 on at least one symptom of re-experience, three symptoms of avoidance, and two symptoms of hyperarousal, was used to identify potential PTSD. The method has shown diagnostic effectiveness when compared to the Structured Clinical Interview for DSM-IV among mothers of childhood cancer survivors. 25 Medical Data. A nurse collected medical data for the children from the medical charts. Data collected from the time around diagnosis, that is, diagnosis, localization, stage of the disease, and risk group, were used to estimate prognosis. This was done by an experienced pediatric oncologist. If the diagnosis was preliminary at T1, that is, not verified by a pathological diagnosis, the diagnosis established at T2 was used to estimate prognosis. Procedure Ethical approval was obtained from the local ethics committees at the respective faculties of medicine. Parents who met the inclusion criteria received written and oral information about the study from a nurse at the respective center (an average of 4.3 days after diagnosis). Parents were informed that participation meant answering questions about emotional support, satisfaction with care, emotional distress, and the child s symptoms. Informed consent was provided over the telephone (an average of 6.8 days after diagnosis) to the first author or a research assistant. Data were collected by one of these persons at T1 (an average of 8.5 days after diagnosis), T2 (an average of 61.0 days after diagnosis), and T3 (an average of days after diagnosis). Permission to contact the parent at T2eT3 was acquired at the end of each interview at T1eT2. The interviewers had no contact with parents besides performing these interviews. For additional information about the procedure, see Pöder et al. 19 Data Analysis All statistical analyses were performed usingspss15.0forwindows(spssinc.,chicago,il).researchquestion1wasansweredby

5 Vol. 40 No. 5 November 2010 Parents Perceptions of Children s Cancer-Related Symptoms 665 descriptive statistics; Cochrane s Q test; and one-way, within-subject analysis of variance (ANOVA) (paired t-test as post hoc). Research Question 2 was answered by Pearson correlation, paired t-test (mothers vs. fathers of the same child), one-way ANOVA, and unpaired t-test. Paired analyses include 90 pairs (88 pairs ¼ custodian parents, two pairs ¼ custodian and stepparent). If less than 20 parents were represented in one of the categories for a category variable, the variable was excluded from data analysis. A P-value of 0.01 was chosen for all analyses, two-tailed for explorative purposes, together with Bonferroni-corrected post hoc analyses when applicable. Results Symptom Burden Table 2 describes the percentage of parents reporting symptoms as prevalent, and if prevalent, as frequent, intense, and distressing at T1eT3. The prevalence ranges from 93% for less hair than usual at T2 to 2% for problems with urination at T2 and T3. At all assessments, lack of appetite, lack of energy, and pain were among the five most prevalent symptoms. At all assessments, difficulty swallowing (54%e48% of n ¼ 56e29), feeling nervous (50%e35% of n ¼ 97e43), pain (60%e43% of n ¼ 173e120), and shortness of breath (54%e38% of n ¼ 28e7) were among the five most distressing symptoms (Table 2). Table 2 Percentage of Parents (n ¼ 214) Reporting Symptoms as Prevalent and Percentage of Parents (n ¼ 4e199) Reporting a Prevalent Symptom as Frequent, Intense, and Distressing at One Week (T1), Two Months (T2), and Four (T3) Months After the Child s Diagnosis Prevalent, % (n Parents) Frequent, % a Intense, % b Distressing, % c Symptoms d T1 T2 T3 T1 T2 T3 T1 T2 T3 T1 T2 T3 Feeling drowsy 85 (181) 55 (117) 45 (97) Pain 81 (173) 61 (131) 56 (120) Feeling sad 80 (171) 56 (119) 47 (101) Lack of energy 78 (168) 57 (122) 50 (108) Lack of appetite 68 (146) 66 (141) 58 (123) Nausea 67 (143) 53 (114) 48 (102) Feeling irritable 67 (143) 56 (119) 49 (105) Worrying 62 (132) 32 (69) 27 (57) Vomiting 60 (129) 37 (80) 34 (73) Lack of concentration 52 (112) 30 (64) 25 (54) Feeling nervous 45 (97) 28 (59) 20 (43) Dry mouth 42 (90) 22 (48) 16 (35) Weight loss 40 (86) 33 (71) 22 (46) na na na Insomnia 37 (80) 28 (59) 22 (46) Headache 37 (80) 28 (59) 21 (45) Cough 37 (79) 35 (75) 22 (46) Sweating 36 (77) 17 (36) 9 (20) Change in food taste 34 (73) 51 (109) 44 (95) na na na Dizziness 34 (72) 10 (22) 11 (24) Constipation 33 (71) 25 (53) 14 (30) na na na Difficulty swallowing 26 (56) 17 (37) 14 (29) Itching 24 (51) 24 (52) 16 (35) I don t look like myself 24 (51) 45 (96) 45 (97) na na na Diarrhea 22 (48) 26 (56) 17 (37) Skin changes 19 (40) 29 (63) 24 (52) na na na Numbness/tingling in 16 (34) 9 (20) 7 (15) hands/feet Mouth sores 15 (32) 27 (58) 18 (39) na na na Shortness of breath 13 (28) 3 (7) 4 (8) Swelling in arms/legs 11 (24) 7 (15) 4 (9) na na na Problems with urination 9 (19) 2 (4) 2 (5) Less hair than usual 9 (19) 93 (199) 91 (194) na na na Hair loss 5 (11) 22 (48) 8 (18) na na na na ¼ not applicable. a Frequent ¼ answers from a lot to almost always. b Intense ¼ answers from moderate to very severe. c Distressing ¼ answers from quite a bit to very much. d Symptoms are presented in a descending order according to symptom prevalence at T1.

6 666 Pöder et al. Vol. 40 No. 5 November 2010 The prevalence of most symptoms, but not diarrhea, itching, lack of appetite, and skin changes, decreased over time (P < 0.01, Table 2). A descending pattern for symptom burden, that is, the number of symptoms and the totals for the MSAS, GDI, PSYCH, and PHYS scales at T1eT3 is shown (P < 0.001) in Table 3. Pairwise comparisons show differences between all assessments for all scales and the number of symptoms (P < 0.01). Parent and Child Characteristics There were moderate to strong associations (T1: r ¼ 0.46e0.69; T2: r ¼ 0.39e0.57; T3: r ¼ 0.60e0.77; P < 0.001) between mothers and fathers (n ¼ 90 pairs) ratings of their child s symptom burden. At T1, mothers reported a higher score than fathers for the subscale PHYS (P < 0.01). At T3, mothers reported more symptoms (P < 0.001) and a higher score for the MSAS total (P < 0.001), the GDI (P < 0.01), and the PHYS (P < 0.001) than fathers (Table 3). At T1, cohabiting parents (n ¼ 191) reported more symptoms (P < 0.01) and a higher score for the subscale PHYS (P < 0.01) compared with parents living alone (n ¼ 23). Parents (n ¼ 214) ratings of the symptom burden do not differ with regard to child sex, presence of siblings, diagnosis, or prognosis. There were differences in child age with regard to the number of prevalent symptoms at T3 (P < 0.01), MSAS total at T1 (P < 0.01) and T3 (P < 0.001), and GDI at T3 (P < 0.01). Post hoc tests revealed the following differences: at T1, parents (n ¼ 37) of adolescents reported a higher score for the MSAS total (M ¼ 1.13) compared with parents of children seven years or younger (zero to three years: n ¼ 49, M ¼ 0.77; four to seven years: n ¼ 70, M ¼ 0.80) (P < 0.01). At T3, parents of adolescents reported more symptoms (M ¼ 10.4 vs. M ¼ 6.9: P < 0.01), a higher score on the MSAS total (M ¼ 0.71 vs. M ¼ 0.34: P < 0.01), and the GDI (M ¼ 0.90 vs. M ¼ 0.44: P < 0.01) compared with parents of children aged zero to three years. Associations, weak to moderate (P < 0.001, Table 4), were shown between ratings of PTSS and the children s symptom burden (Table 4). Parents reporting potential PTSD reported Table 3 Mean Number of Prevalent Symptoms and the Means for the Total MSAS Scale and the GDI, PSYCH, and PHYS Subscales, According to Parents (n ¼ 214) Reports and Mothers (n ¼ 90) and Fathers (n ¼ 90) Reports for the Same Child, at One Week (T1), Two Months (T2), and Four (T3) Months After the Child s Diagnosis T1 T2 T3 MSAS M SD M SD M SD Symptoms Parents a Mothers b Fathers MSAS scales Total Parents a Mothers b Fathers GDI Parents a Mothers c Fathers PSYCH Parents a Mothers Fathers PHYS Parents a Mothers c b Fathers SD ¼ standard deviation. a P < 0.001, i.e., differences between symptom burden over time (one-way, within-subject ANOVA). b P < 0.001, i.e., differences between mothers and fathers of the same child (paired t-test). c P < 0.01, i.e., differences between mothers and fathers of the same child (paired t-test).

7 Vol. 40 No. 5 November 2010 Parents Perceptions of Children s Cancer-Related Symptoms 667 Table 4 Associations Between MSAS 10e18 and PCL-C Scores and the Mean Number of Prevalent Symptoms and the Means for the Total MSAS Scale and the GDI, PSYCH, and PHYS Subscales for Cases vs. Noncases of PTSD a at T2 and T3 T2 T3 Potential PTSD Potential PTSD PTSS (n ¼ 214) Case (n ¼ 61) Noncase (n ¼ 153) PTSS (n ¼ 214) Case (n ¼ 48) Noncase (n ¼ 166) MSAS r M (SD) M (SD) r M (SD) M (SD) Symptoms (5.35) b 9.81 (4.31) (5.74) b 8.12 (4.64) MSAS scales Total (0.59) b 0.56 (0.37) (0.56) b 0.45 (0.37) GDI (0.81) b 0.69 (0.56) (0.68) b 0.59 (0.54) PSYCH (0.91) b 0.64 (0.61) (0.93) b 0.51 (0.56) PHYS (0.83) c 0.78 (0.61) (0.70) c 0.67 (0.58) SD ¼ standard deviation. a Potential PTSD case ¼ fulfilling the symptom criteria for PTSD, according to the PCL-C. b P < 0.001, i.e., differences between potential PTSD case vs. noncase. c P < 0.01, i.e., differences between potential PTSD case vs. noncase. a higher symptom burden than parents who did not report potential PTSD (P < 0.01, Table 4). Discussion This report describes a comprehensive assessment of treatment-related symptoms in children aged 0e18 years undergoing cancer treatment from the perspective of their parents. This information can be used when informing parents about the potential impact of the disease and subsequent treatment on their children during the first months after a cancer diagnosis. The study is rare in its evaluation of prevalence, frequency, intensity, and distress, as perceived by parents of children during cancer treatment. The following symptom areas were shown to cause the most problems for children receiving cancer treatment: emotional distress, fatigue, nutrition, and pain. Pain stands out as the most problematic symptom area. The symptom prevalence and burden decreases over time. Parents ratings of their children s symptom burden and their own PTSS are related at all assessments. Mothers and fathers ratings of their child s symptom burden are associated. The findings support previous findings according to children s self-reports 4,5 and are reasonable considering that the first two assessments cover the induction phase for most treatments. However, some symptoms show a persistent prevalence over time, and among these, lack of appetite is the most prevalent. Answers to open-ended questions suggest that some parents need support with regard to nutritional issues. This is important to consider, as parents, according to earlier findings, perceive lack of appetite as the main cause of their child s eating problems. 26 The results also show that, as expected, because of the impact of treatment, appearance-related symptoms become more prevalent over time. Difficulty swallowing, breathing, and problems with urination are reported as among the most distressing symptoms; however, they have a low prevalence. Taken together, the findings support previous results indicating that the most prevalent symptoms are not necessarily the most distressing and vice versa, either for parents 6,27 or children. 1 The findings support previous results showing that pain is a major problem within pediatric cancer care. 1,2,6,27e32 Pain may cause emotional distress for the person in pain, as well as for close ones, and parents expectancies about a child s pain mediate between children s expectancies about and actual pain. 18 It has been shown that a child s pain may cause acute stress symptoms for the child, mediated by the parents acute stress symptoms. 33 The nature of the relationships between these phenomena remains unknown. It is, however, reasonable to assume that when a parent perceives that a child suffers, it has an impact on the parents, and consequently, on the child s psychosocial function. This is important to consider, especially as findings from this and previous studies 1e6,27,30

8 668 Pöder et al. Vol. 40 No. 5 November 2010 indicate that children on cancer treatment experience a wide range of symptoms. Parents ratings of their children s symptoms and their own PTSS are related, and parents who report potential PTSD report a greater symptom burden for their child than those who do not report potential PTSD. Post-traumatic stress includes symptoms of arousal, for example, being watchful or on guard, which may result in oversensitization to a child s symptoms. It can be speculated whether seeing one s child suffer increases the risk for a parent to develop PTSS and that these symptoms, in turn, may have an impact on parents perceptions of their children s symptoms. It has been shown that parents distress mediates between a child s symptoms and distress. 18,33,34 This may have important clinical implications for parents and children. It is suggested that staff within pediatric cancer care are educated and encouraged to help parents to cope with their children s symptoms. This may have a positive impact on parents ability to care for and support their children. Mothers and fathers ratings of their child s symptom burden were associated, and the mean value differences between their ratings did not indicate a difference of clinical importance. 35 This shows that mothers and fathers perceive their child s symptom experience in a similar way. Four months after diagnosis, parents report a greater symptom burden for adolescents than for the youngest children. This result supports previous findings. 36,37 It should be considered that adolescents express their symptoms in a different way than the youngest children, 38 and that this finding does not necessarily imply that adolescents experience a greater symptom burden than younger children. It can, however, be more difficult to witness an adolescent than a younger child suffer. Parents may consider it more difficult to care for and support an older than a younger child. A better understanding of the relationship between age and symptom burden requires reports by the patients themselves using ageappropriate, validated tools. Methodological Considerations The major strengths of this study are the large and population-based sample, including an equal number of mothers and fathers, allowing generalization of the findings to Swedish parents of children receiving cancer treatment and to parents in other countries with similar health care conditions. The longitudinal design also is a strength, allowing conclusions at different times after diagnosis. However, because of the exclusion criteria, parents of children with a CNS tumor or lymphoma are underrepresented in the sample. This and the fact that a P-value less than 0.01 was used in all data analyses, which increases the risk for Type II errors, should be considered when interpreting the findings. The data have been handled as independent, except in the analyses investigating associations and differences between ratings from mothers and fathers of the same child. The intention was not to report the children s symptoms by proxy but to describe parents perceptions of their children s symptoms; consequently, the data were considered independent. The results do not illustrate the children s experiences or the children s experiences according to their parents, but rather, the parents experiences. A sore mouth or throat, or blisters in the mouth and bleeding gums, often was mentioned by parents in response to the openended questions. However, parents answers to the questionnaire items did not indicate mouth sores as a prevalent symptom. Additionally, the findings show a low prevalence for the symptom hair loss but a high prevalence for the item having less hair than usual, an item that we added according to recommendations by Collins et al. 1 These findings indicate that it is of great importance how questions are posed and illustrate a potential problem with face validity of certain items. The answers from the open-ended questions embrace several problems not assessed by the MSAS (findings not reported in this article). This illustrates a need to add open-ended questions when using a structured instrument, especially if used in clinical care to screen for potential distress. Conclusion The opinions of both the patient and the parent are important in pediatrics, and the results of this study can be used to guide health care professionals in their discussions of

9 Vol. 40 No. 5 November 2010 Parents Perceptions of Children s Cancer-Related Symptoms 669 cancer treatment s adverse effects with patients and families within pediatric oncology. Not only expectations and potential interventions but also the sources of worry should be discussed. Disclosures and Acknowledgments The study was funded by grants from the Swedish Cancer Society, the Swedish Children s Cancer Foundation, and the Swedish Research Council. The authors gratefully acknowledge all the parents who have participated in this study and the nurses and the research assistant who have helped in different parts of the study. References 1. Collins JJ, Byrnes ME, Dunkel IJ, et al. The measurement of symptoms in children with cancer. J Pain Symptom Manage 2000;19:363e Collins JJ, Devine TD, Dick GS, et al. The measurement of symptoms in young children with cancer: the validation of the Memorial Symptom Assessment Scale in children aged J Pain Symptom Manage 2002;23:10e Hedström M, Ljungman G, von Essen L. Perceptions of distress among adolescents recently diagnosed with cancer. J Pediatr Hematol Oncol 2005; 27:15e Hinds PS, Quargnenti A, Bush AJ, et al. An evaluation of the impact of a self-care coping intervention on psychological and clinical outcomes in adolescents with newly diagnosed cancer. Eur J Oncol Nurs 2000;4:6e17; discussion 18e Hinds PS, Quargnenti AG, Wentz TJ. Measuring symptom distress in adolescents with cancer. J Pediatr Oncol Nurs 1992;9:84e Dupuis LL, Milne-Wren C, Cassidy M, et al. Symptom assessment in children receiving cancer therapy: the parents perspective. Support Care Cancer 2009;18:281e Robinson KE, Gerhardt CA, Vannatta K, Noll RB. Parent and family factors associated with child adjustment to pediatric cancer. J Pediatr Psychol 2007;32:400e Pelcovitz D, Libov BG, Mandel F, et al. Posttraumatic stress disorder and family functioning in adolescent cancer. J Trauma Stress 1998;11: 205e Phipps S, Long A, Hudson M, Rai SN. Symptoms of post-traumatic stress in children with cancer and their parents: effects of informant and time from diagnosis. Pediatr Blood Cancer 2005;45: 952e Renk K, Oliveros A, Roddenberry A, et al. The relationship between maternal and paternal psychological symptoms and ratings of adolescent functioning. J Adolesc 2007;30:467e Svavarsdottir EK. Surviving childhood cancer: parents perceptions of their child s health. J Pediatr Oncol Nurs 2005;22:80e Barakat LP, Kazak AE, Gallagher PR, Meeske K, Stuber ML. Post-traumatic stress symptoms and stressful life events predict the long-term adjustment of survivors of childhood cancer and their mothers. J Clin Psychol Med Settings 2000;7: 189e Best M, Streisand R, Catania L, Kazak AE. Parental distress during pediatric leukemia and post-traumatic stress symptoms (PTSS) after treatment ends. J Pediatr Psychol 2001;26:299e Kazak AE, Stuber ML, Barakat LP, et al. Predicting post-traumatic stress symptoms in mothers and fathers of survivors of childhood cancers. J Am Acad Child Adolesc Psychiatry 1998;37:823e Young GS, Mintzer LL, Seacord D, et al. Symptoms of post-traumatic stress disorder in parents of transplant recipients: incidence, severity, and related factors. Pediatrics 2003;111:e725ee Kazak AE, McClure KS, Alderfer MA, et al. Cancer-related parental beliefs: the Family Illness Beliefs Inventory (FIBI). J Pediatr Psychol 2004;29: 531e Steele RG, Dreyer ML, Phipps S. Patterns of maternal distress among children with cancer and their association with child emotional and somatic distress. J Pediatr Psychol 2004;29:507e Liossi C, White P, Franck L, Hatira P. Parental pain expectancy as a mediator between child expected and experienced procedure-related pain intensity during painful medical procedures. Clin J Pain 2007;23:392e Pöder U, Ljungman G, von Essen L. Posttraumatic stress disorder among parents of children on cancer treatment: a longitudinal study. Psychooncology 2008;17:430e Portenoy RK, Thaler HT, Kornblith AB, et al. The Memorial Symptom Assessment Scale: an instrument for the evaluation of symptom prevalence, characteristics and distress. Eur J Cancer 1994;30A: 1326e Cull A, Sprangers M, Bjordal K, Aaronson N, EORTC Quality of Life Study Group. Translation procedure. Brussels, Belgium: EORTC, Weathers FW, Litz BT, Herman DS, Huska JA, Keane TM. The PTSD Checklist (PCL): reliability, validity, and diagnostic utility. Proceedings of the 9th Annual Meeting of the International Society for Traumatic Stress Studies, San Antonio, Texas, October, 1993.

10 670 Pöder et al. Vol. 40 No. 5 November Ruggiero KJ, Del Ben K, Scotti JR, Rabalais AE. Psychometric properties of the PTSD Checklist- Civilian Version. J Trauma Stress 2003;16:495e Blanchard EB, Jones-Alexander J, Buckley TC, Forneris CA. Psychometric properties of the PTSD Checklist (PCL). Behav Res Ther 1996;34: 669e Manne S, Du Hamel K, Gallelli K, Sorgen K, Redd WH. Post-traumatic stress disorder among mothers of pediatric cancer survivors: diagnosis, comorbidity, and utility of the PTSD checklist as a screening instrument. J Pediatr Psychol 1998;23: 357e Skolin I, Wahlin YB, Broman DA, et al. Altered food intake and taste perception in children with cancer after start of chemotherapy: perspectives of children, parents and nurses. Support Care Cancer 2006;14:369e Williams PD, Schmideskamp J, Ridder EL, Williams AR. Symptom monitoring and dependent care during cancer treatment in children: pilot study. Cancer Nurs 2006;29:188e Ljungman G, Gordh T, Sörensen S, Kreuger A. Pain in paediatric oncology: interviews with children, adolescents and their parents. Acta Paediatr 1999;88:623e Jalmsell L, Kreicbergs U, Onelöv E, Steineck G, Henter JI. Symptoms affecting children with malignancies during the last month of life: a nationwide follow-up. Pediatrics 2006;117:1314e Gibson F, Garnett M, Richardson A, Edwards J, Sepion B. Heavy to carry: a survey of parents and healthcare professionals perceptions of cancer-related fatigue in children and young people. Cancer Nurs 2005;28:27e Drake R, Frost J, Collins JJ. The symptoms of dying children. J Pain Symptom Manage 2003;26: 594e Theunissen JM, Hoogerbrugge PM, van Achterberg T, et al. Symptoms in the palliative phase of children with cancer. Pediatr Blood Cancer 2007;49:160e Stoddard FJ, Saxe G, Ronfeldt H, et al. Acute stress symptoms in young children with burns. J Am Acad Child Adolesc Psychiatry 2006;45:87e Dahlquist LM, Pendley JS. When distraction fails: parental anxiety and children s responses to distraction during cancer procedures. J Pediatr Psychol 2005;30:623e Norman GR, Sloan JA, Wyrwich KW. Interpretation of changes in health-related quality of life: the remarkable universality of half a standard deviation. Med Care 2003;41:582e Holdsworth MT, Raisch DW, Frost J. Acute and delayed nausea and emesis control in pediatric oncology patients. Cancer 2006;106:931e Phipps S, Dunavant M, Jayawardene D, Srivastiva DK. Assessment of health-related quality of life in acute in-patient settings: use of the BASES instrument in children undergoing bone marrow transplantation. Int J Cancer Suppl 1999; 12:18e Bleyer A. Young adult oncology: the patients and their survival challenges. CA Cancer J Clin 2007;57:242e255.