Acta Pñ diatr 91: 954± 960. 2002
Predictors of coping strategy selection in paediatric patients MA Landolt 1, M Vollrath2 and K Ribi1 University Children’s Hospital Zurich1 and University of Zurich2, Department of Psychology, Switzerland
Landolt MA, Vollrath M, Ribi K. Predictors of coping strategy selection in paediatric patients. Acta Pædiatr 2002; 91: 954–960. Stockholm. ISSN 0803-5253 Aim: To assess the prevalence of speci c coping strategies and predictors of coping strategy selection in 179 patients (mean age = 10.2 y). The children were investigated one month after the occurrence of an accident (n = 105), diagnosis of cancer (n = 26) or diagnosis of diabetes mellitus type I (n = 48). Results: Patients used a great variety of coping strategies. The most frequent strategies were cognitive avoidance, positive cognitive restructuring and avoidant actions. The strategies of seeking problem-focused support and emotion-focused support were rarely used. Diagnostic category, length of hospital stay, and gender were not associated with coping strategy use. Age, socioeconomic status and functional status of the patient were found to predict coping strategy selection. Younger children made less use of active coping, distraction and seeking support. Patients of lower socioeconomic status used religious coping strategies signi cantly more often, whereas patients with lower functional status used avoidance and support-seeking strategies more often. Conclusion: In this study it was found that paediatric patients used a wide variety of coping strategies, irrespective of diagnosis and gender. Age of the child and functional status were the most important predictors of coping strategy selection. Key words: Child, chronic disease, coping behaviour, injury Markus A Landolt, University Children’s Hospital Zurich, Steinwiesstrasse 75, CH-8032 Zurich, Switzerland (Tel. ‡41 1 266 7396, fax. ‡411 266 7171, e-mail.
[email protected] )
Ó
Coping has been de ned as cognitive and behavioural efforts to manage speci c external and/or internal demands that are appraised as taxing or exceeding the resources of the individual (1). The concept of coping is central to theory, research and clinical practice in the eld of paediatrics. Paediatric populations are faced with a variety of stressful circumstances that require coping responses, including events that contribute to illness or injury, the stress of the illness or injury itself, aversive aspects of treatment, and the correlates or consequences of paediatric conditions (2). How children and adolescents cope with their own illnesses and injuries, as well as with adverse aspects of medical treatment, can signi cantly affect the short- and longterm outcomes of medical interventions (3, 4). In sum, the evidence suggests that problem-focused coping is related to better adjustment, whereas avoidant coping and coping that focuses on the self (self-blame) are related to poorer adjustment. These relations hold across different types of chronic illness and a variety of indicators of adjustment, including both physiologi cal and psychological measures (3, 4). There are remarkably few studies of naturally occurring, spontaneous coping strategies used by ill or injured children (5–9). Comparison of these studies is dif cult because of a lack of agreement among researchers in their de nition of coping, heterogeneous 2002 Taylor & Francis. ISSN 0803-525 3
samples with regard to age and diagnosis, and different time-points of assessment (during and after hospitalization). Different approaches have been used to categorize children’s coping efforts, ranging from categorizing individual strategies (such as distraction and social support) to global dimensions (approach vs. avoidance, problem-focused vs. emotion-focused). In a sample of paediatric oncology patients, Bull and Drotar (5) found that direct action (problem solving, for example) and intrapsychic coping strategies (prayer, wishful thinking) were the most frequent, comprising over three-quarters of the strategies used. Spirito et al. (8) reported that compared to acutely ill children, chronically ill children were less likely to use the avoidance strategies of distraction and wishful thinking and the negative coping strategy of self-criticism. While little is known about the in uence of medical variables on the choice of coping strategies in ill children, the age and gender of the child are well studied and have been found to correlate with coping. Older children have been shown to use a wider repertoire of coping styles, re ecting the fact that children’s cognitive capacities become more sophisticated with age (2, 10). Emotion-focused strategies (such as wishful thinking) and cognitive strategies (problem solving, for example) increase with age (7, 8, 11). In a variety of chronic diseases, girls were found to use more
Coping in paediatri c patients
ACTA PÆDIATR 91 (2002)
955
Table 1. HICUPS scales and examples of items. Scale Active coping Cognitive decision-makin g Direct problem solving Seeking understanding Positive cognitive restructuring Distraction Distracting actions Avoidance Avoidant actions Cognitive avoidance Support seeking Problem focused support Emotion focused support
Example of item “Think about what I could do before I do something” “Do something to solve the problem” “Try to gure out why things like this happen” “Remind myself that things could be worse” “Watch TV” “Try to stay away from the things that make me feel upset” “Wait and hope that things will get better” “Try to solve the problem by talking with my mother or father” “Talk about how I am feeling with my mother or father”
Strategies in italics are composite factors. HICUPS: “How I coped under Pressure Scale”.
coping strategies and to employ emotional regulation and social support more often than boys (7, 9). Boys reported using cognitive restructuring and self-blame more often than girls (9). The aim of this study was twofold. The rst aim was to characterize the prevalence of the use of speci c coping strategies in a large group of paediatric patients with a variety of newly diagnosed acute and chronic conditions. Children were expected to use a wide variety of coping strategies. The second aim was to assess demographic and medical predictors of coping strategy selection. It was hypothesized that age and gender of the child would be associated with selection of speci c coping strategies. According to ndings by Spirito et al. (8), children with serious conditions (such as cancer) were expected to make less use of avoidance strategies than children with readily treatable conditions (such as minor physical injuries).
Patients and methods Patients Subjects were assessed in four different children’s hospitals in the German-speaking sector of Switzerland. Families were recruited consecutively and asked to participate in the study within the rst two weeks after occurrence of an accident or a diagnosis of a chronic illness if the following criteria were met: 1) hospitalization of at least 24 h; 2) new diagnosis of cancer, type 1 diabetes, or occurrence of an accident without severe head injury; 3) age between 6.5 and 14.5 y; 4) uency in German; 5) no evidence of mental retardation. These groups of children were chosen because they differed on key illness-related dimensions including severity of disease, perceived prognosis, course and
degree of interference with daily living. During a period of 24 mo all children who met the criteria for entry in the study were included. Of 239 patients who met these criteria, 60 (16 F, 44 M) did not participate. Mean age of the non-participating group was 10.60 yr (SD = 2.35). Of non-responders, 75% had an accident-related injury, 10% suffered from cancer and 15% had a diagnosis of type 1 diabetes. Comparison of participants and non-responders revealed no signi cant group differences with regard to age (t = 1.15, p = 0.25), gender (w2 = 2.77, p = 0.10), and nationality (w2 = 0.01, p = 0.94). However, there was a tendency for children with accident-related injuries to be overrepresented in the non-responder group (w2 = 5.23, p = 0.07). The main reasons for non-participation were that the study was considered too time-consuming (46.7%) or not suf ciently relevant (15%). In 20% of cases (n = 12), parents reported that their child was not willing to answer questions about his or her coping strategies. The nal sample consisted of 179 patients (response rate 75%). Families who participated were not compensated. Measures Coping. Coping was assessed by means of the “How I Coped Under Pressure Scale” (HICUPS) (12), an instrument that has been used in a variety of studies on coping in children from 8 to 16 y of age. The HICUPS is a 45-item self-report questionnaire based on a four-factor coping model including the factors of active coping, distraction, avoidance and supportseeking. Each factor is composed of several subscales (factors, subscales and item examples are presented in Table 1). Composite factor scores were computed by summing and averaging the corresponding subscales. For the present study, an authorized German translation was used and the format was adapted to an interview situation. One of the two distraction subscales (“physical release of emotions”) was dropped because its items describe vigorous physical activities (e.g. skateboard
956
MA Landolt et al.
riding or roller skating) that are not appropriate for a sample of ill or convalescent children. The scale “expressing emotions” was dropped by the original author of the HICUPS because of low reliability and theoretical considerations. Based on the authors’ clinical experiences, a subscale comprising two items on religious coping was added. Each item was rated on a 4-point Likert scale ranging from 0 (never) to 3 (frequently). Reliability coef cients in the current study were satisfactory and corresponded to those reported by Ayers et al. (12). Composite factors: Active coping a = 0.87; avoidance a = 0.72; seeking support a = 0.77. Subscales: Cognitive decision-making a = 0.68; direct problem solving a = 0.71; seeking understanding a = 0.70; positive cognitive restructuring a = 0.56; distraction a = 0.62; avoidant actions a = 0.58; cognitive avoidance a = 0.65; problem focused support a = 0.60; emotion-focused support a = 0.59; religious coping a = 0.75. Administration of the HICUPS entailed having the children generate one or several situations that had been dif cult or stressful during their illness or injury experience. The interviewers gave some examples of such situations, encompassing both physical and psychological stress, in order to stimulate the child’s memory. Functional status. Functional status with regard to physical activities of daily living was assessed one month after the accident or diagnosis by a single item. The attendant physicians were asked to rate the functional status of patients using a 3-point Likert severity scale: 0 = good functional status (no functional impairment), 1 = moderate functional status (moderate functional impairment), 2 = poor functional status (severe functional impairment). In order to increase interrater reliability, the different grades of functional impairment with regard to physical activities of daily life were de ned in the questionnaire. Since this measure of functional status has not been used before, there are no available reliability and validity data for this rating system. Socioeconomic status. Socioeconomic status (SES) was calculated by means of a 6-point score for both paternal occupation and maternal education. The lowest SES score was 2 points, the highest 12 points. The three social classes were de ned as follows: SES scores 2–5, lower class; SES scores 6–8, middle class; and SES scores 9–12, upper class. This measure was used in a prior study and was shown to be a reliable and valid indicator of SES (13). Procedure The study was approved by the research ethics committees of all hospitals involved in the study. Written informed consent was obtained from all parents. Assessment was carried out one month after
ACTA PÆDIATR 91 (2002)
the occurrence of an accident, or the diagnosis of cancer or diabetes type 1. The children were investigated in a standardized interview by trained graduate students. Duration of the interview ranged from 30 to 60 min. The great majority of these interviews were conducted at the subjects’ homes; a few were done at the hospital. Demographic and medical variables were retrieved from the patients’ records. Statistical analyses All analyses were performed with two-sided tests. In all cases, p < 0.05 was considered signi cant. For categorical comparisons, w2 tests were used. For comparisons of continuous data, Student’s t-tests or one-way analyses of variance (ANOVA) were computed. Because diagnostic groups differed on preliminary analyses with regard to functional status (cf. Table 2), this variable was controlled for comparison of coping strategies between diagnostic groups (ANCOVA). Pearson correlation coef cients were computed to evaluate the association between variables.
Results Sample characteristics Characteristics of the sample are listed in Table 2. Swiss children made up 85% of the sample; 15% of the children originated from Mediterranean countries (Italy, Spain, the former Yugoslavia). There were no differences between diagnostic groups with regard to age, gender and family situation (living with both biological parents vs. other types of families). However, patients with cancer and diabetes had longer periods of hospitalization than patients who had suffered accidents. In addition, the functional status of cancer patients at assessment was signi cantly more impaired. Cancer patients suffered from leukaemia (42.3%), lymphoma (23.1%), brain tumours (11.5%) or other solid tumours (23.1%). Children in the accident group had minor head injuries (47.6%), lower-extremity fractures (22.9%), upper-extremity fractures (15.2%), non-extremity fractures (24.8%), internal injuries (12.4%), or burns (10.5%). Thirty-nine percent of the patients had combined injuries. Mean Injury Severity Score (14) was 8.8 (SD = 7.1; median = 8.0). Five injured children (4.8%) were still hospitalized at the time of assessment, one month after the accident. Association of coping strategies with demographic and illness-related variables As Table 3 shows, age of the patients was signi cantly associated with the use of active coping, distraction, and seeking support. These strategies were more often used by older children. Notably, avoidance and religious coping were not related to the age of the child. No signi cant differences were found between boys and
Coping in paediatri c patients
ACTA PÆDIATR 91 (2002)
957
Table 2. Characteristic s of the sample. All N Gender Female Male Age (y) Mean SD Socioeconomi c status Lower Middle Upper Lives with both biological parents Length of hospitalization (d) Mean SD Functional status Good Moderate Bad Mean SD
Accidents
Diabetes
179
105
48
26
69 110
40 65
18 30
11 15
10.0 2.3
10.5 2.4
10.5 1.9
10 56 31 84
5 31 11 37
0 20 5 21
10.2 2.3 15 107 47 142
w2
Cancer
11.5 8.7
8.5a 7.9
15.0b 7.1
17.5b 9.5
94 64 9 0.49 0.60
50 40 4 0.51a 0.58
42 5 0 0.11b 0.31
2 19 5 1.12c 0.52
F
0.19
p-value
0.91 1.11
0.33
5.71
0.22
1.11
0.57 19.75
