JBUR-3441; No. of Pages 9 burns xxx (2011) xxx–xxx
available at www.sciencedirect.com
journal homepage: www.elsevier.com/locate/burns
Impact of pediatric burn camps on participants’ self esteem and body image: An empirical study Anne Bakker a,*, Peter G.M. Van der Heijden b, Maarten J.M. Van Son c, Rens Van de Schoot b, Nancy E.E. Van Loey a a
Association of Dutch Burn Centres, Beverwijk, The Netherlands Department of Methodology and Statistics, Faculty of Social Sciences, Utrecht University, Utrecht, The Netherlands c Department of Clinical and Health Psychology, Utrecht University, Utrecht, The Netherlands b
article info
abstract
Article history:
Objective: This study focuses on possible effects of specialized summer camps on young
Accepted 10 January 2011
burn survivors’ self esteem and body image. Method: Quantitative as well as qualitative measures was used. To study possible effects, a
Keywords:
pretest–posttest comparison group design with a follow-up was employed. Self-report
Burns
questionnaires were used to measure self esteem and body image in a burn camp group
Children
(n = 83, 8–18 years) and in a comparison group of children with burns who did not attend a
Burn camp
burn camp during the course of the study (n = 90, 8–18 years). Additionally, burn camp
Intervention
participants and parents completed an evaluation form about benefits derived from burn
Body image
camp. Results: A small positive short-term effect of burn camp participation was found on the ‘satisfaction with appearance’ component of body image. Overall, participants and parents showed high appreciation of the burn camps and reported several benefits, particularly concerning meeting other young burn survivors. Conclusions: Albeit statistically modest, this is the first quantitative study to document on a significant short-term impact of burn camp on young burn survivors’ body image. Implications of this result for future research and burn camp organization were discussed, including the strengths of residential camps for young burn survivors. # 2011 Elsevier Ltd and ISBI. All rights reserved.
1.
Introduction
Originating in the US in the 1980s, specialized summer camps for children with burns have been organized for many years. Nowadays, numerous camps for pediatric burn survivors exist throughout the world [1]. Although locations and themes vary, the overall aims seem similar: to provide children with the opportunity to meet other children with burns in a warm and safe environment, to experience success, and to enhance self
esteem through challenging activities [2]. Although most burn camps do not provide structured psychological interventions, they do offer companionship and a challenging program on top of pleasure and fun, and there are believed to be or may be psychosocial benefits from attending camp. Several research initiatives to document potential psychological and social effects for participants have been performed. The outcomes from previous qualitative and quantitative studies are not uniform [1]. Qualitative reports, based on focus groups on burn camps, or evaluation forms from participants,
* Corresponding author at: Association of Dutch Burn Centres, P.O. Box 1015, 1940 EA Beverwijk, The Netherlands. Tel.: +31 251275500; fax: +31 251216059. E-mail address:
[email protected] (A. Bakker). 0305-4179/$36.00 # 2011 Elsevier Ltd and ISBI. All rights reserved. doi:10.1016/j.burns.2011.01.009 Please cite this article in press as: Bakker A, et al. Impact of pediatric burn camps on participants’ self esteem and body image: An empirical study. Burns (2011), doi:10.1016/j.burns.2011.01.009
JBUR-3441; No. of Pages 9
2
burns xxx (2011) xxx–xxx
parents, and staff completed shortly after camp, clearly indicate psychological benefits of attending camp. Positive experiences were for instance reported in self confidence or self esteem related topics, social skills, e.g. developing new relationships or working in a team, and burn scars or appearance related topics, e.g. more confidence to show scars, putting their scars in perspective, integrating scars in overall self image, and enhanced coping with a burn [2–7]. As most quantitative studies that often investigated short-term effects on self esteem in relatively small samples could not confirm these positive statements [5,8,9], there is an ongoing research challenge to document potential benefits of pediatric burn camps [5]. Gaskell [5] proposed possible explanations for quantitative studies’ failure to find significant effects from pediatric burn camps. First, she argued that some children do benefit from burn camp, but others do not, because they may be doing well already before camp. Second, the questionnaires used may have been too general to grasp the particular changes that participants, parents, and staff members observe. Third, one might question whether a weeklong experience can, in the short, have such a substantial impact on a highly internalized complex construct as self esteem. A longer-term follow-up could possibly reveal, if present, a more gradual process. Last, a randomized controlled design may be a more appropriate design to study effects, but such a design would raise ethical issues since a control group would have to wait a year for burn camp participation. The current study adopted some of the aforementioned suggestions and explored the effect of burn camp on self esteem and body image. Self esteem was selected as an outcome to enable comparison with previous research and to study potential longer-term effects. Although global self esteem in children with burns is generally reported to be comparable to norm populations or sometimes even more positive [10,11], suggestions are also made that particular sub domains of self esteem might be impaired [12]. Additionally, we selected ‘‘satisfaction with appearance’’ as a quantitative outcome measure of body image. Body image has been described as the ‘‘inside view’’ people have of their appearance, i.e. a multidimensional concept referring to a package of perceptions, feelings, responses and evaluations about the body [13]. The few available studies describe an overall satisfying body image for children with burns [14,15], but also for instance a relationship between more severe scarring and a more negative body image [15]. Several qualitative studies addressed the positive impact of burn camps on appearance related topics [3,4,7], however, no quantitative data have been reported concerning this topic. Qualitative as well as quantitative measures were employed in the current study, since previous research clearly showed merit of using both methods together. Further, our research design included measurements before, shortly after, and well after burn camp to detect potential short-term and long-term changes. Multiple statistical methods were employed to accurately study possible effects of burn camp. Last, without randomization, an intervention group and a comparison group were composed following the naturalistic line of invitation procedure for Dutch pediatric burn camps. By
means of this comprehensive research approach and the inclusion of another outcome measure besides self esteem, i.e. satisfaction with appearance that may be more susceptible to change, we attempted to grasp potential effects of burn camp participation.
2.
Method
2.1.
Burn camp
2.1.1.
Characteristics Dutch burn camps
In the Netherlands, three camps are held each year in springtime: a burn camp for children (ages 8–11), teenagers (ages 11–14), and adolescents (ages 14–18). The camps vary in length from 5 to 6 days. The Dutch burn camps are financially supported by the Foundation Child and Burn, making contribution for participants very low. All camp staffs consist of one or two coordinators, two cooking staff members, and approximately 10 volunteers. Most volunteers work in burn centres (e.g. medical doctors, physiotherapists, (liaison) nurses), have experienced a burn event themselves, or have for instance a scouting or fire-fighting background or a relative with burns. Central features of the Dutch burn camps for participants include meeting other young burn survivors, and learning from each other and staff members about coping with scars. Various activities allow participants to experience success and extend themselves. Activities such as swimming and body painting have a body image-related background. Having fun and feeling safe are considered of paramount importance.
2.1.2.
Selection procedure Dutch burn camps
The Foundation Child and Burn holds a database containing basically all Dutch burned children treated in one of the three specialized Dutch burn centres. Aftercare nurses from the burn centres, familiar with all children hospitalized in their burn centre, recommend children (8–18 years) from the database to be invited for burn camp. Selection criterion includes having difficulties with respect to the burns and/or the expectation that the child may benefit from burn camp participation. However, having problems is not a prerequisite to be invited and the selection is not based on standard criteria. Each year, approximately half of the invited children accept the invitation. Reasons for declining are diverse, e.g. interference with school or holiday, or not feeling like participating, while in many cases the reason to decline is not known to the organizers. Apart from invited children originating from the burn centres, one or two children, for instance from peripheral hospitals, apply for burn camp on their own initiative.
2.2.
Study participant recruitment and procedures
This study was part of a larger study designed to evaluate the Dutch pediatric burn camps and was conducted in the Netherlands in 2008 and partly in 2009. The Ethics Committee North-Holland, the Netherlands, a legally qualified Dutch Ethics Review Committee, approved this study. All participating children and caregivers signed an informed consent.
Please cite this article in press as: Bakker A, et al. Impact of pediatric burn camps on participants’ self esteem and body image: An empirical study. Burns (2011), doi:10.1016/j.burns.2011.01.009
JBUR-3441; No. of Pages 9 burns xxx (2011) xxx–xxx
2.2.1.
Burn camp group
Written information regarding the research project was sent by regular mail 3 weeks prior to camp to all 2008 camp participants and parents (N = 80). The mailing also included the first questionnaires and a stamped return envelope. Children were promised a gift voucher for their participation in the research project. One week after the initial mailing, brief reminders were sent to encourage children and parents to participate. In the last week before camp, families who had not returned the questionnaires yet, were contacted by telephone to make sure that families willing to participate could still return the questionnaires in time. In sum, 72 families (T0, 90%) completed the first questionnaires. Of the 72 respondents in 2008, only 25% joined a burn camp for their first time. Since we wanted to be able to examine a possible ‘‘first time’’ effect from burn camp, all 2009 new burn camp participants were invited to participate in the research project as well. Of 18 new burn camp participants in 2009 who were eligible to participate, 11 completed the first questionnaire (61%). The 83 final study participants (72 in 2008 and 11 in 2009) did not differ from the 15 non-participants (8 in 2008 and 7 in 2009) in terms of age and gender, but had more frequently attended a burn camp before (66% versus 27%, x2(df = 1) = 8.31, p < 0.01). The second (T1) and third questionnaire (T2) were sent by regular mail, respectively, 1 week and 16 weeks after the last day of camp. Brief postal reminders were sent 1 week after each mailing and resulted in a response rate of 92% at T1 and 93% at T2 compared to T0.
2.2.2.
Comparison group
All children from the Foundation Child and Burn database, who did not join a burn camp in 2008 were invited in 2008 to participate in this study’s comparison group. Seven children could not be invited for various reasons, for instance severe mental disability, and 38 families could not be traced. In sum, 278 of 323 families were sent a study invitation at the same point in time as the 2008 camp participants of the same age. Ninety families (32%) returned the first questionnaire. Of them, 47% had declined an invitation for 2008 burn camp. There were more boys among the non-participants (65%) than among the participants (52%, x2(df = 1) = 4.18, p = 0.04). There were no differences between participants and non-participants in the comparison group regarding age and number of previous burn camp attendances. For T1 and T2, the same procedure as for the burn camp study group was followed. Compared to T0, response rates for the comparison group were 88% at T1 and 85% at T2.
2.3.
Measures
2.3.1.
Burn characteristics
Parents answered questions about the date of the burn event and the etiology of the burns. Participants marked their still present scars on a drawing of a human body. Subsequently, the body was divided in 20 zones [16]. This method provides an indication of both the extent and the visibility of the current scarring. Visible scars were defined as scars on hands, face or neck. Regarding burn camp characteristics, participants endorsed all previous years they had attended a burn camp.
3
Further, children answered how many other children with burns they knew.
2.3.2.
Self esteem
A global perception of self-worth was measured with a Dutch version of the Rosenberg Self Esteem Scale (RSES), a widely used self-report instrument [17]. The RSES consists of 10 items, answered on a four-point Likert scale ranging from strongly agree to strongly disagree. Item scores were summed up to a total scale ranging from 0 to 30, where a higher score represents a higher self-worth. The RSES is developed for adolescents, but has been used in pediatric burn camp studies in children from 6 years onwards [8,9]. Internal consistency in this study was good, with Cronbach’s alpha at T0 of 0.86.
2.3.3.
Body image
The Satisfaction with Appearance Scale (SWAP) was used as an indicator of body image [18]. It is the only questionnaire available that is specifically developed for populations with disfigurements or deformities. It is originally used for adult burn survivors and has not yet been validated for younger populations. However, Pope et al. [15] have used the SWAP in a population of young burn survivors (aged 11–19 years). In their study, young burn survivors reported less satisfaction with the appearance of burned body parts as compared to non-burned body parts and mean burn size was positively associated with overall higher dissatisfaction about appearance [15]. The SWAP consists of 14 items, answered on a seven-point Likert scale ranging from strongly disagree to strongly agree. The SWAP contains four subscales: dissatisfaction with facial body parts, dissatisfaction with non-facial body parts, perceived social impact, and social discomfort. Validity and reliability of the SWAP were reported to be good, internal consistency for the total scale was high (Cronbach’s alpha 0.87) [18]. For the purpose of this study, two investigators translated the original version into Dutch, with approval of the author. The wording was adapted to make sentences shorter, i.e. ‘‘Because of changes in my appearance caused by my burn’’. . . was changed into ‘‘Because of the scars’’. After reaching consensus, the Dutch version was back-translated into English by an independent native speaker, professional translator. Final consensus was reached among all three involved. From this point, the adapted and translated version of the SWAP will be referred to as SWAP-Child-Dutch (SWAP-C-D). Principal component analysis (PCA) did not show the four factor solution that was described by Lawrence et al. [18]. However, comparable to Heinberg et al. [19] who adapted the SWAP for patients with scleroderma, a PCA confirmed the two initially hypothesized components Subjective Dissatisfaction With Appearance and Social Behavioral Impact of Burn Scars. The two factors accounted for 53% of the variance and both factors had an eigenvalue greater than 1. Internal consistencies were good (T0 Cronbach’s alpha Factor Social Behavioral Impact of Burn Scars 0.78 and T0 Cronbach’s alpha Factor Subjective Dissatisfaction With Appearance 0.87). A higher score represents more Social Behavioral Impact, respectively, more Dissatisfaction With Appearance. A squared root (sqrt) transformation was performed on the adapted SWAP-C-D subscales to correct for skewness. This transformation resulted in a normal distribution (post-transformation
Please cite this article in press as: Bakker A, et al. Impact of pediatric burn camps on participants’ self esteem and body image: An empirical study. Burns (2011), doi:10.1016/j.burns.2011.01.009
JBUR-3441; No. of Pages 9
4
burns xxx (2011) xxx–xxx
skewness of Social Behavioral Impact 0.15, min = 0, max = 5; post-transformation skewness of Dissatisfaction With Appearance 0.09, min = 0, max = 7).
2.3.4.
Evaluation burn camp
Evaluation forms, based on Gaskell’s evaluation forms [5] were used to investigate children’s and parents’ opinions about burn camp. The burn camp study group completed the evaluation forms at T1 (sent to home address, 1 week after burn camp1). Questions for children included how much they had enjoyed camp, how much coming to camp had helped them, and how much they had learned. Responses were given on Likert scales ranging from 1 (not at all) to 7 (a lot/very much). In addition, answers of parents are shown concerning their opinion about their child’s personal benefit from camp and the answers of children how coming to burn camp had helped them, and what they had learned from burn camp. Children and parents could give more than one answer on these open-ended questions, so frequencies can exceed the number of participants.
2.4.
Statistical analyses
First, regarding demographic characteristics, the camp group and comparison group were compared with independent samples t-tests (age, years postburn, number of scarred body zones) or Chi square tests (gender, etiology, presence of visible scars, previous burn camp attendance). Second, to investigate short-term and long-term differences in self esteem and body image, paired samples t-tests were performed separately for the burn camp group and the comparison group. Then, new variables were computed for short-term and long-term changes (i.e. T1–T0 and T2–T0, respectively) and multiple regressions were utilized to test whether these changes significantly differed between the burn camp group (1) and the comparison group (0). If at least a trend effect of group appeared, gender, age, number of scarred body zones, visibility of the burns, and number of previous camp attendances were entered to the model. To prevent loss of power, the variables were added separately and not simultaneously. Data were analyzed with Statistical Package for Social Sciences, version 17.0 (SPSS Inc., now IBM, Chicago, 2008). An alpha level of 0.05 was set for all statistical tests, but results were also screened for trends ( p < 0.10), because of a limited sample size and because we expected to find only small effects. Additionally, reliable change indices (RCI’s) [20] were calculated for the RSES and both subscales of the SWAP-C-D (before squared root transformation) to determine the percentage of children whose scores had substantially changed between T0 and T1. For the RSES, SWAP-C-D Social Behavioral Impact, and SWAP-C-D Dissatisfaction With Appearance, reliability coefficients (Cronbach’s alpha) of 0.87, 0.83, 0.88, and standard deviations of 5.1, 6.7, and 9.2 of the comparison group at T0 and deviations of 5.4, 6.8, and 9.6 of the comparison group at T1 were used to determine the standard error of measurement of difference scores. Im1
The comparison group did not complete this burn camp evaluation form.
provement and deterioration were defined as RCI j1.96j, meaning that a change of 6 points or more on the RSES, 8 points or more on SWAP-C-D Social Behavioral Impact, and a change of 10 points or more on SWAP-C-D Dissatisfaction With Appearance were likely to represent a real change (95% C.I.). Subsequently, linear trends were used to test whether the number of children whose scores had substantially improved, remain unchanged, or deteriorated between T0 and T1, differed between the burn camp group and the comparison group. Last, modal responses and percentages of agreement (scores 5–7 on a scale of 1–7, conform Gaskell et al. [3]) were calculated for the questions on how much children had enjoyed camp, how much coming to camp had helped them, and how much they had learned. Answers on open-ended questions on the evaluation forms were categorized according to themes described by Gaskell et al. After discussing the content of these themes and applicability for the current study, two researchers (A.B., N.v.L.) independently categorized the answers of parents on the question what they thought their child had gained from coming to camp. Inter-rater reliability was good (Cohen’s kappa 0.83). In case of disagreement between the raters, consensus was reached through discussion. Subsequently, the other open-ended questions were analyzed by one researcher (A.B.).
2.5.
Drop-out analysis
Children who only completed T0 (n = 13) did not differ from children who completed multiple measurements (n = 160) in terms of age, gender, burn characteristics, number of previous burn camp attendances, and outcome variables as measured on T0.
3.
Results
3.1.
Participants
Table 1 shows demographic, burn, and burn camp characteristics of all study participants. Children in the burn camp group had more body zones with scars, more of them had visible scars, and more of them had attended a burn camp before, compared to children in the comparison group. There were no differences with regard to gender, age, number of years postburn, and etiology of the burns.
3.2.
Quantitative data
3.2.1.
Short-term effects
Paired samples t-tests showed that SWAP-C-D Dissatisfaction With Appearance significantly decreased in the burn camp group (T0 = 2.85 1.55, T1 = 2.50 1.74, t(72) = 2.48, p = 0.02, r = 0.28), but not in the comparison group (T0 = 2.46 1.78, T1 = 2.64 1.72, t(74) = 1.31, p = 0.19). For the RSES, T1 scores did not differ from T0 scores, neither for the camp group (T0 = 23.33 4.81, T1 = 23.16 4.38, t(75) = 0.43, p = 0.67), nor for the comparison group (T0 = 23.20 5.12, T1 = 23.53 5.35, t(74) = 0.70, p = 0.49). The same applied for the scores on SWAP-C-D Social Behavioral Impact, both for the camp group
Please cite this article in press as: Bakker A, et al. Impact of pediatric burn camps on participants’ self esteem and body image: An empirical study. Burns (2011), doi:10.1016/j.burns.2011.01.009
JBUR-3441; No. of Pages 9
5
burns xxx (2011) xxx–xxx
Table 1 – Characteristics burn camp group (n = 83) and comparison group (n = 90). Variable
Burn camp group
Comparison group
Male gender (%) Age (years) Years postburn Etiology: flame/hot fluid/other (%) Number of scarred body zones (0–20) Presence of visible scars (%) No previous burn camp attendance (%)
52 12.3 (2.8) 6.2 (4.1) 36/40/24 5.0 (3.3) 58 35
52 13.1 (2.8) 6.1 (4.0) 29/47/24 3.7 (2.8) 41 76
p 0.96 0.07 0.85 0.55
