Efficacy of a burn-specific cognitive-behavioral group training

burns 41 (2015) 308–316

Available online at www.sciencedirect.com

ScienceDirect journal homepage: www.elsevier.com/locate/burns

Efficacy of a burn-specific cognitive-behavioral group training Annika Seehausen a, Sabine Ripper b, Gu¨nter Germann b, Bernd Hartmann a, Gerhard Wind b, Babette Renneberg c,* a

Department of Hand, Plastic and Reconstructive Surgery, Burn Centre, BG Trauma Centre, Berlin, Germany Department of Hand, Plastic and Reconstructive Surgery, Burn Centre, BG Trauma Centre, Ludwigshafen, Germany c Department of Clinical Psychology and Psychotherapy, Freie Universita¨t Berlin, Berlin, Germany b

article info

abstract

Article history:

Objective: The aim of the present study was to evaluate the efficacy of a newly developed

Accepted 5 July 2014

cognitive-behavioral group training, specifically designed for burn patients. Method: In a multicenter-study data pre- and post treatment and at 6-month follow-up were

Keywords:

obtained from participants of the group program (Intervention group, IG; n = 86) and a

Burn injury

control group who received treatment as usual (TAU; n = 128). Outcome variables of psy-

Group intervention

chological distress, resources and health-related quality of life of both groups were com-

Psychological distress

pared using linear mixed models.

Quality of life

Results: Up to 6 months after group treatment, the IG reported a substantial decline of general symptom severity as well as posttraumatic stress, whereas the TAU group showed no significant change over time. Optimism increased in the IG after group treatment, but not in the TAU group. Regarding overall quality of life both groups showed a gradual improvement over the three assessment points. Conclusion: he newly developed burn-specific cognitive-behavioral group intervention had positive effects on psychological well-being and resources of burn participants. As a consequence, the group intervention has been implemented as inherent part of the regular burn treatment in two rehabilitation centers in Germany. # 2014 Elsevier Ltd and ISBI. All rights reserved.

1.

Introduction

Due to the improvement of intensive care procedures for severely burned patients the probability of survival has increased enormously in the last decades [1]. This development has led to the question of long-term health-related quality of life (HRQoL) of surviving burn patients. The extent of patients’ HRQoL – including not only physical functioning but also mental health and social participation – is a broadly accepted construct in research and clinical practice to * Corresponding author. Tel.: +49 30 838 561 88. E-mail address: [email protected] (B. Renneberg). http://dx.doi.org/10.1016/j.burns.2014.07.006 0305-4179/# 2014 Elsevier Ltd and ISBI. All rights reserved.

evaluate overall rehabilitation outcomes [2]. Recent studies showed that burn patients report high levels of psychosocial resources such as social support, optimism, and self-efficacy [3,4]. However, coping with the consequences of a burn injury, such as pain, functional limitations, aesthetic changes, and intruding memories of the accident is psychologically distressing for many patients. Especially the time after discharge is a challenging phase, because patients have to deal with the physical disabilities and the organization of medical aftercare in their daily lives [5]. Outside the hospital, many burn victims fear negative reactions from the social environment regarding

burns 41 (2015) 308–316

their injured body and therefore withdraw socially [6]. Compared to the general population, increased prevalence rates of post-burn mental disorders are reported among burn patients [7–9]. Rates vary due to various methodical aspects, such as different diagnostic inventories or assessment time points. Studies using structured diagnostic interviews such as Structured Clinical Interview for DSM-IV Axis I Disorders (SCID-I) [10] or Composite International Diagnostic Interview for DSM-IV (CIDI) [11] report particularly high rates of alcohol dependency [7], depressive disorders [7–9], and anxiety disorders, notably posttraumatic stress disorder (PTSD) [7–9]. In a six-month follow-up more than half of the total burn sample (55%) met the criteria for at least one mental disorder according to the SCID interview. Of those with no prior Axis I disorder, 37% developed a mental disorder after the burn injury [7]. Furthermore, pre-existing psychological impairment is often reported for burn samples [12,13] and is proofed to have an impact on perceived health after burn [14]. SCID interviews with burn patients revealed that about 60% of the patients met the criteria for at least one Axis I disorder in pre-burn history [12,13]. Prevalence rates of mental disorders among burn patients indicate that a significant proportion of burn patients would probably benefit from psychological care after discharge. In contrast, only a small percentage of patients receive post-burn psychological care. In a study by Wisely and Tarrier [15], only 6% of burn patients consulted a psychologist or psychiatrist after discharge from acute care. In 6-months post-burn SCID interviews, 27% of the interviewed burn patients reported unequivocal treatment need and 32% indicated probable need for psychological or psychiatric care [16]. However, less than half of patients with unequivocal need received psychiatric care (crisis intervention by a psychiatric nurse, out- or inpatient care in psychiatric unit) and no one in the complete sample received regular psychotherapy after discharge. The gap between the need for psychological aftercare and the actual psychological consultations can partly be explained by the fact that only a minority of burn patients received a psychological screening during medical acute- or aftercare. Thus, the need for psychological assistance might be overlooked [17].Various studies showed that psychological distress, e.g., depressive symptoms and posttraumatic stress are significant predictors for poor long-term HRQoL of burn patients [18–20]. Psychological interventions to support adjustment after injury seem to be promising for improving long-term HRQoL of severely burned patients [19]. Although many studies come to the same conclusion, the offer of specific psychological treatment for burn patients is scarce [16,21]. In Europe, no evaluated burn-specific psychological intervention for adults exists so far. To contribute to the improvement of psychological care for burn victims after discharge, our research group has developed and evaluated a burn-specific cognitive-behavioral group training.

1.1. Development, aims and contents of the group training In a prospective multicenter-study, quality of life and longterm outcome after burn injury were assessed at 5 time points up to five years after the injury in 382 burn patients [18,22]. In addition, qualitative interviews with burn patients and their

309

families were conducted concerning long-term consequences of a burn injury [6]. Based on clinical experience, the results of the longitudinal study, and the information received from the interviews, the training manual specially tailored to the needs of burn survivors was developed [23,24]. The aim of the group treatment is to support participants in coping with the long-term consequences of burn injuries and to prevent psychological distress and social avoidance behavior. The resource-oriented group intervention does not contain trauma-focused approaches and it does not intend to replace individual psychological psychotherapy. Rather, the program aims to reduce the threshold to seek further individual psychological treatment if needed. During acute burn care, many inpatients are unable to leave the bed or need to be temporarily isolated because of the danger posed by multi-resistant germs. Furthermore, the psychological processing of the experienced trauma often sets in later as a delayed response to the accident [25]. Therefore, the time immediately after discharge is a challenging phase, in which many patients would benefit from additional professional support [6,15]. With these aspects in mind, the burnspecific group training described below was designed for all burn patients in the rehabilitation phase or in outpatient aftercare. The program consists of eight sessions with 4–8 participants. Groups were led by one or two psychologists. In addition to psycho-education about possible psychological stress reactions, development of scars, and consequences of social withdrawal, group members were encouraged to share their own experiences on these issues. In role-plays, participants practiced different responses to deal with negative reactions of the social environment on their scars or disfigurement. The program combines sections of social skills training that have been shown to be effective for people with disfigurements [26], programs for prevention of depression [27] and stress management [28]. The composition and adaptation of these cognitive-behavioral techniques were specifically tailored to the reported needs of burn victims. An overview of the program’s contents is provided in Table 1. To evaluate its efficacy, the group intervention was offered to patients at three burn units and two rehabilitation centers in Germany. This paper presents results of the efficacy of the group training.

2.

Methods

2.1.

Recruitment and study design

The study was approved by the Ethics Committee of the State Chamber of Medicine in Rheinland-Pfalz (Germany). The program was conducted at five burn centers in Germany. In the acute care Burn Units in Berlin, Ludwigshafen and Hamburg, the program was offered to outpatients after the end of their acute hospital stay. In the rehabilitation centers in Bad Klosterlausnitz and Passauer Wolf, the program was offered to burn patients who were treated in an inpatient rehabilitation center after acute care. Group leaders were psychologists or psychotherapists who were trained in

310

burns 41 (2015) 308–316

Table 1 – Contents of the group intervention. Session 1

2

3

4

5

6

7

8

Theme Introduction Introduction of the participants and the organizational processes Information on skin and scars Introduction to the relaxation-technique of progressive muscle relaxation (PMR) Expert question and answer session and pain perception Q & A session with an expert about development of skin and scars (with physiotherapist, doctor or nurse of the clinic) Behavioral experiment for pain perception Psychoeducation about dealing with pain PMR Mental stress Psychoeducation about psychological consequences of a burn injury Information on self-help groups and how to gain psychological support PMR Dealing with altered body Sharing problems with disfigurement Development of positive cognitions to scars PMR Coping with stress Psychoeducation about stressors and stress amplifiers Development of individual warning signals and coping strategies for stress PMR Dealing with reactions of others Exchange experiences of reactions of the social environment Role playing for different ways of reacting to unpleasant views or comments PMR Satisfactory social contacts Psychoeducation about social avoidance behavior Development of strategies to signal interpersonal needs Role playing for dealing with questions about the accident PMR Concluding session Clarifying of unanswered questions Reflection of the group program PMR

delivering the group treatment and received supervision on a regular basis. Eligible patients were invited to participate by study assistants or clinical psychologists of the co-operating burnunits. Criteria for inclusion were (a) inpatient or outpatient medical treatment due to burn injury, (b) age between 18 and 65 years, (c) sufficient knowledge of the German language and (d) no acute psychosis, advanced dementia or other severe cognitive impairment. Group assignment followed a sequential design. First, participants of the control group were recruited at the study sites, then participants of the intervention group. The control group received treatment as usual (TAU). TAU consisted of common burn injury aftercare such as physical therapy and scar treatment. Patients of the intervention group (IG) took part in the burn-specific group intervention in addition to TAU. All participating burn patients signed a written informed consent. Participants of the IG completed

questionnaires before and after group intervention as well as at six-month follow-up. Participants of the TAU group completed the questionnaires immediately after signing the consent form, eight weeks later, and again six months after the second assessment. The pre-assessment questionnaires were personally handed out to both groups. Post-assessment questionnaires were given to the IG after the last group session. The TAU group received them by postal mail. The follow-up questionnaires were sent to both groups by postal mail.

2.2.

Outcome measures

In addition to socio-demographic and accident-related questions, German versions of the following validated self-report questionnaires were applied at all three assessment points.

2.2.1.

Psychological distress and quality of life

To assess subjective impairment regarding physical and psychological symptoms, we used the SCL-K-9, a short version of the Symptom Checklist 90-R [29]. The strain due to posttraumatic symptoms was assessed with the PTSD Symptom Scale (PSS) [30,31] which considers the frequency and severity of the three PTSD symptom clusters: intrusion, avoidance, and arousal. Its total severity score is calculated by the sum of severity ratings of the three symptom clusters. Additionally, participants completed the Hospital Anxiety and Depression Scale (HADS) [32] to assess degree of anxiety and depression of adults with physical injuries or diseases. The short version of the WHOQOL questionnaire (WHOQOL-Bref) [33] was applied to assess patients’ perceived quality of life concerning the domains ‘‘physical health’’, ‘‘psychological health’’, ‘‘social relationship’’, and ‘‘environment’’.

2.2.2.

Psychosocial resources

Perceived or anticipated support of a patient’s social surroundings was assessed with the Social Support Questionnaire–short version (F-SozU-K-14) [34]. To assess participants’ optimism, the Life Orientation Test (LOT) [35] was applied. The LOT defines optimism as a general tendency to react to stressful situations with positive outcome expectancies and problem-focused coping strategies.

2.2.3.

Burn severity

Burn severity is measured by the percentage of body surface affected (%TBSA) and the Abbreviated Burn Severity Index (ABSI score) [36]. Information about these medical parameters was obtained from patients’ files.

2.3.

Therapeutic adherence and patient satisfaction

To evaluate the therapeutic adherence of the different group leaders, all sessions were video-taped. Using an especially designed coding scheme, two raters independently assessed the adherence of therapists in the different sessions. The interrater-reliability was high (87%). The average therapeutic adherence was 65%, indicating good adherence [37]. Regarding content, sequence, intervention techniques, work material and time management, the group leaders worked adherent to the manual.

burns 41 (2015) 308–316

After completion of the group program, participants rated how helpful and interesting they thought the individual sessions to be. The overall patient satisfaction with the group treatment was high. 89.3% of the participants regarded the program as ‘‘good or very good’’, 80.4% considered it to be ‘‘very helpful or helpful’’ and 98.2% evaluated the program to be ‘‘very interesting or interesting’’ [3].

2.4.

pooled standard deviations [40]. For multidimensional measures alpha levels were Bonferroni adjusted. Bias due to non-pre-test-equivalence of posttraumatic symptoms in both groups (see Table 3), was controlled by conducting additional repeated measure ANCOVA based on completer sample with PSS pre-test results and ABSI as covariates. Effect sizes (h2) were calculated for between subject effects.

Statistical analyses

Intention to treat (ITT) analyses were used to calculate outcome using separate linear mixed models for each outcome measure. Linear mixed models are appropriate to analyse repeated measures data with many drop-outs because they avoid the loss of subjects due to missing data in any of the time points [38,39]. To avoid biased results due to different burn severity of both groups (see Table 2), the ABSI-score was entered as a covariate. Results are reported for the models that best fitted the data according to Schwarz’s Bayesian Criterion (BIC) fit index. Effect sizes were calculated for estimated means and standard deviations of the ITT sample using Cohen’s formula based on

Table 2 – Sample characteristics and medical parameters of the intervention group (IG) and the treatment as usual control group (TAU).

Age ABSI %TBSA Time since accident (months) Gender Female Male Site of burn event Home Work Leisure time Traffic Suicide attempt Other Vocational status before accident Full time Part time Housewife/man Student/trainee Unemployed Retired Other Marital status Married/in relationship Single Divorced Widowed *

311

p < .05.

IG (n = 86)

TAU (n = 128)

M (SD)

M (SD) (14.6) (2.4) (16.72) (52.7)

t (df) 1.98 (212)* 4.26 (182)* 3.7 (188)* .87 (209)

44 (14.5) 7.5 (2.4) 27 (17.83) 21 (37.7)

40 6 17 15

%

%

45 55

30 70

3.99 (1)*

46 21 18 9 4 2

41 30 22 2 3 2

7.23 (5)

60 12 4 6 8 9 2

60 7 3 13 8 4 7

7.97 (6)

40 40 17 4

49 40 6 5

5.95 (3)

x2(df)

3.

Results

3.1.

Participants

A total of n = 214 burn patients participated in the study. The TAU consisted of 128 burn patients. In addition to treatment as usual, 86 burn patients took part in the burn-specific IG. Table 2 depicts demographic characteristics and medical parameters of the IG and the TAU groups. Participants of the IG were on average 4 years older than the TAU group. The majority of the TAU group was male (70%), whereas gender ratio was more balanced in the IG (55% male, 45% female). On average participants of the IG suffered more severe burn injuries than patients of the TAU group, indicated by a significantly higher ABSI score and a higher percentage of TBSA. As shown in Table 3, resources and psychological impairments were comparable in both groups at pre-assessment with one exception. The IG reported more severe posttraumatic symptoms than the TAU group at pre-assessment. In Table 3 data are provided about average psychological distress, quality of life and psychosocial resources of both groups in comparison to non-clinical samples. Compared to norm samples, burn patients of the IG and the TAU group showed significantly higher scores in HADS-depression, PSS, and SCL-K-9, indicating more severe depressive and posttraumatic symptoms as well as general symptom severity (t(77–121) = 3.21–7.53, all p < .05). Only the level of anxiety, assessed by the HADS-anxiety, was comparable to the norm sample (t(204) = 1.09, p = .28). Quality of life was substantially lower in the IG and the TAU group compared to the nonclinical control sample, indicated by lower WHOQOL scores (t(83–123) = 4.16–5.61, all p < .05). In accordance with other studies of burn patients [3,4], the present burn sample reported high levels of psychosocial resources: the IG as well as the TAU group showed higher levels of optimism (LOT; t(84–124) = 11.14–16.5, all p < .05) and social support (F-SozUK-14, t(84–124) = 3.19–4.01, all p < .05) compared to nonclinical populations.

3.2.

Attrition rate analyses

Attrition rate was considerable, with n = 63 (29.4%) participants not returning post-assessment, and n = 107 (50%) participants not providing follow-up assessment. Attrition rates did not differ between the two groups at post-assessment (x2(1) = 1.03, p = .31) and follow-up assessment (x2(1) = 3.19, p = .07). Fig. 1 summarizes the participant flow. On average, completers of both groups were significantly older (44 years) than non-completers (39 years; t(212) = 2.45,

312

burns 41 (2015) 308–316

Table 3 – Psychosocial distress, quality of life and psychosocial resources of IG and TAU group at pre-assessment and comparison to non-clinical norm samples.

SCL-K-9 PSS total severity WHOQOL overall HADS anxiety HADS depression LOT F-SozU-K-14 * a b c d e f

IG (n = 86)

TAU (n = 128)

M (SD)

M (SD)

t (df)

M (SD)

1.17 (.94) 15.96 (12.59) 53.57 (22.43) 6.61 (4.61) 5.89 (4.66) 27.80 (5.13) 4.19 (.68)

.94 (.79) 10.29 (10.34) 59.48 (20.93) 5.77 (3.77) 4.58 (4.04) 28.38 (4.59) 4.22 (.73)

1.97 (205) 3.45 (195)* 1.94 (206) 1.44 (203) 2.14 (202) .86 (208) .28 (208)

.41 (.51)a 5.31 (7.57)b 67.3c 5.8 (3.2)d 3.4 (2.6)d 21.60 (6.50)e 3.96 (.72)f

Norm samples

p < .05. Norm sample: n = 2057 [29]. Burn sample: n = 144 [22]. Control sample: n = 363 [33]. Control sample: n = 152 [32]. Control sample: n = 51 [35]. Norm sample: n = 2507 [34].

p < .05). Additionally, burn injury was more severe in the completer sample, as indicated by higher ABSI-scores (M(SD)completers = 6.8 (2.73), M(SD)non-completers = 6.05 (2.30); t(128) = 3.98, p < .05). Regarding resources, completers were more optimistic than non-completers (LOT:M(SD)compl(4.35), M(SD)non-completer = 27.06 (5.04), eters = 29.18 t(208) = 3.27, p < .05). Completers and non-completers did not differ in other demographic variables, psychological impairment or resources. Separate attrition rate analyses for the IG and TAU groups revealed comparable results.

3.3.

Results of the mixed model analyses

3.3.1.

Psychological distress and quality of life

Table 4 presents means and standard deviations of the outcome measures including within effect sizes, as well as the coefficients of the interaction effects of the mixed models. Mixed model analyses for the SCL-K-9 revealed a significant interaction effect of group  time. Single object comparison showed a significant decline of subjective impairment by general symptom severity (SCL-K-9) in the intervention group

n = 214 met inclusion criteria and agreed to participate

TAU: n = 128 included at pre assessment

IG: n = 86 included at pre assessment

n = 41 did not reply or could not be reached

n = 22 did not reply or could not be reached

TAU: n = 87 included at post assessment

IG: n = 64 included at post assessment

n = 17 did not reply or could not be reached

n = 27 did not reply or could not be reached

TAU: n = 70 included at follow-up assessment

IG: n = 37 included at follow-up assessment

Fig. 1 – Participant flow-chart.

313

burns 41 (2015) 308–316

Table 4 – Estimated means and standard deviations for outcome measures including within effect sizes and interaction effects of mixed model procedure. Intervention group Estimated mean (SD) SCL-K-9

PSS total severity

HADS anxiety

HADS depression

WHOQOL overall

WHOQOL psychological health

WHOQOL physical health

WHOQOL social relationship

WHOQOL environment

LOT

F-SozU-K-14

Pre Post Follow-up Pre Post Follow-up Pre Post Follow-up Pre Post Follow-up Pre Post Follow-up Pre Post Follow-up Pre Post Follow-up Pre Post Follow-up Pre Post Follow-up Pre Post Follow-up Pre Post Follow-up

1.16 (.93) .85 (1.03) .85 (1.31) 15.49 (13.01) 12.13 (13.75) 12.26 (16.43) 6.60 (4.86) 5.92 (5.27) 6.06 (6.66) 5.72 (4.72) 5.01 (5.07) 4.41 (6.31) 53.40 (23.46) 63.31 (26.31) 64.09 (32.68) 61.06 (21.28) 66.78 (22.94) 65.45 (26.94) 59.32 (21.27) 67.21 (22.95) 68.13 (28.85) 67.12 (23.28) 70.67 (25.24) 67.62 (29.46) 67.87 (16.68) 70.78 (18.36) 71.75 (22.41) 27.35 (5.46) 28.89 (6.04) 28.63 (7.19) 4.20 (.83) 4.27 (.91) 4.15 (1.15)

dwithin

.31* .28* .25* .22* .13 .09 .148 .24* .04* .37* .26* .188 .36* .35* .158 .02 .178 .208 .27* .208 .09 .04

Treatment as usual control group Estimated mean (SD) 1.00 (.92) .99 (.99) .88 (1.10) 10.97 (12.65) 11.21 (13.26) 11.21 (14.37) 5.87 (4.78) 5.86 (5.10) 5.82 (5.63) 5.06 (4.66) 5.26 (5.93) 4.96 (5.41) 57.46 (22.96) 61.23 (25.45) 62.02 (27.32) 65.63 (20.84) 66.10 (22.27) 65.69 (23.43) 63.87 (20.87) 68.32 (22.22) 70.66 (24.46) 68.26 (22.63) 67.00 (24.40) 67.87 (25.72) 69.87 (16.20) 69.55 (17.76) 70.04 (18.95) 28.27 (5.31) 27.65 (5.81) 27.89 (6.19) 4.18 (.81) 4.03 (.88) 4.02 (.98)

Group  time

dwithin

F (df)

p

.01 .12

3.28 (235; 2)

.04*

.02 .02

3.85 (217; 2)

.02*

.00 .01

.7 (225; 2)

.5

.04 .02

1.8 (223; 2)

.17

.168 .18*

2.24 (232; 2)

.11

.07 .12

2.79 (219; 2)

.068

.21* .30*

1.04 (227; 2)

.35

.05 .06

1.74 (223; 2)

.17

.02 .01

1.6 (227; 2)

.21

.09 .15

5.47 (231; 2)

.00*

.18* .178

2.64 (231; 2)

.078

*

p < .05. trend ( p  .08. Note: Higher scores on the HADS, SCL-K-9 and PSS indicate greater impairment. Higher scores on the WHOQOL, LOT and F-SozU-K14 indicate greater quality of life, optimism and social support. 8

(IG) from pre- to post-assessment ( p < .01, d = .31) and from pre- to follow-up assessment ( p < .01, d = .28) with small within group effect sizes. The TAU group reported no substantial change in general symptom severity. Regarding the severity of posttraumatic symptoms (PSS), results showed a significant interaction effect of group  time. The IG reported a substantial reduction in impairment by posttraumatic symptoms from pre- to post-assessment ( p < .05, d = .25) and from pre- to follow-up assessment ( p < .05, d = .22). In contrast, posttraumatic symptoms in the TAU group did not change over time. Because pre-test impairment by posttraumatic symptoms was significantly higher in the IG than in the TAU group (see Table 3), an additional repeated measures ANCOVA with pre-test PSS-score and ABSI-score as covariates was conducted. Results showed a significant main effect of group (F(1) = 6.5, p < .05, h2 = .09): After controlling for pre-test PSS-score and ABSI-score, the IG showed significantly lower impairment by posttraumatic symptoms at post-test and follow-up assessment than the TAU group (M(SE)IG_post = 7.02

(1.6), M(SE)TAU_post = 11.9 (1.0); M(SE)IG_follow-up = 7.8 (1.7), M(SE)TAU_follow-up = 11.7 (1.1)). The result of this follow-up ANCOVA based on the completer sample confirms the previous finding of the ITT mixed model analyses. Regarding anxiety and depression (HADS), separate mixed model analyses with Bonferroni alpha adjustment (number of comparisons = 2, a < .025) revealed neither significant interaction nor main effects (all p > .05). For the separate domains of quality of life (WHOQOL), mixed model analyses with Bonferroni alpha adjustment (number of domains = 5, a < .01) revealed heterogeneous results: Regarding the domains ‘‘overall quality of life’’ and ‘‘physical quality of life’’, both groups reported a substantial increase evidenced by significant main effects of time (overall quality of life: F(232, 2) = 12.05. p < .01; physical quality of life: F(227, 2) = 14.89. p < .01). There were no significant interaction effects. Considering the domain ‘‘psychological quality of life’’, a trend toward an interaction effect of group  time emerged. Single object comparison showed a significant

314

burns 41 (2015) 308–316

improvement from pre- to post-assessment ( p < .01, d = .26) and a trend from pre- to follow-up assessment ( p = .06, d = .18) in the IG. In the TAU group level of psychological quality of life did not change significantly. For the domains ‘‘social relationship’’ as well as ‘‘environment’’ neither significant interaction nor main effects were significant (all p > .05).

3.3.2.

Psychosocial resources

Mixed model analyses for optimism (LOT) showed a significant interaction effect of group  time. In the IG, optimism increased significantly from pre- to post-assessment ( p < .01, d = .27) and showed a trend toward improvement from pre- to follow-up assessment ( p = .06, d = .20). In contrast, optimism in the TAU group did not change significantly. For social support (F-SozU-K14), the trend toward an interaction effect of group  time could be explained by the significant decrease of social support in the TAU group from pre- to post-assessment ( p < .05, d = .18) and a trend towards a decline from pre- to follow-up assessment ( p = .07, d = .17). In the IG, social support remained stable from pre- to postassessment as well as from pre- to follow-up assessment.

4.

Discussion

Overall, our data indicate that the newly developed group intervention led to an improvement of psychological distress and resources of participating burn patients compared to those who received treatment as usual. After group treatment, the intervention group (IG) reported a substantial decline in general symptom severity (SCL-K-9), as well as total severity of posttraumatic symptoms (PSS) compared to patients who received treatment as usual. After the group intervention, participants were substantially more optimistic (LOT), whereas optimism in the TAU group did not change. For social support (F-SozU-K14) a trend indicated that social support remained high in the IG and decreased in the TAU group. For both groups mixed model analysis revealed no significant change in anxiety and depression (HADS). Regarding anxiety this result is not surprising as pre-scores of anxiety in the IG and TAU were already comparable to a non-clinical comparison population (see Table 3). Considering psychological health (WHOQOL), burn patients who participated in the IG group showed a trend towards improvement while psychological health of the TAU group did not change. Both groups reported a gradual improvement in overall and physical quality of life (WHO QOL) over the three assessment points, a finding consistent with other studies concerning long-term outcome after burn injury [8]. As previously reported, the overall patient satisfaction with the group treatment was high [3]. This is particularly noteworthy because in another study nearly one fourth of burn patients seeking help because of physical or psychological problems reported to be dissatisfied with the received services [21]. Although participants of the present study perceived the treatment as helpful for their problems, only small changes in psychopathological symptoms were observed. Considering the purpose of the group intervention, the rather small within effect sizes do not question its benefit. Focus of the group training was on problems that the majority of burn patients

experience such as coping with stress and altered appearance as well as adjusting to everyday life. In case of clinically relevant psychopathology the group training intends to reduce the threshold to seek further psychological treatment. The intervention does not aim to replace individual psychotherapy. Additionally, it should be noted that the group program was offered as an add-on to all the usual care the patients receive. For these reasons, small effect sizes were to be expected. Furthermore, small effects of the group treatment on psychopathological symptoms could also be influenced by mental problems experienced by the participants prior to suffering a burn injury. In future research pre-existing mental disorders should be assessed to investigate to what extent preburn mental health influences the results. It is conceivable that group participation has a positive impact on injury related psychological impairment but not on pre-existing mental disorders. This could explain why depression scores do not change while impairment by posttraumatic stress decreased in the IG. The decline in posttraumatic symptom severity after group participation is particularly noteworthy as no trauma-focused approach such as Eye Movement Desensitization and Reprocessing (EMDR) [41] or exposure therapy [42] were provided. Since various international studies revealed that the degree of posttraumatic stress is predicting poor longterm HRQoL of burn patients [18–20], the reduction of posttraumatic symptoms of group participants is of particular importance and should be investigated in future studies. With two exceptions, participants of the IG and the TAU group were comparable at pre-assessment on accident-related variables, resources and psychological impairments. However, at pre-assessment the IG reported substantial greater impairment by posttraumatic stress and suffered more severe burn injuries than the TAU group. Due to the voluntary group participation, these results suggest that patients with more severe burns and patients suffering from more noticeable posttraumatic stress were more likely to be interested in group participation. To avoid biased results because of non-pre-testequivalence we controlled for burn severity and initial posttraumatic stress. Another specific feature of the current sample is the comparably high rate of female burn patients participating in the group treatment (45%), which is noticeably higher than the average percentage of female burn victims (30%) in Germany [43]. This may reflect an often reported gender difference in health-seeking services in the general population. In Germany, nearly three quarters of the persons consulting outpatient psychotherapy are female [44]. One major limitation of the present study is the considerable attrition rate. However, the rate and characteristics of non-completers in our study is comparable to other studies concerning long-term outcome of burn patients [45]. In a comparison of 15 different longitudinal studies concerning PTSD after severe injury, attrition rates of up to 55% were reported [46]. We dealt with the attrition rate in the present study by calculating ITT outcome analyses using linear mixed models, which are recommended for repeated measure data with many drop-outs [38,39]. Furthermore, we have no data about potential pre-existing psychological impairment that are often reported for burn patients [12–14]. A third limitation of the present study is the non-randomised allocation to the IG and the TAU group. The TAU group was recruited first, the

burns 41 (2015) 308–316

group intervention was offered to all eligible burn patients at a later point. This sequential design could have led to selection effects. Future replications of the efficacy of the psychosocial group treatment should therefore include a randomised comparison group. To investigate whether results of the present paper are specific to the applied techniques of the group intervention or to non-specific group factors such as additional attention or sense of belonging, it would be interesting to include another control group receiving the same amount of group meetings, without the specific therapeutic interventions. Despite these limitations and the rather small within group effect sizes, it can be concluded that the intervention specifically tailored to the needs of burn patients after discharge was well accepted by the participants and had positive effects on psychological well-being and resources of participants. Due to the positive feedback of participating burn patients and the good feasibility of the intervention, the developed group treatment has already been implemented as an inherent part of the regular burn treatment in two rehabilitation centers in Germany.

Conflict of interest The authors do not have any financial or personal relationship with other people or organizations that could have inappropriately influenced the present manuscript.

Acknowledgements The present multicenter-study was funded by the German statutory accident insurance (DGVU; Deutsche Gesetzliche Unfallversicherung; FF-FR0093A). We would like to thank Dipl.-Psych. Bernd Genal, Dipl.Psych. Sabine Bo¨hm (rehabilitation center Passauer Wolf in Bad Griesbach), Dr. Hans Ziegenthaler, Dipl.-Psych. Ulrike Fritzsche, Dipl.-Psych. Antje Hyckel (Moritz clinic, rehabilitation center in Bad Klosterlausnitz), Dr. Christian Braune (BG Trauma Centre, Hamburg, Germany), Dr. Annette Stolle (BG Trauma Centre, Ludwigshafen, Germany) and Dipl.-Psych. Diemut Holtfrerich (BG Trauma Centre, Berlin, Germany) for their cooperation in this study.

references

[1] Voigt PM, Krettek C. Versorgungsstandards in der Verbrennungsmedizin [Standards in burn care]. Unfallchirurg 2009;112:461. ¨ stu¨n TB, Chatterji S, Bickenbach J, Kostanjsek N, [2] U Schneider M. The international classification of functioning, disability and health: a new tool for understanding disability and health. Disabil Rehabil 2003;25:565–71. [3] Seehausen A, Renneberg B. Ein Gruppenbehandlungsprogramm zur Bewa¨ltigung von schweren Brandverletzungen [Group treatment programm for coping with severe burn injuries]. Z Klin Psychol Psych 2012;41:201–10.

315

[4] Ripper S, Stolle A, Seehausen A, Klinkenberg A, Germann G, Bernd Hartmann B, et al. Psychische Folgen schwerer Brandverletzungen. Belastungen und Ressourcen berufsgenossenschaftlich versus nichtberufsgenossenschaftlich Versicherter ein Jahr nach der Brandverletzung [Psychological consequences of severe burn injuries. Strains and resources of BG and non-BG insured one year after the burn injury]. Unfallchirurg 2010;113:915–22. [5] Blakeney PE, Rosenberg L, Rosenberg M, Faber AW. Psychosocial care of persons with severe burns. Burns 2008;34:330–433. [6] Wallis H, Renneberg B, Neumann M, Ripper S, Bastine R. Ressourcen und Belastungsfaktoren nach schweren Brandverletzungen. Eine qualitative Studie zwei Jahre nach dem Unfall [Stress factors after severe burn injuries. A qualitative study two years after the accident]. Verhaltenstherapie und Verhaltensmedizin 2007;28:173–87. [7] Palmu R, Suominen K, Vuola J, Isometsa E. Mental disorders after burn injury: a prospective study. Burns 2011;37:601–9. ¨ ster C, Willebrand M, Ekselius L. Health-related quality of [8] O life 2 years to 7 years after burn injury. J Trauma 2011;71:1435–41. [9] ter Smitten MH, de Graaf R, van Loey NE. Prevalence and comorbidity of psychiatric disorders 1–4 years after burn. Burns 2011;37:753–61. [10] First MB, Spitzer RL, Gibbon M, Williams JBW. Structured clinical interview for DSM-IV axis I disorders, clinician version (SCID-CV). Washington, D.C.: American Psychiatric Press, Inc.; 1996. [11] Robins LN, Wing J, Wittchen HU, Helzer JE, Babor TF, Burke J, et al. The composite international diagnostic interview (CIDI): an epidemiologic instrument suitable for use in conjunction with different diagnostic systems and in different cultures. Arch Gen Psychiatry 1988;45:1069–77. [12] Palmu R, Suominen K, Vuola J, Isometsa E. Mental disorders among acute burn patients. Burns 2010;36:1072–9. [13] Dyster-Aas J, Willebrand M, Wikehult B, Gerdin B, Ekselius L, Major Depression. Posttraumatic stress disorder symptoms following severe burn injury in relation to lifetime psychiatric morbidity. J Trauma 2008;64(5):1349–56. [14] Low AJF, Dyster-Aas J, Willebrand M, Ekselius L, Gerdin B. Psychiatric morbidity predicts perceived burn-specific health 1 year after a burn. Gen Hosp Psychiat 2012;34: 146–52. [15] Wisely JA, Tarrier N. A survey of the need for psychological input in a follow-up service for adult burn-injured patients. Burns 2001;27:801–7. [16] Palmu R, Suominen K, Vuola J, Isometsa E. Psychiatric consultation and care after acute burn injury: a 6-month naturalistic prospective study. Gen Hosp Psychiat 2011;33:16–22. [17] Van Loey NE, Faber AW, Taal LA. A European hospital survey to determine the extent of psychological services offered to patients with severe burns. Burns 2001;27:23–31. [18] Renneberg B, Ripper S, Schulze J, Seehausen A, Weiler M, Wind G, et al. Quality of life and predictors of long-term outcome after severe burn injury. J Behav Med 2013. http:// dx.doi.org/10.1007/s10865-013-9541-6. [19] van Loey NE, van Beeck EF, Faber BW, van de Schoot R, Bremer M. Health-related quality of life after burns: a prospective multicenter cohort study with 18 months follow-up. J Trauma 2012;72:513–20. [20] Corry NH, Klick B, Fauerbach JA. Posttraumatic stress disorder and pain impact functioning and disability after major burn injury. J Burn Care Res 2010;31:13–25. [21] Van Loey NE, Faber AW, Taal LA. Do burn patients need burn specific multidisciplinary outpatient aftercare: research results. Burns 2001;27:103–10.

316

burns 41 (2015) 308–316

[22] Renneberg B, Ripper S, Seehausen A, Germann G. Abschlussbericht zum Forschungsvorhaben: Evaluation und Weiterentwicklung psychotherapeutischer Interventionen fu¨r die Akut-und Rehabilitationsphase nach schweren Brandverletzungen, Teil II. Unvero¨ffentlichter Bericht [Final report of the research project: evaluation and development of psychotherapeutic interventions for acute and rehabilitation phase after severe burns, part II. Unpublished research report]; 2010. [23] Wallis-Simon H, Renneberg B. Psychische Faktoren bei schweren Brandverletzungen: Psychotherapeutische Beitra¨ge zur Verbesserung der Lebensqualita¨t [Psychological factors in severe burn injuries: psychotherapeutic contributions for improving the quality of life]. Psychotherapeutenjournal 2009;2:142–52. [24] Holtfrerich D, Ripper S, Seehausen A, Stolle A, Wallis H, Renneberg B. Ein Gruppenbehandlungsprogramm fu¨r Brandverletzte in der Rehabilitation. Unvero¨ffentlichtes Manual [A group treatment program for burn victims during rehabilitation. Unpublished manual]; 2008. [25] Schneider J, Buchheim P. Integrierte Psychotherapie von psychisch traumatisierten Patienten mit Unfallverletzungen [Integrated psychotherapy for mentally traumatized patients with injuries]. Psychomed 2004;16:216–23. [26] Robinson E, Rumsey N, Partridge J. An evaluation of the impact of social interaction skills training for facially disfigured people. Br J Plast Surg 1996;49:281–9. [27] Schaub A, Roth E, Goldmann U. Kognitiv-psychoedukative Therapie zur Bewa¨ltigung von Depressionen: Ein Therapiemanual [Cognitive-psychoeducational therapy for coping with depression: A treatment manual]. Go¨ttingen: Hogrefe; 2006. [28] Kaluza G. Stressbewa¨ltigung Trainingsmanual zur psychologischen Gesundheitsfo¨rderung [Stress Management: Training Manual for promoting psychological health]. Berlin/Heidelberg: Springer-Verlag; 2004. [29] Klaghofer R, Braehler E. Konstruktion und teststatistische Pru¨fung einer Kurzform der SCL-90-R [Construction and statistical testing of a short form of the SCL-90-R]. Z Klin Psychol Psych 2001;49:115–24. [30] Foa EB, Riggs DS, Dancu CV, Rothbaum BO. Reliability and validity of a brief instrument for assessing post-traumatic stress disorder. J Trauma Stress 1993;6:459–73. [31] Stieglitz RD, Frommberger U, Berger M. Evaluation der deutschen Version der PTSD Symptom Scale (PSS) [Evaluation of the german version of the PSS]. In: Stieglitz RD, editor. Syndromale Diagnostik psychischer Sto¨rungen. Go¨ttingen: Hogrefe; 1998. p. 178–83. [32] Herrmann C, Buss U, Snaith RP. Hospital Anxiety and Depression Scale—Deutsche Version [German version]. Bern: Huber; 1995.

[33] Angermayer MC, Kilian R, Matschinger H. WHOQOL-100 und WHOQOL-BREF. Handbuch fu¨r die deutschsprachige Version der WHO Instrumente zur Erfassung von Lebensqualita¨t [Manual for the German version of the WHO instruments to measure quality of life]. Go¨ttingen: Hogrefe; 2000. [34] Fydrich T, Sommer G, Bra¨hler E. Fragebogen zur sozialen Unterstu¨tzung (F-SozU): Normierung der Kurzform (K-14) [Social Support Questionnaire: Standardization of the short form]. Zeitschrift fu¨r Medizinische Psychologie 2009;18:43–8. [35] Wieland-Eckelmann R, Carver CS. Dispositional coping style, optimism, and coping: a cross-cultural comparison. Zeitschrift fu¨r Differentielle und Diagnostische Psychologie 1990;11:167–84. [36] Tobiasen J, Hiebert JM, Edlich RF. The abbreviated burn severity index. Ann Emerg Med 1982;11:260–2. [37] Barber JP, Triffleman E, Marmar C. Considerations in treatment integrity: implications and recommendations for PTSD research. J Trauma Stress 2007;20:793–805. [38] Chan YH. Biostatistics 301A. Repeated mesaurement analyses (mixed models). Singapore Med J 2004;45: 456–62. [39] Houck PR, Mazumdar S, Koru-Sengul T, Tang G, Mulsant BH, Pollock BG, et al. Estimating treatment effects from longitudinal clinical trial data with missing values: comperative analyses using different methods. Psychiatr Res 2004;129:209–15. [40] Cohen J. Statistical power analysis for the behavioral sciences. 2nd ed. Hillsdale, NJ: Erlbaum; 1988. [41] Shapiro F. Eye movement desensitization and reprocessing (EMDR). Paderborn: Junfermann; 1998. [42] Foa EB, Rothbaum BO. Treating the trauma of rape: cognitive-behavioral therapy for PTSD. New York, NY: Guilford; 1998. [43] Deutsche Gesellschaft fu¨r Verbrennungsmedizin e. V. Verbrennungsstatistik. [database on the Internet]. Date of last access: 10/1/2014. Available from: hhttp://www. verbrennungsmedizin.de/pdf/verbrennungsstatistik-2013. pdfi. [44] Albani CBG, Geyer M, Schmutzer G, Bra¨hler E. Ambulante Psychotherapie in Deutschland aus Sicht der Patienten. Teil 1: Versorgungssituation [Outpatient psychotherapy in Germany from patients perspective. Part 1]. Psychotherapeut 2010;55:503–14. [45] Fauerbach JA, McKibben J, Bienvenu OJ, Magyar-Russell G, Smith MT, Holavanahalli R, et al. Psychological distress after major burn injury. Psychosom Med 2007;69:473–82. [46] Nishi D, Matsuoka Y, Nakajima S, Noguchi H, Kim Y, Kanb S, et al. Are patients after severe injury who drop out of a longitudinal study at high risk of mental disorder. Compre Psychiat 2008;49:393–8.