Pediatr Transplantation 2014: 18: 851–859
© 2014 John Wiley & Sons A/S. Published by John Wiley & Sons Ltd
Pediatric Transplantation DOI: 10.1111/petr.12367
Family strain and its relation to psychosocial dysfunction in children and adolescents after liver transplantation Kaller T, Petersen I, Petermann F, Fischer L, Grabhorn E, Schulz K-H. (2014) Family strain and its relation to psychosocial dysfunction in children and adolescents after liver transplantation. Pediatr Transplant, 18: 851–859. DOI: 10.1111/petr.12367. Abstract: Parental functioning is essential to children’s development. Therefore, this cross-sectional single-center study examined the prevalence of family strain in 181 parents and its associations to psychosocial functioning in their children after LT. Median age at LT was one yr. Mean time elapsed since LT was 5.8 yr. The IFS, and the SDQ were applied to parents. Family strain in the present sample was comparable to that in the German normative group of families with a chronically ill or disabled child, but families of LT recipients showed a significantly higher financial impact, impact on coping, and impact on siblings (p < 0.001). Younger age of patients at survey, a more severe clinical course, child’s restrictions, and financial losses following LT were determined as significant predictors of family strain (R2 = 0.42). Parents reported less family strain after living-related compared with deceased donation. Family strain was significantly correlated to psychosocial dysfunction in children post-LT. Present findings demonstrate a risk of maladjustment to the post-LT condition in families. They emphasize the importance of psychological assessment of parents and patients during transplant and follow-up to ensure the best achievable long-term outcome of patients.
In recent years, pediatric LT has become a standard therapy in children with end stage liver disease and acute liver failure, even in early infancy (1, 2). As medical outcomes continue to improve with excellent results in graft and patient survival, more consideration now is needed on psychosocial functioning of transplant recipients. Few studies have demonstrated emotional and behavioral problems in pediatric LT recipients (3–5). Results from a recent study of our group demonstrated psychosocial problems in 137 pediatric LT recipients (6). Furthermore, current research indicates that children after pediatric LT are at higher risk for developing cognitive deficits compared with the age
Abbreviations: IFS, Impact on Family Scale; LT, liver transplantation; PELD, pediatric end-stage liver disease; PTSD, post-traumatic stress disorder; SDQ, Strengths and Difficulties Questionnaire.
T. Kaller1, I. Petersen2, F. Petermann1, L. Fischer3, E. Grabhorn4 and K.-H. Schulz2 1
Center of Clinical Psychology and Rehabilitation, University Bremen, Bremen, Germany, 2Department of Medical Psychology, University Medical Center Hamburg-Eppendorf, Hamburg, Germany, 3 Department of Hepatobiliary and Transplant Surgery, University Medical Center Hamburg – Eppendorf, Hamburg, Germany, 4Children’s Hospital, University Medical Center Hamburg-Eppendorf, Hamburg, Germany Key words: family strain – distress – quality of life – liver transplantation – parents – children Tanja Kaller, Center of Clinical Psychology and Rehabilitation, University Bremen, Grazer Str. 6, 28359 Bremen, Germany Tel.: 0049 421 218 68610 Fax: 0049 421 218 68629 E-mail: [email protected] K.-H. Schulz and Enke Grabhorn contributed equally. Accepted for publication 29 August 2014
appropriate norm (7–10) or to an age- and gender-matched reference group (11). Nevertheless, still, little is known about the impact of the transplant experience and the ongoing health issues on the child’s family, especially their parents. Evidently, parents are essential in the development of their children and in the setting of transplantation they have to assume a tremendous amount of responsibility. Hence, the better they are able to cope with the burdens of care, the better the child will adapt to the condition of illness (12). Some studies have conceptualized the experience of life-threatening illness in a child as a potentially traumatic event for parents (13–16). More than half of parents with a child who received a liver transplant reported symptoms of PTSD (16, 17). In a metaanalysis, Cabizuca et al. (18) found that almost a quarter of parents of children with chronic diseases – including parents of pediatric LT recipients – fulfilled the criteria for PTSD. Studies indicate high parental and family distress 851
Kaller et al.
in families after pediatric LT (3, 19, 20). Research on chronic conditions in childhood also shows that the child’s disease is a risk factor for health-related quality-of-life impairment in parents and high levels of parental distress (21, 22). These findings are important to consider, because a high level of distress in parents may lead to poor parenting behavior and is associated with the child’s maladjustment to the illness (23). It may also contribute to poorer child health and developmental outcomes. The purpose of this cross-sectional study was to assess family strain in parents of children and adolescents who had undergone LT. We compared subjective impact on family scores to those of a normative sample composed of parents with a chronically ill or disabled child. We expected associations with disease- and family-specific characteristics (severe clinical course, more restrictions in the child and financial losses due to LT, shorter interval after LT, and younger age at survey). Furthermore, we expected that high levels of family strain would be associated with more emotional and behavioral disturbances in the children post-LT. By exploratory analysis we compared the subjective impact on families with a child after deceased and living-related donation. In a semi-structured interview, we identified issues of particular burden. Methods Participants The current sample consisted of parents of 181 children who underwent LT at the University Hospital Hamburg-Eppendorf, Germany, between 1995 and 2009. Inclusion criteria were as follows: (i) residency in Germany, (ii) at least one yr post-LT, and (iii) sufficient German language skills. In total, 191 children were eligible for the study and were recruited between August 2006 and December 2010 during their routine annual medical check-up. Seven parents declined participation, and three appointments were cancelled due to logistic reasons.
Procedures This study was approved by the ethics committee of the medical council of Hamburg, Germany. All participants were informed about the study by letter and phone before their annual medical check-up. Informed consent was obtained from parents. All medical information was obtained from the patients’ charts (see Table 2). Additionally, an index score “Severity of clinical course” was composed of the following six preand post-transplant disease-related variables (dichotomized by 0/1): Days on waiting list and days in hospital post-LT (median split: short vs. long), complications, acute and/or chronic organ rejections, surgical revisions, and re-LT (present vs. not present). The cut-off score for a more severe clinical course was defined by a score >3. PELD scores were not included since most children had received their trans-
852
plant before PELD scores had been in use. Socioeconomic and family characteristics were included as follows: family income, parents’ age, level of education, employment status, marital status, and number of siblings. In addition, a semistructured interview regarding the post-LT situation has been conducted with the parents. Reported relevant issues were, for example, family and financial problems.
Measures Impact on families
We used the German version of the Impact-on-Family Scale (IFS) developed by Stein and Riessmann (24) as a selfreport instrument to measure the effects of chronic conditions and disability in childhood on the family. The German scale (Familien-Belastungs-Questionnaire, [25]) contains 33 Likert-scaled items to assess five dimensions of family strain: daily/social impact (15 items), personal impact/worries about future (five items), financial impact (four items), impact on coping (three items), and impact on siblings (six items). The response scale ranges from 1 to 4 (strongly agree to strongly disagree). Higher scores indicate higher negative impact. A Total score includes all items except for the dimension impact on siblings. The German normative sample (25) included 273 families with a chronically ill or disabled child (25). Psychometric properties are proven with acceptable construct validity and good internal consistency (Cronbach’s a ranging from 0.70 to 0.89) as well as discriminant validity. In the present study, parents of children aged 1–18 yr were assessed with the IFS (n = 181). Behavior and emotional functioning in children
To assess relations between impact on families, and behavioral as well as emotional functioning in children and adolescents post-LT, the SDQ was applied (26). The Questionnaire yields 25 questions and measures prosocial – as well as problem behavior in children between six and 16 yr of age. Assessed dimensions are prosocial behavior, emotional problems, conduct problems, hyperactivity, and problems with peers. A total problem score is derived from all problem scales. The psychometric property is adequate. In the present study, parents of children aged 6–16 yr were assessed with the SDQ (n = 130). Interviews
The semi-structured interviews were conducted by trained psychologists using an interview guide with specified questions. The questions included socio-demographic data, family life, situation in kindergarten or school, leisure activity, contact with peers, experiences around the process of transplantation, special resources, burdens of child, and parents as well as restrictions of the child and medication adherence. Most questions were formulated without predetermined answers, that is, “What has helped you to get through the burdens of transplantation?” In addition, parents had the opportunity to report further issues they missed among questions. The interviews were performed with one or both of the parents during the annual medical follow-up of their child. The responses were recorded by the interviewer in written form. For the present study, those questions were evaluated that were regarded as particularly relevant by the parents: Were there any financial losses due to the LT of your child? Were there any changes in the employment of family caused by the child’s LT – What kind of changes? Were there any problems within the family after LT – What kind of problems? Is your child restricted due to its liver
Family strain after pediatric LT disease resp. the implications of LT – What kind of restrictions?
were obtained regarding IFS scores (t = t = 0.47; p = 0.98 to p = 0.64).
Data analyses
Prediction IFS
Statistical analyses were conducted using the Predictive Analysis SoftWare, version 20.0 (PASW; IBM, Armonk, NY, USA). One sample t-tests were performed for comparison of the impact on families with the normative sample. Effect sizes (d) for differences in means are designated as small (0.20), medium (0.50), and large (0.80) in magnitude (27). Simultaneous multiple regression analysis was performed with the Total score of the IFS as dependent variable. Hypothesized independent variables were as follows: Age of the patient at survey, clinical course, restrictions following LT, financial losses following LT, and time since LT. To control for socio-demographic characteristics, family income and education level of parents were included. Categorical variables were dichotomized. The regression model was controlled for outliers, multicollinearity, stability, and residual statistics including homoscedasticity. Exploratory analysis was conducted using a two-tailed t-test to explore differences between living-related donation and deceased donation concerning family strain. For correlation analyses, Pearson’s r was used. The answers from the interview were dichotomized (yes/no; existent/not existent; appropriate/not appropriate). In this form, they were included as independent variables in the regression analysis (financial losses, restrictions following LT). Responses to open questions were summarized thematically and categorized.
In a simultaneous multiple regression analysis, five disease- and family-related (age of patient at survey, clinical course, restrictions following LT, financial losses following LT, and interval since LT) and two control socioeconomic variables (family income and education of parents) were included to predict the Total Score of the IFS. Age at survey, clinical course, child’s restrictions following LT, and financial losses turned out as significant predictors (Table 4). Younger age at survey, a more severe clinical course, restrictions following LT and more financial losses were associated with a higher Total Score. These variables explained 42% of variance of the impact on family score. In contrast to our hypothesis, interval since LT did not reveal a significant predictor. The results indicate that there was no bias due to family income or education level of the parents.
Results LT sample
Table 1 presents socio-demographic characteristics of study participants. At the time of assessment, LT recipients ranged from 1.1 to 18.3 yr of age with a mean age of 8.5 yr. The mean elapsed time between LT and assessment was 5.8 yr (s.d. = 3.9). Clinical characteristics of patients are presented in Table 2. Mean age by the time of LT was 2.6 yr (s.d. = 3.3). Biliary atresia was the most common diagnosis (n = 96). Parents whose child received a living-related donation were more likely in the group with the highest level of education than parents after a deceased donation (OR = 2.0, CI = 1.1–3.8; v2 = 4.6, p = 0.032). Comparison with the normative sample
We assumed that the level of strain in families with a child who has undergone LT would be comparable to the strain in families with a disabled or chronically ill child (25). Table 3 shows significantly higher impact scores in the IFSdimensions financial impact, coping, and impact on siblings in our sample. In the other dimensions as well as in the Total Score, family strain was comparable to the strain in the normative sample. No differences between fathers and mothers
0.03 to
Relations to psychosocial functioning in children and adolescents after LT
Additionally, significant inter-relations between higher strain in families and more emotional and behavioral disturbances in children and adolescents post-LT were expected. This hypothesis was confirmed by correlational analyses. All domains of the IFS were correlated to the subscales of the SDQ (Table 5). The Total Score of the IFS showed the highest inter-relations with most subscales of the SDQ (r = 0.330 to r = 0.480; p < 0.01). Exploratory analysis IFS
By exploratory analysis, family strain was compared between parents with a child who received a living-related or a deceased donation. The Total Score was significantly higher in parents after deceased than after living-related LT (t (179) = 2, 18; p = 0.031). It should be noted that type of donation and clinical course were interrelated. The risk for a more severe clinical course after a deceased donation was threefold compared to living-related donation (OR = 3.0, CI = 1.6–5.6; v2 = 11.5, p = 0.001). Furthermore, statements of the IFS were identified in which families reported the highest strain. Hence, 66.3% of the sample stated “absolutely true” or “true in most aspects” regarding the item “I am worried about the future of my child (when it has grown up);” 61.4% agreed to the item “Sometimes I feel like we live on a roller coaster: 853
Kaller et al. Table 1. Demographic characteristics of study participants (n = 181) M Child Gender (female) Age at survey (yr) Parents Age Mother (n = 175)* Father (n = 169)* Parents reporting (n = 170)* Mother Father Both Other Marital status (n = 179)* Married or living together Number of siblings (n = 176)* 0 1 2 >2 Household income (month) (n = 153)* 5000 Euro
s.d.
Median
Range
8.5
7.7
4.5
1.1–18.3
36.7 40.2
6.7 6.5
37.0 40.0
21–55 25–61
N
Mother (n = 174)*
%
93
51.4
133 28 8 1
78.2 16.5 4.7 0.6
152
84.9
47 88 27 17
26.0 48.6 14.9 10.5
13 56 54 20 7 5
8.3 36.1 34.8 12.9 4.5 3.2
Father (n = 163)*
Education
N
%
N
%
No graduation Special education school Graduation after 8 or 9 yr of education Graduation after 10 yr of education Graduation after 12 yr of education
7 1 44 70 52
4.0 0.6 25.4 40.2 29.9
5 2 58 56 42
3.1 1.2 35.6 34.4 25.8
Mother (n = 169)*
Father (n = 165)*
Employment
N
%
N
%
Full time Part time Homemaker Other (unemployed, retired)
18 77 61 13
10.7 45.6 36.1 7.7
145 4 1 15
87.9 2.4 0.6 9.1
*Causes for missing data: information was not reported, for example, single mothers did not know the graduation of the father or parents declined to state their family income.
in crisis when my child is acutely ill, OK when things are stable;” 47.5% cannot imagine having further children due to their child’s LT. Age at transplantation was not correlated to the scores of any scale of the IFS (r = 0.00 to r = 0.05; p = 0.97 to p = 0.54). Exploratory analysis interviews
In a semi-structured interview, 40% of the parents reported family problems such as 854
problems with siblings (23.2%), marital or relationship problems (23.2%), child’s restrictions post-LT (33.7%) such as limited capabilities regarding leisure or social life (24%) or medical demands or problems (18%), and financial losses following LT (33.7%), which were mostly due to no return to work of the mother after parental leave (31.8%) or reduced labor time (21.6%) as a consequence of LT. Moreover, 64% would have demanded more psycho-social support by the time of LT.
Family strain after pediatric LT Table 2. Transplant-related characteristics of study participants (n = 181) M Age at LT (yr) Time elapsed since LT (yr) Basic immunosuppression at survey Cyclosporin A Tacrolimus Diagnostic categories Biliary atresia Cholestatic diseases Metabolic diseases Others Type of donation Deceased donation Living related donation* Days on waiting list (n = 175) Days in hospital after LT (n = 158) Rejection†, chronic and acute (n = 167) Re-transplantation† (n = 174) Complications†,‡ (n = 158) Surgical revisions†,§ (n = 153)
s.d. 2.6 5.8
Median
3.3 3.9
103.3 43.7
Range
1 5.0
123.0 25.6
N
%
131 50
72.4 27.6
98 42 17 23
54.4 23.3 9.4 12.8
113 68
62.4 37.6
72 26 66 34
43.1 14.9 41.8 22.2
0–13.3 1.0–15.7
57 39.0
0–762 14–214
*Donors were mothers (n = 34), fathers (n = 31), and others (n = 3). † Until one yr post LT. ‡ For example, cholestasis, cholangitis, portal hypertension. § For example, due to biliary leakage, hepatic artery thrombosis, portal vein thrombosis or stenosis, impaired organ perfusion, intestinal perforation, intraabdominal bleedings.
Table 3. Impact on family scores (IFS) in the sample of families with a child or an adolescent after LT in comparison with a normative sample of families with a disabled or chronically ill child (25)
Dimensions of the IFS
LT-group M (s.d.)
Normative group M (s.d.)
Daily/social impact Personal impact/worries about future Financial impact Impact on siblings Impact on coping Total score
1.95 (0.67) 2.21 (0.67) 1.86 (0.89) 1.84 (0.76) 1.92 (0.65) 1.98 (0.59)
1.98 (0.69) 2.13 (0.71) 1.71 (0.74) 1.55 (0.56) 1.67 (0.69) 1.93 (0.54)
t 0.56 1.59 2.28 4.38 5.11 1.22
p
d
0.58 0.11 0.02
© 2014 John Wiley & Sons A/S. Published by John Wiley & Sons Ltd
Pediatric Transplantation DOI: 10.1111/petr.12367
Family strain and its relation to psychosocial dysfunction in children and adolescents after liver transplantation Kaller T, Petersen I, Petermann F, Fischer L, Grabhorn E, Schulz K-H. (2014) Family strain and its relation to psychosocial dysfunction in children and adolescents after liver transplantation. Pediatr Transplant, 18: 851–859. DOI: 10.1111/petr.12367. Abstract: Parental functioning is essential to children’s development. Therefore, this cross-sectional single-center study examined the prevalence of family strain in 181 parents and its associations to psychosocial functioning in their children after LT. Median age at LT was one yr. Mean time elapsed since LT was 5.8 yr. The IFS, and the SDQ were applied to parents. Family strain in the present sample was comparable to that in the German normative group of families with a chronically ill or disabled child, but families of LT recipients showed a significantly higher financial impact, impact on coping, and impact on siblings (p < 0.001). Younger age of patients at survey, a more severe clinical course, child’s restrictions, and financial losses following LT were determined as significant predictors of family strain (R2 = 0.42). Parents reported less family strain after living-related compared with deceased donation. Family strain was significantly correlated to psychosocial dysfunction in children post-LT. Present findings demonstrate a risk of maladjustment to the post-LT condition in families. They emphasize the importance of psychological assessment of parents and patients during transplant and follow-up to ensure the best achievable long-term outcome of patients.
In recent years, pediatric LT has become a standard therapy in children with end stage liver disease and acute liver failure, even in early infancy (1, 2). As medical outcomes continue to improve with excellent results in graft and patient survival, more consideration now is needed on psychosocial functioning of transplant recipients. Few studies have demonstrated emotional and behavioral problems in pediatric LT recipients (3–5). Results from a recent study of our group demonstrated psychosocial problems in 137 pediatric LT recipients (6). Furthermore, current research indicates that children after pediatric LT are at higher risk for developing cognitive deficits compared with the age
Abbreviations: IFS, Impact on Family Scale; LT, liver transplantation; PELD, pediatric end-stage liver disease; PTSD, post-traumatic stress disorder; SDQ, Strengths and Difficulties Questionnaire.
T. Kaller1, I. Petersen2, F. Petermann1, L. Fischer3, E. Grabhorn4 and K.-H. Schulz2 1
Center of Clinical Psychology and Rehabilitation, University Bremen, Bremen, Germany, 2Department of Medical Psychology, University Medical Center Hamburg-Eppendorf, Hamburg, Germany, 3 Department of Hepatobiliary and Transplant Surgery, University Medical Center Hamburg – Eppendorf, Hamburg, Germany, 4Children’s Hospital, University Medical Center Hamburg-Eppendorf, Hamburg, Germany Key words: family strain – distress – quality of life – liver transplantation – parents – children Tanja Kaller, Center of Clinical Psychology and Rehabilitation, University Bremen, Grazer Str. 6, 28359 Bremen, Germany Tel.: 0049 421 218 68610 Fax: 0049 421 218 68629 E-mail: [email protected] K.-H. Schulz and Enke Grabhorn contributed equally. Accepted for publication 29 August 2014
appropriate norm (7–10) or to an age- and gender-matched reference group (11). Nevertheless, still, little is known about the impact of the transplant experience and the ongoing health issues on the child’s family, especially their parents. Evidently, parents are essential in the development of their children and in the setting of transplantation they have to assume a tremendous amount of responsibility. Hence, the better they are able to cope with the burdens of care, the better the child will adapt to the condition of illness (12). Some studies have conceptualized the experience of life-threatening illness in a child as a potentially traumatic event for parents (13–16). More than half of parents with a child who received a liver transplant reported symptoms of PTSD (16, 17). In a metaanalysis, Cabizuca et al. (18) found that almost a quarter of parents of children with chronic diseases – including parents of pediatric LT recipients – fulfilled the criteria for PTSD. Studies indicate high parental and family distress 851
Kaller et al.
in families after pediatric LT (3, 19, 20). Research on chronic conditions in childhood also shows that the child’s disease is a risk factor for health-related quality-of-life impairment in parents and high levels of parental distress (21, 22). These findings are important to consider, because a high level of distress in parents may lead to poor parenting behavior and is associated with the child’s maladjustment to the illness (23). It may also contribute to poorer child health and developmental outcomes. The purpose of this cross-sectional study was to assess family strain in parents of children and adolescents who had undergone LT. We compared subjective impact on family scores to those of a normative sample composed of parents with a chronically ill or disabled child. We expected associations with disease- and family-specific characteristics (severe clinical course, more restrictions in the child and financial losses due to LT, shorter interval after LT, and younger age at survey). Furthermore, we expected that high levels of family strain would be associated with more emotional and behavioral disturbances in the children post-LT. By exploratory analysis we compared the subjective impact on families with a child after deceased and living-related donation. In a semi-structured interview, we identified issues of particular burden. Methods Participants The current sample consisted of parents of 181 children who underwent LT at the University Hospital Hamburg-Eppendorf, Germany, between 1995 and 2009. Inclusion criteria were as follows: (i) residency in Germany, (ii) at least one yr post-LT, and (iii) sufficient German language skills. In total, 191 children were eligible for the study and were recruited between August 2006 and December 2010 during their routine annual medical check-up. Seven parents declined participation, and three appointments were cancelled due to logistic reasons.
Procedures This study was approved by the ethics committee of the medical council of Hamburg, Germany. All participants were informed about the study by letter and phone before their annual medical check-up. Informed consent was obtained from parents. All medical information was obtained from the patients’ charts (see Table 2). Additionally, an index score “Severity of clinical course” was composed of the following six preand post-transplant disease-related variables (dichotomized by 0/1): Days on waiting list and days in hospital post-LT (median split: short vs. long), complications, acute and/or chronic organ rejections, surgical revisions, and re-LT (present vs. not present). The cut-off score for a more severe clinical course was defined by a score >3. PELD scores were not included since most children had received their trans-
852
plant before PELD scores had been in use. Socioeconomic and family characteristics were included as follows: family income, parents’ age, level of education, employment status, marital status, and number of siblings. In addition, a semistructured interview regarding the post-LT situation has been conducted with the parents. Reported relevant issues were, for example, family and financial problems.
Measures Impact on families
We used the German version of the Impact-on-Family Scale (IFS) developed by Stein and Riessmann (24) as a selfreport instrument to measure the effects of chronic conditions and disability in childhood on the family. The German scale (Familien-Belastungs-Questionnaire, [25]) contains 33 Likert-scaled items to assess five dimensions of family strain: daily/social impact (15 items), personal impact/worries about future (five items), financial impact (four items), impact on coping (three items), and impact on siblings (six items). The response scale ranges from 1 to 4 (strongly agree to strongly disagree). Higher scores indicate higher negative impact. A Total score includes all items except for the dimension impact on siblings. The German normative sample (25) included 273 families with a chronically ill or disabled child (25). Psychometric properties are proven with acceptable construct validity and good internal consistency (Cronbach’s a ranging from 0.70 to 0.89) as well as discriminant validity. In the present study, parents of children aged 1–18 yr were assessed with the IFS (n = 181). Behavior and emotional functioning in children
To assess relations between impact on families, and behavioral as well as emotional functioning in children and adolescents post-LT, the SDQ was applied (26). The Questionnaire yields 25 questions and measures prosocial – as well as problem behavior in children between six and 16 yr of age. Assessed dimensions are prosocial behavior, emotional problems, conduct problems, hyperactivity, and problems with peers. A total problem score is derived from all problem scales. The psychometric property is adequate. In the present study, parents of children aged 6–16 yr were assessed with the SDQ (n = 130). Interviews
The semi-structured interviews were conducted by trained psychologists using an interview guide with specified questions. The questions included socio-demographic data, family life, situation in kindergarten or school, leisure activity, contact with peers, experiences around the process of transplantation, special resources, burdens of child, and parents as well as restrictions of the child and medication adherence. Most questions were formulated without predetermined answers, that is, “What has helped you to get through the burdens of transplantation?” In addition, parents had the opportunity to report further issues they missed among questions. The interviews were performed with one or both of the parents during the annual medical follow-up of their child. The responses were recorded by the interviewer in written form. For the present study, those questions were evaluated that were regarded as particularly relevant by the parents: Were there any financial losses due to the LT of your child? Were there any changes in the employment of family caused by the child’s LT – What kind of changes? Were there any problems within the family after LT – What kind of problems? Is your child restricted due to its liver
Family strain after pediatric LT disease resp. the implications of LT – What kind of restrictions?
were obtained regarding IFS scores (t = t = 0.47; p = 0.98 to p = 0.64).
Data analyses
Prediction IFS
Statistical analyses were conducted using the Predictive Analysis SoftWare, version 20.0 (PASW; IBM, Armonk, NY, USA). One sample t-tests were performed for comparison of the impact on families with the normative sample. Effect sizes (d) for differences in means are designated as small (0.20), medium (0.50), and large (0.80) in magnitude (27). Simultaneous multiple regression analysis was performed with the Total score of the IFS as dependent variable. Hypothesized independent variables were as follows: Age of the patient at survey, clinical course, restrictions following LT, financial losses following LT, and time since LT. To control for socio-demographic characteristics, family income and education level of parents were included. Categorical variables were dichotomized. The regression model was controlled for outliers, multicollinearity, stability, and residual statistics including homoscedasticity. Exploratory analysis was conducted using a two-tailed t-test to explore differences between living-related donation and deceased donation concerning family strain. For correlation analyses, Pearson’s r was used. The answers from the interview were dichotomized (yes/no; existent/not existent; appropriate/not appropriate). In this form, they were included as independent variables in the regression analysis (financial losses, restrictions following LT). Responses to open questions were summarized thematically and categorized.
In a simultaneous multiple regression analysis, five disease- and family-related (age of patient at survey, clinical course, restrictions following LT, financial losses following LT, and interval since LT) and two control socioeconomic variables (family income and education of parents) were included to predict the Total Score of the IFS. Age at survey, clinical course, child’s restrictions following LT, and financial losses turned out as significant predictors (Table 4). Younger age at survey, a more severe clinical course, restrictions following LT and more financial losses were associated with a higher Total Score. These variables explained 42% of variance of the impact on family score. In contrast to our hypothesis, interval since LT did not reveal a significant predictor. The results indicate that there was no bias due to family income or education level of the parents.
Results LT sample
Table 1 presents socio-demographic characteristics of study participants. At the time of assessment, LT recipients ranged from 1.1 to 18.3 yr of age with a mean age of 8.5 yr. The mean elapsed time between LT and assessment was 5.8 yr (s.d. = 3.9). Clinical characteristics of patients are presented in Table 2. Mean age by the time of LT was 2.6 yr (s.d. = 3.3). Biliary atresia was the most common diagnosis (n = 96). Parents whose child received a living-related donation were more likely in the group with the highest level of education than parents after a deceased donation (OR = 2.0, CI = 1.1–3.8; v2 = 4.6, p = 0.032). Comparison with the normative sample
We assumed that the level of strain in families with a child who has undergone LT would be comparable to the strain in families with a disabled or chronically ill child (25). Table 3 shows significantly higher impact scores in the IFSdimensions financial impact, coping, and impact on siblings in our sample. In the other dimensions as well as in the Total Score, family strain was comparable to the strain in the normative sample. No differences between fathers and mothers
0.03 to
Relations to psychosocial functioning in children and adolescents after LT
Additionally, significant inter-relations between higher strain in families and more emotional and behavioral disturbances in children and adolescents post-LT were expected. This hypothesis was confirmed by correlational analyses. All domains of the IFS were correlated to the subscales of the SDQ (Table 5). The Total Score of the IFS showed the highest inter-relations with most subscales of the SDQ (r = 0.330 to r = 0.480; p < 0.01). Exploratory analysis IFS
By exploratory analysis, family strain was compared between parents with a child who received a living-related or a deceased donation. The Total Score was significantly higher in parents after deceased than after living-related LT (t (179) = 2, 18; p = 0.031). It should be noted that type of donation and clinical course were interrelated. The risk for a more severe clinical course after a deceased donation was threefold compared to living-related donation (OR = 3.0, CI = 1.6–5.6; v2 = 11.5, p = 0.001). Furthermore, statements of the IFS were identified in which families reported the highest strain. Hence, 66.3% of the sample stated “absolutely true” or “true in most aspects” regarding the item “I am worried about the future of my child (when it has grown up);” 61.4% agreed to the item “Sometimes I feel like we live on a roller coaster: 853
Kaller et al. Table 1. Demographic characteristics of study participants (n = 181) M Child Gender (female) Age at survey (yr) Parents Age Mother (n = 175)* Father (n = 169)* Parents reporting (n = 170)* Mother Father Both Other Marital status (n = 179)* Married or living together Number of siblings (n = 176)* 0 1 2 >2 Household income (month) (n = 153)* 5000 Euro
s.d.
Median
Range
8.5
7.7
4.5
1.1–18.3
36.7 40.2
6.7 6.5
37.0 40.0
21–55 25–61
N
Mother (n = 174)*
%
93
51.4
133 28 8 1
78.2 16.5 4.7 0.6
152
84.9
47 88 27 17
26.0 48.6 14.9 10.5
13 56 54 20 7 5
8.3 36.1 34.8 12.9 4.5 3.2
Father (n = 163)*
Education
N
%
N
%
No graduation Special education school Graduation after 8 or 9 yr of education Graduation after 10 yr of education Graduation after 12 yr of education
7 1 44 70 52
4.0 0.6 25.4 40.2 29.9
5 2 58 56 42
3.1 1.2 35.6 34.4 25.8
Mother (n = 169)*
Father (n = 165)*
Employment
N
%
N
%
Full time Part time Homemaker Other (unemployed, retired)
18 77 61 13
10.7 45.6 36.1 7.7
145 4 1 15
87.9 2.4 0.6 9.1
*Causes for missing data: information was not reported, for example, single mothers did not know the graduation of the father or parents declined to state their family income.
in crisis when my child is acutely ill, OK when things are stable;” 47.5% cannot imagine having further children due to their child’s LT. Age at transplantation was not correlated to the scores of any scale of the IFS (r = 0.00 to r = 0.05; p = 0.97 to p = 0.54). Exploratory analysis interviews
In a semi-structured interview, 40% of the parents reported family problems such as 854
problems with siblings (23.2%), marital or relationship problems (23.2%), child’s restrictions post-LT (33.7%) such as limited capabilities regarding leisure or social life (24%) or medical demands or problems (18%), and financial losses following LT (33.7%), which were mostly due to no return to work of the mother after parental leave (31.8%) or reduced labor time (21.6%) as a consequence of LT. Moreover, 64% would have demanded more psycho-social support by the time of LT.
Family strain after pediatric LT Table 2. Transplant-related characteristics of study participants (n = 181) M Age at LT (yr) Time elapsed since LT (yr) Basic immunosuppression at survey Cyclosporin A Tacrolimus Diagnostic categories Biliary atresia Cholestatic diseases Metabolic diseases Others Type of donation Deceased donation Living related donation* Days on waiting list (n = 175) Days in hospital after LT (n = 158) Rejection†, chronic and acute (n = 167) Re-transplantation† (n = 174) Complications†,‡ (n = 158) Surgical revisions†,§ (n = 153)
s.d. 2.6 5.8
Median
3.3 3.9
103.3 43.7
Range
1 5.0
123.0 25.6
N
%
131 50
72.4 27.6
98 42 17 23
54.4 23.3 9.4 12.8
113 68
62.4 37.6
72 26 66 34
43.1 14.9 41.8 22.2
0–13.3 1.0–15.7
57 39.0
0–762 14–214
*Donors were mothers (n = 34), fathers (n = 31), and others (n = 3). † Until one yr post LT. ‡ For example, cholestasis, cholangitis, portal hypertension. § For example, due to biliary leakage, hepatic artery thrombosis, portal vein thrombosis or stenosis, impaired organ perfusion, intestinal perforation, intraabdominal bleedings.
Table 3. Impact on family scores (IFS) in the sample of families with a child or an adolescent after LT in comparison with a normative sample of families with a disabled or chronically ill child (25)
Dimensions of the IFS
LT-group M (s.d.)
Normative group M (s.d.)
Daily/social impact Personal impact/worries about future Financial impact Impact on siblings Impact on coping Total score
1.95 (0.67) 2.21 (0.67) 1.86 (0.89) 1.84 (0.76) 1.92 (0.65) 1.98 (0.59)
1.98 (0.69) 2.13 (0.71) 1.71 (0.74) 1.55 (0.56) 1.67 (0.69) 1.93 (0.54)
t 0.56 1.59 2.28 4.38 5.11 1.22
p
d
0.58 0.11 0.02
