Behavioral problems of adolescents with chronic physical illness: A comparison of parent-report and self-report measures

Behavioral problems of adolescents with chronic physical illness: A comparison of parent-report and self-report measures
of 7
All materials on our website are shared by users. If you have any questions about copyright issues, please report us to resolve them. We are always happy to assist you.
Related Documents
  European Child and Adolescent Psychiatry, Vol. 4, Issue 1, January 1995, pp. 14-20 9 Hogrefe & Huber Publishers Behavioral Problems of Adolescents with Chronic Physical Illness: A Comparison of Parent Report and Self Report Measures Mike Stawski*, Judith G. Auerbach**,***, Miriam Barasch**, Yaacov Lerner**, R.obert Zimin****, Mervin S. Miller**** This study reports on the extent of behavior problems in Israeli adolescents suffering from chronic illness. A com- parison was made between parent-reported and self-reported behavioral symptomatology using the Child Behavior Checklist (CBCL) and the Youth Self-Report (YS1L). 103 outpatients, aged 11-16 years, suffering from cystic fibrosis (CF), asthma, or hematological/oncological conditions were assessed. Healthy adolescents and adolescents referred for psychiatric assessment comprised the comparison groups. Parent- and self-reports were significantly positively correlated in each group (all chronically ill children r = .22; Healthy group r = .27; psychiatric group r = .50), but the correlations were particularly low (and non-significant) in younger adolescents with hematologi- cal/oncological conditions or HCF., pointing to the need for physicians to include parents' and adolescents' view- points in their assessments of these adolescents' psychosocial state. The mean number of parent-reported and self- reported behavior problems in the il/ness groups was no different from that of the Healthy group but significantly lower than that of the psychiatric group. Keywords: Chronic Illness, Adolescence, Behavioral Problems, CBCL ntroduction Much effort over the last 20 years has gone into examining the psychological concomitants of chronic physical illness in young peopl e . Overall, it may be concluded that the risk for psychosocial disturbance in chronically ill children and adoles- cents is roughly doubled (Pless & Nolan, 1991; Lavigne & Faier-Routman, 1992). The measured prevalence of psychological dis- turbance is to a degree influenced by the choice of informant, both in clinical and in research set- tings; for example, Costello and Edelbrock (1985) found evidence that pediatricians are sensitive to behavioral and emotional problems which parents identify in their children, but less so to problems the children identify in themselves. The general issue of parent-informant/child-in- formant variance has been rarely addressed until recently, in spite of the fact that it has implications not only for child psychiatric research and practice, but also for clinical pediatrics (Achenbach et al., 1987; Angold et al., 1987; Mokros et al., 1987; Verhulst et al., 1987; Ivens & Rehm, 1988; Phares et al., 1989; Verhulst & van der Ende, 1992). It has been noted that adolescents are indispensable informants about their behavior problems because many of the problems they experience go un- noticed by their parents (Verhulst & van der Ende, 1992). In general, they report more psychological and behavioral symptoms about themselves than * Abarbanel Hospital, Bat Yam, Israel * * Falk Institute for Mental Health and Behavioral Studies, JDC Hill, P.O.B. 3489, Jerusalem 91034, Israel * Department of Behavioral Sciences, Ben Gurion University * * Sheba Hospital, Tel Hashomer, Israel l~eceived: June 18, 1993 Accepted for publication: July 22, 1994 14  Behavioral Problems of dolescents with Chronic Physical Illness their parents report about them, with the symptom count discrepancy increasing with age (Verhulst & van der Ende, 1992). Overall, parent-child correlation is about .28 to .33 for normal samples and about .36 for psychiatrically referred children (Achenbach et al., 1987; Achenbach, 1991 a). There is evidence that the relationship between child and parent reports may differ according to the population assessed and the direction and the extent of the relationship be- tween parent-reported and self-reported behavior problems for adolescents with chronic physical ill- ness is not known; such knowledge would im- prove both research methodology and clinicians' ability to identify especially at-risk adolescents. The principal aim of this study was to investigate this issue in the chronically ill adolescent popula- tion while taking into consideration the effect of age of adolescent on between-informant correla- tion. In addition, the extent of behavioral and emotional problems in ill adolescents as compared to healthy and to psychiatrically ill adolescents was evaluated to determine if parent and youth reports on an Israeli sample of chronically ill adolescents were comparable to those found in other cultures. ethod sample Three groups of adolescents aged 11 through 16 years were examined. The Chronically Ill group was composed of 102 adolescent outpatients of Sheba Hospital, Israel, who were suffering from asthma ( Asthma subgroup), cystic fibrosis ( CF subgroup), or a hematological or oncological con- dition ( Hematology subgroup); of the latter, 62% were suffering from malignant conditions, usually acute lymphocytic leukemia, and 38% from non-malignant conditions. The Psychiatric group consisted of 62 first-dine attenders at out- patient clinics of child and adolescent psychiatry in Israel. The Healthy group consisted of 120 adoles- cents with no known history of medical or psychiatric problems, drawn from a large normal population data base of the Falk Institute and matched for age, sex, and years of parental educa- tion (Zilber et al., 1994). Information on child age, sex ratio, and years of parental education by subject group appears in Table 1. easures a) Child Behavior Checklist A Hebrew translation (Auerbach & Lerner, 1991) of the Child Behavior Checklist (CBCL) (Achen- bach, 1991b) served as the measure of parent- reported behavioral and emotional problems in the adolescents. This questionnaire is designed for rating in a standardized format the behavioral problems and competencies of children aged 4 through 18. Responses to the Social Competence section were not analyzed for the study. One hun- dred and eighteen behavior problem items ask specific questions about difficulties in the behavioral and emotional realms. The total behavior problems score is the sum of all positive responses for a child. The behavior problem items can be grouped into two broad-band categories of Internalizing and Externalizing type problems corresponding to overcontrolled and undercontrolled syndrome groupings. Categorization of items was based on the Table 1. Demographic data. Group Characteristics Number of Number questionnaires Age of Male: Female Mean Years of education Group of returned* subjects Distribution Father Mother subjects CBCL YSR. M s.d. M s.d. M s.d Chronically I11 102 102 97 13.5 1.8 62:41 1.5:1) 12.1 3.6 12.1 3.0 a) H.C.F. 22 22 20 13.7 1.8 13:9 (1.4:1) 13.0 4.0 12.4 2.7 b) Hematology 39 39 37 13.6 1.9 22:I7 (1.2:1) 11.4 3.5 12.0 3.2 c) Asthma 41 41 40 13.2 1.6 27:15 1.8:1) 12.3 3.3 12.0 3.0 Psychiatric 62 62 62 13.2 1.7 43:20 2.1:1) 11.7 3.2 11.8 2.9 Healthy 120 120 120 13.5 1.7 75:45 (1.7:1) 12.2 3.4 12.3 3.0 Note. For each subject, CBCL, YSIL or both were obtained. ~2 analysis revealed no significant differences between groups for mean age, sex distribution or mean years of parents' education (p _< 0.05). 15  M stawslcJ et al 1991 revision of the Child Behavior Profile (Achen- bach, 1991 b). b) Youth Self R.eport A Hebrew version of the Youth Self-Report (YSR.) (Achenbach, 1991 a) was used to obtain information on behavioral and emotional problems of the adolescents as perceived by themselves. The ques- tionnaire is designed for completion by adolescents aged 11 through 18 years. In addition to the Social Competence section, it contains 102 behavior pro- blem items which correspond to CBCL items, with an additional 16 items of a socially desirable nature. The CBCL and YSR were selected for this study because of their reliability and validity, the parallel nature of the two versions, and their wide use in previous studies on children and adolescents which allows for a cross-study comparison. rocedure Adolescent patients and their parents were asked to participate in the study during their regular appoint- ments to their pediatric or psychiatric outpatient departments. They were given information on the purpose of the study, and their consent to participate was obtained. One of the parents was then asked to complete a CBCL form on the adolescent in question, and another form on the sibling closest in age. (Results on the sibling study will be reported in a separate paper.) Simultaneously, the adolescent was asked to complete a YSR form on him-/herself. 81% (N = 230) of CBCL questionnaires were com- pleted by mothers, 15% (N = 43) by fathers, and 4% (N = 11) by either both together or by another caretaker. problem raw scores, between the internalizing scores, and between the externalizing scores, separately for each group. As may be seen in Table 2, parent-report and self-report scores were sig- nificantly correlated in each of the three groups. When the three subgroups of the Chronically IU group were examined separately, only the correla- tions for the Asthma group reached statistical sig- nificance. For all groups, a larger percentage of adolescents reported more symptomatology about themselves than their parents reported about them. The percentages were 78% for the Healthy group, 77% for the Ill group, and 62% for the Psychiatric group. In order to investigate whether the strength of the correlation between CBCL and YSR scores varies as a function of age of adolescent and group membership, Pearson's correlation coefficients were computed separately for those adolescents 13 years and younger and for those 14 years and older. Table 2 presents these correlations. The correla- tions for total behavior problem, externalizing and internalizing scores are significant and moderately high for both younger and older adolescents in the Psychiatric group. For the Healthy group the de- gree of agreement between adolescent and parent decreases with age. Only the correlations between CBCL and YSR total behavior problems scores and externalizing scores in the young adolescent group reach statistical significance. The picture for the Ill group is more complex since type of illness Table 2 Correlations between CBCL and YSR totalbehav- ior problem scores, internalizing scores and externalizing scores Groups Total Behavior Internalizing Externalizing Problem Scores Scores Scores Chronically IU .22* Younger .04 Older .43** Results a) HCF -.07 Younger -.24 Older .21 Because of the possibility (Perrin et al., 1991) that b) Hematology .20 the somatic symptoms of the CBCL might bias ounger -.25 results in ill populations, all analyses were repeated Older .56** with the somatically-related items being excluded c) Asthma .44** Younger .42* (see appendix). No results were substantially af- Older .46 fected; therefore, the results of these analyses are Psychiatric .50** not reported here. Younger .51 * To examine the extent of correlation between Older .47** Healthy .27** parents' reports of behavior problems and their ounger .36** adolescent children's self-reported behavior prob- Older .19 lems, Pearson correlation coefficients were calcu- lated between the CBCL and YSR total behavior .32** .32** .08 .06 .43 .54* -.01 -.31 -.18 -.26 .25 .60 .31 .26 -.21 -.01 .59* .46* .49** .35* .45* .20 .50* .58* .45** .52** .41 .52* .51 .54** .24** .35** .31 .48** .20 .25 p < .05 p < .01 6  Behavioral Problems of dolescents with Chronic Physical Illness affects both the direction and size of the correla- tion. The correlation between CBCL and YSIL scores for young adolescents with CF and hema- tology conditions are negatively, albeit non-sig- nificantly, correlated while the correlations in the younger Asthma group are positive and similar to those found in the comparison groups with statis- tically significant correlations for total behavior problems and internalizing problems. For older adolescents, the correlations are positive and the size of correlation in the Hematology and Asthma groups is similar to that in the Psychiatric group. The correlations in the CF group failed to reach significance in either age group perhaps because of the small number of subjects in each group. Groups differences on total behavior problem, externalizing and internalizing scores on the CBCL and YSR were analyzed using normalized scores instead of raw scores (see Table 3). The CBCL and YSR total behavior problem scores for the Chroni- cally Ill group did not differ from the Healthy group and both were significandy less than the scores for the Psychiatric group (CBCL: F(2,281) = 30.5, p = .000; YSR: F(2,276) = 17.9, p = .000). For the CBCL and YSP,. internalizing scores, the Chroni- cally II1 and Healthy groups each differed sig- nificandy from the Psychiatric group (CBCL: F(2,281) = 23.0, p = .000; F(2,276) = 20.4, p = Table 3 Means and standard deviations of CBCL and YSR. total behavior problems, internalizing and externalizing T scores by group. Problem Chronically IH Psychiatric Healthy Scores Total C.E HematologyAsthma Total CBCL Mean 52.1 55.0 51.5 51.2 64.4 49.8 SD 12.4 11.9 13.9 11.2 15.0 8.4 N 102 22 39 41 62 120 YSR. Mean 51.7 52.7 52.0 51.0 60.5 52.0 SD 9.5 7.5 10.2 10.0 12.7 10.0 N 97 20 37 40 62 120 Internalizing CBCL Mean 54.5 57.4 53.3 54.0 65.8 51.1 SD 15.3 16.3 13.4 14.1 18.7 10.4 YSR. Mean 52.8 52.6 53.5 52.4 62.1 51.1 SD 11.1 8.9 12.5 10.9 14.3 9.6 Externalizing CBCL Mean 49.7 52.7 48.5 49.5 58.3 49.9 SD 10.0 11.3 10.0 9.2 13.8 9.1 YSR. Mean 50.0 52.6 49.3 49.3 56.2 51.6 SD 8.8 10.5 8.0 8.6 12.1 10.1 .000). On both the CBCL and YSR, the externaliz- ing scores of the Psychiatric group were significandy higher than the scores for the Chronically IU and Healthy groups (CBCL: F(2,281) = 15.5, p = .000; YSIL: F(2,276) = 7.3, p = .001). The pattern of significant differences did not change when the groups were split into younger and older adoles- cents. For all comparisons, significant differences were due to the higher scores of the Psychiatric group. iscussion The principal aim of this study was to assess the relation between parent-reported and self- reported behavioral and emotional symptoms in chronically ill adolescents, in view of the fact that these young people are generally thought to be at increased risk for psychological disturbance. It was found that the degree of correlation between chronically ill adolescents' reports and their parents' reports was similar to that found in our Healthy comparison group and in a 1987 meta- analysis of a large number of studies (Achenbach et al., 1987) on children and adolescents. Like Ver- hulst van der Ende (1992), the majority of adolescents in each group reported more problems about themselves than their parents did about themselves. The importance of examining data from a devel- opmental perspective was highlighted by the find- hag that the degree and direction of the correlations between parent report and self report differed by age of adolescent. Between-informant correlations were lower for younger than for older adolescents suffering from CF or from hematological/on- cological conditions, with the correlations for the younger adolescents in fact being negative. These adolescents differed from those suffering from asth- ma and psychiatric disturbances (for whom there was no age gradient) and from the Healthy adoles- cent control group (for whom the age gradient went in the opposite direction). The finding for the Heal- thy group is in accordance with some (e. g., Achen- bach, 1991 a) though not all (e. g., Verhulst van der Ende, 1992) previous work. The findings for the ill adolescents, however, were surprising. Only one previous study (Edelbrock et al., 1986) has demonstrated a positive age gradient for between- rater correlations through adolescence and the sam- ple for this study was psychiatrically disturbed chil- dren and adolescents. The significance of a positive 17  M Stawski et al age gradient for chronically-ill adolescents, which has not previously been reported, is unclear. A pos- sible explanation is that the normal separation from parents that occurs in adolescence and which is in- voked as the explanation for increased parent- report/self-report discrepancy with age (Verhulst & van der Ende, 1992), often does not occur in adoles- cents with life-threatening illness (C.F. and hema- tology/oncology), thus maintaining the parents in a position of relative expertise about their offsprings' behavioral symptoms. Even so, the size of correla- tion is such as to indicate that any psychosocial as- sessment of the adolescent must include his own viewpoint, and this is particularly crucial for younger adolescents with life-threatening illnesses. This finding gains more salience when seen against the background of evidence of pediatricians' failure to identify the majority of children and adolescents with psychosocial problems (Cadman et al., 1987; Costello et al., 1988; Gortmaker et al., 1990). Although it was not the main thrust of this study, the opportunity to compare chronically ill Israeli adolescents to their healthy and psychiatri- cally-referred age-mates was taken in order to as- sess whether our findings would replicate findings reported on similar groups in other cultures. The finding of the chronically ill adolescents being in- distinguishable from the healthy ones on the CBCL and YSR total behavior problem scores while having significantly fewer behavior prob- lems than psychiatrically-referred adolescents con- tradicts the conclusions of most recent reviews on the subject (e. g., Eiser, 1985; Pless & Nolan, 1991; Lavigne & Faier-Routman, 1992), which have found that children and adolescents with chronic illnesses are at greater risk than healthy adolescents of developing behavioral/emotional problems though the majority do not do so. One possible explanation for this contradiction is the fact that all the ill adolescents in the present study sample had ready access to psychosocial counseling as part of their clinic service may have acted to reduce somewhat the incidence of behavior problems as compared to epidemiological samples on which the above reviews based their conclusions. If the CBCL and YSR had been completed when these chronically ill adolescents first came to medical at- tention and before psychosocial intervention, their level of reported behavior problems might have been significantly higher than that of the general population. Generalizability of the findings and conclusions of the study could be called into question because of the grouping together of different illnesses and because of the sample source. As for the first of these two issues, though there is some evidence (e. g., Rutter et al., 1970) that diagnosis has a major influence on the prevalence of disturbance, more recent evidence (WaUander et al., 1988) suggests that this may not be the case. However the issue cannot be considered closed, and thus caution must be exercised when ex- trapolating from our findings. Concerning the issue of sample source, two fac- tors must be considered. The first, which was dis- cussed above, is that all the ill adolescents were hos- pital outpatient attenders, and thus might not be representative of their diagnostic groupings and children who have just come to medical attention. The second factor which needs to be taken into account is the extent to which cultural factors may play a role in the development of emotional/behav- ioral problems in the chronically ill. There is evidence of cultural differences in the way chronic illnesses are perceived (Groce & Zola, 1993) and these differences in perception may influence the degree to which behavioral/emotional problems will develop as a result of these conditions. In summary, it was found that the correlation between parent-report and self-report of behavior- al symptoms in chronically ill adolescents is fairly low, especially among young adolescents suffering from life-threatening illnesses. Thus it is important to obtain data on these youths' psychological status both from themselves and from their parents. cknowledgment We wish to thank the staff of the pediatric pul- monology and hematology services of Sheba Hos- pital. R~sum~ Cette &ude rapporte les probl~mes de comporte- ment d'adolescents isratliens souffrant de maladie chronique. Une comparaison fut faite entre les rap- ports &ablis par les parents et la symptomatologie comportementale auto-tvalute au moyen de la Child Behavior Checklist (CBCL) et la Youth Self-Report (YSR). 103 patients de la consultation ~gts de 11 ~ 16 ans souffrant de mucovicidose (CF) d'asthme ou de pathologie h~matologique ou on- cologique furent ~valu~e. Des adolescents sains et 18
Similar documents
View more...
Related Search
We Need Your Support
Thank you for visiting our website and your interest in our free products and services. We are nonprofit website to share and download documents. To the running of this website, we need your help to support us.

Thanks to everyone for your continued support.

No, Thanks

We need your sign to support Project to invent "SMART AND CONTROLLABLE REFLECTIVE BALLOONS" to cover the Sun and Save Our Earth.

More details...

Sign Now!

We are very appreciated for your Prompt Action!