Brief Report: Development and Pilot Testing of a Coping Questionnaire for Children and Adolescents With Chronic Health Conditions
Institute and Policlinics of Medical Psychology, University Clinic Hamburg-Eppendorf
All correspondence concerning this article should be addressed to Corinna Petersen, Center of Psychosocial Medicine, Institute and Policlinics of Medical Psychology, University Clinic Hamburg-Eppendorf, Martinistr. 52, S 35, 20246 Hamburg, Germany. E-mail: copeters{at}uke.uni-hamburg.de.
 |
Abstract |
|---|
 Top Abstract Development of the CODI...Pilot-Testing of the CODIResultsDiscussionReferences |
|---|
Objective The current paper describes the development, pilot testing, and item reduction process of a coping questionnaire for children and adolescents with chronic health conditions. Methods A pilot test with 188 children and adolescents was conducted in six European countries. Included in the test were children and adolescents (ages ranging 8–18 years) with various chronic health conditions—asthma, epilepsy, diabetes, arthritis, atopic dermatitis, cerebral palsy, or cystic fibrosis. Based on the focus groups with the children and adolescents and on expert consulting, items were developed and translated into the participants’ respective languages. Data were analyzed according to predefined psychometric and content criteria. Results Analyses resulted in a selection of 29 out of 50 items for the final coping questionnaire with six domains: Acceptance
= .83, Avoidance = .72, Cognitive–Palliative = .69, Distance = .70, Emotional Reaction = .82, and Wishful Thinking = .81. Conclusions This study developed a short reliable international questionnaire to assess coping strategies of children and adolescents with chronic health conditions. Ongoing research will investigate the validity of this new coping questionnaire. Key words: adolescents; children; chronic condition; coping; pediatric.
The way in which children and adolescents cope with chronic health conditions is considered as an increasingly important predictor of health in clinical and psychosocial research (Boekaerts & Röder, 1999; Fields, L., & Prinz, R. J. (1997)
). Questionnaires for children and adolescents were developed to assess their coping with general stress, health-related stress, or specific symptoms such as pain. With regard to the pediatric field, the KIDCOPE (Spirito, Stark, Gil, & Tyc, 1995) and the Coping Responses Inventory (Ebata & Moos, 1991) are measures that were tested in different clinical populations (e.g., rheumatic diseases, depression, diabetes). Current coping measures consist of items that were either selected according to a theoretical framework or generated from interviews with children in a specific country (Connor-Smith, Saltzman, Thomsen, & Wadsworth, 2001). However, no coping questionnaire for children and adolescents with chronic health conditions has been developed in a simultaneous cross-national approach, starting with the perspective of children and adolescents. Such a measure is needed because a cross-national approach is a first step to guarantee the equivalence of questionnaire content and structure across countries. The advantage of such a measure is the possibility of cross-national comparisons of coping strategies and, consequently, a better understanding of how children deal with their illness. So far, few cross-cultural comparison studies have been conducted. Thus, no final conclusion can be drawn whether coping differences across cultures exist. The work presented here was based on the European Union–funded project DISABKIDS (Bullinger, Schmidt, Petersen, & DISABKIDS Group, 2002). In this study, quality-of-life questionnaires were developed with a stepwise procedure: development, testing, and implementation. As an ancillary project within this study, a coping measure for children and adolescents (ages 8–18 years) with chronic health conditions was developed following the DISABKIDS methodology. The present report describes the development, pilot testing, and psychometric analyses of the coping questionnaire. |
Development of the CODI Coping Questionnaire |
|---|
|
TopAbstractDevelopment of the CODI...Pilot-Testing of the CODIResultsDiscussionReferences |
|---|
Focus Groups
The aim of the focus group work that took place simultaneously in all seven countries was to collect information about children’s and adolescents’ perceptions of their illnesses and their views concerning their quality of lives. A focus-group manual delivered the general outline. Individual interviews were conducted as a second option if not enough children participated in a focus group or if the child could not take part in a group because he or she was too young or had cognitive deficits. Potential participants were identified by a physician. Their parents were contacted by letter and informed about the aim. If they were willing to participate, an appointment was suggested. Inclusion criteria were as follows: the consent to participate, an age between 4 and 18 years old, the ability to understand questions, and a chronic health condition. The particular conditions were chosen because of the expertise of the participating centers and the aim to include a variety of health conditions representing aspects of a chronic health condition (e.g. restriction in mobility, visibility, severity). A total of 154 children and adolescents either took part in focus groups (n = 122) or were interviewed (n = 32)—that is, 37 children with asthma; 18, diabetes; 20, epilepsy; 28, atopic dermatitis; 23, arthritis; 10, cystic fibrosis; and 18, cerebral palsy. Geographically, 16 children from Austria participated; 11, France; 54, Germany (two centers); 27, Greece; 23, the Netherlands; 15, Sweden; and 10, the United Kingdom. The groups were composed of up to 6 children or adolescents of similar ages with the same chronic health condition. They were conducted in a quiet room and moderated by a researcher experienced in working with children. The aim was to stimulate a discussion about the children’s perception of their illness and their quality of lives. The moderator posed direct questions only when the discussion stopped; at that point, questions asked included those such as "How does your illness affect you at school/home?" and "What kind of things keep you healthy?" The focus group discussions or interviews took from 30 min up to 90 min and were audiotaped. For item generation, statements of the children and adolescents were transcribed and documented. Any statements referring to coping were grouped together. All statements were then translated into English, sent to the study center, and merged into one cross-national statement pool. In sum, 310 statements were categorized as belonging to the construct of coping with a chronic health condition.
Item Generation
Items for the new questionnaire were derived from focus group data. Items of other questionnaires were not included, because the applied methodology focuses on a bottom-up development approach starting with the children’s perspectives. The 310 statements were included in a card-sorting procedure with the aim to reduce the number of items and to produce coping scales. Statements were printed out, affixed on pieces of card, and read aloud to three researchers experienced in the field of pediatric psychology. If the statements shared a common feature, they were put onto the same pile. This procedure was continued until every statement was put on a particular pile. Statements were removed if they were repeated in the statement pool or semantically equivalent. For each pile, a term was chosen that characterized the content of the statements. The card-sorting process resulted in a selection of 50 statements for the pilot test version of the coping questionnaire. In accordance with their content, the statements were grouped to eight coping domains and rewritten to suit an item format. The coping strategies categorized a priori included one general question referring to how well the child copes with the chronic health condition and the following:
- Spiritual Support: "I believe that faith in God helps me."
- Optimism: "I am optimistic about my illness."
- Acceptance: "I have got used to my illness."
- Activities: "I do things that make me happy."
- Self-Disclosure: "I talk with other people about my illness."
- Expressing Negative Feelings: "I cry."
- Distancing: "I try to ignore my illness."
- Cognitive Restructuring: "I think of worse situations."
Afterward, the English items were forward translated into Dutch, German, Greek, and Swedish (forward translation) by linguists or psychologists and furthermore backward translated into English again to check the comparibility of the language versions. The coping questionnaire was named CODI (Coping with a Disease).
|
Pilot-Testing of the CODI |
|---|
|
TopAbstractDevelopment of the CODI...Pilot-Testing of the CODIResultsDiscussionReferences |
|---|
The CODI questionnaire was pilot tested in a cross-sectional study in six European countries: Austria, Germany, Greece, Sweden, the Netherlands, and United Kingdom. An agreed-on standardized pilot-test manual was followed in the participating centers.
Sample
The coping questionnaire was given to children and adolescents who were being treated in participating centers. The inclusion criteria were (a) the presence of a chronic health condition (asthma, arthritis, epilepsy, cerebral palsy, diabetes mellitus, atopic dermatitis, or cystic fibrosis) diagnosed according to the international classification system (ICD-10) by a physician, (b) an age between 8 and 18 years, (c) the ability to understand questions, and (d) a signed consent form. Physicians were asked to identify the respective children. The centers were obligated to include children with asthma because, according to the DISABKIDS proposal, one condition had be tested across countries, and each center had contact with this patient group. In addition, each country could choose up to three other conditions.
Procedure
Eligible participants were contacted from the centers’ patient files with an information letter that included consent forms; other patients were contacted directly during visits to specialist clinics. If children or adolescents and their parents agreed to participate in the DISABKIDS study, they were invited to the clinic to fill out a set newly developed quality-of-life questionnaires and sociodemographic information. Because the development of the CODI was an ancillary project of the DISABKIDS study, the participants (or their parents) received the CODI only as a take-home questionnaire in a prepaid envelope with the polite request to send it back within 2 weeks.
Instruments
The newly developed CODI questionnaire contained 50 items organized in the eight scales agreed on in the item-generation exercise. The items were expressed as statements in the present tense and in the first person. The instruction for the CODI was "Think of situations when you have been bothered or stressed because of your illness. Below you will find a list of ways in which kids may deal with their feelings in these situations. Please tell us how often you usually do the things or have this kind of thoughts related to your illness."
The response format was a frequency-related 5-point Likert scale ranging from never to always. The overall rating of coping in general ranged on a 5-point scale from very well to not very well at all. In addition, six open-ended questions were added concerning the child’s or adolescent’s general impression about the acceptability of the questionnaire—that is, if the questions and answer categories were understandable or if anything was missing or should be changed. Also obtained was information about socioeconomic status and clinical variables—gender, age, diagnosis, child age at diagnosis, comorbidity, child mental development (parent rating: normal, slow, delayed), years of schooling, days absent from school/kindergarten.
Psychometric Analyses
After having completed a plausibility check, data analyses for the pilot test were carried out using the Statistical Product and Service Solutions (SPSS)(Version 11; Windows) and the Multitrait Analysis Program for psychometric testing on the item and scale level (Hays, Hayashi, Carson, & Ware, 1988). If out-of-range or implausible values were entered into the database, they were recoded as missing values (0.16%). If two answers were coded, one answer was randomly picked (1.03%). The entire sample of children and adolescents was described in terms of socioeconomic status and medical characteristics. Answers from open-ended questions about the CODI were reviewed for problems, such as difficulties in understanding items. Descriptive item statistics included means, standard deviations, percentage of missing items, skew, and changes in the reliability index (Cronbach’s coefficient) if the item was deleted from the scale. Corrected item-scale correlations were calculated. Participants were grouped into two age groups (8–12, 13–18 years) according to a group consensus based on clinical and empirical experience. Mann–Whitney tests were used to explore item-response differences between those groups. Descriptive scale statistics included range, means, standard deviations, and floor and ceiling effects. Scale intercorrelations were examined with the Pearson coefficient. The criteria for item selection were as follows: corrected item-scale-correlation, percentage of missing values, and expert opinion. Reliability (Cronbach’s coefficient) as well as floor and ceiling effects were reassessed for the reduced questionnaire version. Calculated for each scale was a scale fit, the percentage of items that correlate higher with their own than with another scale. The dimensionality of the questionnaire was explored with principal component analyses using the varimax with the Kaiser normalization method for rotation.
|
Results |
|---|
|
TopAbstractDevelopment of the CODI...Pilot-Testing of the CODIResultsDiscussionReferences |
|---|
In total, 332 questionnaires were handed out, and 188 children and adolescents returned the coping measure, for a response rate of 57%. Included in the data analysis were 86 girls (46%) and 102 boys (54%), with a mean age of 12.69 years (SD = 2.63): 38 children with arthritis; 69, asthma; 16, atopic dermatitis; 24, cystic fibrosis; 13, cerebral palsy; 21, diabetes mellitus; 7, epilepsy. Of the respondents, 93 were recruited in Germany; 29, Netherlands; 5, United Kingdom; 29, Greece; 13, Sweden; 19, Austria. France did not participate in this stage of the CODI development due to organizational problems. In sum, 33% of the children had more than one type of illness, and 84% were rated by the parent as showing a normal mental development. In general, 45 (25%) children and adolescents thought that the coping questionnaire was very good; 119 (66%), good; 16 (9%), not good in general.
Item and Scale Characteristics
Of the 50 items, 13 had either poor corrected item-scale correlation, below r = .30, or the internal consistency increased if the item was deleted from that scale. Two items had a problematic percentage of missing values above 5%. The distribution of most of the items was skewed. Age differences (8–12 versus 13–18 years) were predominantly found for items dealing with self-disclosure—for example, talking to peers. The a priori defined Spiritual Support scale showed a high percentage of floor (29%) and ceiling (13%) effects. The highest negative scale correlation was found between the Negative Feelings and Acceptance scale (r = –.54, p = .01); the highest positive correlation was between the Cognitive Restructuring and Self-Disclosure scale (r = .28, p = .01).
Item Reduction
On the basis of the pilot-test results, 13 items with a low corrected item-scale correlation were omitted. Consecutively, one item with a high percentage of missing values was selected for deletion. Finally, experts consented on the further omission of seven items. This decision was based on clinical and conceptual considerations. An item was selected for omission if more than two experts voted to omit it. The group decided to reformulate the item "I don’t think about my illness" in a positive way. In sum, the item reduction resulted in a CODI version with 29 items. An exploratory factor analysis was conducted with the reduced item pool. The aim was to analyze the dimensionality of the 29 items selected for the final CODI version. The structure of the coping questionnaire and the items with the factor loadings are depicted in Table I.
|
The explanatory factor analysis revealed eight factors. Factors 7 and 8 were combined for the final version because the seventh factor contained only two items and the eighth factor one item. These items were reassigned to the factor with the second-highest loading. Two items did not show a high factor loading. These items were reassigned to a scale and are candidates for further omission. The final six scales of the CODI questionnaire are as follows:
- Acceptance: "I am able to manage my illness."
- Avoidance: "I try to ignore my illness."
- Cognitive–Palliative: "I believe that faith in God helps me."
- Distance: "I don’t care about my illness."
- Emotional Reaction: "I cry."
- Wishful Thinking: "I want to stop having my illness."
A higher score on a scale is associated with a high endorsement of the construct—for example, "Distance." The reliability coefficients (internal consistency, ) of the final coping scales ranged from .69 to .83, and the scale fit values ranged from 95% to 100%. Pronounced ceiling effects were detected for Wishful Thinking and pronounced floor effects for the Emotional Reaction scale.
|
Discussion |
|---|
|
TopAbstractDevelopment of the CODI...Pilot-Testing of the CODIResultsDiscussionReferences |
|---|
The growing interest of health care providers and researchers in valid and sensitive outcome measurement has been a driving force for the development of instruments in the pediatric area. The study presented here is intended to contribute to this effort by providing a sound measure for the assessment of coping with a chronic health condition. The final coping questionnaire contains 29 items and six coping strategies with acceptable scale properties for the revised version. In contrast to other existing measures, the CODI was developed cross-nationally and designed for children and adolescents with chronic health conditions, regardless of the specific type of condition. Several limitations regarding the study need to be addressed. First, the number of patients across all health conditions is acceptable, but it is low within the different types of health conditions and countries. Due to such small sample sizes per chronic health condition, differences with regard to country or type of disease effects could not be examined. Although it has been addressed in the literature whether children with different health conditions show different coping strategies (Compas, Connor-Smith, Saltzman, Thomsen, & Wadsworth, 2001
), findings are so far inconclusive. Another critical point is the heterogeneity of samples included in the present study—that is, in terms of numbers per country and health condition. While the heterogeneity causes problems with regard to the generalizability of the findings, it can also be advantageous in terms of sampling a broad range of important strategies. In addition, with increasing age and growing cognitive ability, children employ different sets of coping strategies. Factor analyses thus need to be replicated separately for different age groups. Also, the information about the coping strategies of the children and adolescents are based on self-reporting, which may differ from actual behavior. With regard to the discussion of state and trait theory, the CODI measure focuses on strategies frequently used—that is, habitual responses. Whether these assess state or trait concepts has to be decided empirically, examining stability or variation of coping with regard to specific situations. Future steps will involve a field test of the new measure in a large international sample in which the validity and psychometric performance of the CODI questionnaire will be tested with confirmatory factor analysis. Then, age and potential health condition effects need to be examined. |
Acknowledgements |
|---|
The DISABKIDS study is funded by the European Commission (QLG5-CT-2000–00716). The DISABKIDS project (www.disabkids.de) cooperates closely with the KIDSCREEN project (www.kidscreen.de).
The DISABKIDS Group comprises a coordinating group (Prof. Monika Bullinger, Dr. Corinna Petersen, Dr. Silke Schmidt, Institute of Medical Psychology, University Clinic Hamburg-Eppendorf) and study centers in seven countries: Prof. Michael Quittan, Dr. Othmar Schuhfried, and Dr. Nilouparak Hachemian, Department of Physical Medicine and Rehabilitation, University of Vienna, Austria; Dr. Marie Claude Simeoni and Dr. Audrey Clement, Department of Public Health, University Hospital of Marseille, France; Dr. Ute Thyen. and Dipl.-Psych. Esther Müller-Godeffroy, Department of Pediatrics, Medical University of Lübeck, Germany; Dr. Athanasios Vidalis, assistant professor John Tsanakas, Dr. Elpis Hatziagorou, and Dr. Paraskevi Karagianni, Department of Psychiatry, Hippocratio Hospital, Greece; Dr. Hendrik Koopman and Dr. Rolanda Baars, Department of Pediatrics, Leiden University Medical Center, Netherlands; Dr. John Eric Chaplin, Department of Pediatrics, University Hospital Lund, Sweden; Prof. Mick Power, Dr. Clare Atherton, and Dr. Peter Hoare, Department of Psychiatry, University of Edinburgh, United Kingdom.
Received September 28, 2003; revision received December 9, 2003; revision received February 24, 2004; accepted February 29, 2004
|
References |
|---|
|
TopAbstractDevelopment of the CODI...Pilot-Testing of the CODIResultsDiscussionReferences |
|---|
Boekaerts, M., & Röder, I. (1999). Stress, coping, and adjustment in children with chronic disease: A review of the literature. Disability and Rehabilitation, 21(7), 311–337.[CrossRef][Web of Science][Medline]
Bullinger, M., Schmidt, S., Petersen, C., & DISABKIDS Group. (2002). Assessing quality of life of children with chronic health conditions and disabilities: A European approach. International Journal of Rehabilitation Research, 25, 197–206.
Compas, B. E., Connor-Smith, J. K., Saltzman, H., Thomsen, A. H., & Wadsworth, M. E. (2001). Coping with stress during childhood and adolescence: Problems, progress, and potential in theory and research. Psychological Bulletin, 127(1), 87–127.[CrossRef][Web of Science][Medline]
Connor-Smith, J. K., Compas, B. E., Wadsworth, M. E., Thomsen, A. H., & Saltzman, H. (2000). Responses to stress in adolescence: Measurement of coping and involuntary stress responses. Journal of Consulting and Clinical Psychology, 68(6), 976–992.[CrossRef][Web of Science][Medline]
Ebata, A., & Moos, R. (1991). Coping and adjustment in distressed and healthy adolescents. Journal of Applied Developmental Psychology, 12, 35–54.
Fields, L., & Prinz, R. J. (1997). Coping and adjustment during childhood and adolescents. Clinical Psychology Review, 17(8), 937–976.[CrossRef][Web of Science][Medline]
Hays, R. D., Hayashi, T., Carson, S., & Ware, J. E. (1988). User’s guide for the Multitrait Analysis Program (MAP) (Rand Corporation Report No. N-2786-RC 1988). Santa Monica, CA: Rand.
Spirito, A., Stark, L. J., Gil, K. M., & Tyc, V. L. (1995). Coping with everyday and disease-related stressors by chronically ill children and adolescents. Journal of the American Academy of Child and Adolescent Psychiatry, 34(3), 283–290.[CrossRef][Web of Science][Medline]
Wallander, J. L., & Varni, J. W. (1998). Effects of pediatric chronic physical disorders on child and family adjustment. Journal of Child Psychology and Psychiatry, 39(1), 29–46.[CrossRef][Web of Science][Medline]
CiteULike 
Connotea 
Del.icio.us What's this?
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||