Health Promotion International, Vol. 14, No. 3, 251-260,
September 1999
© Oxford University Press 1999
Health survey research on children with physical disabilities in Canada
Department of Public Health Sciences, University of Toronto, Toronto, Ontario, 1 Bloorview MacMillan Centre, Toronto, Ontario and 2 Graduate Program in Clinical Psychology, University of New Brunswick, St John, New Brunswick, Canada
Address for correspondence: Ilze Kalnins Department of Public Health Sciences McMurrich Building University of Toronto Toronto, Ontario Canada M5S 1A8 E-mail: ilze.kalnins{at}utoronto.ca
 |
SUMMARY |
|---|
 TOP SUMMARY INTRODUCTIONRESULTSDISCUSSIONACKNOWLEDGEMENTSREFERENCES |
|---|
Knowledge about health behaviours of children with physical disabilities is important in health promotion efforts to prevent the development of secondary conditions. To this end, we discuss the feasibility of including children with physical disabilities, e.g. spina bifida and cerebral palsy, in mainstream children's health behaviour surveys, e.g. the international Health Behaviour of School-aged Children: A WHO Cross-national Survey (HBSC). Our findings showed that the survey questions about lifestyles, peers, family and schools were relevant to these children. However, recruitment of a representative sample and survey administration required considerable adaptation to recommended protocol. We propose that for children with moderate or severe disabilities, who have low reading skills or motor problems, special substudies linked to national health behaviour surveys should be carried out. However, for a more population-based approach, the utility of routinely including questions about physical disabilities and other health problems in national and international health behaviour surveys should be investigated.
Key words: children; health behaviours; lifestyles; physical disabilities
|
INTRODUCTION |
|---|
|
TOPSUMMARYINTRODUCTIONRESULTSDISCUSSIONACKNOWLEDGEMENTSREFERENCES |
|---|
Advances in medicine have enabled more children with physical disabilities, e.g. cerebral palsy, spina bifida, rheumatoid arthritis and muscular dystrophy to survive into adulthood. However, longer lifespans have also placed these children at risk of secondary or additional conditions including heart disease, stroke, respiratory problems and emotional disorders, which result in deterioration of health status, functional capacity and quality of life (Marge, 1988
; Office for Disabled Persons, 1988, 1990; Blum, 1991, 1992; Pope and Tarlow, 1991; Pope, 1992). Many secondary conditions are related to health behaviours including tobacco, alcohol and drug use, unhealthy diets, and insufficient physical activity (Bullock, 1993; Connor-Kuntz et al., 1995; Kessler and Klein, 1995).
Given the risk that children with physical disabilities face from the possible development of secondary conditions, understanding their health behaviour and its psychological and social correlates is important (Wood, 1994). Comparisons of children with and without physical disabilities on standard health indicators would provide important new information to identify areas in which health promotion interventions should be implemented. Our research has demonstrated that children with physical disabilities are less likely than their counterparts without physical disabilities to smoke, drink and use drugs. But, they are more likely to have unhealthy eating patterns, and engage in sedentary leisure activities, e.g. watching television at the expense of regular physical exercise (Steele et al., 1996). Children with physical disabilities have markedly more positive attitudes towards parents, teachers and classmates (Stevens et al., 1996). These differences suggest that health promotion interventions may need special tailoring to be relevant to children with physical disabilities.
Among children without physical disabilities, information about health behaviour and its social and psychological correlates is collected routinely by means of national and international surveys (e.g. King and Coles, 1992; King et al., 1996). The information obtained is used to monitor changes in trends in health behaviour, and to identify important foci for the development of health promotion interventions at individual, community and policy levels. Children with physical disabilities are routinely excluded from such mainstream health behaviour surveys. This exclusion may reflect a lack of awareness that lifestyle issues are important to them. It may also reflect a perception that motor, visual or verbal limitations would preclude their participation.
As part of our research programme on the health behaviour of children with physical disabilities, we have studied the feasibility of having them participate in a mainstream international children's health behaviour survey known as the Health Behaviour in School-aged Children: A WHO Cross-national Survey (HBSC) (Wold et al., 1994). The purpose of this survey is to assess health behaviour and its social and psychological correlates among 11- to 16-year-old children. Since its inception in 1982, four separate surveys have been carried out at 4-year intervals in an increasing number of countries. In 1993— 1994, 24 countries in Europe participated as well as Israel and Canada. For the current round in 1997—1998, more countries are expected to join. The administration of the HBSC is guided by a common protocol to which all countries must adhere (Wold et al., 1994). Approximately 1300 children in each age group participate. The data bank for the HBSC study is at the University of Bergen.
The HBSC has two primary goals. The first is to develop national-level information on health and health behaviour of young people and to monitor trends over time. This is important for identifying foci for health promotion interventions and policy development. The second is to form international networks of interdisciplinary research groups interested in studying children's health behaviour. To this end, research findings are disseminated through national and international reports (e.g. King and Coles, 1992; King et al., 1996), papers in academic and professional journals, and conferences.
In this paper, we describe the research process of including children with physical disabilities in this survey as a special adjunct group to the Canadian HBSC survey in 1993—1994. We describe the adaptations made in sampling, survey content and administration format in order to do so. We also discuss the potential biases introduced by these adaptations and their implications for a routine inclusion of children with physical disabilities in mainstream health behaviour survey research.
|
RESULTS |
|---|
|
TOPSUMMARYINTRODUCTIONRESULTSDISCUSSIONACKNOWLEDGEMENTSREFERENCES |
|---|
Sampling
The recommended HBSC procedure is cluster sampling in which the cluster is the school class. In the 1993—1994 survey, classes were to be selected systematically from a list of all classes or from a list of all schools in each country, or region, in which the survey was administered. Sample size was set at a minimum of 1300 children in each of the three age groups 11—12, 13—14 and 15—16 years of age.
For our study, we only attempted to include children with physical disabilities residing in the province of Ontario. Even so, the recommended sampling procedure was not possible. Within the education system in Ontario, children with physical disabilities receive schooling in a variety of locales including special orthopaedic classes in the regular school system, special classes in a treatment centre, or they may be completely integrated into a regular classroom. No central registry exists. Thus, neither cluster sampling, as suggested by the HBSC protocol, nor random sampling were possible.
To locate children who could participate in the survey, a partnership was formed with the Ontario Association of Children's Rehabilitation Services (OACRS) which includes 19 rehabilitation centres: Bloorview MacMillan Centre in Metropolitan Toronto and 18 centres located throughout the province. Children were recruited from Bloorview Macmillan Centre and from 16 of the 18 provincial centres. Bloorview MacMillan Centre is the largest paediatric rehabilitation centre in Ontario providing comprehensive in-patient and out-patient services. Even within this facility, the process of recruiting children for the study from among those who use the Centre's rehabilitation services was complex. To arrange for children to complete the survey during their regular clinic visits, assistance was required from several departments including the School Health Support Services, Health Data Resources, and the Outpatient staff of the various treatment services of the institution.
Through its outreach programmes, Bloorview MacMillan Centre provided the necessary links with the Metropolitan Toronto School Board. Once children were located in various types of schools and classes (e.g. special classes in the regular school system, integrated regular classrooms), consent had to be sought hierarchically from the School Board, principal of the school, the classroom teacher, and finally the parents and children. The classroom teacher facilitated obtaining consent from parents and children, and the scheduling of survey administration dates and times.
The rehabilitation centres outside of Metropolitan Toronto conducted recruitment locally. Surveys and detailed instructions were mailed to each centre and followed up by telephone. The procedures for recruitment and consent approximated those followed at Bloorview MacMillan Centre.
Through these recruitment procedures, 319 children (11—16 years of age) with physical disabilities from across the province of Ontario participated in the HBSC survey. Of these, 188 children were recruited at Bloorview MacMillan Centre and 131 children from the 16 other centres. The physical disabilities of the children included: spina bifida, cerebral palsy, muscular dystrophy, amputations, arthritis and a variety of other conditions (Table 1). The physical disability level of the children was moderate to severe as seen by their need for assistance in daily living (Table 2).
|
|
Few children refused to participate during their clinic visit at Bloorview MacMillan Centre. Reasons for non-participation included illness, lack of time during the clinic visit and lack of interest. Within the schools in Metropolitan Toronto, agreement to participate among children in the special orthopaedic classes was 65%, while among those integrated into regular classes it was 43%. Reasons for non-participation included lack of interest or the child did not return the consent form, poor English comprehension, busy adjusting to a new school, illness, change of address, and the disability was not a physical disability. No information about refusal rates or reasons for refusal were available from the OACRS centres because no-one could undertake to follow up families who did not respond to an invitation to participate or who did not return a consent form.
In summary, recruitment of children was complex and the sample obtained can be considered a convenience sample, although the range and proportion of disabilities reflects what is seen in rehabilitation settings. The sample excluded children with a physical disability who were fully integrated in regular classrooms and who rarely, if ever, used services offered by the rehabilitation centres.
Relevance of survey questions
The HBSC survey includes questions on: demographic characteristics (age, gender, family structure, mother's and father's occupation, place of living); health behaviour (tobacco, alcohol and drug use, dental hygiene, eating habits, physical activity, and behaviour associated with injuries); health status (perceptions of health, psychosomatic complaints); and the social and psychological milieu in the family, school and among peers (Wold et al., 1994; King et al., 1996).
We sought expert opinion on the relevance of the questions for children with physical disabilities from medical and rehabilitation personnel at Bloorview MacMillan Centre, and from the education personnel of the Toronto Board of Education Committee on the Needs of Children with Physical Disabilities. In addition, we questioned the children who completed the survey. Overall, professionals working with children who have physical disabilities judged that all questions were relevant. Initially, some experts doubted the relevance of asking about smoking, drinking and drug use. However, queries by some children about why there were no questions about drugs, e.g. ‘ice’ and ‘angel snow’ indicated that children with physical disabilities know about drugs, although they have much less experience with them (Steele et al., 1996).
Six questions were added to the survey to gain information about the disability including: name of the disability, its cause, age of onset, help needed from other persons or special equipment or techniques used to perform daily living routines (Table 2
), and frequency of using a wheelchair locking device when riding in a car or van. An examination of missing responses showed that 5.6% of the children did not know the name of their disability, 7.5% the cause and 7.8% the age of onset. For the other questions, non-responses were in the order of 0.6—2.0%.
Clearly questions included in regular health behaviour survey research are relevant to children with physical disabilities. Almost all children reported that they enjoyed participating and found the questions interesting. The attitudes expressed are reflected in one child's comment that, ‘You asked all the questions you would ask if you really wanted to know about someone’.
Survey administration
The HBSC protocol stipulates that the survey must be administered in the classroom during school hours by specially instructed personnel, e.g. teachers, school nurses, guidance counsellors or school psychologists. Survey administrators are reminded not to read the children's answers even when children ask for help in understanding a question. Anonymity must be ensured through procedures, e.g. having children seal their completed surveys in plain envelopes.
Physical disability can be associated with low level reading skills, an inability to turn the pages of the survey questionnaire, or place a check mark beside the response item. Therefore, the survey was prepared in large, bold print. Three formats were developed for survey administration: written, interview and interview—intervenor. The format used for each child was determined by the survey administrator in consultation with the teacher or health care professional familiar with the child.
In the written format, children completed the survey independently in the same manner as children without a physical disability. The interview format was developed for children who had less than grade 6 reading level and/or whose motor skills did not permit them to place a check mark beside their response. In this format, the survey administrator first read the question and the possible responses to the child. The child would then indicate the desired response verbally or non-verbally by pointing. For children unable to point to their desired response, the responses to the question were read twice. The first reading was to familiarize the child with the possible responses to the question. On the second reading, the child would indicate, usually with an eye blink, or a nod, the desired response. This was verified by reading back the response for reconfirmation. The interview—intervenor format was developed for children who communicate with the aid of an assistant. The procedure was the same as in the interview format except that during the second reading of the possible responses to the question children relayed their response to an intervenor who then informed the survey administrator.
The percentage of children completing the survey in written format was 64% (n = 203), interview format 32% (n = 103) and interview— intervenor 0.6% (n = 2). The format was unknown for 3% (n = 11) of the participants. In the majority of cases, children who completed the survey using the interview format did so because of low reading skills.
For 101 of the surveys administered by the project co-ordinator, the length of time for survey completion was noted. The mean length of time required was 49 min for the written format, 54 min for the interview and 60 min for the interview—intervenor.
Table 3 summarizes the questions on which there was a difference among children completing the written and interview formats. The two children using the intervenor—interview formats, as well as 11 children for whom format was unknown, were excluded from the analysis.
|
Table 3
summarizes the 13 out of 106 HBSC survey questions (12%) for which statistically significant differences in response pattern were found. On these questions, children using the interview format tended to present themselves in a more negative light. More of them reported taking pills when they have coughs (38 versus 23%) and stomach aches (48 versus 31%). With respect to lifestyles, they said they watch videos 4 h or more per week (24 versus 10% for the combined categories 4—6 h, 7—9 h and 10+ h). They seldom or never eat raw vegetables (19 and 32% versus 24 and 13%, respectively). They reported finding it difficult or very difficult to talk to same sex peers (22 and 15% versus 12 and 9%, respectively). More of them stated that at school other children do not want to spend time with them, and they end up alone once a week or more often (7 and 11% versus 4 and 5%, respectively). The greatest differences were found among the questions about school. Children using the interview format were less likely to think that at age 16 they would prepare to go on to university (36 versus 67%). They also expressed negative attitudes to school in that more of them agreed or strongly agreed with the statement that students are treated too severely or strictly at their school (28 and 4% versus 10 and 4%, respectively), and that teachers expect too much of them (19 and 7% versus 8 and 6%, respectively). Although an equal number of children in both groups strongly agreed or agreed that they feel they belong at school, a larger proportion of those using the interview format disagreed or strongly disagreed (10 and 4% versus 6 and 3%, respectively). On the positive side, more of the children using the interview format reported that they feel very happy with life (61 versus 44%), although there is little difference between the two groups when the response ‘quite happy’ is included. With respect to body image, fewer of them indicated a wish to change anything about their bodies (68 versus 51%).
In summary, in the interview format, children tended to give answers that cast them in a more negative light in comparison with their counterparts using the written format. This tendency may be associated with degree of disability, because children using the interview format also had more severe disabilities. The correlation between format and degree of disability as measured by a summative composite score on needing personal assistance was r = 0.21, p < 0.0002, and on needing technical assistance was r = 0.18, p < 0.0016.
|
DISCUSSION |
|---|
|
TOPSUMMARYINTRODUCTIONRESULTSDISCUSSIONACKNOWLEDGEMENTSREFERENCES |
|---|
From our work, we have concluded that children with physical disabilities can be included in mainstream health behaviour survey research. However, several issues concerning sampling, survey questions and administration format must be considered.
With respect to sampling, the issue is how best to recruit a representative sample. In this study, we used the network of rehabilitation service providers. Our approach was labour intensive and expensive. It also did not result in a representative sample of all children with physical disabilities at a population level because we were unable to recruit children with physical disabilities who rarely, if ever, visit clinics or other service organizations. Through the service network, we were also unable to include children who are completely integrated into regular classrooms. Our approach seems most suitable for special studies of children with moderate to severe physical disabilities that are carried out as adjuncts to mainstream health behaviour surveys.
For a population-based approach, we are investigating the usefulness of including questions about health status, including physical disabilities in the 1997—1998 HBSC survey. Three countries, Canada, Finland and Scotland will include the following two questions: ‘Do you have a long-time illness or medical condition? (some examples are asthma, cerebral palsy, hearing problems, diabetes, epilepsy, allergy)’. For children who respond in the affirmative, an open-ended follow-up question asks ‘What is your illness or medical condition? (If you have more than one, please include them all)’. The experience of researchers in Finland shows that about 20% of children participating in the HBSC survey can be expected to respond to these questions (Välimaa et al., 1995). If so, we can expect about 1000—1500 children in each of these countries to indicate that they have a health problem. To what extent children with physical disabilities will be represented remains to be seen.
Our study showed that the questions included in the HBSC survey were perceived as appropriate and relevant by children with physical disabilities. An unanticipated problem was the low reading skills of some of the children. Of the 32% of children who used the interview format to complete the survey, the majority did so because their reading skills were below grade 6 level. Low level reading skills will probably be an enduring issue in research with children with moderate to severe physical disabilities. It reinforces the need for conducting special health behaviour surveys as adjuncts to the regular surveys that are conducted nationally or internationally.
Our study also demonstrated that further development is needed of questions related to the diagnosis, cause and age of onset of the physical disability. Non-response rates to these questions were 5.6, 7.5 and 7.8%, respectively. This is consistent with research which has shown that children with a physical disability often lack knowledge about their disability (Anderson and Klarke, 1982; Blum, 1991). Non-responses may also reflect the fact that the majority of children in this study had spina bifida or cerebral palsy which are congenital conditions with multiple effects. Thus, children may have difficulty thinking about a specific cause or point in time when they became aware of having a disability. Apart from the challenges raised for the design of suitable survey questions to assess diagnosis, cause and age of onset, our findings raise the question of whether children need more explicit information about their disability, so that they can take a greater role in self-care.
We found some differences in response patterns between children using the written and interview survey administration formats. The overall trend was that children who used the interview format, in which answers could not be anonymous, gave a more negative view of themselves than did children who used the anonymous written format. To some extent, this difference in responses between the two formats may reflect the children's disability levels. Correlation coefficients showed that children using the interview formats had higher levels of disability than children using the written format. Thus, it is not unreasonable to conclude that the more negative response pattern of the children using the interview format was influenced by their level of disability. For example, greater levels of disability may be implicated in the greater use of medication reported by children using the interview format. Less frequent eating of raw vegetables can be related to the difficulties children with a physical disability have in chewing or swallowing hard, crunchy foods. Similarly, more video watching, less involvement with friends, and lower academic aspirations might be expected among children with more severe physical disabilities. Some kind of social desirability factor may, however, have biased children using the interview mode to say that they are very happy about their lives, or that there is nothing they wish to change about their body.
It is likely that interview formats will be necessary to enable children with moderate to severe physical disabilities to participate in health behaviour survey research. Thus, further research is needed to clarify the existence and type of bias introduced into children's answers by the lack of anonymity of this administration format. Specifically, the findings suggest that future research should be carried out on the effects of format through a randomized controlled study.
In summary, our experience has shown that children with moderate to severe physical disabilities can participate in survey research designed to study health behaviour and its social and psychological correlates among children without physical disabilities. However, the resources needed should not be underestimated and multiple approaches may be needed to gain a representative database. For children with low reading skills or motor problems that preclude them from turning pages or marking their responses, it seems that special substudies linked to national health behaviour surveys will be necessary. However, for a more population-based approach, further investigation is needed around the utility of routinely including questions about physical disabilities and other health problems in national and international health behaviour surveys. It may be that only children who are mildly affected would be able to participate. Even that would be useful if combined with the more specialized approaches described in this paper. As was noted in the Introduction, health behaviour among children having a physical disability is important to understand, and monitor, if we wish to reduce their risk of secondary conditions, and promote health and wellness throughout their life.
|
ACKNOWLEDGEMENTS |
|---|
|
TOPSUMMARYINTRODUCTIONRESULTSDISCUSSIONACKNOWLEDGEMENTSREFERENCES |
|---|
We are indebted and grateful to the children who participated in this study. We appreciate the Metropolitan Toronto School Boards, principals and teachers for granting us permission to conduct the survey in the schools and for the expert input of the Metropolitan Toronto Committee on the Needs of Students with Physical Disabilities. We are also grateful to the Easter Seal Society and School Health Support Services and other staff at Bloorview MacMillan Centre for their support. This research was funded by Bloorview Childrens Hospital Foundation and the Easter Seal Research Institute.
HBSC is a WHO/EURO cross-national study. International co-ordinators of the 1993—1994 study: Chris Smith, Health Promotion Wales; and Bente Wold, University of Bergen Norway. This study was an adjunct to the 1993—1994 HBSC survey in Canada carried out by Alan King and Wendy Warren, Social Program Evaluation Group, Queen's University, and Mary Johnston, Health Canada.
|
REFERENCES |
|---|
|
TOPSUMMARYINTRODUCTIONRESULTSDISCUSSIONACKNOWLEDGEMENTSREFERENCES |
|---|
Anderson, E. M. and Klarke, L. (1982) Disability in Adolescence. Methuen, London.
Blum, R. W. (1991) Overview of transition issues for youth with disabilities. Pediatrician, 18, 101—104.[Medline]
Blum, R. W. (1992) Chronic illness and disability in adolescence. Journal of Adolescent Health, 13, 364—368.[Web of Science][Medline]
Bullock, D. (1993) The physically disabled substance abuser. In Howard, B. M., Harrison, S., Carver, V. and Lightfoot, L. (eds) Alcohol and Drug Problems: A Practical Guide for Counsellors. Addiction Research Foundation, Toronto.
Connor-Kuntz, F. J., Dummer, G. M. and Paciorek, M. J. (1995) Physical education and sports participation of children and youth with spina bifida myelomeningocele. Adapted Physical Activity Quarterly, 12, 228—238.
Kessler, D. T. and Klein, M. A. (1995) Drug use patterns and risk factors of adolescents with physical disabilities. The International Journal of the Addictions, 30, 1243—1270.[Web of Science][Medline]
King, A., Wold, B., Tudor-Smith, C. and Harel, Y. (1996) The Health of Youth: A Cross-National Study (HBSC). World Health Organization, Copenhagen.
King, A. J. C. and Coles, B. (1992) The Health of Canada's Youth: Views and Behaviours of 11-, 13- and 15-year-olds from 11 Countries. Minister of Supply and Services. Catalogue no. H39-239/1992E, Ottawa, Canada.
Marge, M. (1988) Health promotion for people with disabilities: Moving beyond rehabilitation. American Journal of Health Promotion, 2, 29—35.
Office for Disabled Persons (1988) Statistical Profile of Disabled Persons in Ontario Volume I. Queen's Printer for Ontario, Toronto, Ontario, Canada.
Office for Disabled Persons (1990) Statistical Profile of Disabled Persons in Ontario Volume II. Queen's Printer for Ontario, Toronto, Ontario, Canada.
Pope, A. M. (1992) Preventing secondary conditions. Mental Retardation, 30, 347—354.[Web of Science][Medline]
Pope, A. M. and Tarlow, A. R. (1991) Disability in America: Toward a National Agenda for Prevention. National Academy Press, Washington, DC.
Steele, C. A., Kalnins, I. V., Jutai, J. W., Stevens, S. E., Bortolussi, J. A. and Biggar, D. (1996) Lifestyle health behaviours of 11- to 16-year-old youth with physical disabilities. Health Education Research, 11, 173—186.
Stevens, S. E., Steele, C. A., Jutai, J. W., Kalnins, I. V., Bortolussi, J. A. and Biggar, D. (1996) The psychosocial health of 11- to 16-year-old youth with physical disabilities. Journal of Adolescent Health, 19, 157—164.[Web of Science][Medline]
Välimaa, R., Kepler, K. and Yeganegi, N. (1995) Koettu terveys ja onnellisuus sekä sairaudet koululaisten arjessa. In Kannas, L. (ed.) Koululaisten kokema terveys, hyvinvointi ja kouluviihtyuvyys. Opetushallitus, Helsinki.
Wold, B., Aaro, L. E. and Smith, C. (1994) Health Behaviours in School-aged Children: A WHO Cross-national Survey (HBSC), Research Protocol for the 1993—94 Study. Research Center for Health Promotion, University of Bergen, Bergen, Norway.
Wood, D. W. (1994) Preventing Secondary Conditions Associated with Spina Bifida or Cerebral Palsy: Proceedings and Recommendations of a Symposium. Spina Bifida Association of America, Washington, DC.
CiteULike 
Connotea 
Del.icio.us What's this?
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||