Statistics from Altmetric.com
Spina bifida (SB) is a serious congenital anomaly with an incidence of 0.4–1.0 per 1000 births in the USA.1Universal folic acid supplementation will reduce, but not entirely eliminate neural tube defects.2 The sequelae of SB include decreased mobility, bladder and bowel incontinence,1 and neurosensory disabilities as well as social disadvantages such as decreased community acceptance and lack of a supportive peer network.3-5 In addition, SB has major financial and employment implications for affected families.4
Improved operative interventions for SB (and particularly for hydrocephalus) during the neonatal period became available after 1960 and highlighted the need for the development of treatment guidelines. The therapeutic pendulum has swung considerably; from recommendations for active therapy regardless of the level of lesion6-8; to a selective treatment based on deficits assessed by a neurophysical examination (NPE)7-11; and back to aggressive therapy for virtually all patients.1 This latter approach follows concerns about the predictive accuracy of the NPE, the ethics of selective practices,12-15 and the survival of non-operated infants.13-15
Given the uncertainty about the best management for an individual infant with SB, it has been suggested that “parental directives for care” might resolve some of the dilemmas that arise when health caregivers make judgements about their patients' future quality of life.16 This suggestion is in keeping with an increasingly heard request for more parental involvement in medical decisions for critically ill newborns.17 Nonetheless, in order to make rational decisions, parents need accurate prognostic information; the NPE is used commonly to predict future physical and cognitive function.18 Moreover, in clinical practice, limitations in mobility, bladder and bowel function, and cognitive development are often used to predict an individual's future health related quality of life (HRQL), despite the absence of empirical data on the predictors of HRQL in this population.
In our work with parents of children with spina bifida, it was suggested that factors other than NPE might predict children's HRQL. One hypothesised factor was level of parental hopefulness. Advisors from the Spina Bifida and Hydrocephalus Association of Ontario agreed with the importance of a construct such as hope in modifying outcomes. The construct of hope has been described in the nursing, psychological, and sociological literature.19-31 It has been suggested that hope is essential for mental health,19-21 for coping and adapting,22 and for sustaining and energising a mother's maternal work in the care of her disabled child.23 Hope is increasingly recognised as a disease modifier.24-26 We therefore hypothesised that parental hope is an important determinant of HRQL in SB.
The objectives of our study were to evaluate the relative contributions of neonatal and current physical function, and of parental hope, to the HRQL of children and adolescents with SB.
Following initial surgery, all children with SB in Ontario are referred to a regional, publicly funded specialist treatment centre for ongoing care. We obtained patient lists from two such regional centres. Chedoke-McMaster Hospitals serves the Hamilton-Wentworth region of Ontario and follows approximately 90 SB patients; the Bloorview MacMillan Centre serves the South Central region of Ontario and follows approximately 650 SB patients. The study received ethical approval from these two institutions and from the Hospital for Sick Children in Toronto. A random sample was taken of all patients in the age groups 5–12 years and 13–20 years; their families received a postal enquiry as described below. This random selection ensured a representative sample of patients whose characteristics would cover the full range of disability and functional impairment.
Parents were asked for permission to retrieve their child's neurophysical examination findings from hospital records. In addition, parents were asked to complete a demographic data collection form, a form to describe current function, and the Miller Hope Scale. Parents of children aged 5–12 years were requested to complete the HRQL questionnaire on behalf of their child; adolescents aged 13–20 years completed this questionnaire themselves. Parental completion of the Hope Scale, and child completion of the HRQL occurred concurrently. To maximise response rates, we completed three mailings accompanied by two telephone calls which were spread over three months after the initial telephone call.
Demographic data and current functional status
Demographic data included family income, place of residence (urban or rural), and family composition. In addition, data were obtained on the current level of physical function, including mobility, and degree of bladder and bowel independence. Finally, the number of previous medical interventions was recorded, including all shunt revisions, and the number of surgical operations the child had received.
Parental hope: the Miller Hope Scale (MHS)
This 40 item scale was devised by Miller and Powers20and validated in 522 adults. This instrument has good measurement properties including test–retest reliability. In a pilot group of parents of both 5–12 year old children and 13–20 year olds with spina bifida, we found that the MHS was easily administered. In both age groups, the MHS was completed by the same parent who provided the data on current function and demographics.
Quality of Life Scales (HRQL)
This self administered reliable and validated age specific instrument was devised following the general methodology outlined by Kirshner and Guyatt,32 and applied to HRQL.33We have previously reported the methodological development of this instrument.34 The items for this instrument were generated through open ended interviews with families of children with SB. The items were then reduced, and validated by accepted techniques.33 Two age specific questionnaires were developed: for children aged 5–12 years (44 items), and for adolescents aged 13–20 years (47 items). Items from several domains were included.
NEWBORN NEUROPHYSICAL EXAMINATION (NPE)
Data on the NPE were retrieved from the newborn hospital record where available. Details of the NPE included head circumference and percentile at birth, and presence or absence of other congenital anomalies (that is, chromosomal anomalies and congential anomalies not typically associated with spina bifida). Level of the spinal cord lesion was judged from the hospital record by references to sensory function, muscle innervation level, muscle power, and the presence or absence of an anal wink.1 In a pilot study of 15 cases, there was 100% agreement of the information retrieved from the hospital record, as assessed independently by two individuals, a research assistant and a physician. For the entire study sample, two physicians independently rated the level of lesion for each study patient. (One physician is a neonatologist (HMK) and the other a tertiary care academic general paediatrician (PCP).) Inter-rater agreement was 85%. A third physician, a developmental paediatrician (PLR), arbitrated the 15% of cases where HMK and PCP did not agree.
Statistical analysis was performed using SAS (SAS Institute Inc., Cary, North Carolina). We performed two separate forward selection regression models in each age group. A linear least squares regression model was used. Terms with a significance level less than 0.05 were deemed to be statistically significant, and were added to the model.
The first analysis in each age group was termed the “birth status” model. This examined the extent to which the current HRQL of the child could be predicted from two sets of variables: the neurological status at birth using the NPE, and parental hope measured by the MHS.
The second regression model was termed the “current function” model. This analysis asked how much of the variability in the HRQL could be explained by the current physical function of the child, and by parental hope.
We calculated post hoc that the respective sample sizes of 60 and 121 had provided us with 80% power at the 5% level for detecting an association between the MHS measure of parental hope and an independent variable if the strength of the correlations were 0.33 and 0.23, respectively.35
The response rate was 71% (137 of 194 families) in the 5–12 year age group. The results of 121 matching NPEs (88%) could be obtained. The remainder of the children were either born out of province or had inadequate details recorded. Current physical function was determined in all 137 children and 131 responses were received for the parental MHS.
In the 13–20 year age group, the response rate was 54% (74 of 138 families). The NPE was available in 60 of these.
CHILD AND FAMILY CHARACTERISTICS
Table 1 outlines the socioeconomic characteristics of the families in the two age groups. Table 2 presents details of the children's characteristics including the child's physical function as assessed by the families and by teenagers themselves, for bowel and bladder function, degree of mobility, and the number of operative procedures and shunt procedures.
The full details of the final regression models are presented by age group, in Appendix . Further detailed results are presented in the tables. Table 3 summarises the results for the birth status model for each age group. Table 4 summarises the results for the current function model.
Birth status model
This model explained 26% (total R2) of the variance of HRQL in children aged 5–12 years and 23% in adolescents aged 13–20 years. In both age groups, the explanation provided by the parental MHS was far greater than that provided by any other variable.
Current function model
In the younger age group, this model explained 47% of the variance associated with HRQL. The variance was mainly explained by MHS, bladder function, and the number of surgical operations. In the older age group, this model explained 31% of the variance associated with HRQL, with MHS and bowel function contributing most importantly.
In this study, we found a striking association between the child's HRQL and parental hope. In contrast, physical limitations which clinicians commonly use to predict HRQL appeared to be less important. Our data are consistent with the view that some of the ethical concerns surrounding the prediction of outcome for newborn infants with SB can only be solved by a better understanding of family desires, problems, and coping strategies.16
Previous work has shown that perceptions of HRQL in SB and other disabilities vary considerably between different observers. Cadmanet al reported that parents of children with and without disabilities perceive physical and mental handicaps very differently from health care professionals.36 Other investigators have shown that in contrast to physicians who care for children with SB, parents do not view those children as being sick.5 37 In the present study, we have shown for the first time that differences in HRQL for children and adolescents can be attributed, at least in part, to differences in the degree of parental hope. Taken together, these findings suggest that health professionals should exercise extreme caution when attempting to predict an individual child's HRQL based on physical symptoms and signs. As patients and their families are the best judges of their own HRQL, we do not support the recent approach by Takayanagi and Suruga38 to determine quality of life in neonatal surgical cases, as it is based on the assessment of medical providers rather than of parents.
Hope is a construct that has been extensively explored in the nursing, psychological, and sociological literature, and reliable and valid scales for its measurement have been devised, including the one used in this study.20 The Miller Hope Scale has been found to be correlated with religiosity, spiritual well being, hope, and other positive mood states21; the coping effectiveness22; social support and self esteem27 29; resolution of psychosocial issues and the perception that one is able to control one's life28; and psychological well being.19 20 Larson23 has found that hopefulness is critical for sustaining and energising a mother's maternal work in the care of her disabled child. Hope is increasingly recognised as a disease modifier,24-26although we are only beginning to understand how hope (and hopelessness) affect biological processes.
This study has limitations. We used proxy responses from parents to judge HRQL for children aged 5–12 years, while adolescents rated their own HRQL. Importantly, whether or not a parental proxy was used to determine HRQL, a strong association was found between parental hope and HRQL. Another limitation to the interpretation of the relation between parental hope and HRQL results from the cross sectional nature of the study. It is possible that parental hope is a reflection of the experiences of the family, or that parental hope directly influences the child's HRQL. The latter interpretation seems plausible, given that the construct of hope has been linked with positive psychological factors as noted above. Parents caring for a child with a physical disability may well empower their children with a similar sense of hopefulness, which may result in an improved perception of their HRQL.
It is not known whether hope is a static trait or a dynamic state. If hope is a trait, that is, a stable personality or temperamental characteristic, it follows that hope is likely to be a true determinant of HRQL. If hope is a dynamic state, it may potentially be responsive to targeted interventions. It has already been shown that a prospective nursing intervention reduced maladjustment in children with chronic disorders.39 It is conceivable that nursing provision had in some way fostered hope. Wake and Miller have described several hope inspiring nursing strategies for treating hopelessness.30 31 Furthermore, it is not known what the determinants of hope might be. It is possible that factors such as family income, family structure, or parents' level of education may in some way determine degree of parental hopefulness. Although several of these factors were measured in our study, we feel that determinants and dynamics of hope need new and prospective testing in future studies.
We would argue that more attention to the understanding and nurturing of hope is warranted in the long term care of children with SB and probably also in other childhood disorders. The importance of hope has been recognised by nurses and psychologists. It should be equally recognised by physicians caring for children.
The authors wish to thank Ms Bonnie Charbonneau of the Spina Bifida and Hydrocephalus Association (Canada), and Dr David Cadman for their insights; and Drs G Guyatt, D Streiner, and B Schmidt (McMaster University) for advice on the manuscript. This work was supported by grant number 02943 from the Ministry of Health Ontario and from Bloorview Children's Hospital Foundation.
The final models are as follows.
5–12 year age group: HRQL = 0.32 + 0.48 (parental MHS) + 0.02 (level of lesion)
13–20 year age group: HRQL = 0.41 + 0.56 (parental MHS)
5–12 year age group: HRQL = 0.74 + 0.42 (parental MHS) − 0.01 (number of operations) − 0.07 (bladder function) − 0.01 (age)
13–20 year age group: HRQL = 0.53 + 0.52 (parental MHS) − 0.04 (bowel function)
If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.