Article Text
PDF
Abstract
Background Children surviving after spina bifida repair often have significant disability, the consequences of which may be more profound in low-income countries. The aim of this cross-sectional study was to measure quality of life (QOL) reported by children with spina bifida in Uganda, and to define factors associated with QOL.
Methods QOL was measured using both the Health Utilities Index (HUI3) Tool and a visual analogue scale (VAS) marked from 0 to 10. In keeping with the WHO definition of QOL, further analysis was conducted using subjective QOL scores (using the VAS). Multivariate regression was used to investigate the association between VAS scores and prespecified variables: age, sex, hydrocephalus, mobility, urinary continence, school attendance and family size.
Results Sixty two of 68 surviving children aged 10–14 were able to complete all aspects of the assessment. There was poor correlation between the VAS and HUI3 Tool (Pearson correlation 0.488). On multivariate regression, the following variables were associated with a significant change in the 10-point VAS (change in score; 95% CI): male sex (−1.45; −2.436 to −0.465), urinary continence (1.681; 0.190 to 3.172), large family size (−1.775; −2.773 to −0.777) and hydrocephalus (−1.382; −2.374 to −0.465).
Conclusions Urinary continence and family size are potentially modifiable, the former by simple and inexpensive medical management. Enhanced investment in community-based rehabilitation and support is urgently needed. Delivery of family planning services is a national priority in Uganda, and should be discussed with families as part of holistic care.
- Neurodisability
- Neurosurgery
- Tropical Paediatrics
- Congenital Abnorm
- Patient perspective
Statistics from Altmetric.com
What is already known?
The WHO defines quality of life (QOL) as an individual’s perception of their position in life, in the context of their culture and value systems.
Spina bifida can profoundly affect physical, social, neurocognitive and psychological function, but the impact on QOL has not been well described in a low-income country.
The impact of mobility, continence and hydrocephalus (among other factors) on QOL in spina bifida has been investigated with variable results, using different scoring tools.
What this study adds?
Scores from the objective QOL tool (Health Utilities Index 3) and carer predictions of QOL correlated poorly with the child’s self-reported score on a visual analogue scale.
Urinary incontinence, large family size, male sex and the presence of hydrocephalus were associated with poorer subjective QOL in this cohort.
Simple measures must be employed to improve continence and address the subject of family planning in this population, which may in turn improve QOL.
Introduction
Open spina bifida (myelomeningocele) is a congenital malformation of the spinal cord, classified as a neural tube defect. There is marked geographic variation, but the average worldwide prevalence of neural tube defects has been estimated at one case per 1000 births.1 Even with prompt surgical repair of the spinal lesion, most infants will have residual neurological deficit, bowel and bladder dysfunction. The development of hydrocephalus has been described in 50%–90% of cases,2 and orthopaedic deformities are common.
The WHO defines quality of life (QOL) as an individual’s perception of his or her position in life in the context of the culture and value systems in which he or she lives and in relation to his or her goals, expectations, standards and concerns.3
Spina bifida can profoundly affect physical, social, neurocognitive and psychological function.4 5 The impact on QOL may be more significant in a low-income country where most families have severely limited resources and little access to healthcare facilities. This has not been well described in the current literature, but there is increasing recognition of the need to engage with the important global health issue of childhood disability.6
Ten-year survival of children following surgical repair at CURE Children’s Hospital of Uganda (CCHU) has been reported elsewhere.7 The aim of this cross-sectional study was to ascertain the QOL of surviving children, and to determine whether this was influenced by mobility, urinary continence, hydrocephalus, sex, family size and school attendance. We also investigated the relationship between attendance at follow-up, clean intermittent catheterisation (CIC) adherence and urinary continence to determine whether children who were regular attenders at clinic were more likely to be performing CIC (and more likely to be dry).
Methods
Study design and participants
Eligible patients were identified from the CCHU electronic database according to the following criteria: diagnosis of myelomeningocele, initial presentation at CCHU between December 2000 and December 2004, and age younger than 6 months at the time of surgical repair. In order to facilitate home visits, the study cohort was restricted to children from 16 districts in the southeastern region where the hospital is located.7
Families of all patients thought to be alive were telephoned or visited by a research assistant between October 2014 and January 2015. Surviving children were invited to attend CCHU on a specified date, accompanied by a caregiver. They were offered reimbursement of travel costs, provision of meals and overnight accommodation if required.
Ethical approval was granted by the Institutional Review Board of CCHU. Written consent for participation in the research and publication of the results was obtained from all caregivers. Children also completed an assent form.
All children who were able to communicate underwent a qualitative semistructured interview, covering a range of topics, including school attendance, continence, clean intermittent catheterisation (CIC and QOL. This was conducted in the child’s first language by a healthcare worker who had been trained in the use of the questionnaire. A translator was used if required. A similar interview occurred with the caregiver in a separate room, again in their first language or translated.
Answers were written down at the time of the interview, and transcribed at the end of the day. Where there was disagreement between child and caregiver responses, for example, relating to continence, the ‘worst’ response was selected.
QOL tools
We considered a number of tools previously used in QOL studies of children with spina bifida.8–10 However, most were rejected because the questionnaires were too time consuming, contained linguistically complex questions or culturally irrelevant concepts.
We selected the Health Utilities Index (HUI3) Tool11 because it is simple and quick to administer, and has been previously validated in children with spina bifida (though not in a low-income setting).8 12 The HUI3 is designed to describe health-related QOL, and includes eight dimensions of health status, each composed of five or six levels. The calculated output can range from 1.0 (perfect health) to a minimum of −0.36 (a health state worse than death).
The output of HUI3 is a calculation based on scores in several predominantly ‘objective’ domains, including vision, communication, hearing and mobility. In addition to the HUI3 Tool, we obtained a subjective response by asking the children to indicate their QOL on a visual analogue scale (VAS) marked from 0 to 10.13–15 Caregivers were asked to complete the HUI3 Tool on behalf of the child, and to report their perception of the child’s QOL according to the VAS.
Statistical analysis
The Pearson correlation coefficient was used to assess the relationship between QOL scores. A multivariate regression model was built using QOL as the dependent variable. We chose to use the subjective VAS score for this analysis, as it more accurately represents the child’s perception, in keeping with the WHO definition of QOL.3 Furthermore, the HUI3 domains included mobility, which we were investigating as an independent variable.
Independent variables with p<0.1 on univariate analysis were included in the multivariate model. A p <0.05 was considered a significant association. Plots of residuals confirmed assumptions of linearity and normality.
The following independent variables were investigated for possible association with subjective QOL: age, sex, hydrocephalus (managed either endoscopically or by ventriculoperitoneal shunt),7 mobility, urinary continence, school attendance and family size. The choice of variables was guided by current literature,9 10 13 15 though there is limited evidence from a low-income setting.
Age was included in the analysis as a continuous variable. School attendance and the presence of hydrocephalus were treated as binary (yes/no) variables. Mobility was divided into three categories: unable to walk, walking with aids and walking independently. During interviews, the child and caregiver were asked separately to describe their urinary continence as always dry, mostly dry, mostly wet or always wet. For the purpose of the statistical analysis, the responses were simplified into two categories (always or mostly dry vs always or mostly wet). Family size was treated as a binary variable: six children or fewer versus seven children or more.
The associations between urinary continence, CIC and clinic attendance were investigated using Fisher’s exact test. All analyses were performed using SPSS V. 22 (IBM SPSS Statistics, IBM).
Results
Study population
Sixty six of 68 surviving children attended CCHU for assessment, all aged between 10 and 14 years. Contact was made with the other two survivors (one male and one female) but in both cases the father declined to bring them to the hospital.
Patient characteristics are summarised in table 1. Interviews were conducted in 12 languages.
Patient characteristics
QOL scores
Four children were unable to complete the assessment due to deafness and/or learning disability. The mean VAS score for the 62 children who were able to answer was 7.19 (95% CI 6.61 to 7.78). The median score was 8, and 14 children (21%) stated that their VAS score was 10/10.
All 66 caregivers responded on behalf of the children. The mean VAS score was 6.58 (95% CI 5.95 to 7.20). The median score was 6; 13 caregivers (20%) gave a score of 10/10.
There was poor correlation between the child’s reported QOL and the caregiver’s perception using the VAS (Pearson correlation 0.133); thus, where children were not able to answer we did not impute the caregiver’s reported value. This left missing data in four cases, and the QOL analysis was conducted using the complete data on 62 children.
The mean HUI3 score reported by 63 children was 0.575 (95% CI 0.493 to 0.658). The mean HUI3 score reported by 66 caregivers was 0.549 (95% CI 0.465 to 0.633). As expected, given the predominantly objective nature of the HUI3 measures, these scores were highly correlated (Pearson correlation 0.848).
The HUI3 scores were not highly correlated with the VAS scores. The correlation between HUI3 and VAS score for children was 0.488. The correlation between the HUI3 score completed by caregivers and the child’s reported VAS score was 0.380.
QOL analysis
The following factors were investigated for an association with the child’s subjective QOL as reported using the VAS: age, sex, hydrocephalus, mobility, urinary continence, family size and school attendance. On univariate analysis, all but age appeared to be associated with QOL (p<0.1) and were included in the multivariate analysis. The results are presented in table 2.
Multiple regression analysis of factors predicting subjective quality of life in children with spina bifida
In the multiple regression model, the following factors were significantly associated with lower subjective QOL scores: male sex, urinary incontinence (wet or mostly wet), large family size (seven or more children) and hydrocephalus. The model explained 45% of the variability in QOL scores between children (R2=0.454).
Urinary continence and CIC
There was a strong association between urinary continence and the practice of CIC: 8 of 17 children (47%) performing CIC, and 3 of 49 (6%) not performing CIC, were dry or mostly dry (p<0.0005). The 13 children who had attended clinic in the last 2 years were more likely to be continuing with CIC (p<0.0005) and more likely to be dry (p<0.0005).
Discussion
The WHO defines QOL as an individual’s perception of his or her position in life in the context of the culture and value systems in which he or she lives and in relation to his or her goals, expectations, standards and concerns.3 In keeping with this definition, we chose to measure subjective QOL of 62 Ugandan children aged 10–14 with spina bifida, using a VAS (marked from 0 to 10). The mean score was 7.19, with 14 children stating that their QOL scored 10.
There was poor correlation between the VAS score and results obtained using the largely objective HUI3 Tool, which supports the assertion that calculating QOL based on an individual’s abilities does not accurately reflect their experience. Furthermore, there was no independent association between subjective QOL and the ability to walk, though ambulation is included as a domain within the HUI score. This is in keeping with the work of Kirpalani et al, who reported that parental hope was more strongly associated with QOL than either predicted or current physical deficits.16
We identified four independent variables associated with subjective QOL: sex, hydrocephalus, urinary continence and family size. These four factors explained 45% of the variability in QOL scores. Encouragingly, both urinary continence and family size are amenable to modification.
Female sex
There was a strong association between female sex and higher VAS scores. This has not previously been reported. Our findings could support a hypothesis that females are more resilient in challenging circumstances, or perhaps resigned to them (text box 1). In support of this, a study of QOL among primary healthcare attendees aged 18–84 in Central Uganda17 found significantly lower physical health scores among females, yet no statistically significant difference in overall QOL ratings or general satisfaction with health. An alternative explanation may relate to higher expectations of male children (actual or perceived).
During the interviews, girls appeared satisfied with their life if they were loved and included, and able to go to school, even if they experienced some ‘challenges’ such as urinary incontinence.
‘Though I am always wet, friends laugh at me and I am unable to do [some] activities, I remain a happy girl. I appreciate life the way it is.’
Eleven-year-old girl, living with both parents and five siblings, attending school. Quality of life (QOL) score 10/10.
‘In most cases I am happy because I am at least able to do some activities like other children and I am loved by my parents and other family members… but I am also unhappy when neighbours and some other people laugh at me because of my condition, and when I get sick.’
Thirteen-year-old girl, living with both parents and seven siblings, attends boarding school for children with disabilities. QOL score 8/10.
‘I am somewhat happy, but my problem is that of urine incontinence - my other siblings go to school and I am left at home.’
Eleven-year-old girl living with both parents and eleven siblings, not attending school because of urinary incontinence. QOL score 6/10.
Hydrocephalus
The presence of hydrocephalus was associated with poorer VAS scores. A previous small study in San Diego9 found that shunted hydrocephalus was significantly associated with poorer self-reported physical health summary scores, but other studies have not replicated this finding.13 18 Hydrocephalus has been associated with increased functional dependence19 and poorer neurocognitive outcomes.20 A significant association between executive function and subjective QOL was reported in young adults with spina bifida and hydrocephalus, and was a stronger predictor than functional independence.14
Urinary continence
Urinary incontinence was predictive of poorer VAS scores. This association has not been consistently reported in literature from Europe or the USA.9 10 21 There are conflicting reports concerning the impact of continence on self-esteem.18 22 The consequences of urinary incontinence will be amplified in a low-income country, where the cost of nappies is prohibitive, and where laundry and personal care are more challenging. This issue has been noted in relation to school attendance in Uganda.23 Of 36 children in our cohort who were not attending school, 15 (42%) explained that this was due to incontinence, and in some cases, the associated bullying.
CCHU trains caregivers in the technique of CIC, and a free supply of catheters and oxybutynin is provided by the International Federation for Spina Bifida and Hydrocephalus to families at every visit. However, most families (74%) had abandoned CIC. There is growing recognition that optimal medical management of the neurogenic bladder from birth can preserve renal function and achieve safe urinary continence in the majority of children.24 Thus, it is unsurprising that children who reported that they were performing CIC were significantly more likely to be dry or mostly dry. Clinic attendance encourages continued practice of CIC and allows provision of materials.
Family size
Coming from a family with seven or more children was strongly associated with poorer subjective QOL. Large family size has been cited as one of the drivers of increasing poverty in rural Eastern Uganda.25 Larger families will have fewer resources to meet the additional needs of a child with spina bifida, but there is also an indirect cost, as caregivers have less time to devote to income-generating activities.26 Neglect of children with spina bifida has been observed on home visits both in this study and in others.7 26 27 A sense of belonging within the family may be more important than formal education and independence in determining the QOL of African children with disabilities.26
Family size is potentially modifiable. In December 2014, Uganda’s Ministry of Health launched the Uganda Family Planning Costed Implementation Plan, 2015–2020, which aims to reduce Uganda’s vast unmet need for family planning.28 In order to deliver this, all healthcare providers have a role to play in promoting change in attitudes and behaviour, especially in rural populations.
Strengths and limitations
This study surveyed QOL and related factors in 66 of 68 surviving children with operated spina bifida, reflecting an excellent participation rate of 97%. There are a number of limitations. We have established association and not causation. Most importantly, we did not have a norm for comparison with unaffected Ugandan children. Without a reference group, it is not possible to comment on the significance of the subjective QOL scores obtained, and particularly the fact that they seem surprisingly high. There is potential for bias given that we were only able to survey children who had survived beyond 10 years; those who had already died may well have experienced poorer QOL. We only included children in southeastern Uganda; significant cultural and socioeconomic differences are recognised between the regions of Uganda. The definition of this cohort resulted in a narrow age range, which precludes exploration of the association between age and QOL. It is recognised that increasing age adversely affects QOL, perhaps as older teenagers develop increasing insight into their limitations as compared with their peers.12 We were unable to identify a suitable disease-specific measure of QOL. Previous studies investigating QOL in individuals with spina bifida have used a VAS.13–15
While this model explains a significant portion of the variation in subjective QOL scores (R2=0.454), over 50% remains unexplained. The findings relating to sex and family size suggest that other social factors may be important in determining QOL for these children, and these warrant further investigation.
Conclusion
Improving the QOL of children with disabilities in low-income countries should be a priority. The WHO definition emphasises the individual’s perception of his or her position in life. In keeping with this, we found a poor correlation between the score generated by a predominantly objective QOL tool (HUI3) and the child’s experience reported using a VAS.
Of the four factors associated with VAS score, both urinary continence and family size are to some extent modifiable, and future work should be directed towards improving continence and delivering education regarding family planning to new mothers, followed by measurement of the benefits in terms of QOL. Greater investment in community-based rehabilitation and support for families coping with a disabled child is urgently required.
Acknowledgments
The authors would like to acknowledge the assistance of Hellen Titin, Ezra Bamulikulwaki, Rebecca Muduwa, Moses Mukalo and Melissa Kovacs in the conduct of this study.
Dr Helen J Sims-Williams has been awarded the Thomas Watts Eden Paediatric Fellowship by the Royal College of Physicians (London). Patient follow-up was supported through Dr Benjamin C Warf’s John D. and Catherine T. MacArthur Fellowship award. Neither funder had any involvement in study design, data collection, data analysis, manuscript preparation or publication decisions.
References
Footnotes
Contributors HJS-W conceived and designed the study, conducted data collection, performed data analysis, drafted the initial manuscript, revised and approved the final manuscript as submitted.
BCW was the operating neurosurgeon, provided oversight for the study and editing of the manuscript, and approved the final manuscript as submitted.
HPS-W conceived and designed the study, conducted data collection, reviewed and revised the manuscript, and approved the final manuscript as submitted.
EMK conceived and designed the study, conducted data collection, critically reviewed the manuscript and approved the final manuscript as submitted.
Competing interests None declared.
Patient consent Obtained.
Ethics approval Institutional review board of CURE Children’s Hospital Uganda.
Provenance and peer review Not commissioned; externally peer reviewed.
Data sharing statement Additional data concerning survival of these children have already been published. Further data concerning neurological examination, family structure and quality of life are available. Data concerning renal outcomes are due to be published.