Quality of Life of Children With Spinal Muscular Atrophy and Their Caregivers: A Chinese Cross-sectional Study

Background: Spinal muscular atrophy (SMA) is an autosomal recessive motor neuron disease leading to multiple organs dysfunction, which can impair the quality of life (QoL) of patients and family. We aimed to evaluate QoL of children with SMA and their caregivers, and to assess the associated factors in a Chinese cross-sectional study. Methods: 101 caregivers and children aged 0-14 years with SMA were recruited from a children hospital. 26 children had type I SMA, 56 type II and 19 type III. Children’s QoL was measured by the Pediatric Quality of Life Inventory 3.0 Neuromuscular Module (PedsQL NMM) that was lled by caregivers. Caregiver’s QoL was measured by the Pediatric Quality of Life Inventory Family Impact Module (PedsQL FIM). Information on sociodemographic, disease-specic characteristics, and treatments were collected by the proxy-reported questionnaire. Two independent t-tests and one-way ANOVA were applied to compare differences in scores of QoL across subgroups. Results: Children with type III SMA had higher total score of PedsQL NMM, and scores in neuromuscular disease and family resources domain than those with type I and type II SMA (p < 0.001). Caregivers of children with type III SMA reported higher scores in domains of physical, emotional, social, and cognitive function of PedsQL FIM than those of children with types I and II (p < 0.05). In addition, disease-related characteristics (e.g. limited mobility, motor degeneration, skeleton deformity, and digestive system dysfunction) and respiratory support were associated with lower scores of PedsQL NMM and PedsQL FIM (p < 0.05). Exercise training, multidisciplinary team management and using the medication nusinersen were associated with higher scores of QoL in both PedsQL NMM and FIM (p < 0.05). Conclusion: Our study has demonstrated the factors that may impair or improve QoL of children patients with SMA and their parents. Particularly, QoL was relatively poor in children with type I and type II SMA as well as in their caregivers than with III on in their to enhance impacts and decisions in different subtypes of SMA, which indicated that the differences in quality of life of patients has reached the threshold of clinical relevance, which worth attention from health professionals. And we found there is no statistical difference in neuromuscular disease domain between children with type I and type II, but the effect size was moderate, which indicated that there was a clinical difference in neuromuscular disease domain between children with type I and with type II. Our results showed that the physical, emotional, societal, and cognition items in the Family Impact Module were effective for comparing patients with different types of SMA. Patients with SMA type III tend to have higher scores than the other two types, which is related to the patients’ motor ability. The effect size SMA type III compared with type I and type II also indicated that there was a large clinical difference in the above functional domains. Most patients with SMA type III are able to walk alone, while patients with type II and type I cannot. However, because patients with type III can move their limbs freely and complete various functional exercises by themselves, their caregivers put less time and energy into their care, resulting in signicantly higher family inuence module scores in patients with type III than patients with types I and II. Our research showed that there was no signicant difference among SMA types in communication, worry, daily activities, and family relations domains, which may be caused by the characteristics of China’s national social conditions, similar to the communication function outcomes seen in the PedsQL NMM. family lower


Introduction
Spinal muscular atrophy (SMA) is a rare, autosomal-recessive neuromuscular disease caused by genetic deletion or mutation in the survival of motor neuron 1 (SMN1) gene on chromosome 5q13, which results in reduced levels of the survival of motor neuron (SMN) protein, causing muscle weakness and atrophy [1][2][3][4][5]. SMA is traditionnally divided into ve clinical subtypes (Type 0, I, II, III, IV) based on the age of symptom onset highest motor milestones archived [6,7]. SMA Type 0 is the most severe subtype that onsets during prenatal period and the live birth survives less than one month after birth. SMA Type I is the most common and most severe subtype that usually presents before 6 months of age. The patient never learns to sit up independently, and his/her life expectancy is seldom more than two years without respiratory support. Type II usually onsets between 6 and 18 months of age. Patients can sit independently but never be able to walk. They can usually live into adulthood. Symptoms of SMA type III normally present after age 18 months; patients can acquire independent ambulation, although some patients may lose the ability to walk in adulthood due to the progressive nature of the disease. Their life spans are almost identical to that of the general population. SMA Type IV is rarest and with lowest morbidity and mortality, which occurs after 20 years of age. Their life span is similar to patients with Type III [8,9]. SMN protein de ciency is detrimental to the functioning of other tissues including skeletal muscle, heart, autonomic and enteric nervous, metabolic/endocrine (e.g. pancreas), lymphatic, and reproductive systems [10][11][12]. Therefore, multiple organ system dysfunction may occur in SMA as the disease progresses, requiring daily medical care and nursing support for daily activities as well as for the long-term co-management of several medical devices [13,14]. As the advanced clinical treatments have been emerged and developed in the last decades, the morbidity and mortality of SMA patients have been decreased and their life span has been signi cantly extended [15][16][17]. Accordingly, the quality of life (QoL) of SMA patients is supposed to be improved.
Quality of life (QoL) is a multidimensional concept, and is an important patient-reported outcome measure in clinical research and practice. Based on the de nition by the World Health Organization (WHO). QoL is de ned as "an individual's perception of their position in the life in the context of the culture in which they live and in relation to their goals, expectations, standards and concerns" [18]. Investigating QoL in the clinical setting is important to understand patient's experience, to evaluate the effectiveness of treatments, and optimize the personized therapy plan, which ultimately improve patient's wellbeing [19,20].
In the area of pediatric neurology research, attentions mostly have been focused on developing the novel technologies and pharmaceuticals for children with greater than 0.70 [23].
The PedsQL FIM was designed to assess the impact of pediatric chronic health conditions on caregivers' QoL and family functions in the past month. The PedsQL FIM includes six scales measuring caregivers' self-reported functioning: physical functioning (six items), emotional functioning ( ve items), social functioning (four items), cognitive functioning ( ve items), communication (three items), and worry ( ve items). Additionally, the PedsQL FIM further explored two scales for primary caregivers reporting family functioning: daily activities (three items) and family relationships ( ve items). A ve-point response scale is utilized (scale: 0 = never a problem; 4 = always a problem). The item scores were reversely graded and linearly converted into a 0-100 scale (0 = 100, 1 = 75, 2 = 50, 3 = 25, 4 = 0) with higher scores indicating better functioning (or less negative impacts). In initial validation studies, PedsQL FIM scales demonstrated Cronbach's coe cient alpha scores > 0.70 [24].

Statistical analysis
Descriptive analysis was applied to describe the characteristics of the study population. Continuous variables were presented as means ± standard deviations (SDs), and categorical variables were shown as frequencies with percentages. Two independent samples t-test was used to compare the differences in the scores of QoL between two subgroups; while one-way ANOVA was used to compare the differences in the scores of QoL across more than two subgroups. In addition, we used the Cohen's effect size to assess the clinical relevance in terms of the between-groups difference. Effect size (Cohen's d) was calculated as: the absolute value of the difference in mean scores divided by the largest SD, and interpreted as (d): 0.2 ≤ d < 0.5 small difference, 0.5 ≤ d < 0.8 moderate, and d ≥ 0.8 large [25].
All tests were two-tailed and the statistical signi cance was indicated by p < 0.05. All the analyses were conducted by the SPSS version 22 (IBM Corp. NY, USA).

Results
Characteristics of the study population Table 1 presented the characteristics of socio-demographics andcharacteristics of disease and treatment of our study population. 101 caregivers with SMA children were involved in our study. 52 (51%) were female. All caregivers completed the PedsQL FIM questionnaire. Because the PedsQL NMM was only applicable for children aged from 2 to 18 years, a total of 87 caregivers completed NMM scale in our study. 46 (45.5%) patients were unable to sit alone; 39 (38.6%) able to sit alone but unable to walk alone; and 16 (15.8%) patients able to walk freely. 60 (59.4%) of patients had motor degeneration problems; 74 (73.3%) had skeletal deformity; and 23(22.8%) had a digestive system dysfunction. 39 (33.7%) patients sought for help regarding medical services, and 13 patients under multidisciplinary team (MDT) health management. Overall, 45(44.6%) of patients received exercise training in daily life; 26 (25.7%) received nutritional support; and 19 (18.8%) received respiratory support. Only two children in our study received scoliosis surgery, and nine children were using medication nusinersen.   Table 3 presented the differences in average scores of PedsQL FIM across SMA subtypes. The average total score and average domain scores of physical/emotional/social/cognitive functioning were higher in patients with SMA Type III compared with those in patients with type I or II (p < 0.05). The largest effect size was seen in the average domain scores of physical functioning when comparing parent self-reported physical functioning between subgroup type III and I (d = 1.04). There were no statistically signi cant differences in the above-mentioned average scores between type I and II (p > 0.05). Differences in the average scores of PedsQL NMM across disease-speci c characteristics Table 4 presented the average score of PedsQL NMM and the difference across disease-speci c characteristics. Children who were unable to walk alone and unable to sit and children who were able to sit but unable to walk were reported by their parents with relatively low total QoL score and low scores of neuromuscular disease domain, and family resources domain compared with parents with patients who were able to walk independently (p < 0.05). Particularly, comparing children who were unable to sit nor walk with those who were able to walk, effect sizes were 1.19, 1.35 and 1.07 in terms of the differences in the average total score, the average score of neuromuscular disease domain and of family resources domain. The average total score, and the average scores of neuromuscular diseases domain and family resource domain were signi cantly higher in children without motor degeneration than those children with motor degeneration (p < 0.01); the effect size ranged from 0.29 to 0.64. The average total score, and the average scores of neuromuscular diseases domain and of family resource domain were signi cantly higher in children without motor skeleton deformity than those with skeleton deformity (p < 0.05); the effect size ranged from 0.46 to 0.49. The average total scores, the average score of neuromuscular disease domain and of family resources domain were higher among children with normal digestive system function than those with digestive system dysfunction (p < 0.01); the effect size ranged from 0.68 to 1.06. Differences in the average scores of PedsQL FIM across disease-speci c characteristics Table 5 showed the average scores of PedsQL FIM across disease-speci c characteristics and effect sizes. Parents with children who were unable to walk alone reported lower QoL in average total score and the average scores of physical, emotional, social, cognitive domains compared with those with children who were able to walk independently (p < 0.05); the largest effect size was 1.40 regarding the comparison between non-sitter and walker. The average score of physical and emotional domains were signi cantly lower reported by parents whose children had skeletal deformity than those whose children had no deformity (p < 0.05).
Regarding digestive system dysfunction, the average total score and the average scores of physical, emotional, social, cognitive and daily activities domains were signi cantly lower among parents whose children had the dysfunction than those whose children did not have the dysfunction (p < 0.01); the effect sizes ranged from 0.56-1.26. Differences in the average scores of PedsQL NMM across the clinical treatments Table 6 demonstrated the average scores of parent-reported QoL measured by PedsQL NMM. Regarding the exercise training, the average total score and the average score of neuromuscular disease were signi cant higher in children who had the training than those who did not have (p < 0.001); the effect sizes were 0.60 and 0.75, respectively. Regarding the respiratory support, the average total score and the average score of neuromuscular disease were signi cant lower in children who had the support than those who did not have (p < 0.001); the effect sizes were 0.57 and 0.79, respectively. Regarding MDT, signi cant differences in total score, neuromuscular disease, and family resources domain scores were seen between the MDT health management and control groups.
Children accepted medical services but not under MDT were got signi cantly higher scores in total score, neuromuscular disease, and family resources domain than those without medical services (p < 0.05). And children under MDT have a signi cantly difference in total score, neuromuscular disease, and family resources domain than those without medical services groups (p < 0.05) or those accept medical services but not under MDT (p < 0.05). Effect sizes ranged from 0.13 to 0.90.

Discussion
Our study found that quality of life of patients and caregivers differed across SMA types,disease-related and treatment characteristics. A signi cant difference among SMA types presented in the domain of neuromuscular diseases on the PedsQL NMM, with type III having the highest score and type I the lowest score.
By contrast, Meaghann et al. found there was no signi cant difference among SMA types in the eld of neuromuscular diseases in the proxy-report of the PedsQL NMM [22]. A reasonable explanation for this nding is that in our study the sample size was larger than in the previous study and the children's physical function was worse in severe types, indirectly re ecting the severity of the disease. Compared with patients with SMA types II and I, patients with type III had signi cantly higher scores in family resources, which is in accordance with the previous study [22]. In contrast, our results showed there was no signi cant difference among SMA types in the communication domain. We infer that these negative results may be related to social and humanistic characteristics in China -that is, all patients received care and concern from their relatives, friends, community, and medical staff. This result further indicates that all members of Chinese society are concerned with children's health. In terms of the total score of PedsQL NMM, patients with type III showed a signi cant difference compared with types II and I. These results are consistent with previous reports indicating that SMA types re ect the physical function of patients.
Additionally, we also found the effect sizes were large or moderate in different subtypes of SMA, which indicated that the differences in quality of life of patients has reached the threshold of clinical relevance, which worth attention from health professionals. And we found there is no statistical difference in neuromuscular disease domain between children with type I and type II, but the effect size was moderate, which indicated that there was a clinical difference in neuromuscular disease domain between children with type I and with type II.
Our results showed that the physical, emotional, societal, and cognition items in the Family Impact Module were effective for comparing patients with different types of SMA. Patients with SMA type III tend to have higher scores than the other two types, which is related to the patients' motor ability. Disease-related characteristics of patients with SMA such as skeletal malformation, digestive dysfunction, motor degeneration, and current motor ability obviously in uence the PedsQL NMM and PedsQL FIM results. SMA is a disease often involving dysfunction of multiple systems, as seen in the reduction in the distribution of SMN protein to the body's multiple organs and tissues [10, 11]. As SMA is a progressive disease, some complications such as skeletal malformation, digestive disorders, respiratory disorders, and motor function degeneration appear gradually. Our ndings showed that the total score, neuromuscular disease score, and family score of PedsQL NMM were signi cantly lower in patients with these complications than in those without them. Our results support the ndings from previous study -that is, motor function is reduced in patients with skeletal deformities, which in turn affects all aspects of the patient's life [26]. Digestive disorders make malnutrition a higher risk in patients, which may cause fatal or serious disorders [27,28]. The degeneration in motor ability and current movement ability can indirectly re ect the patient's survival condition, and all of the disease-related characteristics can have a signi cant effect on QoL. In the PedsQL FIM, the physical and emotional functions of the caregivers of patients with skeletal malformations were worse than those of the caregivers of patients without skeletal malformations. The skeletal malformations of patients with SMA often include scoliosis, muscle contracture, dislocation of hip joint, and other conditions, which often lead to the degeneration or loss of motor ability [29]. Faced with this situation, caregivers have to increase the daily nursing level of patients to achieve a high QoL. For so long caregivers' physical health and mental health are greatly affected. Dysphagia and other digestive disorders often appear in the last stage of SMA, which seriously affects the patients' physical health, and reduces their caregivers' QoL and family relationships. Individual physical and psychological functions may be signi cantly affected by mobility, and the caregivers of patients who can walk freely require less care and energy than the caregivers of patients who have lost the motor ability to walk. This means that when caregivers concentrate their energy on their children, the caregivers have frequent and serious limits to their own lives and increased stress on their own bodies, which in turn can lead to the onset of anxiety, fear, depression, and other physical or mental side-effects. In addition to the statistical signi cance, effect sizes in terms of the above-mentioned domains were moderate or large, which indicated disease-related characteristics may impact quality of life of patients and caregivers, which warranted attention from clinical professionals during their practice. respiratory support have lower neuromuscular disease and family resources scores compared with those without respiratory support. One possible explanation for this unexpected result is that patients receive respiratory support when their condition has deteriorated, which in turn seriously affects their QoL and simultaneously requires caregivers to spend more energy in managing daily life. Many studies have con rmed that respiratory support is helpful in improving pulmonary function as SMA is a progressive disease involving multiple system dysfunction, including respiratory complications, and patients with type I often die of respiratory failure. Based on our ndings, we suggest that patients with SMA type I should adopt mechanical ventilation strategies early to maintain respiratory function effectively, preventing respiratory complications and pulmonary infections, and subsequently improving their QoL.
Recently, nusinersen became the rst effective drug treatment for SMA available in China. As previous research noted, nusinersen treatment can improve the motor, respiratory, digestive, and other system functions in patients [39][40][41][42]. In our study, six caregivers of patients who were receiving nusinersen treatment complete the PedsQL NMM. Although the questionnaire results for these six patients showed that the neuromuscular diseases and family resource scores were higher than those of patients who did not use nusinersen, the caregivers of these six patients reported that drug treatment effectively improved the patients' QoL. We cannot make de nite conclusions on whether their QoL was improved by nusinersen use because of the small sample size. These patients also received exercise training, so it is possible that a combination of the two factors increased QoL. The sample size should be increased in follow-up studies and should include an extended course of nusinersen treatment to verify whether this treatment improves patients' QoL.
Health management has been de ned as an integrated approach to assess, guide, and intervene in the risk factors affecting the health of a group or an individual, based on the modern concept of health and the new medical models and with the theories, techniques, and means of modern medicine and Our study had several limitations. First, the sample size included in this study is modest, re ecting the rarity of the disease, and the statistical effectiveness of our analysis is limited as well. A second limitation of this study is that patients were only sampled from central and east China. Given the lack of QoL survey data on patients with SMA in other regions, the study may not be representative of the QoL of patients with SMA across China. Nevertheless, our results may be generalized to patients from the areas mentioned above. Third, this study was a cross-sectional cohort study with data collected at a single time point, which precluded drawing conclusions regarding causality. Further, the study re ected the in uence of disease-related characteristics on the QoL of children and the initial effect of medical measures only. Finally, although the PedsQL is designed to be used for all types of disease, it has not been speci cally validated for proxy-reported use in children with SMA.
To sum up, further longitudinal follow-up of these patients should be conducted and QoL indicators should be used to investigate the effectiveness of clinical treatments for childhood SMA. Furthermore, future studies need larger sample sizes and should pursue multicenter enrollment to involve patients with SMA from all of China to obtain a database that re ects the QoL of patients across the country. In addition, with the continuous progress in the diagnosis and treatment of SMA, larger patient sample sizes, especially in the drug treatment group, should be included in subsequent studies to con rm the effectiveness of treatment approaches and to make the results more generalizable.

Conclusion
In summary, he more severe the SMA disorder was, the lower were the scores on the PedsQL NMM and PedsQL FIM as reported by the patients' caregivers, and the poorer was the patients' QoL