2.1 Study design and population
A multicenter cross-sectional study was conducted from October 2023 to March 2024. The questionnaire was sent to 9 tertiary children’s and comprehensive hospitals in different regions of China by sharing the link to the questionnaire online. These regions included eastern, central and western China. The managers of the neurology units in the survey gave their consent before the survey. 226 children and adolescents with epilepsy and their caregivers were recruited.
Participants were recruited using convenience sampling. The inclusion criteria comprised of: (1) children and adolescents ages ranging from 8 to 18 years and diagnosed with epilepsy by neurologists according the 2014 International League Against Epilepsy criteria for 2 months or longer[31]; (2) caregivers older than 18 years who provided care for children and adolescents with epilepsy, only one of the caregivers who was present with the child during the period of data collection was included in the study; (3) both child and caregiver were able to read and understand the questionnaire independently or with the help of the researcher; (4) willing to participate the survey. The exclusion criteria were: (1) children and adolescents with intellectual impairment, an IQ score lower than 80 on the Wechsler Intelligence Scale; (2) caregivers were diagnosed with severe medical conditions, cognitive impairment, or mental illness.
The sample size of this study was determined by the number of parameters tested. A minimum ratio of sample size to parameters of 10:1 was recommended. 17 parameters were observed in the study. Considering the invalid response rate of 15%, the required sample size was 196. In the end, a total of 226 patients were recruited.
2.2 Instrument
2.2.1 Sociodemographic characteristics
The self-designed questionnaire was used to collect basic sociodemographic characteristics of children and adolescents and their families. The data included age, gender, region of residence, whether it is a one-child family, average monthly household income, height and weight. These were mainly collected by the caregivers.
2.2.2 Disease-related characteristics
Disease-related characteristics involved age of seizure onset in years, disease duration, seizure type, whether it is drug-resistant epilepsy, frequency of seizures in recent 3 months, and comorbidities. These were mainly collected by medical records and self-report of caregivers.
2.2.3 Stigma
The Kilifi Stigma Scale for Epilepsy (KSSE) is a three-point Likert 15-item scoring scale[32], which was used to measure stigma. The Chinese version of KSSE were translated and validated by Song et al[33]. The translated version exhibits good reliability and validity (Cronbach's α=0.93). This scale is the only validated scale of stigma for epilepsy patients in China. The total score ranges from 0 to 30. A score above the 66th percentile of the total score indicates perceived stigma, while a score below the 66th percentile shows the absence of stigma.
2.2.4 Psychological distress
The Depression Anxiety and Stress Scale (DASS-21) is a shortened version of the full 42-item DASS developed by Lovibond and Lovibond[34], which is a common instrument for screening psychological distress symptoms during the past week. The Chinese version of the DASS-21 showed good reliability and validity in Chinese samples (The total Cronbach's α=0.91)[35, 36]. This scale comprises 21 items grouped into three subscales: depression, anxiety, and stress. Each item is rated on a 4-point Likert scale from 0 (not true) to 3 (very true). The scores are summed and a higher score indicates a higher level of psychological distress.
2.2.5 Physical Activity (PA)
The Physical Activity Rating Scale-3 (PARS-3), a three-question scale, was used to measure the level of physical activity of participants in terms of intensity, duration and frequency of physical activity[37]. Rating of each item on a scale of 1 to 5 and the total score for physical activity were computed using the following equation: intensity × (duration - 1) × frequency, with a range of 0-100, and was further classified into three levels according to the score: high (>42), moderate (20-42), and low (0-19) physical activity. The scale demonstrated strong reliability and validity in Chinese individuals (Cronbach's α=0.86)[28], with a excellent test-retest reliability (r = 0.82)[38].
2.3 Ethical Statement
Participation in this study was completely voluntary and anonymous. Researchers obtained participants and their caregivers informed consent before completing the questionnaire. This study was approved by the Medical Ethical Committee of Children’s Hospital of Chongqing Medical University (No.2023-370).
2.4 Data collection methods
A multicenter cross-sectional survey study was conducted from October 2023 to March 2024 in China. A total of 9 hospitals in 8 cities of 6 provinces were investigated (at least ten valid questionnaires in each hospitals were distributed) in China. The questionnaire was distributed to pediatric epilepsy specialist clinics and specialty wards. The participants were recruited using a convenient sampling method. The researchers were trained in advance and then provided participants with unified instructions, instructions for completing the questionnaire, and links to access the questionnaire. Researchers obtained their informed consent before completing the questionnaire in face-to-face interviews. For children and adolescents, consent was obtained from the child's caregiver. Questionnaires were completed separately by children and adolescents and their caregivers. If the questionnaire was incomplete, the respondent was asked to complete the missing items immediately. 239 children and adolescents with epilepsy and caregivers was selected. A total of 13 patients declined to participate due to lack of interest (five patients), time (four patients), or other unspecified reasons (four patients). In the end, 226 (94.6%) participants were recruited and completed the survey. There were no statistically significant differences in the distribution of age, gender, or seizure type between respondents and non-respondents.
2.5 Statistical analysis
The SPSS version 26 was used to analyze all data. Normally distributed measurement data were described by mean and standard deviation (SD), whereas non-normally distributed measurement data were described by median and quartile. The categorical variables were stated as the frequency and percentage. One-way ANOVA and the S-N-K test were used to assess continuous data, while the Chi-square test was used to assess categorical variables.
All analyses were adjusted for age, gender, one-child family, place of residence, average monthly household income, and seizure type. We used multivariate linear regression to determine the relationship among stigma, PA, and psychological distress. Andrew F Hayes’ PROCESS macro v.3.5 was used to develop the mediation model. Psychological distress (depression, anxiety, and stress) was the dependent variable (Y), PA (intensity, duration, and frequency) was separately the mediating variable (M), and stigma was the independent variable (X).
The combination of the direct and indirect effects of stigma on psychological distress was the total impact (path c). After adjusting for PA, the direct effect (path c’) was the effect of stigma on psychological distress. The indirect effect (path ab) mediated the relationship between stigma and psychological distress. In addition, the PA components were investigated independently as mediating variables. We used the PROCESS macro’s bootstrap approach. The bootstrap method produces 95% bias-corrected CI for these effects from 5000 re-sample of the data. CI that do not contain zero indicate a significant effect[39].