Addendum: Supplementary Information of Questionnaire
Parental education
Paternal and maternal education attainments were assessedseparately by multiple choice including elementary school, junior high school, senior high school, college/university, and graduate school. Parental education levels were further dichotomized according to the length of school years. In Taiwan, it generally requires 12 years to complete the prerequisite study before entering university or college. Given a strong correlation found between paternal and maternal educational attainment (Spearman σ = 0.667, p < 0.001), we included only maternal education in the subsequent analyses.
Living situation
Living situation was defined by the family members residing in the same household. The choices included “living with both parents”, “living with single parent”, “living with other family relatives who are not parents”, and “living in the dormitory”. The last two categories were recombined into one group named as “not living with parents” in the analysis.
Family relationship
Participants were asked to score their family relationship with a 5-point scale consisting of 1 (very distant), 2 (somewhatdistant), 3 (neutral), 4 (somewhat close), and 5 (very close). The answers were regrouped into three categories: distant (scores 1 and 2), neutral (score 3), and close (scores 4 and 5). This variable remained as an ordinal variable in the regression models when analyzing the socio-demographiccorrelates of adolescent self-rated health.
Tutorial class
Hours spent in after-school tutorial classes were assessed by multiple choices including “less than 2 hours”, “2-3 hours”, and “more than 3 hours”. The answers were regrouped into those with 3 or less than 3 hours, and the others with more than 3 hours.
Chronic illness
Participants were asked to report whether they had diseases which required regular clinical follow-up, such as asthma, diabetes mellitus, congenital heart disease, etc.
School location
Schools were categorizedinto urban, suburban, and rural according to the urbanicity of the municipalities where they were located.
Clinical visits in the past three months
Participants were asked to report the times of clinical consultation in the past three months. They were allowed to choose from “none”, “1-3 times”, “4-6 times”, “7-9 times”, and “more than 10 times”. The last three categories were recombined into one group named as “more than 3 times” in the analysis.
Self-rated health status
Two questions were used to assess the self-rated health status: “How do you feel your overall health is at this moment?” and “How do you feel your overall health is compared to your peers that are the same age?”Answers consisted of 1 (very poor/much worse), 2 (poor/worse), 3 (neutral/equal), 4 (good/better), and 5 (very good/much better). The alpha coefficient for the scale was 0.77. The sums of scores for both questions were regrouped into three categories: “poor health status” (scores from 2 to 4), “intermediate health status” (scores from 5 to 7), and “good health status” (scores from 8 to 10).