Supplementary Material for: Secular Trends in the Incidence, Prevalence, and Medications for Epilepsy from 2007 to 2015 in Taiwan: A Nationwide Population-Based Study

<b><i>Background:</i></b> Patients with epilepsy have a higher mortality rate than the general population. Up-to-date estimates of epilepsy incidence, prevalence, and medication use are critical to assist policymaking. <b><i>Methods:</i></b> Using the National Taiwan Insurance Research Database, the standardized incidence and prevalence of epilepsy were estimated in each calendar year from 2007 to 2015. We used the incident cases of epilepsy to analyze the change in prescribing patterns from 2007 to 2015. Joinpoint regression was used to estimate secular trends. <b><i>Results:</i></b> From 2007 to 2015, the age- and sex-standardized incidence decreased from 0.72 (95% confidence interval [CI] 0.70–0.73) to 0.54 (95% CI 0.53–0.55) per 1,000 person-years, giving an annual percentage change (APC) of −2.73 (<i>p</i> &lt; 0.05). Among patients younger than 20 years, the incidence did not change significantly. The age- and sex-standardized prevalence decreased from 6.94 (95% CI 6.90–6.98) to 6.86 (95% CI, 6.82–6.89) per 1,000 people, giving an APC of −0.31 (<i>p</i> &lt; 0.05). However, the prevalence increased in the 35- to 49- and 50- to 64-year age-groups. The most common first-line anticonvulsant was phenytoin in 2007 and valproate in 2015. The use of levetiracetam, clobazam, and valproate increased during the study period, with APCs of 25.48% (95% CI 19.97–31.24), 6.41 (3.09–9.85), and 2.83 (1.51–4.16), respectively. The use of carbamazepine, phenytoin, and topiramate decreased; the APCs were −23.86% (95% CI −25.25 to −22.44), −6.61 (−8.40 to −4.79), and −4.29% (−7.87 to −0.57), respectively. <b><i>Conclusions:</i></b> The overall prevalence and incidence of epilepsy decreased slightly from 2007 to 2015. The prescribed first-line anticonvulsant also changed over time.

背景：癫痫患者的死亡率高于普通人群。最新的癫痫发病率、患病率及药物使用情况估算数据，对辅助政策制定至关重要。
方法：本研究依托中国台湾地区全民健康保险研究数据库（National Taiwan Insurance Research Database），估算了2007至2015年各历年癫痫的标准化发病率与患病率。我们以癫痫新发病例为研究对象，分析了2007至2015年的处方模式变化。采用Joinpoint回归（Joinpoint regression）估计长期趋势。
结果：2007至2015年，校正年龄与性别的标准化发病率由每1000人年0.72（95%置信区间（confidence interval, CI）0.70–0.73）降至0.54（95% CI 0.53–0.55），年均变化百分比（annual percentage change, APC）为-2.73（p < 0.05）。在20岁以下患者中，发病率无显著变化。校正年龄与性别的标准化患病率由每1000人6.94（95% CI 6.90–6.98）降至6.86（95% CI 6.82–6.89），年均变化百分比为-0.31（p < 0.05）。但35~49岁及50~64岁年龄组的患病率呈上升趋势。2007年最常用的一线抗惊厥药（anticonvulsant）为苯妥英，2015年则为丙戊酸盐。研究期间，左乙拉西坦、氯巴占及丙戊酸盐的使用量呈上升趋势，年均变化百分比分别为25.48%（95% CI 19.97–31.24）、6.41（95% CI 3.09–9.85）及2.83（95% CI 1.51–4.16）。卡马西平、苯妥英及托吡酯的使用量呈下降趋势，年均变化百分比分别为-23.86%（95% CI -25.25 ~ -22.44）、-6.61（95% CI -8.40 ~ -4.79）及-4.29%（95% CI -7.87 ~ -0.57）。
结论：2007至2015年，癫痫的总体患病率与发病率均小幅下降。一线抗惊厥药的处方选择也随时间发生了变化。