(safety population), described within the medical review. Next, we used Thomson Reuters’ Incidence and Prevalence Database,3 primary epidemiology literature, and summary reports to estimate the number of US patients who were potentially eligible for treatment (target population). We then calculated the ratio of safety population to the target population, categorizing therapeutics as having a ratio in the top quartile (ie, larger safety population relative to target population) or below the 75th percentile. The threshold ratio for categorization in the top quartile was 0.005, implying that the equivalent (or more) of 0.5% of the total number of US patients who were potentially eligible for treatment were used for the FDA’s therapeutic safety analysis. We repeated our original multivariable model, including the 7 novel therapeutic characteristics and features of their regulatory approval, as well as expected length of treatment and safety population:target population ratio. Neither characteristic was statistically associated with increased risk of postmarket safety events. For expected length of treatment, using short-term treatment as the reference, the incidence rate ratio (IRR) for intermediate treatment was 0.67 (95% CI, 0.35-1.29), whereas the IRR for long-term treatment was 0.88 (95% CI, 0.34-2.29). For the safety population:target population ratio, using the top quartile as the reference, the IRR for therapeutics below the 75th percentile was 1.17 (95% CI, 0.632.16). Further research should consider additional novel therapeutic characteristics or features of their regulatory approval that may be potential predictors of an increased risk of postmarket safety events.

中文翻译：

---

美国医生短缺或过剩——答复

（安全人群），在医学评论中描述。接下来，我们使用汤森路透的发病率和患病率数据库、3 主要流行病学文献和总结报告来估计可能符合治疗条件的美国患者（目标人群）的数量。然后我们计算了安全性人群与目标人群的比率，将治疗归类为比例在前四分位数（即相对于目标人群更大的安全性人群）或低于第 75 个百分位数。前四分位数的分类阈值为 0.005，这意味着 FDA 的治疗安全性分析使用了相当于（或更多）可能符合治疗条件的美国患者总数的 0.5%。我们重复了我们原来的多变量模型，包括 7 种新的治疗特性和监管批准的特征，以及预期的治疗时间和安全人群：目标人群比例。这两个特征都与上市后安全事件的风险增加没有统计学相关性。对于预期的治疗时间，以短期治疗为参考，中间治疗的发生率比（IRR）为 0.67（95% CI，0.35-1.29），而长期治疗的 IRR 为 0.88（95% CI，0.34-2.29）。对于安全性人群：目标人群比例，以最高四分位数为参考，低于 75% 的治疗的 IRR 为 1.17（95% CI，0.632.16）。