肿瘤靶向药物不良反应发生率的Meta分析

目的探讨肿瘤患者使用靶向药物治疗后引发的不良反应发生率。方法从建库至2020年7月1日检索中国期刊全文数据库、中文期刊全文数据库、万方数据知识服务平台等数据库,获取现有的不同类型肿瘤靶向药物及不良反应发生率的相关文献,根据纳入和排除标准选取临床研究文献,采用ReviewManager 5.3软件对肿瘤靶向药物不良反应发生率进行Meta分析。结果共检索出相关文献568篇,通过浏览标题和摘要后筛选68篇文献精读,剔除重复研究和与结局指标无关的文献,最终纳入文献8篇,其中共568例受试者接受过药物不良反应评价。采用Cochrane风险偏倚评估工具对纳入的文献进行风险评估,本研究中偏倚检测提示无显著偏倚。Meta分析结果显示,分别有3篇文献、7篇文献、7篇文献的临床结局涉及到恶心呕吐、腹泻、皮肤不良反应,各结局研究间的异质性较显著,I2值分别为66%、53%、64%,故以随机效应模型(random effects models,REM)进行合并。Meta分析结果显示,上述结局合并发生率及其95%置信区间(95%CI)分别为7.26(2.94~17.91)、1.00(0.74~1.36)、1.01(0.76~1.33)。结论靶向药物可引发一定占比的肿瘤患者发生药物不良反应,应重视肿瘤靶向药物治疗引发的不良反应并对其严重性进行有效评估,提高肿瘤患者的生命质量和治疗依从性,继而提升疗效。     

Repairing damaged reputations through targeted poverty alleviation: Evidence from private companies’ strategies to deal with negative media coverage

When negative media coverage causes reputational crises, companies must find suitable tools to repair their reputation and reverse their negative image. As a CSR activity with political- and livelihood-related implications, targeted poverty alleviation may be an effective tool. Using data on negative media coverage of Chinese A-share private listed companies, we examine whether companies engage in targeted poverty alleviation in response to reputational crises caused by negative media coverage. We find that negative media coverage leads private companies to engage more actively and intensively in targeted poverty alleviation because of the significant increase in public attention to the bad news. These companies must urgently rebuild their positive image using targeted poverty alleviation to resolve their public opinion crisis. Further analyses suggest that original and in-depth negative media coverage is more likely to cause companies’ active participation in targeted poverty alleviation. In addition, negative media coverage is more likely to lead companies to engage in targeted poverty alleviation when they are in heavily polluting industries or face greater pressure from external investors. Finally, we find that active involvement in targeted poverty alleviation helps companies improve their market reputation and thus effectively manage public relations crises caused by negative media coverage.     

Targeting predictors in random forest regression

Random forest (RF) regression is an extremely popular tool for analyzing high-dimensional data. Nonetheless, its benefits may be lessened in sparse settings due to weak predictors, and a pre-estimation dimension reduction (targeting) step is required. We show that proper targeting controls the probability of placing splits along strong predictors, thus providing an important complement to RF’s feature sampling. This is supported by simulations using finite representative samples. Moreover, we quantify the immediate gain from targeting in terms of the increased strength of individual trees. Macroeconomic and financial applications show that the bias–variance trade-off implied by targeting, due to increased correlation among trees in the forest, is balanced at a medium degree of targeting, selecting the best 5%–30% of commonly applied predictors. Improvements in the predictive accuracy of targeted RF relative to ordinary RF are considerable, up to 21%, occurring both in recessions and expansions, particularly at long horizons.