Racial and Ethnic Bias in Risk Prediction Models for Colorectal Cancer Recurrence When Race and Ethnicity Are Omitted as Predictors.
Racial and Ethnic Bias in Risk Prediction Models for Colorectal Cancer Recurrence When Race and Ethnicity Are Omitted as Predictors.
复制标题
DOI：
10.1001/jamanetworkopen.2023.18495
复制
发表时间：
2023-06-01
影响因子：
13.8
通讯作者：
Bansal, Aasthaa
中科院分区：
文献类型：
--
作者：
Khor, Sara;Haupt, Eric C.;Hahn, Erin E.;Lyons, Lindsay Joe L.;Shankaran, Veena;Bansal, Aasthaa
在结直肠癌复发风险预测模型中，省略种族和民族是否与种族和民族偏见有关？在这项4230例结直肠癌患者的预后研究中，建立了4个预测癌症术后复发风险的模型并进行了验证。明确考虑种族和民族作为预测因子，改善了种族和少数民族患者的模型预测性能，并在多个性能测量中增加了算法公平性。这些发现表明，实施简单地忽略种族和民族的临床风险模型可能会导致对种族和少数民族群体的预测准确性较差，这可能会导致不适当的护理建议，最终导致健康差异。这项预后研究考察了在结直肠癌(CRC)复发的临床风险模型中将种族和民族作为预测因素是否与种族偏见有关，种族偏见的定义是模型准确性的差异，可能导致小规模种族和民族群体之间的不平等待遇。在临床风险预测算法中将种族和民族作为预测因子已受到越来越多的审查，但仍然缺乏实证研究，以确定算法中简单地省略种族和民族是否最终会影响小型化种族和民族群体患者的决策。为了检验在结直肠癌复发风险算法中将种族和民族作为预测因子是否与种族偏见有关，种族偏见被定义为可能导致不平等治疗的模型准确性的种族和民族差异。这项前瞻性预后研究使用了南加州大型综合医疗保健系统的数据，这些患者在2008至2013年间接受了初步治疗，并进行了随访，直至2018年12月31日。分析了2021年1月至2022年6月的数据。四个Cox比例风险回归预测模型被用来预测从监测开始到癌症复发的时间：(1)明确排除种族和民族作为预测因素的种族中立模型，(2)包括种族和民族的种族敏感模型，(3)临床预测因素与种族和民族之间双向交互作用的模型，以及(4)按种族和民族分开的模型。通过模型校正、判别能力、假阳性率和假阴性率、阳性预测值(PPV)和阴性预测值(NPV)来评估算法的公平性。研究队列包括4230名患者(平均年龄65.3[12.5]岁；女性2034[48.1%]；亚洲、夏威夷或太平洋岛民490[11.6%]；黑人或非裔美国人554[13.1%]；西班牙裔937[22.1%]；非西班牙裔白人2249[53.1%])。种族中立模型在种族和少数民族亚组中的校正、NPV和假阴性率比非西班牙裔白人个体更差(例如，西班牙裔患者的假阴性率：12.0%[95%CI，6.0%-18.6%]；非西班牙裔白人患者：3.1%[95%CI，0.8%-6.2%])。加入种族和民族作为预测因素，改善了校正斜率、辨别能力、PPV和假阴性率的算法公平性(例如，西班牙裔患者的假阴性率：9.2%[95%CI，3.9%-14.9%]；非西班牙裔白人患者：7.9%[95%CI，4.3%-11.9%])。包括种族交互作用项或使用种族分层模型并不能改善模型的公平性，这可能是因为子组中的样本量较小。在这项关于癌症复发风险算法中种族偏见的预后研究中，删除种族和民族作为预测因子会恶化多项测量中的算法公平性，这可能会导致对属于小规模种族和民族群体的患者的不适当护理建议。临床算法开发应包括公平标准的评估，以了解消除种族和民族对健康不平等的潜在后果。
Is omitting race and ethnicity as a predictor in colorectal cancer recurrence risk prediction models associated with racial and ethnic bias? In this prognostic study with 4230 patients with colorectal cancer, 4 prediction models for risk of postoperative cancer recurrence were developed and validated. Explicitly considering race and ethnicity as a predictor improved model predictive performance among racial and ethnic minority patients and increased algorithmic fairness in multiple performance measures. These findings suggest that implementing clinical risk models that simply omit race and ethnicity may result in worse prediction accuracy for racial and ethnic minority groups that may lead to inappropriate care recommendations that ultimately contribute to health disparities. This prognostic study examines whether including race and ethnicity as a predictor in a clinical risk model for colorectal cancer (CRC) recurrence is associated with racial bias, defined as differences in model accuracy that could lead to unequal treatment among minoritized racial and ethnic groups. Including race and ethnicity as a predictor in clinical risk prediction algorithms has received increased scrutiny, but there continues to be a lack of empirical studies addressing whether simply omitting race and ethnicity from the algorithms will ultimately affect decision-making for patients of minoritized racial and ethnic groups. To examine whether including race and ethnicity as a predictor in a colorectal cancer recurrence risk algorithm is associated with racial bias, defined as racial and ethnic differences in model accuracy that could potentially lead to unequal treatment. This retrospective prognostic study was conducted using data from a large integrated health care system in Southern California for patients with colorectal cancer who received primary treatment between 2008 and 2013 and follow-up until December 31, 2018. Data were analyzed from January 2021 to June 2022. Four Cox proportional hazards regression prediction models were fitted to predict time from surveillance start to cancer recurrence: (1) a race-neutral model that explicitly excluded race and ethnicity as a predictor, (2) a race-sensitive model that included race and ethnicity, (3) a model with 2-way interactions between clinical predictors and race and ethnicity, and (4) separate models by race and ethnicity. Algorithmic fairness was assessed using model calibration, discriminative ability, false-positive and false-negative rates, positive predictive value (PPV), and negative predictive value (NPV). The study cohort included 4230 patients (mean [SD] age, 65.3 [12.5] years; 2034 [48.1%] female; 490 [11.6%] Asian, Hawaiian, or Pacific Islander; 554 [13.1%] Black or African American; 937 [22.1%] Hispanic; and 2249 [53.1%] non-Hispanic White). The race-neutral model had worse calibration, NPV, and false-negative rates among racial and ethnic minority subgroups than non-Hispanic White individuals (eg, false-negative rate for Hispanic patients: 12.0% [95% CI, 6.0%-18.6%]; for non-Hispanic White patients: 3.1% [95% CI, 0.8%-6.2%]). Adding race and ethnicity as a predictor improved algorithmic fairness in calibration slope, discriminative ability, PPV, and false-negative rates (eg, false-negative rate for Hispanic patients: 9.2% [95% CI, 3.9%-14.9%]; for non-Hispanic White patients: 7.9% [95% CI, 4.3%-11.9%]). Inclusion of race interaction terms or using race-stratified models did not improve model fairness, likely due to small sample sizes in subgroups. In this prognostic study of the racial bias in a cancer recurrence risk algorithm, removing race and ethnicity as a predictor worsened algorithmic fairness in multiple measures, which could lead to inappropriate care recommendations for patients who belong to minoritized racial and ethnic groups. Clinical algorithm development should include evaluation of fairness criteria to understand the potential consequences of removing race and ethnicity for health inequities.
登录
查看更多内容
影响因子：
39.2
作者：
Chapman, Christina Hunter;Schechter, Clyde B.;Cadham, Christopher J.;Trentham-Dietz, Amy;Gangnon, Ronald E.;Jagsi, Reshma;Mandelblatt, Jeanne S.
通讯作者：
Mandelblatt, Jeanne S.
影响因子：
3
作者：
Hassett MJ;Ritzwoller DP;Taback N;Carroll N;Cronin AM;Ting GV;Schrag D;Warren JL;Hornbrook MC;Weeks JC
通讯作者：
Weeks JC
影响因子：
4.2
作者：
Ahn, Jaeil;Harper, Sam;Luta, George
通讯作者：
Luta, George
影响因子：
2.5
作者：
Derose, Stephen F.;Contreras, Richard;Jacobsen, Steven J.
通讯作者：
Jacobsen, Steven J.
DOI：
10.1016/s0140-6736(20)32076-6
发表时间：
2020-10-10
期刊：
Lancet (London, England)
影响因子：
--
作者：
Cerdeña JP;Plaisime MV;Tsai J
通讯作者：
Tsai J