Haub School of Environment and Natural Resources, University of Wyoming, Laramie.
College of Health Solutions, Arizona State University, Phoenix.
JAMA Netw Open. 2021 Nov 1;4(11):e2132388. doi: 10.1001/jamanetworkopen.2021.32388.
Prostate cancer screening and diagnosis exhibit known racial and ethnic disparities. Whether these disparities persist in prostate magnetic resonance imaging (MRI) utilization after elevated prostate-specific antigen (PSA) results is poorly understood.
To assess potential racial and ethnic disparities in prostate MRI utilization following elevated PSA results.
DESIGN, SETTING, AND PARTICIPANTS: This cohort study of 794 809 insured US men was drawn from deidentified medical claims between January 2011 and December 2017 obtained from a commercial claims database. Eligible participants were aged 40 years and older and received a single PSA result and no prior PSA screening or prostate MRI claims. Analysis was performed in January 2021.
Multivariable logistic regression was used to examine associations between elevated PSA results and follow-up prostate MRI. For patients receiving prostate MRI, multivariable regressions were estimated for the time between PSA and subsequent prostate MRI. PSA thresholds explored included PSA levels above 2.5 ng/mL, 4 ng/mL, and 10 ng/mL. Analyses were stratified by race, ethnicity, and age.
Of 794 809 participants, 51 500 (6.5%) had PSA levels above 4 ng/mL; of these, 1524 (3.0%) underwent prostate MRI within 180 days. In this sample, mean (SD) age was 59.8 (11.3) years (range 40-89 years); 31 350 (3.9%) were Asian, 75 935 (9.6%) were Black, 107 956 (13.6%) were Hispanic, and 455 214 (57.3%) were White. Compared with White patients, Black patients with PSA levels above 4 ng/mL and 10 ng/mL were 24.1% (odds ratio [OR], 0.78; 95% CI, 0.65-0.89) and 35.0% (OR, 0.65; 95% CI, 0.50-0.85) less likely to undergo subsequent prostate MRI, respectively. Asian patients with PSA levels higher than 4 ng/mL (OR, 0.76; 95% CI, 0.58-0.99) and Hispanic patients with PSA levels above 10 ng/mL (OR, 0.77; 95% CI, 0.59-0.99) were also less likely to undergo subsequent prostate MRI compared with White patients. Black patients between ages 65 and 74 years with PSA above 4 ng/mL and 10 ng/mL were 23.6% (OR, 0.76; 95% CI, 0.64-0.91) and 43.9% (OR, 0.56; 95% CI, 0.35-0.91) less likely to undergo MRI, respectively. Race and ethnicity were not significantly associated with mean time between PSA and MRI.
Among men with elevated PSA results, racial and ethnic disparities were evident in subsequent prostate MRI utilization and were more pronounced at higher PSA thresholds. Further research is needed to better understand and mitigate physician decision-making biases and other potential sources of disparities in prostate cancer diagnosis and management.
前列腺癌筛查和诊断存在已知的种族和民族差异。在 PSA 升高后，前列腺磁共振成像（MRI）的利用是否存在这些差异尚不清楚。
评估 PSA 升高后前列腺 MRI 利用的潜在种族和民族差异。
设计、地点和参与者：本队列研究从商业索赔数据库中获得的 2011 年 1 月至 2017 年 12 月期间的匿名医疗索赔中抽取了 794809 名美国男性参保者。合格的参与者年龄在 40 岁及以上，接受过单次 PSA 检测，且无既往 PSA 筛查或前列腺 MRI 检测。分析于 2021 年 1 月进行。
多变量逻辑回归用于检查 PSA 升高结果与后续前列腺 MRI 之间的关联。对于接受前列腺 MRI 的患者，对 PSA 与后续前列腺 MRI 之间的时间进行了多变量回归估计。分析中探索的 PSA 阈值包括 PSA 水平高于 2.5ng/mL、4ng/mL 和 10ng/mL。分析按种族、族裔和年龄进行分层。
在 794809 名参与者中，51500 名（6.5%）PSA 水平高于 4ng/mL；其中，1524 名（3.0%）在 180 天内接受了前列腺 MRI。在这个样本中，平均（SD）年龄为 59.8（11.3）岁（范围 40-89 岁）；31350 名（3.9%）为亚洲人，75935 名（9.6%）为黑人，107956 名（13.6%）为西班牙裔，455214 名（57.3%）为白人。与白人患者相比，PSA 水平高于 4ng/mL 和 10ng/mL 的黑人患者接受后续前列腺 MRI 的可能性分别降低了 24.1%（比值比[OR]，0.78；95%CI，0.65-0.89）和 35.0%（OR，0.65；95%CI，0.50-0.85）。PSA 水平高于 4ng/mL 的亚洲患者（OR，0.76；95%CI，0.58-0.99）和 PSA 水平高于 10ng/mL 的西班牙裔患者（OR，0.77；95%CI，0.59-0.99）接受后续前列腺 MRI 的可能性也低于白人患者。PSA 水平高于 4ng/mL 和 10ng/mL 的 65-74 岁黑人患者接受 MRI 的可能性分别降低了 23.6%（OR，0.76；95%CI，0.64-0.91）和 43.9%（OR，0.56；95%CI，0.35-0.91）。种族和族裔与 PSA 和 MRI 之间的平均时间无显著相关性。
在 PSA 升高结果的男性中，种族和民族差异在后续前列腺 MRI 利用中明显存在，且在较高的 PSA 阈值下更为明显。需要进一步研究以更好地了解和减轻医生决策中的偏见和其他潜在的前列腺癌诊断和管理方面的差异来源。
