Impact of Guideline-Discordant Treatment on Cost and Health Care Utilization in Older Adults with Early-Stage Breast Cancer

Oncologist. 2019 Jan;24(1):31-37. doi: 10.1634/theoncologist.2018-0076. Epub 2018 Aug 17.

Abstract

Background: National Comprehensive Cancer Network (NCCN) guideline-based treatment is a marker of high-quality care. The impact of guideline discordance on cost and health care utilization is unclear.

Materials and methods: This retrospective cohort study of Medicare claims data from 2012 to 2015 included women age ≥65 with stage I-III breast cancer receiving care within the University of Alabama at Birmingham Cancer Community Network. Concordance with NCCN guidelines was assessed for treatment regimens. Costs to Medicare and health care utilization were identified from start of cancer treatment until death or available follow-up. Adjusted monthly cost and utilization rates were estimated using linear mixed effect and generalized linear models.

Results: Of 1,177 patients, 16% received guideline-discordant treatment, which was associated with nonwhite race, estrogen receptor/progesterone receptor negative, human epidermal growth receptor 2 (HER2) positive, and later-stage cancer. Discordant therapy was primarily related to reduced-intensity treatments (single-agent chemotherapy, HER2-targeted therapy without chemotherapy, bevacizumab without chemotherapy, platinum combinations without anthracyclines). In adjusted models, average monthly costs for guideline-discordant patients were $936 higher compared with concordant (95% confidence limits $611, $1,260). For guideline-discordant patients, adjusted rates of emergency department visits and hospitalizations per thousand observations were 25% higher (49.9 vs. 39.9) and 19% higher (24.0 vs. 20.1) per month than concordant patients, respectively.

Conclusion: One in six patients with early-stage breast cancer received guideline-discordant care, predominantly related to undertreatment, which was associated with higher costs and rates of health care utilization. Additional randomized trials are needed to test lower-toxicity regimens and guide clinicians in treatment for older breast cancer patients.

Implications for practice: Previous studies lack details about types of deviations from chemotherapy guidelines that occur in older early-stage breast cancer patients. Understanding the patterns of guideline discordance and its impact on patient outcomes will be particularly important for these patients. This study found 16% received guideline-discordant care, predominantly related to reduced intensity treatment and associated with higher costs and rates of health care utilization. Increasing older adult participation in clinical trials should be a priority in order to fill the knowledge gap about how to treat older, less fit patients with breast cancer.

摘要

背景。美国国家综合癌症网络 (NCCN) 基于指南的治疗是优质医疗的标志。不遵守指南对费用和卫生服务利用的影响尚不明确。

材料和方法。该 2012 年至 2015 年期间联邦医疗保险 (Medicare) 理赔数据的回顾性队列研究包括在伯明翰癌症社区网络中的阿拉巴马大学接受治疗的 ≥65 岁的 I–III 期乳腺癌女性。我们针对治疗方案评估了对 NCCN 指南的遵从性。从开始治疗癌症至患者死亡或可行随访,我们确定了联邦医疗保险费用和卫生服务利用情况。利用线性混合效应和广义线性模型,我们对校正后的每月费用和利用率进行了估算。

结果。在 1 177 名患者中,16% 的患者接受了不遵守指南的治疗,这与非白色人种、雌激素受体/孕激素受体阴性、人表皮生长因子受体 2 (HER2) 阳性以及晚期癌症相关。不遵守指南的治疗主要与降低强度的治疗(单药化疗、无化疗的 HER2 靶向治疗、无化疗的贝伐珠单抗治疗、无蒽环类药物的铂类联合治疗)相关。在校正后的模型中,不遵守指南的患者的平均每月费用比遵守指南的患者高 $936(95% 置信界限为 $611,$1 260)。对于不遵守指南的患者,每月每一千名观察对象中的校正后的急诊就诊率和住院率分别比遵守指南的患者高 25%(49.9 与 39.9)和 19%(24.0 与 20.1)。

结论。六分之一的早期乳腺癌患者接受了不遵守指南的治疗,这主要与治疗不足有关,而治疗不足与较高的费用和卫生服务利用率有关。为了测试较低毒性的方案以及指导临床医生治疗乳腺癌老年患者,需要进行更多的随机试验。

对临床实践的提示

先前的研究缺乏有关在早期乳腺癌老年患者中出现的化疗指南偏差类型的详细信息。了解不遵守指南的模型及其对患者结果的影响对于这些患者而言尤为重要。本研究发现,16% 的患者接受了不遵守指南的治疗,这主要与降低强度的治疗以及较高的费用和卫生服务利用率相关。为了弥补关于如何治疗身体虚弱的乳腺癌老年患者的知识缺口,应优先考虑提高老年人在临床试验中的参加率。

Keywords: Breast cancer; Guidelines; Health care utilization; Medicare costs; Treatment.

Publication types

  • Research Support, Non-U.S. Gov't
  • Research Support, U.S. Gov't, P.H.S.

MeSH terms

  • Aged
  • Breast Neoplasms / economics*
  • Cohort Studies
  • Female
  • Humans
  • Neoplasm Staging
  • Patient Acceptance of Health Care / statistics & numerical data*
  • Retrospective Studies