Scholarlink

Objectives <p>This study aimed to evaluate the validity of both generic and cancer-specific preference-based measures (PBMs) in patients undergoing cancer treatment, providing solid evidence for the optimal selection and implementation of these PBMs in oncology clinical practice settings.</p> Methods <p>A cross-sectional survey was conducted among patients diagnosed with one of the six most common cancers in China, recruited from hospitals in Harbin, China, between August 2022 and December 2023. Participants completed two generic PBMs (EQ-5D-5L and SF-6Dv2) and two cancer-specific PBMs (Quality of Life Utility-Core 10 Dimension [QLU-C10D] and Functional Assessment of Cancer Therapy Eight Dimension [FACT-8D]). Convergent validity was assessed using pairwise polychoric correlations between the four measures. Known-groups validity was evaluated by examining the measures’ ability to discriminate between patients based on cancer stage, with effect sizes (ES) calculated. Agreement between the utility scores of the four measures was assessed using intraclass correlation coefficients (ICC).</p> Results <p>A total of 827 cancer patients completed the survey, including 270 with lung cancer, 96 with stomach cancer, 88 with liver cancer, 71 with oesophageal cancer, 142 with colorectal cancer, and 160 with breast cancer. The proportions in full health were 25.8%, 13.7%, 7.3%, and 2.4% for EQ-5D-5L, SF-6Dv2, QLU-C10D, and FACT-8D, respectively. Convergent validity analysis revealed that the symptom dimensions of cancer-specific PBMs had lower correlations with generic PBMs compared to their functional dimensions. Known-groups validity showed QLU-C10D was most effective in distinguishing cancer stages (highest ES across subgroups), whereas EQ-5D-5L performed poorest (ES = 0.33); SF-6Dv2 was most sensitive to ECOG differences (ES = 1.37), with all instruments robust for ECOG stratification. Agreement results indicated that the agreement between generic PBMs and cancer-specific PBMs was the lowest, compared to the agreement within each category.</p> Conclusions <p>Cancer-specific PBMs captured stage- and symptom-related burden more comprehensively, whereas generic PBMs were more informative for functional status and cross-cancer comparability. These findings underline the importance of balancing cancer-specific and generic PBMs through purpose- and site-specific combinations to optimize utility assessment and support decision-making in oncology.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Cancer-specific or generic preference-based measures in oncology? A validity comparison across six commonly diagnosed cancers

  • 期刊论文

  • 发表日期：

    2026年3月8日

查看全文
  • Yiyin Cao,
  • Sakhno Alina,
  • Zitong Li,
  • Haofei Li,
  • Jiaxuan Shi,
  • Runhong Li,
  • Weidong Huang,
  • Hongjuan Yu

摘要

Objectives

This study aimed to evaluate the validity of both generic and cancer-specific preference-based measures (PBMs) in patients undergoing cancer treatment, providing solid evidence for the optimal selection and implementation of these PBMs in oncology clinical practice settings.

Methods

A cross-sectional survey was conducted among patients diagnosed with one of the six most common cancers in China, recruited from hospitals in Harbin, China, between August 2022 and December 2023. Participants completed two generic PBMs (EQ-5D-5L and SF-6Dv2) and two cancer-specific PBMs (Quality of Life Utility-Core 10 Dimension [QLU-C10D] and Functional Assessment of Cancer Therapy Eight Dimension [FACT-8D]). Convergent validity was assessed using pairwise polychoric correlations between the four measures. Known-groups validity was evaluated by examining the measures’ ability to discriminate between patients based on cancer stage, with effect sizes (ES) calculated. Agreement between the utility scores of the four measures was assessed using intraclass correlation coefficients (ICC).

Results

A total of 827 cancer patients completed the survey, including 270 with lung cancer, 96 with stomach cancer, 88 with liver cancer, 71 with oesophageal cancer, 142 with colorectal cancer, and 160 with breast cancer. The proportions in full health were 25.8%, 13.7%, 7.3%, and 2.4% for EQ-5D-5L, SF-6Dv2, QLU-C10D, and FACT-8D, respectively. Convergent validity analysis revealed that the symptom dimensions of cancer-specific PBMs had lower correlations with generic PBMs compared to their functional dimensions. Known-groups validity showed QLU-C10D was most effective in distinguishing cancer stages (highest ES across subgroups), whereas EQ-5D-5L performed poorest (ES = 0.33); SF-6Dv2 was most sensitive to ECOG differences (ES = 1.37), with all instruments robust for ECOG stratification. Agreement results indicated that the agreement between generic PBMs and cancer-specific PBMs was the lowest, compared to the agreement within each category.

Conclusions

Cancer-specific PBMs captured stage- and symptom-related burden more comprehensively, whereas generic PBMs were more informative for functional status and cross-cancer comparability. These findings underline the importance of balancing cancer-specific and generic PBMs through purpose- and site-specific combinations to optimize utility assessment and support decision-making in oncology.