Princess Margaret Cancer Centre, Toronto, ON, Canada
Elizabeth Hall , Mindy Liang , Emily Tam , Judy Chen , Chenchen Tian , Matthew A Campbell , Kathryn Bucci , Lin Lu , Brandon C Tse , Dennis Zheng , Lauren Wong , Samantha Sarabia , Sabrina Yeung , Gursharan Gill , Andrea Perez-Cosio , M. Catherine Brown , Wei Xu , Geoffrey Liu , Doris Howell
Background: A patient’s functional status is a key outcome variable used to both measure and improve quality of care in cancer survivors. However, comprehensive research tools such as the HAQ-DI and WHODAS instruments pose 35 questions altogether, increasing patient reporting burden. We evaluated whether a BLESS could be developed with high sensitivity of screening questions followed by questions relevant to specific physical function domain, through a branching logic algorithm. Methods: Adult cancer clinic outpatients at Princess Margaret Cancer Centre used tablet technology to complete the HAQ-DI, WHODAS, EQ-5D-3L and PRO-ECOG. BLESS was developed as an algorithm that used PRO-ECOG/EQ-5D-3L to screen for appropriate domains of HAQ-DI/WHODAS to query. Sensitivity/specificity of BLESS screeners to HAQ-DI/WHODAS items were reported. BLESS derived physical function scores were also compared to scores generated by the full version of WHODAS/HAQ-DI. Results: Of 407 patients,median age was 62 (range 20-93) years, 51% female, 75% Caucasian, with a median EQ-5D-3L index score of 0.83 (0.31-1.00), and WHODAS of 6.0 (0-37). Of cancer sites, 14% had breast, 11% GI, 11% GU, 18% head/neck, 10% thoracic, 18% hematologic, and 15% gynecologic cancers. 34% were stage III-IV; 72% were treated for curative intent. Using the sum of mobility, self-care, and usual activities dimensions of the EQ-5D-3L to dichotomize patients as with or without difficulty, we found an area under the receiver operating curves that was > 90% when comparing to recommended WHODAS and HAQ cut-offs for significant physical dysfunction. Using BLESS, 38% of our clinic outpatients need answer only 5 questions, generating derived-WHODAS scores with no median difference when compared to reported scores; sensitivity of screeners for each WHODAS/HAQ-DI item ranged from 92-100%. Overall, median number of questions asked was 9. Conclusions: By focusing on relevance, BLESS maintained high sensitivity with a dramatic decrease in question burden compared to traditional research surveys for physical function. BLESS is suitable for use in immediate, routine screening of chronic care outcomes.
Disclaimer
This material on this page is ©2024 American Society of Clinical Oncology, all rights reserved. Licensing available upon request. For more information, please contact licensing@asco.org
Abstract Disclosures
2022 ASCO Quality Care Symposium
First Author: Sarah A Birken
2022 ASCO Annual Meeting
First Author: Elizabeth Adams
2023 ASCO Annual Meeting
First Author: Christopher K. Daugherty
2019 ASCO Quality Care Symposium
First Author: Michael T. Halpern