Comorbidities among patients with breast cancer during COVID-19: Agreement between patient-reported data and electronic medical records.

Authors

null

Martine C. Maculaitis

Cerner Enviza, Malvern, PA

Martine C. Maculaitis, Xianchen Liu, Jeffrey A. Thompson, Alexandra Berk, Angelina Massa, Marisa C. Weiss, Benjamin Li, Samantha K. Kurosky, Lynn McRoy

Organizations

Cerner Enviza, Malvern, PA, Pfizer Inc., New York, NY, Invitae, San Francisco, CA, Breastcancer.org, Ardmore, PA

Research Funding

Pharmaceutical/Biotech Company
Pfizer Inc.

Background: Comorbidities are a major cause of complications in cancer patients and can increase the risk of severe illness from coronavirus disease 2019 (COV). Yet, in real-world studies of cancer patients, comorbidities are often not well captured in electronic medical records (EMR); self-reported comorbidities may be limited by recall error. Combining and comparing self-reported and EMR data may help identify key data gaps in comorbidity diagnosis. We aimed to estimate self-reported and EMR-documented comorbidities and examine agreement between these data sources in US patients with early-stage (eBC) or metastatic (mBC) breast cancer. Methods: From March 30 to July 6, 2021, patients (aged ≥18 years) who self-reported a BC diagnosis (no current Stage 0 or ductal carcinoma in situ) and provided consent were recruited via Ciitizen, a patient-mediated health records and real-world evidence platform, and patient advocacy groups to complete a cross-sectional online survey. EMR data from Invitae?s Ciitizen platform, covering November 1, 2019-September 28, 2021, were collected. The datasets were then linked; in preliminary assessment, depression (DEP), anxiety (ANX), and COV were the only comorbidities sufficiently populated to enable agreement analysis. DEP, ANX, and COV prevalence rates in EMR and survey data were computed. Agreement between data sources was estimated using Cohen?s kappa. Results: Overall, 542 patients in the linked sample were included in the analyses. A majority was female (99%), aged ≥50 years (52%), and diagnosed with mBC (53%). Patients were similarly distributed by US geographic region. DEP, ANX, and COV prevalence rates were 40%, 50%, and 6% in EMR and 20%, 16% and 10% in survey, respectively (Table). The kappa values for agreement were.5 (DEP),.3 (ANX), and.8 (COV). Conclusions: Mental health conditions and COV were prevalent in BC patients during the pandemic. It is important for oncologists to consider that DEP and ANX are often underreported among patients with BC. Agreement between data sources was low for mental health conditions. Self-report and EMR data may thus provide complementary information on comorbidities.

Estimated prevalence and agreement.



DEP


ANX


COV


Total
eBC
mBC
Total
eBC
mBC
Total
eBC
mBC
EMR Only, n (%)
119 (22)
42 (16)
77 (27)
192 (35)
76 (30)
116 (40)
0 (0)
0 (0)
0 (0)
Self-report Only, n (%)
10 (2)
2 (1)
8 (3)
8 (1)
5 (2)
3 (1)
20 (4)
9 (4)
11 (4)
Both, n (%)
96 (18)
49 (19)
47 (16)
80 (15)
43 (17)
37 (13)
33 (6)
10 (4)
23 (8)
Either, n (%)
225 (42)
93 (36)
132 (46)
280 (52)
124 (49)
156 (54)
53 (10)
19 (7)
34 (12)
EMR Prevalence, n (%)
215 (40)
91 (36)
124 (43)
272 (50)
119 (47)
153 (53)
33 (6)
10 (4)
23 (8)
Self-report Prevalence, n (%)
106 (20)
51 (20)
55 (19)
88 (16)
48 (19)
40 (14)
53 (10)
19 (8)
34 (12)
Kappa*
.46
.58
.35
.26
.34
.21
.75
.67
.79

Note. N = 542 (Total), N = 255 (eBC), and N = 287 (mBC). *Values >.60 indicate adequate agreement.

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 Details

Meeting

2022 ASCO Quality Care Symposium

Session Type

Poster Session

Session Title

Poster Session B

Track

Palliative and Supportive Care,Technology and Innovation in Quality of Care,Quality, Safety, and Implementation Science

Sub Track

Real-World Evidence

Citation

J Clin Oncol 40, 2022 (suppl 28; abstr 409)

DOI

10.1200/JCO.2022.40.28_suppl.409

Abstract #

409

Poster Bd #

F12

Abstract Disclosures