Citations

Abstract Background Electronic health record–linked portals may improve health-care quality for patients with cancer. Barriers to portal access and use undermine interventions that rely on portals to reduce cancer care disparities. This study examined portal access and persistence of portal use and associations with patient and structural factors before the implementation of 3 portal-based interventions within the Improving the Management of symPtoms during And following Cancer Treatment (IMPACT) Consortium. Methods Portal use data were extracted from electronic health records for the 12 months preceding intervention implementation. Sociodemographic factors, mode of accessing portals (web vs mobile), and number of clinical encounters before intervention implementation were also extracted. Rurality was derived using rural-urban commuting area codes. Broadband access was estimated using the 2015-2019 American Community Survey. Multiple logistic regression models tested the associations of these factors with portal access (ever accessed or never accessed) and persistence of portal use (accessed the portal ≤20 weeks vs ≥21 weeks in the 35-week study period). Results Of 28 942 eligible patients, 10 061 (35%) never accessed the portal. Male sex, membership in a racial and ethnic minority group, rural dwelling, not working, and limited broadband access were associated with lower odds of portal access. Younger age and more clinical encounters were associated with higher odds of portal access. Of those with portal access, 25% were persistent users. Using multiple modalities for portal access, being middle-aged, and having more clinical encounters were associated with persistent portal use. Conclusion Patient and structural factors affect portal access and use and may exacerbate disparities in electronic health record–based cancer symptom surveillance and management.

Abstract Background Physical activity is associated with decreased breast cancer recurrence and mortality, as well as fewer treatment-related symptoms. Nevertheless, most breast cancer survivors do not meet physical activity guidelines. The purpose of this manuscript is to characterize physical activity trends over time in breast cancer survivors. Methods Mayo Clinic Breast Disease Registry participants received surveys at baseline and at 1 and 4 years after diagnosis; breast cancer recurrence and/or metastatic disease were exclusion criteria. Participants were considered to be meeting guidelines if they self-reported at least 150 minutes of moderate-intensity (eg, fast walking) and/or strenuous (eg, jogging) physical activity per week. Statistical analyses include analysis of covariance methods, paired t tests, conditional logistic regression models, and McNemar tests of homogeneity. Results A total of 171 participants were included in the analysis. The amount of total physical activity decreased over time (P = .07). Mild-intensity physical activity (eg, easy walking) decreased most over time (P = .05). Among participants aged 18-49 years, mild-intensity (P = .05) and moderate-intensity (P = .02) physical activity decreased over time. Strenuous-intensity physical activity levels decreased over time among participants with a normal body mass index (P = .002) and with obesity (P = .01). Conclusions We found a trend-level decrease in total physical activity over time, driven mostly by a decrease in mild-intensity physical activity. Young breast cancer survivors are especially likely to reduce their physical activity over time. Further research on implementing physical activity guidelines in clinical practice is warranted.

Abstract Embedded pragmatic clinical trials (ePCTs) play a vital role in addressing current population health problems, and their use of electronic health record (EHR) systems promises efficiencies that will increase the speed and volume of relevant and generalizable research. However, as the number of ePCTs using EHR-derived data grows, so does the risk that research will become more vulnerable to biases due to differences in data capture and access to care for different subsets of the population, thereby propagating inequities in health and the healthcare system. We identify 3 challenges—incomplete and variable capture of data on social determinants of health, lack of representation of vulnerable populations that do not access or receive treatment, and data loss due to variable use of technology—that exacerbate bias when working with EHR data and offer recommendations and examples of ways to actively mitigate bias.

e13552 Background: Real-world data from Electronic Health Records (EHR) have been widely used for patient identification to build study cohorts for clinical research. Traditionally, diagnosis codes in the EHR, such International Classification of Diseases (ICD), are used to identify the target patients. However, the accuracy of this approach is dependent on the accuracy of ICD coding, with potential errors especially for tumor types that are frequent locations for metastases (which may contribute to mis-coding). In this study, we attempted to develop a Machine Learning (ML) based approach on EHR data to improve the accuracy of identification of patients with lung cancer. Methods: We used survey respondents in the Enhanced, EHR-facilitated Cancer Symptom Control (E2C2, NCT03892967) cluster-randomized trial at Mayo Clinic as our initial pan-cancer cohort. E2C2 includes adults receiving Medical Hematology/Oncology care for a solid or liquid tumor at Mayo Clinic. We collected cancer diagnoses from the individually abstracted Mayo Clinic Cancer Registry to annotate the cancer type for the patients. Lung cancer related ICD-9 (162.X) and ICD-10 (C34.X) codes were used to build a search query on the Mayo’s EHR to find target patients from the E2C2 cohort, and to investigate the ICD-based lung cancer patient identification performance. Diagnosis, radiation oncology treatment (CPT 77261 - 77799), and antineoplastic drug administration data were collected from EHR as variables. Logistic Regression (LR), Support vector machine (SVM), Random Forest (RF), and eXtreme Gradient Boosting (XGB) were selected to build models for lung cancer patient identification. 10-fold cross-validation was implemented to assess the models. Precision, Recall, F1 Score, and area under the curve (AUC) were selected to measure the performance. Results: We collected 13,893 patients with a specific cancer diagnosis, and 1,394 were identified as having lung cancer. The identification performance across different methods is shown. The ICD-based method only had 0.65 precision. It means we collected a lot of false positive cases (other cancer but no lung cancer patient), as we mentioned in the background. SVM gets the best precision results, but its recall and F1 score are not good enough. XGB shows the best F1 Score and AUC, which also means this method achieved the best and most balanced performance. Conclusions: In this study, we found that XGB-based methods achieved the best identification performance for lung cancer. In the future, we will investigate if this is also true for the identification of other cancer types. [Table: see text]

Background. Mortality risk assessment before kidney transplantation (KT) is imperfect. An emerging risk factor for death in nontransplant populations is physiological age as determined by the application of artificial intelligence to the electrocardiogram (ECG). The aim of this study was to examine the relationship between ECG age and KT waitlist mortality. Methods. We applied a previously developed convolutional neural network to the ECGs of KT candidates evaluated 2014 to 2019 to determine ECG age. We used a Cox proportional hazard model to examine whether ECG age was associated with waitlist mortality. Results. Of the 2183 patients evaluated, 59.1% were male, 81.4% were white, and 11.4% died during follow-up. Mean ECG age was 59.0 ± 12.0 y and mean chronological age at ECG was 53.3 ± 13.6 y. After adjusting for chronological age, comorbidities, and other characteristics associated with mortality, each increase in ECG age of >10 y than the average ECG age for patients of a similar chronological age was associated with an increase in mortality risk (hazard ratio 3.59 per 10-y increase; 95% confidence interval, 2.06-5.72; P < 0.0001). Conclusions. ECG age is a risk factor for KT waitlist mortality. Determining ECG age through artificial intelligence may help guide risk-benefit assessment when evaluating candidates for KT.

Abstract We previously described the hypotheses, outcomes, design, and analysis for E2C2, a pragmatic stepped-wedge trial to assess an intervention to improve symptom control in patients with cancer. Subsequent consideration of the design and cohort led to the addition of a second primary hypothesis. This article describes and presents the rationale for this second hypothesis. This addendum also details a revised analytic approach, necessitated by inconsistencies in the original analytic plan. The design, outcomes, and other aspects of the protocol remain unchanged.

419 Background: The patient portal is part of an electronic health record (EHR) that allows patients to communicate with their healthcare team. The portal also provides a platform for patients to receive and complete symptom surveys that can be directly integrated into the EHR, allowing clinical care teams to monitor symptoms and provide cancer symptom management. The Improving Management of symPtoms during And following Cancer Treatment (IMPACT) consortium, supported by the National Cancer Institute’s Cancer MoonshotSM, aims to improve symptom control for cancer patients through assessment and symptom management interventions deployed via the EHR. This initiative presents an opportunity to examine portal enrollment and variation in use, factors critical to successful implementation, especially among groups that have high cancer symptom burden. To this end, we examine: 1) relationships between portal enrollment prior to the launch of IMPACT interventions and neighborhood broadband access, demographic, and social characteristics; and 2) frequency of pre-intervention portal use for any purpose among enrollees. Methods: Data are derived from two of three IMPACT research centers. Enrollment in and frequency of portal use, mode of accessing the portal (web vs. phone), social, demographic, and cultural factors were extracted from the EHR. Rural Urban Commuting Area (RUCA) codes were used to classify population density and degree of rurality. Broadband access was estimated using 2015-2019 American Community Survey estimates matched to zip codes from enrolled IMPACT patients and classified as a patient’s residence being in a community with high (≥85% of households) or low (< 85% of households) access. Bivariate comparisons and adjusted odds ratios were used to describe all associations. Results: Forty-seven percent of patients (22,596/48,034) were enrolled in the portal prior to the intervention. Patients in zip codes with low broadband access and those who were men, > 65 years old, not White, of Hispanic ethnicity, or disabled or not employed had significantly lower odds of being enrolled in the portal. If enrolled, 21% (n = 4825) used the portal at least once a week. Less variation was found in the average frequency of portal use, but patients younger than 40, and those who were Black, disabled, unemployed, or those who used a mobile device to access the portal had the lowest odds of accessing it at least once a week. Conclusions: Significant disparities in portal enrollment exist across demographic groups and among those with limited broadband access. Among those enrolled, most used the portal less than once a week. Fewer differences in frequency of use were observed by sociodemographic factors. Improving portal enrollment and frequency of use may be critical for symptom management interventions deployed via patient portals.

Abstract Purpose Breast cancer patients are at risk of losing function during and after treatment. Unfortunately, little is known about the individual and additive effects of specific treatment, disease-related, and demographic factors that may contribute to functional decline. This manuscript reports the results of a multi-center study to evaluate the effects of these factors on function. Methods In this cross-sectional study, women with a history of breast cancer referred to cancer rehabilitation clinics were administered the PROMIS Cancer Function Brief 3D Profile to evaluate function in the domains of physical function, fatigue, and social participation. Clinical and demographic information, including treatment history and disease status, was recorded by clinicians. Patients were analyzed in two groups: those with active disease on antineoplastic treatment, and those with no evidence of disease (NED). A multivariable model was constructed to detect associations between clinical and demographic factors. Results In patients with NED, the presence of chemotherapy-induced peripheral neuropathy (CIPN) was strongly associated with reduced function in all three domains. In those with active disease, having brain metastases was significantly associated with reduced function in all domains and CIPN with reduced physical function. Radiation was associated with improved function in both cohorts. Conclusions Among women seeking rehabilitative care, CIPN and the presence of brain metastases were most strongly associated with a decline in function. The effects of radiation on function were unexpected and may be partially explained by the treatment’s role in symptom management. Clinicians who treat breast cancer should consider a patient’s functional status when providing supportive care.

Abstract Physical activity (PA) is associated with improvement in breast cancer treatment-related symptoms and survival, yet most breast cancer survivors do not meet national PA guidelines. This study aimed to identify characteristics of participants that were associated with an increased likelihood of meeting PA guidelines. Adults with breast cancer seen at Mayo Clinic (Rochester, MN) were surveyed regarding their PA participation, and those who self-reported at least 150 minutes of moderate and/or strenuous aerobic PA weekly on average were considered to be “meeting guidelines”. Three thousand participants returned PA data. Younger age, completion of the survey 7-12 years after diagnosis, absence of recurrence, no bilateral mastectomy, absence of metastatic disease, and lower BMI at the time of survey completion were associated with PA participation (P &lt; .05 in univariate and multivariate analyses). Findings were similar when a threshold of 90 minutes was applied. These results may inform the development of targeted PA-facilitating interventions.

BackgroundMeasuring function with valid and responsive tools in patients with cancer is essential for driving clinical decision‐making and for the end points of clinical trials. Current patient‐reported outcome measurements of function fall short for many reasons. This study evaluates the responsiveness of the Patient‐Reported Outcomes Measurement Information System (PROMIS) Cancer Function Brief 3D Profile, a novel measure of function across multiple domains.MethodsTwo hundred nine participants across five geographically distinct tertiary care centers completed the assessment and pain rating at two outpatient cancer rehabilitation clinic visits. Patients and providers completed a global rating of change measure at the second visit to indicate whether the patient was improving or worsening in function. Multiple response indices and linear models measured whether the measure was responsive to self‐reported and clinician‐rated changes over time. Correlations between changes in function and changes in anchors (pain rating and performance status) were also calculated.ResultsFunction as measured by the PROMIS Cancer Function Brief 3D Profile changed appropriately as both patients and clinicians rated change. Small to moderate effect sizes supported the tool's responsiveness. Function was moderately correlated with pain and more strongly correlated with performance status, and changes in function corresponded with changes in anchor variables. No floor/ceiling effect was found.ConclusionsThe PROMIS Cancer Function Brief 3D Profile is sensitive to changes over time in patients with cancer. The measure may be useful in clinical practice and as an end point in clinical trials.Lay summary We gave patients a questionnaire by which they told their physicians how well they were functioning, including how fatigued they were. This study tested that questionnaire to see whether the scores would change if patients got better or worse.