Physicians' Hierarchy of Tumor Biomarkers for Optimizing Chemotherapy in Breast Cancer Care.

BACKGROUND: Tumor biomarkers are regularly used to guide breast cancer treatment and clinical trial enrollment. However, there remains a lack of knowledge regarding physicians' perspectives towards biomarkers and their role in treatment optimization, where treatment intensity is reduced to minimize toxicity. METHODS: Thirty-nine academic and community oncologists participated in semi-structured qualitative interviews, providing perspectives on optimization approaches to chemotherapy treatment. Interviews were audio-recorded, transcribed, and analyzed by 2 independent coders utilizing a constant comparative method in NVivo. Major themes and exemplary quotes were extracted. A framework outlining physicians' conception of biomarkers, and their comfortability with their use in treatment optimization, was developed. RESULTS: In the hierarchal model of biomarkers, level 1 is comprised of standard-of-care (SoC) biomarkers, defined by a strong level of evidence, alignment with national guidelines, and widespread utilization. Level 2 includes SoC biomarkers used in alternative contexts, in which physicians expressed confidence, yet less certainty, due to a lack of data in certain subgroups. Level 3, or experimental, biomarkers created the most diverse concerns related to quality and quantity of evidence, with several additional modulators. CONCLUSION: This study demonstrates that physicians conceptualize the use of biomarkers for treatment optimization in successive levels. This hierarchy can be used to guide trialists in the development of novel biomarkers and design of future trials.

Predicting Incident Adenocarcinoma of the Esophagus or Gastric Cardia Using Machine Learning of Electronic Health Records.

BACKGROUND & AIMS: Tools that can automatically predict incident esophageal adenocarcinoma (EAC) and gastric cardia adenocarcinoma (GCA) using electronic health records to guide screening decisions are needed. METHODS: The Veterans Health Administration (VHA) Corporate Data Warehouse was accessed to identify Veterans with 1 or more encounters between 2005 and 2018. Patients diagnosed with EAC (n = 8430) or GCA (n = 2965) were identified in the VHA Central Cancer Registry and compared with 10,256,887 controls. Predictors included demographic characteristics, prescriptions, laboratory results, and diagnoses between 1 and 5 years before the index date. The Kettles Esophageal and Cardia Adenocarcinoma predictioN (K-ECAN) tool was developed and internally validated using simple random sampling imputation and extreme gradient boosting, a machine learning method. Training was performed in 50% of the data, preliminary validation in 25% of the data, and final testing in 25% of the data. RESULTS: K-ECAN was well-calibrated and had better discrimination (area under the receiver operating characteristic curve [AuROC], 0.77) than previously validated models, such as the Nord-Trøndelag Health Study (AuROC, 0.68) and Kunzmann model (AuROC, 0.64), or published guidelines. Using only data from between 3 and 5 years before index diminished its accuracy slightly (AuROC, 0.75). Undersampling men to simulate a non-VHA population, AUCs of the Nord-Trøndelag Health Study and Kunzmann model improved, but K-ECAN was still the most accurate (AuROC, 0.85). Although gastroesophageal reflux disease was strongly associated with EAC, it contributed only a small proportion of gain in information for prediction. CONCLUSIONS: K-ECAN is a novel, internally validated tool predicting incident EAC and GCA using electronic health records data. Further work is needed to validate K-ECAN outside VHA and to assess how best to implement it within electronic health records.

Patient and physician decision-making dynamics in overactive bladder care: A mixed methods study.

AIMS: Overactive bladder management includes multiple therapeutic options with comparable efficacy but a range of administration modalities and side effects, creating an ideal setting for shared decision-making. This study investigates patient and physician health beliefs surrounding decision-making and expectations for overactive bladder with the aim of better understanding and ultimately improving decision-making in overactive bladder care. METHODS: Patient and physician participants completed a questionnaire followed by a semi-structured interview to assess health beliefs surrounding decision making and expectations for overactive bladder treatment. The semi-structured interview guide, developed in an iterative fashion by the authors, probed qualities of overactive bladder therapies patients and physicians valued, their process of treatment selection, and their experiences with therapies. RESULTS: Patients (n = 20) frequently cited treatment invasiveness, efficacy, and safety as the most important qualities that influenced their decision when selecting overactive bladder therapy. Physicians (n = 12) frequently cited safety/contraindications, convenience, cost/insurance, and patient preference as the most important qualities. In our integration analysis, we identified four key themes associated with decision making in overactive bladder care: frustration with inaccessibility of overactive bladder treatments, discordant perception of patient education, diverging acceptability of expected outcomes, and lack of insight into other parties' decisional priorities and control preferences. CONCLUSIONS: While both patients and physicians desire to engage in a shared decision-making process when selecting therapies for overactive bladder, this process is challenged by significant divergence between patient and physician viewpoint across key domains.

Primary care provider attitudes about and tendency to use non-recommended surveillance tests after curative breast cancer treatment.

PURPOSE: Little is known about the factors contributing to the receipt of non-recommended surveillance testing among early-stage breast cancer survivors. We assessed primary care providers (PCP) attitudes about and tendency to order non-recommended surveillance testing for asymptomatic early-stage breast cancer survivors post-adjuvant chemotherapy. METHODS: A stratified random sample of PCPs identified by early-stage breast cancer survivors were surveyed (N = 518, 61% response rate). PCPs were asked how likely they would be to order bone scans, imaging and/or tumor marker testing using a clinical vignette of an early-stage asymptomatic patient where these tests are non-recommended. A composite tendency to order score was created and categorized by tertiles (low, moderate, high). PCP-reported factors associated with high and moderate tendency to order non-recommended testing (vs. low) were estimated using multivariable, multinomial logistic regression. RESULTS: In this sample, 26% reported a high tendency to order non-recommended surveillance tests during survivorship for early-stage breast cancer survivors. PCPs who identified as family practice physicians and PCPs reporting more confidence in ordering surveillance testing were more likely to report a high tendency to order non-recommended testing (vs. low) ((aOR family practice 2.09, CI 1.2, 3.8; aOR more confidence 1.9, CI 1.1, 3.3). CONCLUSIONS: In this population-based sample of PCPs caring for breast cancer survivors, over a quarter of PCPs reported they would order non-recommended surveillance testing for asymptomatic early-stage breast cancer survivors. Efforts to better support PCPs and disseminate information about appropriate surveillance for cancer survivors are warranted.

Patient Cognitive Status and Physician Recommendations for Cardiovascular Disease Treatment: Results of Two Nationwide, Randomized Survey Studies.

BACKGROUND: Clinical guidelines recommend that older patients (65+) with mild cognitive impairment (MCI) and early-stage dementia receive similar guideline-concordant care after cardiovascular disease (CVD) events as those with normal cognition (NC). However, older patients with MCI and dementia receive less care for CVD and other conditions than those with NC. Whether physician recommendations for guideline-concordant treatments after two common CVD events, acute myocardial infarction (AMI) and acute ischemic stroke (stroke), differ between older patients with NC, MCI, and early-stage dementia is unknown. OBJECTIVE: To test the influence of patient cognitive status (NC, MCI, early-stage dementia) on physicians' recommendations for guideline-concordant treatments for AMI and stroke. DESIGN: We conducted two parallel, randomized survey studies for AMI and stroke in the US using clinical vignettes where the hypothetical patient's cognitive status was randomized between physicians. PARTICIPANTS: The study included cardiologists, neurologists, and generalists who care for most patients hospitalized for AMI and stroke. MAIN MEASURES: The primary outcome was a composite quality score representing the number of five guideline-concordant treatments physicians recommended for a hypothetical patient after AMI or stroke. KEY RESULTS: 1,031 physicians completed the study (58.5% response rate). Of 1,031 respondents, 980 physicians had complete information. After adjusting for physician factors, physicians recommended similar treatments after AMI and stroke in hypothetical patients with pre-existing MCI (adjusted ratio of expected composite quality score, 0.98 [95% CI, 0.94, 1.02]; P = 0.36) as hypothetical patients with NC. Physicians recommended fewer treatments to hypothetical patients with pre-existing early-stage dementia than to hypothetical patients with NC (adjusted ratio of expected composite quality score, 0.90 [0.86, 0.94]; P < 0.001). CONCLUSION: In these randomized survey studies, physicians recommended fewer guideline-concordant AMI and stroke treatments to hypothetical patients with early-stage dementia than those with NC. We did not find evidence that physicians recommend fewer treatments to hypothetical patients with MCI than those with NC.

Trends in hospice referral timing and location among individuals dying of ovarian cancer: persistence of missed opportunities.

OBJECTIVE: To evaluate trends, racial disparities, and opportunities to improve the timing and location of hospice referral for women dying of ovarian cancer. METHODS: This retrospective claims analysis included 4258 Medicare beneficiaries over age 66 diagnosed with ovarian cancer who survived at least 6 months after diagnosis, died between 2007 and 2016, and enrolled in a hospice. We examined trends in timing and clinical location (outpatient, inpatient hospital, nursing/long-term care, other) of hospice referrals and associations with patient race and ethnicity using multivariable multinomial logistic regression. RESULTS: In this sample, 56% of hospice enrollees were referred to a hospice within a month of death, and referral timing did not vary by patient race. Referrals were most commonly inpatient hospital (1731 (41%) inpatient, 703 (17%) outpatient, 299 (7%) nursing/long-term care, 1525 (36%) other), with a median of 6 inpatient days prior to hospice enrollment. Only 17% of hospice referrals were made in an outpatient clinic, but participants had a median of 1.7 outpatient visits per month in the 6 months prior to hospice referral. Referral location varied by patient race, with non-Hispanic black people experiencing the most inpatient referrals (60%). Hospice referral timing and location trends did not change between 2007 and 2016. Compared with individuals referred to a hospice in an outpatient setting, individuals referred from an inpatient hospital setting had more than six times the odds of a referral in the last 3 days of life (OR=6.5, 95% CI 4.4 to 9.8) versus a referral more than 90 days before death. CONCLUSION: Timeliness of hospice referral is not improving over time despite opportunities for earlier referral across multiple clinical settings. Future work delineating how to capitalize on these opportunities is essential for improving the timeliness of hospice care.

New directions in cancer and aging: State of the science and recommendations to improve the quality of evidence on the intersection of aging with cancer control.

BACKGROUND: The global population of older cancer survivors is growing. However, the intersections of aging-related health risks across the cancer control continuum are poorly understood, limiting the integration of aging into cancer control research and practice. The objective of this study was to review the state of science and provide future directions to improve the quality of evidence in 6 priority research areas in cancer and aging. METHODS: The authors identified priority research areas in cancer and aging through an evidence-based Research Jam process involving 32 investigators and trainees from multiple disciplines and research centers in aging and cancer; then, they conducted a narrative review of the state of the science and future directions to improve the quality of evidence in these research areas. Priority research areas were defined as those in which gaps in scientific evidence or clinical practice limit the health and well-being of older adults with cancer. RESULTS: Six priority research areas were identified: cognitive and physical functional outcomes of older cancer survivors, sampling issues in studies of older cancer survivors, risk and resilience across the lifespan, caregiver support and well-being, quality of care for older patients with cancer, and health disparities. Evidence in these areas could be improved through the incorporation of bias reduction techniques into longitudinal studies of older cancer survivors, novel data linkage, and improved representation of older adults in cancer research. CONCLUSIONS: The priority research areas and methodologies identified here may be used to guide interdisciplinary research and improve the quality of evidence on cancer and aging.

Physician Perspectives on Reducing Curative Cancer Treatment Intensity for Populations Underrepresented in Clinical Trials.

BACKGROUND: Historically, clinical trials involved adding novel agents to standard of care to improve survival. There has been a shift to an individualized approach with testing less intense treatment, particularly in breast cancer where risk of recurrence is low. Little is known about physician perspectives on delivering less intense treatment for patients who are not well represented in clinical trials. METHODS: Open-ended, individual qualitative interviews with medical oncologists explored their perspectives on trials that test less intense treatment for patients with cancer, with a focus on breast cancer. Interviews were audio-recorded and transcribed. Four independent coders utilized a content analysis approach to analyze transcripts using NVivo. Major themes and exemplary quotes were extracted. RESULTS: Of the 39 participating physicians, 61.5% felt comfortable extrapolating, 30.8% were hesitant, and 7.7% would not feel comfortable extrapolating trial outcomes to underrepresented populations. Facilitators of comfort included the sentiment that "biology is biology" (such that the cancer characteristics were what mattered), the strength of the evidence, inclusion of subset analysis on underrepresented populations, and prior experience making decisions with limited data. Barriers to extrapolation included potential harm over the patient's lifetime, concerns about groups that had minimal participants, application to younger patients, and extending findings to diverse populations. Universally, broader inclusion in trials testing lowering chemotherapy was desired. CONCLUSIONS: The majority (92%) of physicians reported that they would de-implement treatment for patients poorly represented in clinical trials testing less treatment, while expressing concerns about applicability to specific subpopulations. Further work is needed to increase clinical trial representation of diverse populations to safely and effectively optimize treatment for patients with cancer. TRIAL REGISTRATION: NCT03248258.

Exploring Variation in the Receipt of Recommended Active Surveillance for Men with Favorable-Risk Prostate Cancer.

PURPOSE: Men on active surveillance for favorable-risk prostate cancer do not receive all the recommended testing. Reasons for variation in receipt are unknown. MATERIALS AND METHODS: We combined prospective registry data from the Michigan Urological Surgery Improvement Collaborative, a collaborative of 46 academic and community urology practices across Michigan, with insurance claims from 2014 to 2018 for men on active surveillance for favorable-risk prostate cancer. We defined receipt of recommended surveillance according to the collaborative's low-intensity criteria as: annual prostate specific antigen testing and either magnetic resonance imaging or prostate biopsy every 3 years. We assessed receipt of recommended surveillance among men with ≥36 months of followup (246). We conducted multilevel analyses to examine the influence of the urologist, urologist and primary care provider visits, and patient demographic and clinical factors on variation in receipt. RESULTS: During 3 years of active surveillance, just over half of men (56.5%) received all recommended surveillance testing (69.9% annual prostate specific antigen testing, 72.8% magnetic resonance imaging/biopsy). We found 19% of the variation in receipt was attributed to individual urologists. While increasing provider visits were not significantly associated with receipt, older men were less likely to receive magnetic resonance imaging/biopsy (≥75 vs <55 years, adjusted odds ratio 0.07; 95% confidence interval 0.01-0.81). CONCLUSIONS: Nearly half of men on active surveillance for favorable-risk prostate cancer did not receive all recommended surveillance. While urologists substantially influenced receipt of recommended testing, exploring how to leverage patients and their visits with their primary care providers to positively influence receipt appears warranted.

The Association Between Mild Cognitive Impairment Diagnosis and Patient Treatment Preferences: a Survey of Older Adults.

BACKGROUND: Older patients (65+) with mild cognitive impairment (MCI) receive less guideline-concordant care for cardiovascular disease (CVD) and other conditions than patients with normal cognition (NC). One potential explanation is that patients with MCI want less treatment than patients with NC; however, the treatment preferences of patients with MCI have not been studied. OBJECTIVE: To determine whether patients with MCI have different treatment preferences than patients with NC. DESIGN: Cross-sectional survey conducted at two academic medical centers from February to December 2019 PARTICIPANTS: Dyads of older outpatients with MCI and NC and patient-designated surrogates. MAIN MEASURES: The modified Life-Support Preferences-Predictions Questionnaire score measured patients' preferences for life-sustaining treatment decisions in six health scenarios including stroke and acute myocardial infarction (range, 0-24 treatments rejected with greater scores indicating lower desire for treatment). KEY RESULTS: The survey response rate was 73.4%. Of 136 recruited dyads, 127 (93.4%) completed the survey (66 MCI and 61 NC). The median number of life-sustaining treatments rejected across health scenarios did not differ significantly between patients with MCI and patients with NC (4.5 vs 6.0; P=0.55). Most patients with MCI (80%) and NC (80%) desired life-sustaining treatments in their current health (P=0.99). After adjusting for patient and surrogate factors, the difference in mean counts of rejected treatments between patients with MCI and patients with NC was not statistically significant (adjusted ratio, 1.08, 95% CI, 0.80-1.44; P=0.63). CONCLUSION: We did not find evidence that patients with MCI want less treatment than patients with NC. These findings suggest that other provider and system factors might contribute to patients with MCI getting less guideline-concordant care.

What patients with eosinophilic esophagitis may not share with their providers: a qualitative assessment of online health communities.

BACKGROUND: Little is known about how patients with eosinophilic esophagitis (EoE) experience their symptoms, receive care, and cope with their disease. Patients commonly seek peer support from online communities, which provide insights on unmet needs and barriers to care. We performed a qualitative analysis of electronic health forums to characterize patient-to-patient conversations about EoE symptoms and the experience of disease. METHODS: We identified three publicly accessible electronic health forums hosting EoE communities. Conversation threads posted between July 2018 and June 2020 were coded using emergent and a priori codes based on the THRIVE conceptual framework of coping with chronic illness. RESULTS: Of 659 threads (4,933 posts) collected over two years, a random sample of 240 threads (30 per 3-month quarter) were selected for analysis. Thematic saturation was reached after 172 threads. Patient experience of EoE was driven by their perspectives in four key domains: (i) perception of EoE as episodic rather than chronic, (ii) treatment choices, (iii) personal definitions of success in the disease, and (iv) views of providers. CONCLUSION: Online health communities are a valuable and unfiltered source of patient perspectives that can be used to understand patient needs and goals. EoE patients interpret their disease as sporadic events and lack reliable sources of knowledge, which may influence how patients prioritize treatment. If providers are to succeed in providing high-quality EoE care, they need to equip themselves with evidence-based knowledge, engage in shared decision making, and look outside of clinical settings to recognize barriers to disease management.

Trends and racial disparities in aggressive end-of-life care for a national sample of women with ovarian cancer.

BACKGROUND: The clinical landscape has moved toward less aggressive end-of-life care for women with ovarian cancer. However, whether there has been a decline in the use of aggressive end-of-life services is unknown. The authors evaluated current national trends and racial disparities in end-of-life care among women with ovarian cancer using the Surveillance, Epidemiology, and End Results-Medicare-linked data set. METHODS: In total, 7756 Medicare beneficiaries aged >66 years with ovarian cancer who died between 2007 and 2016 were identified. The authors examined trends and racial disparities in late hospice or no hospice use, >1 emergency department (ED) visit, intensive care unit admission, >1 hospitalization, terminal hospitalization, chemotherapy, and invasive and/or life-extending procedures using multivariable logistic regression. RESULTS: The median hospice length of stay did not change over time; however, women were increasingly admitted to the intensive care unit and had multiple ED visits in the last month of life (P < .001). Not enrolling in hospice at the end of life and terminal hospitalizations decreased over time (P < .001). Non-White women were more likely to receive aggressive end-of-life care, particularly for hospital-related utilization and life-extending procedures, whereas non-Hispanic Black women were more likely to have >1 ED visit (odds ratio, 2.04; 95% CI, 1.57-2.64) or life-extending procedures (odds ratio, 1.89; 95% CI, 1.45-2.48) compared with non-Hispanic White women. CONCLUSIONS: Despite clinical guidelines and increasing emphasis on reducing aggressive end-of-life care, the use of aggressive end-of-life care for women with ovarian cancer persists, and care is most aggressive for non-White women.

Improving the delivery of team-based survivorship care after primary breast cancer treatment through a multi-level intervention: a pilot randomized controlled trial.

PURPOSE: We developed and tested a multi-level intervention, ConnectedCancerCare (CCC), which includes a tailored website and appointment reminder system for women with early-stage breast cancer and a provider summary letter sent to their medical oncologist and primary care provider to improve the delivery of team-based survivorship care. METHODS: We conducted a pilot randomized controlled trial to establish the feasibility and acceptability of CCC. Women diagnosed with stages 0-II breast cancer within one year of completing primary treatment were randomized to CCC (intervention) or a static online survivorship care plan (control). Participants completed baseline and 3-month follow-up surveys online. Post-trial interviews with 5 PCPs, 6 oncology providers, and 8 intervention patients were conducted. RESULTS: Of the 160 eligible women invited to participate, 66 completed the baseline survey and were randomized (41%) and 54 completed a follow-up survey (83%). Participants in the intervention arm found the CCC content to be acceptable, with 82% reporting it was easy to use and 86% reporting they would recommend it to other patients. Women randomized to CCC (vs. control) more often reported scheduling a PCP follow-up visit (64% vs. 42%), communicating with their PCP about provider roles (67% vs. 18%), and higher mean team-based cancer care knowledge scores (3.7 vs. 3.4). CONCLUSION: Deploying CCC in medical oncology practices was feasible, and the intervention content was acceptable. CCC shows promise for improving patient knowledge and patient-provider communication about provider roles in team-based cancer care and encouraging patients to engage with their PCP early in the survivorship period.

Understanding quality and equity: patient experiences with care in older adults diagnosed with hematologic malignancies.

BACKGROUND: Oncology settings increasingly use patient experience data to evaluate clinical performance. Given that older patients with hematologic malignancies are a high-risk population, this study examined factors associated with patient-reported health care experiences during the first year of their cancer diagnosis. METHODS: Cross-sectional study using the 2000-2015 SEER-CAHPS® data to examine patient experiences of Medicare enrollees with a primary diagnosis of leukemia or lymphoma. The primary outcomes were three CAHPS assessments: overall care, personal doctor, and health plan overall. We estimated case-mix adjusted and fully adjusted associations between factors (i.e., clinical and sociodemographic) and the CAHPS outcomes using bivariate statistical tests and multiple linear regression. RESULTS: The final sample included 1,151 patients, with 431 diagnosed with leukemia and 720 diagnosed with lymphoma (median time from diagnosis to survey 6 months). Patients who completed the survey further apart from the diagnosis date reported significantly higher adjusted ratings of care overall (ß .39, p = .008) than those closer to diagnosis. American Indian/Alaska Native, Asian, and Pacific Islander patients had lower adjusted ratings of care overall (ß - .73, p = .003) than Non-Hispanic white patients. Multimorbidity was significantly associated with higher adjusted personal doctor ratings (ß .26, p = .003). CONCLUSIONS: Unfavorable patient experiences among older adults diagnosed with hematologic malignancies warrant targeted efforts to measure and improve care quality. Future measurement of experiences of cancer care soon after diagnosis, coupled with careful sampling of high-priority populations, will inform oncology leaders and clinicians on strategies to improve care for high-risk, high-cost populations.

Primary Care Providers' Perceptions About Participating in Low-Risk Prostate Cancer Treatment Decisions.

BACKGROUND: Primary care provider's (PCP) role in cancer care is expanding and may include supporting patients in their treatment decisions. However, the degree to which PCPs engage in this role for low-risk prostate cancer is unknown. OBJECTIVE: Characterize PCP perceptions regarding their role in low-risk prostate cancer treatment decision-making. DESIGN: Cross-sectional, national survey. MAIN MEASURES: For men with low-risk prostate cancer, PCP reports of (1) confidence in treatment decision-making (high vs. low); (2) intended participation in key aspects of active surveillance treatment decision-making (more vs. less). KEY RESULTS: A total of 347 from 741 eligible PCPs responded (adjusted response rate 56%). Half of respondent PCPs (50.3%) reported high confidence about engaging in low-risk prostate cancer treatment decision-making. The odds of PCPs reporting high confidence were greater among those in solo practice (vs working with > 1 PCP) (OR 2.18; 95% CI 1.14-4.17) and with higher volume of prostate cancer patients (> 15 vs. 6-10 in past year) (OR 2.16; 95% CI 1.02-4.61). PCP report of their intended participation in key aspects of active surveillance treatment decision-making varied: discussing worry (62.4%), reviewing benefits (48.5%) and risks (41.8%), and reviewing all treatment options (34.2%). PCPs who reported high confidence had increased odds of more participation in all aspects of active surveillance decision-making: reviewing all treatment options (OR 3.11; 95% CI 1.82-5.32), discussing worry (OR 2.12; 95% CI 1.28-3.51), and reviewing benefits (OR 3.13; 95% CI 1.89-5.16) and risks (OR 3.20; 95% CI 1.91-5.36). CONCLUSIONS: The majority of PCPs were confident about engaging with patients in low-risk prostate cancer treatment decision-making, though their intended participation varied widely across four key aspects of active surveillance care. With active surveillance being considered for other low-risk cancers (such as breast and thyroid), understanding factors influencing PCP involvement will be instrumental to supporting team-based cancer care.

Continuity of care and receipt of aggressive end of life care among women dying of ovarian cancer.

OBJECTIVE: To evaluate the association between post-diagnosis continuity of care and receipt of aggressive end of life care among women dying of ovarian cancer. METHODS: This retrospective claims analysis included 6680 Medicare beneficiaries over age 66 with ovarian cancer who survived at least one year after diagnosis, had at least 4 outpatient evaluation and management visits and died between 2000 and 2016. We calculated the Bice-Boxerman Continuity of Care Index (COC) for each woman, and split COC into tertiles (high, medium, low). We compared late or no hospice use, >1 emergency department (ED) visit, intensive care unit (ICU) admission, >1 hospitalization, terminal hospitalization, chemotherapy, and invasive and/or life extending procedures among women with high or medium vs. low COC using multivariable adjusted logistic regression. RESULTS: In this sample, 49.8% of women received aggressive care in the last month of life. Compared to women with low COC, women with high COC had 66% higher odds of chemotherapy (adjusted OR 1.66 CI 1.23-2.24) in the last two weeks of life. Women with high COC also had 16% greater odds of not enrolling in hospice compared to women with low COC (adjusted OR 1.16 CI 1.01-1.33). COC was not associated with late enrollment in hospice, hospital utilization, or aggressive procedures. CONCLUSIONS: COC at the end of life is complicated and may pose unique challenges in providing quality end of life care. Future work exploring the specific facets of continuity associated with quality end of life care is needed.

Disparities in risk perception of thyroid cancer recurrence and death.

BACKGROUND: To the authors' knowledge, studies regarding risk perception among survivors of thyroid cancer are scarce. METHODS: The authors surveyed patients who were diagnosed with differentiated thyroid cancer from the Surveillance, Epidemiology, and End Results registries of Georgia and Los Angeles County (2632 patients; 63% response rate). The analytic cohort was defined by a ≤5% risk of disease recurrence and mortality (1597 patients). Patients estimated their recurrence and mortality risks separately (increments of 10% and endpoints of ≤5% and ≥95%). Both outcomes were dichotomized between reasonably accurate estimates (risk perception of ≤5% or 10%) versus overestimation (risk perception of ≥20%). Multivariable logistic regression was used to identify factors associated with risk overestimation, and the relationships between overestimation and both worry and quality of life were evaluated. RESULTS: In the current study sample, 24.7% of patients overestimated their recurrence risk and 12.5% overestimated their mortality risk. A lower educational level was associated with overestimating disease recurrence (≤high school diploma: odds ratio [OR], 1.64 [95% CI, 1.16-2.31]; and some college: OR, 1.36 [95% CI, 1.02-1.81]) and mortality (≤high school diploma: OR, 1.86 [95% CI, 1.18-2.93]) risk compared with those attaining at least a college degree. Hispanic ethnicity was found to be associated with overestimating recurrence risk (OR, 1.44, 95% CI 1.02-2.03) compared with their white counterparts. Worry about recurrence and death was found to be greater among patients who overestimated versus those who had a reasonably accurate estimate of their risk of disease recurrence and mortality, respectively (P < .001). Patients who overestimated mortality risk also reported a decreased physical quality of life (mean T score, 43.1; 95% CI, 41.6-44.7) compared with the general population. CONCLUSIONS: Less educated patients and Hispanic patients were more likely to report inaccurate risk perceptions, which were associated with worry and a decreased quality of life.

Use of gabapentinoid medications among US adults with cancer, 2005-2015.

BACKGROUND: Gabapentinoid use for long-term cancer pain control may be problematic, given unclear mechanisms of action and increased concerns for physical dependence. The purpose of this report is to examine trends of gabapentinoid use among US adults with cancer from 2005 to 2015. METHODS: We conducted a serial, cross-sectional study using data from the Medical Expenditure Panel Survey (MEPS). We performed multiple logistic regression to examine the annual percentages of gabapentinoid users, which were adjusted for age, sex, and US region of residence. The amount of gabapentinoid prescriptions filled in 2015 was also estimated. RESULTS: The adjusted percentage of gabapentinoid users in 2015 was 5.60% (3.79%, 7.41%), 2.39 times greater than the percentage in 2005 (p < .001). By 2015, the number of gabapentinoid prescriptions had grown to approximately 3.52 million (2.40 million, 4.65 million). CONCLUSION: We observed greater than a twofold increase in the trend of gabapentinoid medication use among US adults with cancer. Investigations on the long-term efficacy of gabapentinoids for complex pain syndromes, and mitigation of risks, is essential to guide informed clinical management and keep patients safe.

Understanding the engagement of key decision support persons in patient decision making around breast cancer treatment.

BACKGROUND: Patients with breast cancer involve multiple decision support persons (DSPs) in treatment decision making, yet little is known about DSP engagement in decision making and its association with patient appraisal of the decision process. METHODS: Patients newly diagnosed with breast cancer reported to Georgia and Los Angeles Surveillance, Epidemiology, and End Results registries in 2014-2015 were surveyed 7 months after their diagnosis. The individual most involved in each respondent's decision making (the key DSP) was surveyed. DSP engagement was measured across 3 domains: 1) informed about decisions, 2) involved in decisions, and 3) aware of patient preferences. Patient decision appraisal included subjective decision quality (SDQ) and deliberation. This study evaluated bivariate associations with chi-square tests between domains of DSP engagement and independent DSP variables. Analysis of variance and multivariable logistic regression were used to compare domains of DSP engagement with patient decision appraisal. RESULTS: In all, 2502 patients (68% response rate) and 1203 eligible DSPs (70% response rate) responded. Most DSPs were husbands/partners or daughters, were white, and were college graduates. Husbands/partners were more likely to be more informed, involved, and aware (all P values < .01). English- and Spanish-speaking Latinos had a higher extent of (P = .02) but lower satisfaction with involvement (P < .01). A highly informed DSP was associated with higher odds of patient-reported SDQ (odds ratio, 1.46; 95% confidence interval, 1.03-2.08; P = .03). A highly aware DSP was associated with higher odds of patient-reported deliberation (odds ratio, 1.83; 95% confidence interval, 1.36-2.47; P < .01). CONCLUSIONS: In this population-based study, informal DSPs were engaged with and positively contributed to patients' treatment decision making. To improve decision quality, future interventions should incorporate DSPs.

Primary care provider-reported involvement in breast cancer treatment decisions.

BACKGROUND: Treatment decisions for patients with early-stage breast cancer often involve discussions with multiple oncology providers. However, the extent to which primary care providers (PCPs) are involved in initial treatment decisions remains unknown. METHODS: A stratified random sample of PCPs identified by newly diagnosed patients with early-stage breast cancer from the Georgia and Los Angeles Surveillance, Epidemiology, and End Results registries were surveyed (n = 517; a 61% response rate). PCPs were asked how frequently they discussed surgery, radiation, and chemotherapy options with patients; how comfortable they were with these discussions; whether they had the necessary knowledge to participate in decision making; and what their confidence was in their ability to help (on 5-item Likert-type scales). Multivariate logistic regression was used to identify PCP-reported attitudes associated with more PCP participation in each treatment decision. RESULTS: In this sample, 34% of PCPs reported that they discussed surgery, 23% discussed radiation, and 22% discussed chemotherapy options with their patients. Of those who reported more involvement in surgical decisions, 22% reported that they were not comfortable having a discussion, and 17% did not feel that they had the necessary knowledge to participate in treatment decision making. PCPs who positively appraised their ability to participate were more likely to participate in all 3 decisions (odds ratio [OR] for surgery, 6.01; 95% confidence interval [CI], 4.16-8.68; OR for radiation, 8.37; 95% CI, 5.16-13.58; OR for chemotherapy, 6.56; 95% CI, 4.23-10.17). CONCLUSIONS: A third of PCPs reported participating in breast cancer treatment decisions, yet gaps in their knowledge about decision making and in their confidence in their ability to help exist. Efforts to increase PCPs' knowledge about breast cancer treatment options may be warranted.