Publications

BACKGROUND: Guidelines recommend shared decision making (SDM) for mammography screening for women ≥ 75 and not screening women with < 10-year life expectancy. High-quality SDM requires consideration of women's breast cancer (BC) risk, life expectancy, and values but is hard to implement because no models simultaneously estimate older women's individualized BC risk and life expectancy.

METHODS: Using competing risk regression and data from 83,330 women > 55 years who completed the 2004 Nurses' Health Study (NHS) questionnaire, we developed (in 2/3 of the cohort, n = 55,533) a model to predict 10-year non-breast cancer (BC) death. We considered 60 mortality risk factors and used best-subsets regression, the Akaike information criterion, and c-index, to identify the best-fitting model. We examined model performance in the remaining 1/3 of the NHS cohort (n = 27,777) and among 17,380 Black Women's Health Study (BWHS) participants, ≥ 55 years, who completed the 2009 questionnaire. We then included the identified mortality predictors in a previously developed competing risk BC prediction model and examined model performance for predicting BC risk.

RESULTS: Mean age of NHS development cohort participants was 70.1 years (± 7.0); over 10 years, 3.1% developed BC, 0.3% died of BC, and 20.1% died of other causes; NHS validation cohort participants were similar. BWHS participants were younger (mean age 63.7 years [± 6.7]); over 10-years 3.1% developed BC, 0.4% died of BC, and 11.1% died of other causes. The final non-BC death prediction model included 21 variables (age; body mass index [BMI]; physical function [3 measures]; comorbidities [12]; alcohol; smoking; age at menopause; and mammography use). The final BC prediction model included age, BMI, alcohol and hormone use, family history, age at menopause, age at first birth/parity, and breast biopsy history. When risk factor regression coefficients were applied in the validation cohorts, the c-index for predicting 10-year non-BC death was 0.790 (0.784-0.796) in NHS and 0.768 (0.757-0.780) in BWHS; for predicting 5-year BC risk, the c-index was 0.612 (0.538-0.641) in NHS and 0.573 (0.536-0.611) in BWHS.

CONCLUSIONS: We developed and validated a novel competing-risk model that predicts 10-year non-BC death and 5-year BC risk. Model risk estimates may help inform SDM around mammography screening.

CONTEXT: During acute health decompensations for seriously ill patients, emergency clinicians often determine the intensity end-of-life care. Little is known about how emergency clinicians conduct these conversations, especially among those who have received serious illness communication training.

OBJECTIVES: To determine the self-reported practice patterns of code status conversations by emergency clinicians with and without serious illness communication training.

METHODS: A cross-sectional survey was conducted among emergency clinicians with and without a recent evidence-based, serious illness communication training tailored for emergency clinicians. Emergency clinicians were included from two academic medical centers. A five-point Likert scale ("very unlikely" to "very likely" to ask) was used to assess the self-reported likelihood of asking about patients' preferences for medical procedures and patients' values and goals.

RESULTS: Among 161 respondents (71% response rate), 77 (48%) received the training. A total of 70% of emergency clinicians reported asking about procedure-based questions, and only 38% reported asking about patient's values regarding end-of-life care. For value-based questions, statistically significant differences were observed between emergency clinicians who underwent the training and those who did not in four of the seven questions asked (e.g., the higher odds of exploring the patient's life priorities [adjusted OR = 4.34, 95% CI = 1.95-9.65, P-value < 0.001]). No difference was observed in the self-reported rates of all procedure-based questions between the two groups.

CONCLUSION: Most emergency clinicians reported asking about procedure-based questions, and some asked about patient's value-based questions. Clinicians with recent serious illness communication training may ask more about some values and priorities.

BACKGROUND: Serious illness conversations may lead to care consistent with patients' goals near the end of life. The emergency department could serve as an important time and location for these conversations.

AIM: To determine the feasibility of an emergency department-based, brief motivational interview to stimulate serious illness conversations among seriously ill older adults by trained nurses.

DESIGN: A pre-/post-intervention study.

SETTINGS/PARTICIPANTS: In an urban, tertiary care, academic medical center and a community hospital from January 2021 to January 2022, we prospectively enrolled adults ⩾50 years of age with serious illness and an expected prognosis <1 year. We measured feasibility outcomes using the standardized framework for feasibility studies. In addition, we also collected the validated 4-item Advance Care Planning Engagement Survey (a 5-point Likert scale) at baseline and 4-week follow-up and reviewing the electronic medical record for documentation related to newly completed serious illness conversations.

RESULTS: Among 116 eligible patients who were willing and able to participate, 76 enrolled (65% recruitment rate), and 68 completed the follow-up (91% retention rate). Mean patient age was 64.4 years (SD 8.4), 49% were female, and 58% had metastatic cancer. In all, 16 nurses conducted the intervention, and all participants completed the intervention with a median duration of 27 min. Self-reported Advance Care Planning Engagement increased from 2.78 pre to 3.31 post intervention (readiness to "talk to doctors about end-of-life wishes," p < 0.008). Documentation of health care proxy forms increased (62-70%) as did Medical Order for Life Sustaining Treatment (1-11%) during the 6 months after the emergency department visit.

CONCLUSION: A novel, emergency department-based, nurse-led brief motivational interview to stimulate serious illness conversations is feasible and may improve advance care planning engagement and documentation in seriously ill older adults.

BACKGROUND: Emergency department (ED) visits are common at the end-of-life, but the identification of patients with life-limiting illness remains a key challenge in providing timely and resource-sensitive advance care planning (ACP) and palliative care services. To date, there are no validated, automatable instruments for ED end-of-life screening. Here, we developed a novel electronic health record (EHR) prognostic model to screen older ED patients at high risk for 6-month mortality and compare its performance to validated comorbidity indices.

METHODS: This was a retrospective, observational cohort study of ED visits from adults aged ≥65 years who visited any of 9 EDs across a large regional health system between 2014 and 2019. Multivariable logistic regression that included clinical and demographic variables, vital signs, and laboratory data was used to develop a 6-month mortality predictive model-the Geriatric End-of-life Screening Tool (GEST) using five-fold cross-validation on data from 8 EDs. Performance was compared to the Charlson and Elixhauser comorbidity indices using area under the receiver-operating characteristic curve (AUROC), calibration, and decision curve analyses. Reproducibility was tested against data from the remaining independent ED within the health system. We then used GEST to investigate rates of ACP documentation availability and code status orders in the EHR across risk strata.

RESULTS: A total of 431,179 encounters by 123,128 adults were included in this study with a 6-month mortality rate of 12.2%. Charlson (AUROC (95% CI): 0.65 (0.64-0.69)) and Elixhauser indices (0.69 (0.68-0.70)) were outperformed by GEST (0.82 (0.82-0.83)). GEST displayed robust performance across demographic subgroups and in our independent validation site. Among patients with a greater than 30% mortality risk using GEST, only 5.0% had ACP documentation; 79.0% had a code status previously ordered, of which 70.7% were full code. In decision curve analysis, GEST provided greater net benefit than the Charlson and Elixhauser scores.

CONCLUSIONS: Prognostic models using EHR data robustly identify high mortality risk older adults in the ED for whom code status, ACP, or palliative care interventions may be of benefit. Although all tested methods identified patients approaching the end-of-life, GEST was most performant. These tools may enable resource-sensitive end-of-life screening in the ED.

Background. Clinicians need to find decision aids (DAs) useful for their successful implementation. Therefore, we aimed to conduct an exploratory study to learn primary care clinicians' (PCPs) perspectives on a mammography DA for women ≥75 to inform its implementation. Methods. We sent a cross-sectional survey to 135 PCPs whose patients had participated in a randomized trial of the DA. These PCPs practiced at 1 of 11 practices in Massachusetts or North Carolina. PCPs were asked closed-ended and open-ended questions on shared decision making (SDM) around mammography with women ≥75 and on the DA's acceptability, appropriateness, and feasibility. Results. Eighty PCPs participated (24 [30%] from North Carolina). Most (n = 69, 86%) thought that SDM about mammography with women ≥75 was extremely/very important and that they engaged women ≥75 in SDM around mammography frequently/always (n = 49, 61%). Regarding DA acceptability, 60% felt the DA was too long. Regarding appropriateness, 70 (89%) thought it was somewhat/very helpful and that it would help patients make more informed decisions; 55 (70%) would recommend it. Few (n = 6, 8%) felt they had other resources to support this decision. Regarding feasibility, 53 (n = 67%) thought it would be most feasible for patients to receive the DA before a visit from medical assistants rather than during or after a visit or from health educators. Most (n = 62, 78%) wanted some training to use the DA. Limitations. Sixty-nine percent of PCPs in this small study practiced in academic settings. Conclusions. Although PCPs were concerned about the DA's length, most found it helpful and informative and felt it would be feasible for medical assistants to deliver the DA before a visit. Implications. Study findings may inform implementation of this and other DAs.

Background: The Gail, Breast Cancer Surveillance Consortium (BCSC), and Tyrer-Cuzick breast cancer risk prediction models are recommended for use in primary care. Calculating breast cancer risk is particularly important for women in their 40s when deciding on mammography, with some guidelines recommending screening for those with 5-year risk similar to women age 50 (≥1.1%). Yet, little is known about risk estimate agreement among models for these women. Materials and Methods: Four hundred nine Boston-area women 40-49 years of age completed a risk questionnaire before a primary care visit to compute their breast cancer risk. The kappa statistic was used to examine when (1) Gail and BCSC agreed on 5-year risk ≥1.1%; (2) Gail estimated 5-year risk ≥1.7% and Tyrer-Cuzick estimated 10-year risk ≥5% (guideline thresholds for recommending prevention medications); and when (3) Gail and Tyrer-Cuzick agreed on lifetime risk ≥20% (threshold for breast MRI using Tyrer-Cuzick). Results: Participant mean age was 44.1 years, 56.7% were non-Hispanic white, and 7.8% had a first-degree relative with breast cancer. Of 266 with breast density information to estimate both Gail and BCSC, the models agreed on 5-year risk being ≥1.1% for 36 women, kappa = 0.34 (95% confidence interval: 0.23-0.45). Gail and Tyrer-Cuzick estimates led to agreement about prevention medications for 8 women, kappa 0.41 (0.20-0.61), and models agreed on lifetime risk ≥20% for 3 women, kappa 0.08 (-0.01 to 0.16). Conclusions: There is weak agreement on breast cancer risk estimates generated by risk models recommended for primary care. Using different models may lead to different clinical recommendations for women in their 40s.

To learn how to improve telemedicine for adults >65, we asked primary care clinicians ("PCPs") affiliated with one large Boston-area health system their views on using telemedicine (which included phone-only or video visits) with adults >65 during the COVID-19 pandemic. In open-ended questions, we asked PCPs to describe any challenges or useful experiences with telemedicine and suggestions for improving telemedicine as part of a larger web-based survey conducted between September 2020 and February 2021. Overall, 163/383 (42%) PCPs responded to the survey. Of these, 114 (70%) completed at least one open-ended question, 85% were non-Hispanic white, 59% were female, 75% were community-based, and 75% were in practice >20 years. We identified three major themes in participants' comments including the need to optimize telemedicine; integrate telemedicine within primary care; and that PCPs had disparate attitudes towards telemedicine for older adults. To optimize telemedicine, PCPs recommended more effective digital platforms, increased utilization of home medical equipment (e.g., blood pressure cuffs), and better coordination with caregivers. For integration, PCPs recommended targeting telemedicine for certain types of visits (e.g., chronic disease management), enabling video access, and reducing administrative burdens on PCPs. As for PCP attitudes, some felt telemedicine enhanced the doctor-patient relationship, improved the patient experience, and improved show rates. Others felt that telemedicine visits were incomplete without a physical exam, were less rewarding, and could be frustrating. Overall, PCPs saw a role for telemedicine in older adults' care but felt that more support is needed for these visits than currently offered.

BACKGROUND: Randomized controlled trials show that certain axillary surgical practices can be safely deescalated in older adults with early-stage breast cancer. Hospital volume is often equated with surgical quality, but it is unclear whether this includes performance of low-value surgeries. We sought to describe how utilization of two low-value axillary surgeries has varied by time and hospital volume.

METHODS: Women aged ≥ 70 years diagnosed with breast cancer from 2013 to 2016 were identified in the National Cancer Database. The outcomes of interest were sentinel lymph node biopsy (SLNB) in cT1N0 hormone receptor-positive cancer patients and axillary lymph node dissection (ALND) in cT1-2N0 patients undergoing breast-conserving surgery with ≤ 2 pathologically positive nodes. Time trends in procedure use and multivariable regression with restricted cubic splines were performed, adjusting for patient, disease, and hospital factors.

RESULTS: Overall, 83.4% of 44,779 women eligible for omission of SLNB underwent SLNB and 20.0% of 7216 patients eligible for omission of ALND underwent ALND. SLNB rates did not change significantly over time and remained significantly different by age group (70-74 years: 93.5%; 75-79 years: 89.7%, 80-84 years: 76.7%, ≥ 85 years: 48.9%; p < 0.05). ALND rates decreased over the study period across all age groups included (22.5 to 16.9%, p < 0.001). In restricted cubic splines models, lower hospital volume was associated with higher likelihood of undergoing SLNB and ALND.

CONCLUSIONS: ALND omission has been more widely adopted than SLNB omission in older adults, but lower hospital volume is associated with higher likelihood of both procedures. Practice-specific deimplementation strategies are needed, especially for lower-volume hospitals.