Using natural language processing to identify the status of homelessness and housing instability among serious illness patients from clinical notes in an integrated healthcare system.

Background: Efficiently identifying the social risks of patients with serious illnesses (SIs) is the critical first step in providing patient-centered and value-driven care for this medically vulnerable population. Objective: To apply and further hone an existing natural language process (NLP) algorithm that identifies patients who are homeless/at risk of homeless to a SI population. Methods: Patients diagnosed with SI between 2019 and 2020 were identified using an adapted list of diagnosis codes from the Center for Advance Palliative Care from the Kaiser Permanente Southern California electronic health record. Clinical notes associated with medical encounters within 6 months before and after the diagnosis date were processed by a previously developed NLP algorithm to identify patients who were homeless/at risk of homelessness. To improve the generalizability to the SI population, the algorithm was refined by multiple iterations of chart review and adjudication. The updated algorithm was then applied to the SI population. Results: Among 206 993 patients with a SI diagnosis, 1737 (0.84%) were identified as homeless/at risk of homelessness. These patients were more likely to be male (51.1%), age among 45-64 years (44.7%), and have one or more emergency visit (65.8%) within a year of their diagnosis date. Validation of the updated algorithm yielded a sensitivity of 100.0% and a positive predictive value of 93.8%. Conclusions: The improved NLP algorithm effectively identified patients with SI who were homeless/at risk of homelessness and can be used to target interventions for this vulnerable group.

Evolving Advance Care Planning in a Health Ecosystem: The Kaiser Permanente Experience.

BACKGROUND: Advance care planning is an integral part of supporting patients through serious illness and end-of-life care. PROBLEM: Several components of advance care planning may be too inflexible to account for patients' changing disease and evolving goals as serious illness progresses. Health systems are starting to implement processes to address these barriers, though implementation has varied. PROPOSED SOLUTION: In 2017, Kaiser Permanente introduced Life Care Planning (LCP), incorporating advance care planning dynamically into concurrent disease management. LCP provides a framework for identifying surrogates, documenting goals, and eliciting patient values across disease progression. LCP provides standardized training to facilitate communication and utilizes a centralized section within the electronic health record for longitudinal documentation of goals. OUTCOMES: More than 6000 physicians, nurses, and social workers have been trained in LCP. Over one million patients have engaged in LCP since its inception, with over 52% of patients age 55+ having a surrogate designated. There is evidence of high treatment concordance with patients' desired wishes (88.9%), with high rates of advance directive completion as well (84.1%).

Improving breast cancer survivors' knowledge using a patient-centered intervention.

BACKGROUND: Low-income, minority women with breast cancer experience a range of barriers to receiving survivorship information. Our objective was to test a novel, patient-centered intervention aimed at improving communication about survivorship care. METHODS: We developed a wallet card to provide oncologic and follow-up care survivorship information to breast cancer patients. We used a prospective, pre-post design to assess the intervention at a safety net hospital. The intervention was given by a patient navigator or community health worker. RESULTS: Patient knowledge (n = 130) of personal cancer history improved from baseline pretest to 1 week after the intervention for stage (66-93%; P < .05), treatment (79-92%; P < .05), and symptoms of recurrence (48-89%; P < .05), which was retained at 3 months. The intervention reduced the number of patients who were unsure when their mammogram was due (15-5% at 1 week and 6% at 3 months; P < .05). Nearly 90% reported they would be likely to share their survivorship card with their providers. CONCLUSION: A patient-centered survivorship card improved short-term recall of key survivorship care knowledge and seems to be effective at reducing communication barriers for this population. Further studies are warranted to assess long-term retention and the impact on receipt of appropriate survivorship follow-up care.

Closing the loop: an interactive action-research conference format for delivering updated medical information while eliciting Latina patient/family experiences and psychosocial needs post-genetic cancer risk assessment.

A patient/family-centered conference was conducted at an underserved community hospital to address Latinas' post-genetic cancer risk assessment (GCRA) medical information and psychosocial support needs, and determine the utility of the action research format. Latinas seen for GCRA were recruited to a half-day conference conducted in Spanish. Content was partly determined from follow-up survey feedback. Written surveys, interactive discussions, and Audience Response System (ARS) queries facilitated the participant-healthcare professional action research process. Analyses included descriptive statistics and thematic analysis. The 71 attendees (41 patients and 27 relatives/friends) were primarily non-US born Spanish-speaking females, mean age 43 years. Among patients, 73 % had a breast cancer history; 85 % had BRCA testing (49 % BRCA+). Nearly all (96 %) attendees completed the conference surveys and ARS queries; ≥48 % participated in interactive discussions. Most (95 %) agreed that the format met their personal interests and expectations and provided useful information and resources. Gaps/challenges identified in the GCRA process included pre-consult anxiety, uncertainty about reason for referral and expected outcomes, and psychosocial needs post-GCRA, such as absorbing and disseminating risk information to relatives and concurrently coping with a recent cancer diagnosis. The combined action research and educational conference format was innovative and effective for responding to continued patient information needs and addressing an important data gap about support needs of Latina patients and family members following genetic cancer risk assessment. Findings informed GCRA process improvements and provide a basis for theory-driven cancer control research.

Social-cognitive aspects of underserved Latinas preparing to undergo genetic cancer risk assessment for hereditary breast and ovarian cancer.

OBJECTIVES: As Latinos are a growing ethnic group in the United States, it is important to understand the socio-cultural factors that may be associated with cancer screening and prevention in this population. The socio-cultural factors that may affect preparedness to undergo genetic cancer risk assessment (GCRA) deserve particular attention. The pre-GCRA period can provide insight into variables that may influence how medically underserved Latinas, with limited health resources and access, understand hereditary cancer information and subsequently implement cancer risk management recommendations. This study explores social, cognitive and cultural variables in Latinas prior to undergoing GCRA. METHODS: The study sample consisted of low-income, underserved Latinas referred for GCRA because of a personal and/or family history of breast or ovarian cancer. Acculturation, cancer-specific fatalism, self-efficacy and social support were assessed prior to GCRA. RESULTS: Fifty Latinas (mean age=40.1+/-7.7) completed instruments; 86% had invasive cancer, 78% spoke primarily Spanish and 61% were of Mexican ancestry. Low levels of acculturation (n=50, mean=9.0+/-5.8) and cancer-specific fatalism (n=43, mean=5.6+/-3.2), but relatively high self-efficacy (n=49, mean=40.9+/-7.8) and social support (n=49, mean=37.3+/-8.7) were reported. Cancer-specific fatalism and self-efficacy were inversely correlated (r=-0.47, p=0.002). Those over age 38 at the time of cancer diagnosis reported higher acculturation (mean=11.4+/-7.2, p=0.02) and social support (mean=40.5+/-1.2, p=0.05). CONCLUSIONS: These findings suggest that medically underserved Latinas may already possess some of the necessary skills to successfully approach the GCRA process, but that special attention should be given to cultural factors.