Bolstering Advance Care Planning Measurement Using Natural Language Processing

Despite its growth as a clinical activity and research topic, the complex dynamic nature of advance care planning (ACP) has posed serious challenges for researchers hoping to quantitatively measure it. Methods for measurement have traditionally depended on lengthy manual chart abstractions or static documents (e.g., advance directive forms) even though completion of such documents is only one aspect of ACP. Natural language processing (NLP), in the form of an assisted electronic health record (EHR) review, is a technological advancement that may help researchers better measure ACP activity. In this article, we aim to show how NLP-assisted EHR review supports more accurate and robust measurement of ACP. We do so by presenting three example applications that illustrate how using NLP for this purpose supports (1) measurement in research, (2) detailed insights into ACP in quality improvement, and (3) identification of current limitations of ACP in clinical settings.

Structural Barriers to Well-grounded Advance Care Planning for the Seriously Ill: a Qualitative Study of Clinicians' and Administrators' Experiences During a Pragmatic Trial

BACKGROUND

Advance Care Planning (ACP) comprises an iterative communication process aimed at understanding patients' goals, values, and preferences in the context of considering and preparing for future medical treatments and decision making in serious illness. The COVID pandemic heightened patients' and clinicians' awareness of the need for ACP.

OBJECTIVE

Our goal was to explore the experiences of clinicians and administrators in the context of an intervention to improve ACP during the COVID pandemic.

DESIGN

Qualitative interview study.

PARTICIPANTS

Clinicians and administrators across five sites that participated in the ACP-COVID trial.

APPROACH

We conducted semi-structured, qualitative interviews examining the context and approach to ACP. Interviews were analyzed using template analysis to systematically organize the data and facilitate review across the categories and participants. Templates were developed with iterative input and line-by-line review by the analytic team, to reach consensus. Findings were then organized into emergent themes.

KEY RESULTS

Across 20 interviews (4 administrators, 16 clinicians) we identified three themes related to how participants thought about ACP: (1) clinicians have varying views of what constitutes ACP; (2) the health system critically shapes ACP culture and norms; and (3) the centrality of clinicians' affective experience and own needs related to ACP. Varying approaches to ACP include a forms-focused approach; a discussion-based approach; and a parental approach. System features that shape ACP norms are (1) the primacy of clinician productivity measures; (2) the role of the EHR; and (3) the culture of quality improvement.

CONCLUSIONS

Despite high organizational commitment to ACP, we found that the health system channeled clinicians' ACP efforts narrowly on completion of forms, in tension with the ideal of well-grounded ACP. This resulted in a state of moral distress that risks undermining confidence in the process of ACP and may increase risk of harm for patients, family/caregivers, and providers.

TRIAL REGISTRATION

ClinicalTrials.gov Identifier: NCT04660422.

Forms or Free-Text? Measuring Advance Care Planning Activity Using Electronic Health Records

Advance care planning (ACP) discussions seek to guide future serious illness care. These discussions may be recorded in the electronic health record by documentation in clinical notes, structured forms and directives, and physician orders. Yet, most studies of ACP prevalence have only examined structured electronic health record elements and ignored data existing in notes. We sought to investigate the relative comprehensiveness and accuracy of ACP documentation from structured and unstructured electronic health record data sources. We evaluated structured and unstructured ACP documentation present in the electronic health records of 435 patients with cancer drawn from three separate healthcare systems. We extracted structured ACP documentation by manually annotating written documents and forms scanned into the electronic health record. We coded unstructured ACP documentation using a rule-based natural language processing software that identified ACP keywords within clinical notes and was subsequently reviewed for accuracy. The unstructured approach identified more instances of ACP documentation (238, 54.7% of patients) than the structured ACP approach (187, 42.9% of patients). Additionally, 16.6% of all patients with structured ACP documentation only had documents that were judged as misclassified, incomplete, blank, unavailable, or a duplicate of a previously entered erroneous document. ACP documents scanned into electronic health records represent a limited view of ACP activity. Research and measures of clinical practice with ACP should incorporate information from unstructured data.

Video Intervention and Goals-of-Care Documentation in Hospitalized Older Adults: The VIDEO-PCE Randomized Clinical Trial

Importance

Despite the benefits of goals-of-care (GOC) communication, many hospitalized individuals never communicate their goals or preferences to clinicians.

Objective

To assess whether a GOC video intervention delivered by palliative care educators (PCEs) increased the rate of GOC documentation.

Design, Setting, and Participants

This pragmatic, stepped-wedge cluster randomized clinical trial included patients aged 65 years or older admitted to 1 of 14 units at 2 urban hospitals in New York and Boston from July 1, 2021, to October 31, 2022.

Intervention

The intervention involved PCEs (social workers and nurses trained in GOC communication) facilitating GOC conversations with patients and/or their decision-makers using a library of brief, certified video decision aids available in 29 languages. Patients in the control period received usual care.

Main Outcome and Measures

The primary outcome was GOC documentation, which included any documentation of a goals conversation, limitation of life-sustaining treatment, palliative care, hospice, or time-limited trials and was obtained by natural language processing.

Results

A total of 10 802 patients (mean [SD] age, 78 [8] years; 51.6% male) were admitted to 1 of 14 hospital units. Goals-of-care documentation during the intervention phase occurred among 3744 of 6023 patients (62.2%) compared with 2396 of 4779 patients (50.1%) in the usual care phase (P < .001). Proportions of documented GOC discussions for Black or African American individuals (865 of 1376 [62.9%] vs 596 of 1125 [53.0%]), Hispanic or Latino individuals (311 of 548 [56.8%] vs 218 of 451 [48.3%]), non-English speakers (586 of 1059 [55.3%] vs 405 of 863 [46.9%]), and people living with Alzheimer disease and related dementias (520 of 681 [76.4%] vs 355 of 570 [62.3%]) were greater during the intervention phase compared with the usual care phase.

Conclusions and Relevance

In this stepped-wedge cluster randomized clinical trial of older adults, a GOC video intervention delivered by PCEs resulted in higher rates of GOC documentation compared with usual care, including among Black or African American individuals, Hispanic or Latino individuals, non-English speakers, and people living with Alzheimer disease and related dementias. The findings suggest that this form of patient-centered care delivery may be a beneficial decision support tool.

Trial Registration

ClinicalTrials.gov Identifier: NCT04857060.

Facilitators of palliative care quality improvement team cohesion: Lessons from a seven-site implementation project in India

BACKGROUND

The Palliative Care: Promoting Access and Improvement of the Cancer Experience (PC-PAICE) initiative is a team-based, palliative care (PC) quality improvement (QI) project working to promote high-quality PC in India. As a PC QI initiative, PC-PAICE implementation relied upon building interdisciplinary teams, providing the ideal context for understanding facilitators of team cohesion that compelled clinical, organizational, and administrative team members to work together. There is an opportunity to leverage the intersection between QI implementation and organizational theory to inform and improve implementation science.

PURPOSE

As a subaim of a larger implementation evaluation, we aimed to identify facilitators of team cohesion within QI implementation context.

METHODOLOGY

A quota sampling approach captured the perspectives of 44 stakeholders across three strata (organizational leaders, clinical leaders, and clinical team members) from all seven sites through a semistructured interview guide informed by the Consolidated Framework for Implementation Research (CFIR). We used a combination of inductive and deductive approaches informed by organizational theory to identify facilitators.

RESULT

We identified three facilitators of PC team cohesion: (a) balancing formalization and flexibility around team roles, (b) establishing widespread awareness of the QI project, and (c) prioritizing a nonhierarchical organizational culture.

PRACTICE IMPLICATIONS

Leveraging CFIR to analyze PC-PAICE stakeholder interviews created a data set conducive to understanding complex multisite implementation. Layering role and team theory to our implementation analysis helped us identify facilitators of team cohesion across levels within the team (bounded team), beyond the team (teaming), and surrounding the team (culture). These insights demonstrate the value of team and role theories in implementation evaluation efforts.

Reaching Ambulatory Older Adults with Educational Tools: Comparative Efficacy and Cost of Varied Outreach Modalities in Primary Care

BACKGROUND

Providing patients with access to health information that can be obtained outside of an office visit is an important part of education, yet little is known about the effectiveness of outreach modalities to connect older adults to online educational tools. The objective was to identify the effectiveness and cost of outreach modalities providing online information about advance care planning (ACP) for older adults.

METHODS

Six different outreach modalities were utilized to connect patients to online educational tools (ACP video decision aids). Participants were 13,582 patients aged 65 and older of 185 primary care providers with appointments over a 30-month period within a large health system in the greater New York City area. Main outcome measures were number of online video views and costs per outreach for each modality.

KEY RESULTS

There were 1150 video views for 21,407 remote outreach events. Text messages, sent to the largest volume of patients (8869), had the highest outcome rate (9.6%) and were the most economical ($0.09). Characterization of phone calls demonstrated 21.7% engagement in the topic of ACP but resulted in minimal video views (<1%) and incurred the highest cost per outreach ($2.88). In-office handouts had negligible results (<1%).

CONCLUSIONS

Text was the most cost-effective modality to connect older adults to an online educational tool in this pragmatic trial, though overall efficacy of all modalities was low.

Creating KidneyPal: A Specialty-Aligned Palliative Care Service for People with Kidney Disease

BACKGROUND

Patients with kidney disease have notable unmet palliative care needs and represent an underserved population for specialty palliative care teams.

INTERVENTION

We designed a specialty-aligned interprofessional palliative care service called KidneyPal that is aimed at improving delivery of palliative care to patients with kidney disease through focus groups and iterative improvement cycles.

MEASURES

We iteratively measured the development of KidneyPal through clinical process metrics: percent of the inpatient nephrology census followed by KidneyPal, patient demographics, consult origin, clinician feedback, and self-reported team interventions.

OUTCOMES

KidneyPal saw 314 unique patients from January 2019 to January 2021. The majority of consultations came from nephrology services though the source of consultation changed over time. We consulted on an average of 13.5% of the entire inpatient nephrology patient hospital census with highest involvement with patients on the inpatient nephrology hemodialysis service (mean of 29.9%). KidneyPal was rated highly by surveyed nephrology clinicians and provided high rates of psychosocial support and goals of care interventions.

LESSONS LEARNED

The creation of KidneyPal led to us to serve a new cohort of patients with specialty palliative care. We grew over time to serve the full range of patients with kidney disease as defined by our nephrology service lines. We succeeded in doing so by embedding in nephrology and building relationships with those caring for people with kidney disease while tailoring our service and interventions over time.

Study protocol for Video Images about Decisions to Improve Ethical Outcomes with Palliative Care Educators (VIDEO-PCE): a pragmatic stepped wedge cluster randomised trial of older patients admitted to the hospital

INTRODUCTION

Despite the known benefit to patients and families, discussions about goals, values and preferences for medical care in advancing serious illness often do not occur. Many system and clinician factors, such as patient and clinician reticence and shortage of specialty palliative care teams, contribute to this lack of communication. To address this gap, we designed an intervention to promote goals-of-care conversations and palliative care referrals in the hospital setting by using trained palliative care educators and video decision aids. This paper presents the rationale, design and methods for a trial aimed at addressing barriers to goals-of-care conversations for hospitalised adults aged 65 and older and those with Alzheimer's disease and related Dementias, regardless of age.

METHODS AND ANALYSIS

The Video Image about Decisions to Improve Ethical Outcomes with Palliative Care Educators is a pragmatic stepped wedge, cluster randomised controlled trial, which aims to improve and extend goals-of-care conversations in the hospital setting with palliative care educators trained in serious illness communication and video decision aids. The primary outcome is the proportion of patients with goals-of-care documentation in the electronic health record. We estimate that over 9000 patients will be included.

ETHICS AND DISSEMINATION

The Institutional Review Board (IRB) at Boston Medical Center will serve as the single IRB of record for all regulatory and ethical aspects of this trial. BMC Protocol Number: H-41482. Findings will be presented at national meetings and in publications. This trial is registered at ClinicalTrials.gov.

TRIAL REGISTRATION NUMBER

NCT04857060; ClinicalTrials.gov.

Video Images about Decisions for Ethical Outcomes in Kidney Disease (VIDEO-KD): the study protocol for a multi-centre randomised controlled trial

INTRODUCTION

Older patients with advanced chronic kidney disease (CKD) often are inadequately prepared to make informed decisions about treatments including dialysis and cardiopulmonary resuscitation. Further, evidence shows that patients with advanced CKD do not commonly engage in advance care planning (ACP), may suffer from poor quality of life, and may be exposed to end-of-life care that is not concordant with their goals. We aim to study the effectiveness of a video intervention on ACP, treatment preferences and other patient-reported outcomes.

METHODS AND ANALYSIS

The Video Images about Decisions for Ethical Outcomes in Kidney Disease trial is a multi-centre randomised controlled trial that will test the effectiveness of an intervention that includes a CKD-related video decision aid followed by recording personal video declarations about goals of care and treatment preferences in older adults with advancing CKD. We aim to enrol 600 patients over 5 years at 10 sites.

ETHICS AND DISSEMINATION

Regulatory and ethical aspects of this trial include a single Institutional Review Board mechanism for approval, data use agreements among sites, and a Data Safety and Monitoring Board. We intend to disseminate findings at national meetings and publish our results.

TRIAL REGISTRATION NUMBER

NCT04347629.

Association of an Advance Care Planning Video and Communication Intervention With Documentation of Advance Care Planning Among Older Adults: A Nonrandomized Controlled Trial

IMPORTANCE

COVID-19 has disproportionately killed older adults and racial and ethnic minority individuals, raising questions about the relevance of advance care planning (ACP) in this population. Video decision aids and communication skills training offer scalable delivery models.

OBJECTIVE

To assess whether ACP video decision aids and a clinician communication intervention improved the rate of ACP documentation during an evolving pandemic, with a focus on African American and Hispanic patients.

DESIGN, SETTING, AND PARTICIPANTS

The Advance Care Planning: Communicating With Outpatients for Vital Informed Decisions trial was a pre-post, open-cohort nonrandomized controlled trial that compared ACP documentation across the baseline pre-COVID-19 period (September 15, 2019, to March 14, 2020), the COVID-19 wave 1 period (March 15, 2020, to September 14, 2020), and an intervention period (December 15, 2020, to June 14, 2021) at a New York metropolitan area ambulatory network of 22 clinics. All patients 65 years or older who had at least 1 clinic or telehealth visit during any of the 3 study periods were included.

MAIN OUTCOMES AND MEASURES

The primary outcome was ACP documentation.

RESULTS

A total of 14 107 patients (mean [SD] age, 81.0 [8.4] years; 8856 [62.8%] female; and 2248 [15.9%] African American or Hispanic) interacted with clinicians during the pre-COVID-19 period; 12 806 (mean [SD] age, 81.2 [8.5] years; 8047 [62.8%] female; and 1992 [15.6%] African American or Hispanic), during wave 1; and 15 106 (mean [SD] 80.9 [8.3] years; 9543 [63.2%] female; and 2535 [16.8%] African American or Hispanic), during the intervention period. Clinicians documented ACP in 3587 patients (23.8%) during the intervention period compared with 2525 (17.9%) during the pre-COVID-19 period (rate difference [RD], 5.8%; 95% CI, 0.9%-7.9%; P = .01) and 1598 (12.5%) during wave 1 (RD, 11.3%; 95% CI, 6.3%-12.1%; P < .001). Advance care planning was documented in 447 African American patients (30.0%) during the intervention period compared with 233 (18.1%) during the pre-COVID-19 period (RD, 11.9%; 95% CI, 4.1%-15.9%; P < .001) and 130 (11.0%) during wave 1 (RD, 19.1%; 95% CI, 11.7%-21.2%; P < .001). Advance care planning was documented for 222 Hispanic patients (21.2%) during the intervention period compared with 127 (13.2%) during the pre-COVID-19 period (RD, 8.0%; 95% CI, 2.1%-10.9%; P = .004) and 82 (10.2%) during wave 1 (RD, 11.1%; 95% CI, 5.5%-14.5%; P < .001).

CONCLUSIONS AND RELEVANCE

This intervention, implemented during the evolving COVID-19 pandemic, was associated with higher rates of ACP documentation, especially for African American and Hispanic patients.

TRIAL REGISTRATION

ClinicalTrials.gov Identifier: NCT04660422.

Natural Language Processing to Assess Palliative Care and End-of-Life Process Measures in Patients With Breast Cancer With Leptomeningeal Disease

Background:

Palliative care consultation during serious life-limiting illness can reduce symptom burden and improve quality of care. However, quantifying the impact of palliative care is hindered by the limitations of manual chart review and administrative coding.

Objectives:

Using novel natural language process (NLP) techniques, we examined associations between palliative care consultations and performance on nationally endorsed metrics for high-quality end-of-life (EOL) care in patients with leptomeningeal disease (LMD) secondary to metastatic breast cancer.

Methods:

Patients with breast cancer with LMD were identified using administrative billing codes and NLP review of magnetic resonance imaging reports at 2 tertiary care centers between 2010 and 2016. Next, NLP was used to review clinical notes to (1) determine the presence of palliative care consultations and (2) determine the performance of process measures associated with high-quality EOL care, including discussions of goals of care, code status limitations, and hospice. Associations between palliative care consultation and documentation of EOL process measures were assessed using logistic regression.

Results:

We identified 183 cases of LMD. Median age was 56 (interquartile range [IQR]: 46-64) years and median survival was 150 days (IQR: 67-350). Within 6 months of diagnosis, 88.5% of patients had documentation of ≥1 process measure, including discussions of goals of care (63.4%), code status limitations (62.8%), or hospice (72.1%). Palliative care consultation was a predictor of subsequent documentation of goals of care (odds ratio [OR], 3.15; 95% confidence interval [CI], 1.58-6.27) and hospice discussions (OR, 4.61; 95% CI, 2.12-10.03).

Conclusion:

Palliative care involvement is associated with increased performance of EOL process measures in patients with breast cancer with LMD.

Natural Language Processing to Assess End-of-Life Quality Indicators in Cancer Patients Receiving Palliative Surgery

BACKGROUND

Palliative surgical procedures are frequently performed to reduce symptoms in patients with advanced cancer, but quality is difficult to measure.

OBJECTIVE

To determine whether natural language processing (NLP) of the electronic health record (EHR) can be used to (1) identify a population of cancer patients receiving palliative gastrostomy and (2) assess documentation of end-of-life process measures in the EHR.

DESIGN/SETTING

Retrospective cohort study of 302 adult cancer patients who received a gastrostomy tube at a single tertiary medical center.

MEASUREMENTS

Sensitivity and specificity of NLP compared to gold standard of manual chart abstraction in identifying a palliative indication for gastrostomy tube placement and documentation of goals of care discussions, code status determination, palliative care referral, and hospice assessment.

RESULTS

Among 302 cancer patients who underwent gastrostomy, 68 (22.5%) were classified by NLP as having a palliative indication for the procedure compared to 71 patients (23.5%) classified by human coders. Human chart abstraction took >2600 times longer than NLP (28 hours vs. 38 seconds). NLP identified the correct patients with 95.8% sensitivity and 97.4% specificity. NLP also identified end-of-life process measures with high sensitivity (85.7%-92.9%,) and specificity (96.7%-98.9%). In the two months leading up to palliative gastrostomy placement, 20.5% of patients had goals of care discussions documented. During the index hospitalization, 67.7% had goals of care discussions documented.

CONCLUSIONS

NLP offers opportunities to identify patients receiving palliative surgical procedures and can rapidly assess established end-of-life process measures with an accuracy approaching that of human coders.