Large multisite clinical trials studying decision-making when facing serious illness require an efficient method for abstraction of advance care planning (ACP) documentation from clinical text ...documents. However, the current gold standard method of manual chart review is time-consuming and unreliable.
To evaluate the ability to use natural language processing (NLP) to identify ACP documention in clinical notes from patients participating in a multisite trial.
Patients with advanced cancer followed in three disease-focused oncology clinics at Duke Health, Mayo Clinic, and Northwell Health were identified using administrative data. All outpatient and inpatient notes from patients meeting inclusion criteria were extracted from electronic health records (EHRs) between March 2018 and March 2019. NLP text identification software with semi-automated chart review was applied to identify documentation of four ACP domains: (1) conversations about goals of care, (2) limitation of life-sustaining treatment, (3) involvement of palliative care, and (4) discussion of hospice. The performance of NLP was compared to gold standard manual chart review.
435 unique patients with 79,797 notes were included in the study. In our validation data set, NLP achieved F1 scores ranging from 0.84 to 0.97 across domains compared to gold standard manual chart review. NLP identified ACP documentation in a fraction of the time required by manual chart review of EHRs (1-5 minutes per patient for NLP, vs. 30-120 minutes for manual abstraction).
NLP is more efficient and as accurate as manual chart review for identifying ACP documentation in studies with large patient cohorts.
Patients and their families struggle with myriad choices concerning
medical treatments that frequently precede death. Advance directives have
been proposed as a tool to facilitate end-of-life ...decision making, yet frequently
fail to achieve this goal. In the context of the case of a man with metastatic
cancer for whom an advance directive was unable to prevent a traumatic death,
I review the challenges in creating and implementing advance directives, discuss
factors that can affect clear decision making; including trust, uncertainty,
emotion, hope, and the presence of multiple medical providers; and offer practical
suggestions for physicians. Advance care planning remains a useful tool for
approaching conversations with patients about the end of life. However, such
planning should occur within a framework that emphasizes responding to patient
and family emotions and focuses more on goals for care and less on specific
treatments.
Summary Background Palliative oxygen therapy is widely used for treatment of dyspnoea in individuals with life-limiting illness who are ineligible for long-term oxygen therapy. We assessed the ...effectiveness of oxygen compared with room air delivered by nasal cannula for relief of breathlessness in this population of patients. Methods Adults from outpatient clinics at nine sites in Australia, the USA, and the UK were eligible for enrolment in this double-blind, randomised controlled trial if they had life-limiting illness, refractory dyspnoea, and partial pressure of oxygen in arterial blood (PaO2 ) more than 7·3 kPa. Participants were randomly assigned in a 1:1 ratio by a central computer-generated system to receive oxygen or room air via a concentrator through a nasal cannula at 2 L per min for 7 days. Participants were instructed to use the concentrator for at least 15 h per day. The randomisation sequence was stratified by baseline PaO2 with balanced blocks of four patients. The primary outcome measure was breathlessness (0–10 numerical rating scale NRS), measured twice a day (morning and evening). All randomised patients who completed an assessment were included in the primary analysis for that data point (no data were imputed). This study is registered, numbers NCT00327873 and ISRCTN67448752. Findings 239 participants were randomly assigned to treatment (oxygen, n=120; room air, n=119). 112 (93%) patients assigned to receive oxygen and 99 (83%) assigned to receive room air completed all 7 days of assessments. From baseline to day 6, mean morning breathlessness changed by −0·9 points (95% CI −1·3 to −0·5) in patients assigned to receive oxygen and by −0·7 points (−1·2 to −0·2) in patients assigned to receive room air (p=0·504). Mean evening breathlessness changed by −0·3 points (−0·7 to 0·1) in the oxygen group and by −0·5 (−0·9 to −0·1) in the room air group (p=0·554). The frequency of side-effects did not differ between groups. Extreme drowsiness was reported by 12 (10%) of 116 patients assigned to receive oxygen compared with 14 (13%) of 108 patients assigned to receive room air. Two (2%) patients in the oxygen group reported extreme symptoms of nasal irritation compared with seven (6%) in the room air group. One patient reported an extremely troublesome nose bleed (oxygen group). Interpretation Since oxygen delivered by a nasal cannula provides no additional symptomatic benefit for relief of refractory dyspnoea in patients with life-limiting illness compared with room air, less burdensome strategies should be considered after brief assessment of the effect of oxygen therapy on the individual patient. Funding US National Institutes of Health, Australian National Health and Medical Research Council, Duke Institute for Care at the End of Life, and Doris Duke Charitable Foundation.
IMPORTANCE: Family caregivers of patients with chronic critical illness experience significant psychological distress. OBJECTIVE: To determine whether family informational and emotional support ...meetings led by palliative care clinicians improve family anxiety and depression. DESIGN, SETTING, AND PARTICIPANTS: A multicenter randomized clinical trial conducted from October 2010 through November 2014 in 4 medical intensive care units (ICUs). Adult patients (aged ≥21 years) requiring 7 days of mechanical ventilation were randomized and their family surrogate decision makers were enrolled in the study. Observers were blinded to group allocation for the measurement of the primary outcomes. INTERVENTIONS: At least 2 structured family meetings led by palliative care specialists and provision of an informational brochure (intervention) compared with provision of an informational brochure and routine family meetings conducted by ICU teams (control). There were 130 patients with 184 family surrogate decision makers in the intervention group and 126 patients with 181 family surrogate decision makers in the control group. MAIN OUTCOMES AND MEASURES: The primary outcome was Hospital Anxiety and Depression Scale symptom score (HADS; score range, 0 best to 42 worst; minimal clinically important difference, 1.5) obtained during 3-month follow-up interviews with the surrogate decision makers. Secondary outcomes included posttraumatic stress disorder experienced by the family and measured by the Impact of Events Scale-Revised (IES-R; total score range, 0 best to 88 worst), discussion of patient preferences, hospital length of stay, and 90-day survival. RESULTS: Among 365 family surrogate decision makers (mean age, 51 years; 71% female), 312 completed the study. At 3 months, there was no significant difference in anxiety and depression symptoms between surrogate decision makers in the intervention group and the control group (adjusted mean HADS score, 12.2 vs 11.4, respectively; between-group difference, 0.8 95% CI, −0.9 to 2.6; P = .34). Posttraumatic stress disorder symptoms were higher in the intervention group (adjusted mean IES-R score, 25.9) compared with the control group (adjusted mean IES-R score, 21.3) (between-group difference, 4.60 95% CI, 0.01 to 9.10; P = .0495). There was no difference between groups regarding the discussion of patient preferences (intervention, 75%; control, 83%; odds ratio, 0.63 95% CI, 0.34 to 1.16; P = .14). The median number of hospital days for patients in the intervention vs the control group (19 days vs 23 days, respectively; between-group difference, −4 days 95% CI, −6 to 3 days; P = .51) and 90-day survival (hazard ratio, 0.95 95% CI, 0.65 to 1.38, P = .96) were not significantly different. CONCLUSIONS AND RELEVANCE: Among families of patients with chronic critical illness, the use of palliative care–led informational and emotional support meetings compared with usual care did not reduce anxiety or depression symptoms and may have increased posttraumatic stress disorder symptoms. These findings do not support routine or mandatory palliative care–led discussion of goals of care for all families of patients with chronic critical illness. TRIAL REGISTRATION: clinicaltrials.gov Identifier: NCT01230099
Recruitment of health research participants through social media is becoming more common. In the United States, 80% of adults use at least one social media platform. Social media platforms may allow ...researchers to reach potential participants efficiently. However, online research methods may be associated with unique threats to sample validity and data integrity. Limited research has described issues of data quality and authenticity associated with the recruitment of health research participants through social media, and sources of low-quality and fraudulent data in this context are poorly understood.
The goal of the research was to describe and explain threats to sample validity and data integrity following recruitment of health research participants through social media and summarize recommended strategies to mitigate these threats. Our experience designing and implementing a research study using social media recruitment and online data collection serves as a case study.
Using published strategies to preserve data integrity, we recruited participants to complete an online survey through the social media platforms Twitter and Facebook. Participants were to receive $15 upon survey completion. Prior to manually issuing remuneration, we reviewed completed surveys for indicators of fraudulent or low-quality data. Indicators attributable to respondent error were labeled suspicious, while those suggesting misrepresentation were labeled fraudulent. We planned to remove cases with 1 fraudulent indicator or at least 3 suspicious indicators.
Within 7 hours of survey activation, we received 271 completed surveys. We classified 94.5% (256/271) of cases as fraudulent and 5.5% (15/271) as suspicious. In total, 86.7% (235/271) provided inconsistent responses to verifiable items and 16.2% (44/271) exhibited evidence of bot automation. Of the fraudulent cases, 53.9% (138/256) provided a duplicate or unusual response to one or more open-ended items and 52.0% (133/256) exhibited evidence of inattention.
Research findings from several disciplines suggest studies in which research participants are recruited through social media are susceptible to data quality issues. Opportunistic individuals who use virtual private servers to fraudulently complete research surveys for profit may contribute to low-quality data. Strategies to preserve data integrity following research participant recruitment through social media are limited. Development and testing of novel strategies to prevent and detect fraud is a research priority.
Cultural beliefs and values are thought to account for differences between African Americans and whites in the use of advance directives and beliefs about hospice care, but few data clarify which ...beliefs and values explain these differences. Two hundred five adults aged 65 and older who received primary care in the Duke University Health System were surveyed. The survey included five scales: Hospice Beliefs and Attitudes, Preferences for Care, Spirituality, Healthcare System Distrust, and Beliefs About Dying and Advance Care Planning. African Americans were less likely than white subjects to have completed an advance directive (35.5% vs 67.4%, P<.001) and had less favorable beliefs about hospice care (Hospice Beliefs and Attitudes Scale score, P<.001). African Americans were more likely to express discomfort discussing death, want aggressive care at the end of life, have spiritual beliefs that conflict with the goals of palliative care, and distrust the healthcare system. In multivariate analyses, none of these factors alone completely explained racial differences in possession of an advance directive or beliefs about hospice care, but when all of these factors were combined, race was no longer a significant predictor of either of the two outcomes. These findings suggest that ethnicity is a marker of common cultural beliefs and values that, in combination, influence decision‐making at the end of life. This study has implications for the design of healthcare delivery models and programs that provide culturally sensitive end‐of‐life care to a growing population of ethnically diverse older adults.
Correspondence to Dr Alec Petersen, Internal Medicine Residency Program, Brigham and Women's Hospital, Boston, MA 02115, USA; AWPETERSEN@BWH.HARVARD.EDU Introduction Every patient admitted to the ...hospital, scheduled for a procedure or facing a life-limiting illness potentially confronts a decision about cardiopulmonary resuscitation (CPR). Despite their importance and frequency, resuscitation or code status discussions (CSD) are often not included in broader serious illness conversations (SIC) or ignored altogether.1 CSDs for patients with serious, life-limiting illness should be incorporated into comprehensive serious illness care delivery, which includes discussions about advance care planning and goals of care at every stage of illness; ideally, for most patients, this will occur early in the disease trajectory.1 Yet, even when conversations occur, health systems frequently do not capture code status in an accurate, retrievable, timely and consistent manner.2 Failing to understand, document and act on patients’ preferences may lead to harm, like other medical errors. CPR, cardiopulmonary resuscitation; CSD, code status discussion; EHR, electronic health record; ICU, intensive care unit; MOLST, Medical Orders for Life-Sustaining Treatment. Only one in four patients with incurable metastatic cancer were found to have SICs in the outpatient setting.11 Delayed SICs in patients have significant consequences: unwanted CPR, increased hospital length of stay and unnecessary health costs.3 Of course, inpatient clinicians must revisit patients’ goals and preferences at admission and other key transition points.