Context: Clinicians deciding whether to refer a patient or family to specialty palliative care report facing high levels of uncertainty. Most research on medical uncertainty has focused on prognostic uncertainty. As part of a pediatric palliative referral intervention for oncology teams we explored how uncertainty might influence palliative care referrals.
Objectives: To describe distinct meanings of the term “uncertainty” that emerged during the qualitative evaluation of the development and implementation of an intervention to help oncologists overcome barriers to palliative care referrals.
Methods: We conducted a phenomenological qualitative analysis of “uncertainty” as experienced and described by interdisciplinary pediatric oncology team members in discussions, group activities and semistructured interviews regarding the introduction of palliative care.
Results: We found that clinicians caring for patients with advanced cancer confront seven broad categories of uncertainty: prognostic, informational, individual, communication, relational, collegial, and inter-institutional. Each of these kinds of uncertainty can contribute to delays in referring patients to palliative care.
Conclusion: Various types of uncertainty arise in the care of pediatric patients with advanced cancer. To manage these forms of uncertainty, providers need to develop strategies and techniques to handle professionally challenging situations, communicate bad news, manage difficult interactions with families and colleagues, and collaborate with other organizations.
On March 28, 2020, the Office of Civil Rights at the Department of Health and Human Services (HHS) opened investigations into recently released critical care crisis triage protocols. Disability rights advocates are urging Congress to prohibit crisis triage based on “anticipated or demonstrated resource-intensity needs, the relative survival probabilities of patients deemed likely to benefit from medical treatment, and assessments of pre- or post-treatment quality of life.”
Objectives: Research suggests that clinicians are not very accurate at prognosticating in palliative care. The ‘horizon effect’ suggests that accuracy ought to be better when the survival of patients is shorter. The aim of this study was to determine the accuracy of specialist palliative care clinicians at identifying which patients are likely to die within 72 hours.
Design In a secondary data analysis of a prospective observational study, specialist palliative care doctors and nurses (in a hospice and a hospital palliative care team) provided survival predictions (yes/no/uncertain) about which patients would die within 72 hours.
Results: Survival predictions were obtained for 49 patients. A prediction from a nurse was obtained for 37/49 patients. A prediction from a doctor was obtained for 46/49 patients. In total, 23 (47%)/49 patients actually died within 72 hours of assessment. Nurses accurately predicted the outcome in 27 (73%)/37 cases. Doctors accurately predicted the outcome in 30 (65%)/46 cases. When comparing predictions given on the same patients (27 [55%]/49), nurses were slightly better at recognising imminent death than doctors (positive predictive value (the proportion of patients who died when the clinician predicted death)=79% vs 60%, respectively). The difference in c-statistics (nurses 0.82 vs doctors 0.63) was not significant (p=0.13).
Conclusion: Even when patients are in the terminal phase and close to death, clinicians are not very good at predicting how much longer they will survive. Further research is warranted to improve prognostication in this population.
We developed a predictive score system for 30-day mortality after palliative radiotherapy by using predictors from routine electronic medical record. Patients with metastatic cancer receiving first course palliative radiotherapy from 1 July, 2007 to 31 December, 2017 were identified. 30-day mortality odds ratios and probabilities of the death predictive score were obtained using multivariable logistic regression model. Overall, 5,795 patients participated. Median follow-up was 39.6 months (range, 24.5–69.3) for all surviving patients. 5,290 patients died over a median 110 days, of whom 995 (17.2%) died within 30 days of radiotherapy commencement. The most important mortality predictors were primary lung cancer (odds ratio: 1.73, 95% confidence interval: 1.47–2.04) and log peripheral blood neutrophil lymphocyte ratio (odds ratio: 1.71, 95% confidence interval: 1.52–1.92). The developed predictive scoring system had 10 predictor variables and 20 points. The cross-validated area under curve was 0.81 (95% confidence interval: 0.79–0.82). The calibration suggested a reasonably good fit for the model (likelihood-ratio statistic: 2.81, P = 0.094), providing an accurate prediction for almost all 30-day mortality probabilities. The predictive scoring system accurately predicted 30-day mortality among patients with stage IV cancer. Oncologists may use this to tailor palliative therapy for patients.
Background: When patients are likely to die in the coming hours or days, families often want prognostic information. Prognostic uncertainty and a lack of end-of-life communication training make these conversations challenging.
Aim: The objective of this study is to understand how clinicians and the relatives/friends of patients at the very end of life manage uncertainty and reference time in prognostic conversations.
Design: Conversation analysis of audio-recorded conversations between clinicians and the relatives/friends of hospice inpatients.
Setting/participants: Experienced palliative care clinicians and relatives/friends of imminently dying hospice inpatients. Twenty-three recorded conversations involved prognostic talk and were included in the analysis.
Results: Requests for prognostic information were initiated by families in the majority of conversations. Clinicians responded using categorical time references such as ‘days’, allowing the provision of prognostic estimates without giving a precise time. Explicit terms such as ‘dying’ were rare during prognostic discussions. Instead, references to time were understood as relating to prognosis. Relatives displayed their awareness of prognostic uncertainty when requesting prognostic information, providing clinicians with ‘permission’ to be uncertain. In response, clinicians often stated their uncertainty explicitly, but presented evidence for their prognostic estimates, based on changes to the patient’s function previously discussed with the family.
Conclusion: Prognostic uncertainty was managed collaboratively by clinicians and families. Clinicians were able to provide prognostic estimates while being honest about the related uncertainty, in part because relatives displayed their awareness of uncertainty within their requests. The conversation analytic method identified contributions of both clinicians and families, and identified strategies based on real interactions, which could inform communication training.
Objective: To develop a proposal for a 2-year mortality prognostic approach for patients with advanced chronic conditions based on the palliative care need (PCN) items of the
NECesidades PALiativas (NECPAL) CCOMS-ICO V.3.1 2017 tool.
Methods: A phase 1 study using three components based on the NECPAL items: (1) a rapid review of systematic reviews (SRs) on prognostic factors of mortality in patients with advanced chronic diseases and PCNs; (2) a clinician and statistician experts' consensus based on the Delphi technique on the selection of mortality prognostic factors; and (3) a panel meeting to discuss the findings of components (1) and (2).
Results: Twenty SRs were included in a rapid review, and 50% were considered of moderate quality. Despite methodological issues, nutritional and functional decline, severe and refractory dyspnoea, multimorbidity, use of resources and specific disease indicators were found to be potentially prognostic variables for mortality across four clinical groups and end-of-life (EoL) trajectories: cancer, dementia and neurologic diseases, chronic organ failure and frailty. Experts’ consensus added ‘needs’ identified by health professionals. However, clinicians were less able to discriminate which NECPAL items were more reliable for a ‘general’ model. A retrospective cohort study was designed to evaluate this proposal in phase 2.
Conclusions: We identified several parameters with prognostic value and linked them to the tool’s utility to timely identify PCNs of patients with advanced chronic conditions in all settings of care. Initial results show this is a clinical and feasible tool, that will help with clinical pragmatic decision-making and to define services.
Context: Universal screening to identify vulnerable patients who may receive limited benefits from life-sustaining treatments can facilitate palliative care in dialysis populations.
Objectives: We aimed to develop prediction models for 1-year mortality in peritoneal dialysis patients.
Methods: This prospective cohort study included 401 adult Taiwanese prevalent peritoneal dialysis patients (average age 56.2 ± 14 years). In addition to obtaining clinical characteristics and laboratory data, the primary care nurses evaluated the “surprise question” and “palliative care screening tool” for each patient in March 2015. Multivariate logistic regression models were conducted to predict the primary outcome of 1-year all-cause mortality.
Results: There were 34 (8.5%) patients who died during the first year of follow-up. Patients allocated to the “not surprised” group according to the surprise question and those who received a score = 4 on the palliative care screening tool had increased odds of death [odds ratio 24.68 (95% CI 10.66 - 57.13) and 12.18 (95% CI 5.66 - 26.21), respectively]. We also developed a clinical risk model for 1-year mortality that included sex, dialysis vintage, coronary artery disease, malignancy, normalized protein nitrogen appearance, white blood cell count, and serum albumin and sodium levels. Integrating the surprise question, palliative care screening tool, and clinical risk model exhibited good discrimination with an area under the receiver operating characteristic curve of 0.95. Kaplan-Meier analysis showed worse survival in high risk patients predicted by the integrated model (log-rank P<.001).
Conclusion: screening with the use of the integrated measurement can identify high-risk peritoneal dialysis patients. This approach may facilitate palliative care interventions for at-risk the subpopulations.
Background: General practitioners’ (GPs) play a central role in facilitating end-of-life discussions with older patients nearing the end-of-life. However, prognostic uncertainty of time to death is one important barrier to initiation of these discussions.
Objective: To explore GPs’ perceptions of the feasibility and acceptability of a risk prediction checklist to identify older patients in their last 12 months of life and describe perceived barriers and facilitators for implementing end-of-life planning.
Methods: Qualitative, semi-structured interviews were conducted with 15 GPs practising in metropolitan locations in New South Wales and Queensland between May and June 2019. Data were analysed thematically.
Results: Eight themes emerged: accessibility and implementation of the checklist, uncertainty around checklist’s accuracy and usefulness, time of the checklist, checklist as a potential prompt for end-of-life conversations, end-of-life conversations not an easy topic, end-of-life conversation requires time and effort, uncertainty in identifying end-of-life patients and limited community literacy on end-of-life. Most participants welcomed a risk prediction checklist in routine practice if assured of its accuracy in identifying which patients were nearing end-of-life.
Conclusions: Most participating GPs saw the value in risk assessment and end-of-life planning. Many emphasized the need for appropriate support, tools and funding for prognostic screening and end-of-life planning for this to become routine in general practice. Well validated risk prediction tools are needed to increase clinician confidence in identifying risk of death to support end-of-life care planning.
OBJECTIVE: The "surprise question" ("Would you be surprised if this patient died in the next year?") has been shown to be predictive of 12-month mortality in multiple populations, but has not been studied in gynecologic oncology (GO) patients. We sought to evaluate the prognostic performance of the surprise question in GO patients among physician and non-physician providers.
METHODS: GO providers at two tertiary care centers were asked the surprise question about a cohort of their patients undergoing chemotherapy or radiation. Demographic and clinical information was chart abstracted. Mortality data were collected at one year; relative risk of death at one year based on response to the surprise question was then calculated.
RESULTS: 32 providers (12 MDs, 7 APPs, 13 RNs) provided 942 surprise question assessments for 358 patients. Fifty-seven % had ovarian cancer and 54% had recurrent disease. Eighty-three (24%) patients died within a year. Patients whose physician answered "No" to the surprise question had a 43% one-year mortality (compared to 10% for "Yes"). Overall RR of 12-month mortality for "No" was 3.76 (95% CI 2.75-5.48); this association remained significant in all provider types. Among statistically significant predictors of 12-month mortality (including recurrent disease and >2 prior lines of chemotherapy), the surprise question had the highest RR.
CONCLUSIONS: The surprise question is a simple, one question tool that effectively identifies GO patients increased risk of 12-month mortality. The surprise question could be used to identify patients for early referral to palliative care and initiation advance care planning.
Prediction of short-term mortality in elderly patients with heart failure (HF) would be useful for clinicians when discussing HF management or palliative care.
A prospective multicenter cohort study was conducted between July 2014 and July 2018. A total of 504 consecutive elderly patients (age = 75 years) with HF (mean age 85 years, 50% women) were enrolled. We used a multiple logistic regression analysis with stepwise variable selection to select predictive variables and to determine weighted point scores. After analysis, the following variables predicted short-term mortality and comprised the risk score: previous HF admission (3 points), New York Heart Association III or IV (2 points), body mass index < 17.7 kg/m2 (4 points), serum albumin < 3.5 g/dL (9 points), and left ventricular ejection fraction < 50% (2 points). The c-statistic was 0.820. We compared mortality in low-risk (0-6 points, n = 188), intermediate-risk (7-13 points, n = 241), and high-risk (14-20 points, n = 75) groups. A total of 43 (8.5%) patients died within 6 months after discharge. Mortality was significantly higher in groups with higher scores (low-risk group, 0.5%; intermediate-risk group, 9.1%; high-risk group, 26.7%; P < 0.001).
We developed a predictive model for 6-month mortality in elderly patients with HF. This risk score could be useful when discussing advanced HF therapies, palliative care, or hospice referral with patients.
Prognosis after severe brain injury is highly uncertain, and decisions to withhold or withdraw life-sustaining treatment are often made prematurely. These decisions are often driven by a desire to avoid a situation where the patient becomes 'trapped' in a condition they would find unacceptable. However, this means that a proportion of patients who would have gone on to make a good recovery, are allowed to die. I propose a shift in practice towards the routine provision of aggressive care, even in cases where the probability of survival and acceptable recovery is thought to be low. In conjunction with this shift, I argue in favour of a presumption towards withdrawing life-sustaining treatment, including artificial nutrition and hydration, when it becomes clear that a patient will not recover to a level that would be acceptable to them. I then respond to three potential objections to this proposal.
Survival estimates are very important to patients with terminal cancer. The C-reactive protein (CRP)/albumin ratio is associated with cancer outcomes. However, few studies have investigated the dose-response association in terminal cancer patients. Therefore, we aimed to evaluate the association between the CRP/albumin ratio and mortality in terminal cancer patients using a longitudinal analysis. We retrospectively investigated the electronic medical records of 435 inpatients with terminal cancer admitted to the palliative care unit of Yeouido St. Mary's Hospital between October 8, 2015, and January 17, 2018. In total, 382 patients with terminal cancer were enrolled in the study. The serum CRP/albumin ratio measured at admission had a linear dose-response relationship with the risk of death among the terminal cancer patients (P for linearity = .011). The multivariate analyses showed that the CRP/albumin ratio was an independent prognostic factor (Model 1, CRP/albumin ratio >48.53 × 10-4: HR = 2.68, 95% CI = 1.82–3.93; Model 2, tertile 2: HR = 1.91, 95% CI = 1.31–2.82 and tertile 3: HR = 3.66, 95% CI = 2.24–5.97). The relationship between a high CRP/albumin ratio and poor survival was a flat L-shape for survival time with an inflection point at approximately 15 days, while the relationship was not significant in terminal cancer patients who survived beyond 30 days. This study demonstrated that high CRP/albumin ratios are significantly and independently associated with the short-term survival prognosis of terminal cancer patients within 30 days.
As artificial intelligence (AI) spreads across clinical specialties, its potential to revolutionise health care at the individual and population levels has placed it alongside genomics as one of the frontiers in medicine. The promise that AI could help health systems and clinicians optimise patient care in core domains of diagnosis, prognosis, and treatment drives widespread interest and investment.
BACKGROUND: Predicting death in a cohort of clinically diverse, multi-condition hospitalized patients is difficult. This frequently hinders timely serious illness care conversations. Prognostic models that can determine 6-month death risk at the time of hospital admission can improve access to serious illness care conversations.
OBJECTIVE: The objective is to determine if the demographic, vital sign, and laboratory data from the first 48 h of a hospitalization can be used to accurately quantify 6-month mortality risk.
DESIGN: This is a retrospective study using electronic medical record data linked with the state death registry.
PARTICIPANTS: Participants were 158,323 hospitalized patients within a 6-hospital network over a 6-year period.
MAIN MEASURES: Main measures are the following: the first set of vital signs, complete blood count, basic and complete metabolic panel, serum lactate, pro-BNP, troponin-I, INR, aPTT, demographic information, and associated ICD codes. The outcome of interest was death within 6 months.
KEY RESULTS: Model performance was measured on the validation dataset. A random forest model-mini serious illness algorithm-used 8 variables from the initial 48 h of hospitalization and predicted death within 6 months with an AUC of 0.92 (0.91-0.93). Red cell distribution width was the most important prognostic variable. min-SIA (mini serious illness algorithm) was very well calibrated and estimated the probability of death to within 10% of the actual value. The discriminative ability of the min-SIA was significantly better than historical estimates of clinician performance.
CONCLUSION: min-SIA algorithm can identify patients at high risk of 6-month mortality at the time of hospital admission. It can be used to improved access to timely, serious illness care conversations in high-risk patients.
BACKGROUND: Accurate awareness of the prognosis is an important factor in the treatment decision of patients with advanced cancer; however, prognostic disclosure is still subject to debate because it can reduce patient's satisfaction and increase depression.
AIM: The purpose of this study is to assess whether patients' prognostic awareness is associated with decreased quality of life (QoL) or increased depressive mood in patients with advanced cancer.
DESIGN AND PARTICIPANTS: In this cohort study, 386 patients with advanced cancer were recruited across 3 periods from December 2016 to August 2018. The outcome of this study was a change in QoL and depression according to the patients' prognostic awareness at baseline, 3 months, and 6 months.
RESULTS: This study found significant differences in changes of QoL based on patients' prognostic awareness. From baseline to 3 months, emotional functioning (P = .039), pain (P = .042), existential well-being (P = .025), and social support (P = .038) subscale scores improved significantly more in those with lack of prognostic awareness. Over 6 months, the group without prognostic awareness improved significantly in terms of physical functioning (P = .037), emotional functioning (P = .002), nausea/vomiting (P = .048), and constipation (P = .039) subscale scores and existential well-being scores (P = .025). No significant difference between the groups was found in terms of depression.
CONCLUSION: Accurate prognostic awareness may pose harm and may provide no additional benefits in terms of QoL and mood among patients with advanced cancer for a short period of time.
Context: Dyspnea is one of the most distressing symptoms for terminally ill cancer patients and a predictor of poor prognosis. Identification of simple clinical signs, such as heart rate, indicating clinical course of each patient is of value.
Objectives: To explore the potential association between heart rate and reversibility of the symptom, treatment response to palliative intervention, and survival in terminally ill cancer patients with dyspnea at rest.
Methods: This is a secondary analysis of a multicenter prospective cohort study of patients with advanced cancer to validate multiple prognostic tools. In the patients with dyspnea at rest at the baseline, we examined a potential association between heart rate and the reversibility of dyspnea and refractoriness to palliative treatment using logistic regression analysis. Survivals were compared using the Cox proportional hazards model among four groups with different levels of the heart rate (=74, 75–84, 85–97, and =98).
Results: A total of 2298 patients were enrolled, and 418 patients (18%) had dyspnea at rest. Reversibility of dyspnea was significantly higher in the patients with lower heart rate (P for trend = 0.008), and the refractoriness to palliative treatment tended to be higher in the patients with higher heart rate (P for trend = 0.101). The median survival for each heart rate quartile groups was significantly higher in the lower heart rate group (24 vs. 21 vs. 14 vs. 9 days; heart rate =74, 75–84, 85–97, and =98, respectively; log-rank P < 0.001).
Conclusion: Heart rate may help clinicians to make the prediction of the patient's clinical course more accurate.
Context: Increasing emphasis on patient-centered care has led to highlighted importance of shared decision making, which better aligns medical decisions with patient care preferences. Effective shared decision making in metastatic breast cancer (MBC) treatment requires prognostic understanding, without which patients may receive treatment inconsistent with personal preferences.
Objectives: To assess MBC patient and provider perspectives on the role of prognostic information in treatment decision making.
Methods: We conducted semi-structured interviews with MBC patients and community oncologists and separate focus groups involving lay navigators, nurses, and academic oncologists. Qualitative analysis utilized a content analysis approach that included a constant comparative method to generate themes.
Results: Of 20 interviewed patients with MBC, 30% were African American. Academic oncologists were mostly women (60%), community oncologists were all Caucasian, and nurses were all women and 28% African American. Lay navigators were all African American and predominately women (86%). Five emergent themes were identified. (1) Most patients wanted prognostic information but differed in when they wanted to have this conversation, (2) Emotional distress and discomfort was a critical reason for not discussing prognosis, (3) Religious beliefs shaped preferences for prognostic information, (4) Health care professionals differed on prognostic information delivery timing, and (5) Providers acknowledged that an individualized approach taking into account patient values and preferences would be beneficial.
Conclusion: Most MBC patients wanted prognostic information, yet varied in when they wanted this information. Understanding why patients want limited or unrestricted prognostic information can inform oncologists' efforts toward shared decision making.
Study Design: Retrospective study.
Objective: The purpose of the study was to examine survival after surgery for a metastatic spinal tumor using prognostic factors in the new Katagiri score.
Summary of Background Data: surgery for spinal metastasis can improve quality of life and facilitate treatment of the primary cancer. However, choice of therapy requires identification of prognostic factors for survival, and these may change over time due to treatment advances. The new Katagiri score for the prognosis of skeletal metastasis includes classification of the primary tumor site and the effects of chemotherapy and hormonal therapy.
Methods: The subjects were 201 patients (127 males, 74 females) who underwent surgery for spinal metastases at 6 facilities in the Nagoya Spine Group. Age at surgery, gender, follow-up, metastatic spine level, primary cancer, new Katagiri score (including primary site, visceral metastasis, laboratory data, performance status (PS), and chemotherapy) and survival were obtained from a prospectively maintained database.
Results: Posterior decompression (n = 29) and posterior decompression and fixation with instrumentation (n = 182) were performed at a mean age of 65.9 (range, 16-85) years. Metastasis was present in the cervical (n = 19, 10%), thoracic (n = 155, 77%), and lumbar (n = 26, 13%) spine, and sacrum (n = 1, 1%). In multivariate analysis, moderate growth (HR 2.95, 95% CI, 1.27–7.89, P < 0.01) and rapid growth (HR 4.71, 95% CI, 2.78–12.31, P < 0.01) at the primary site; nodular metastasis (HR 1.53, 95% CI, 1.07–3.85, P < 0.01) and disseminated metastasis (HR 2.94, 95% CI, 1.33–5.42, P < 0.01); and critical laboratory data (HR 3.15, 95% CI, 2.06–8.36, P < 0.01) and poor PS (HR 2.83, 95% CI, 1.67–4.77, p < 0.01) were significantly associated with poor survival.
Conclusion: accurate prognostic factors are important in deciding the treatment strategy in patients with spinal metastasis, and our identification of these factors may be useful for these patients.
Level of Evidence: 3
BACKGROUND: News of cancer progression is critical to setting accurate prognostic understanding, which guides patients' treatment decision making. This study examines whether religious belief in miracles modifies the effect of receiving news of cancer progression on change in prognostic understanding.
METHODS: In a multisite, prospective cohort study, 158 patients with advanced cancer, whom oncologists expected to die within 6 months, were assessed before and after the visit at which scan results were discussed. Before the visit, religious belief in miracles was assessed; after the visit, patients indicated what scan results they had received (cancer was worse vs cancer was stable, better, or other). Before and after the visit, prognostic understanding was assessed, and a change score was computed.
RESULTS: Approximately 78% of the participants (n = 123) reported at least some belief in miracles, with almost half (n = 73) endorsing the strongest possible belief. A significant interaction effect emerged between receiving news of cancer progression and belief in miracles in predicting change in prognostic understanding (b = -0.18, P = .04). Receiving news of cancer progression was associated with improvement in the accuracy of prognostic understanding among patients with weak belief in miracles (b = 0.67, P = .007); however, among patients with moderate to strong belief in miracles, news of cancer progression was unrelated to change in prognostic understanding (b = 0.08, P = .64).
CONCLUSIONS: Religious belief in miracles was highly prevalent and diminished the impact of receiving news of cancer progression on prognostic understanding. Assessing patients' beliefs in miracles may help to optimize the effectiveness of "bad news" scan result discussions.