The Uncertain Science of Predicting Death

Maintaining High-Touch in High-Tech Digital Health Monitoring and Multi-Omics Prognostication: Ethical, Equity, and Societal Considerations in Precision Health for Palliative Care

Advances in digital health, systems biology, environmental monitoring, and artificial intelligence (AI) continue to revolutionize health care, ushering a precision health future. More than disease treatment and prevention, precision health aims at maintaining good health throughout the lifespan. However, how can precision health impact care for people with a terminal or life-limiting condition? We examine here the ethical, equity, and societal/relational implications of two precision health modalities, (1) integrated systems biology/multi-omics analysis for disease prognostication and (2) digital health technologies for health status monitoring and communication. We focus on three main ethical and societal considerations: benefits and risks associated with integration of these modalities into the palliative care system; inclusion of underrepresented and marginalized groups in technology development and deployment; and the impact of high-tech modalities on palliative care's highly personalized and "high-touch" practice. We conclude with 10 recommendations for ensuring that precision health technologies, such as multi-omics prognostication and digital health monitoring, for palliative care are developed, tested, and implemented ethically, inclusively, and equitably.

Developing an Interpretable Machine Learning Model to Predict in-Hospital Mortality in Sepsis Patients: A Retrospective Temporal Validation Study

BACKGROUND

Risk stratification plays an essential role in the decision making for sepsis management, as existing approaches can hardly satisfy the need to assess this heterogeneous population. We aimed to develop and validate a machine learning model to predict in-hospital mortality in critically ill patients with sepsis.

METHODS

Adult patients fulfilling the definition of Sepsis-3 were included at a large tertiary medical center. Relevant clinical features were extracted within the first 24 h in ICU, re-classified into different genres, and utilized for model development under three strategies: "Basic + Lab", "Basic + Intervention", and "Whole" feature sets. Extreme gradient boosting (XGBoost) was compared with logistic regression (LR) and established severity scores. Temporal validation was conducted using admissions from 2017 to 2019.

RESULTS

The final cohort included 24,272 patients, of which 4013 patients formed the test cohort for temporal validation. The trained and fine-tuned XGBoost model with the whole feature set showed the best discriminatory ability in the test cohort with AUROC as 0.85, significantly higher than the XGBoost "Basic + Lab" model (0.83), the LR "Whole" model (0.82), SOFA (0.63), SAPS-II (0.73), and LODS score (0.74). The performance in varying subgroups remained robust, and predictors, such as increased urine output and supplemental oxygen therapy, were crucially correlated with improved survival when interpretability was explored.

CONCLUSIONS

We developed and validated a novel XGBoost-based model and demonstrated significantly improved performance to LR and other scores in predicting the mortality risks of sepsis patients in the hospital using features in the first 24 h.

Development of Machine Learning Algorithms Incorporating Electronic Health Record Data, Patient-Reported Outcomes, or Both to Predict Mortality for Outpatients With Cancer

PURPOSE

Machine learning (ML) algorithms that incorporate routinely collected patient-reported outcomes (PROs) alongside electronic health record (EHR) variables may improve prediction of short-term mortality and facilitate earlier supportive and palliative care for patients with cancer.

METHODS

We trained and validated two-phase ML algorithms that incorporated standard PRO assessments alongside approximately 200 routinely collected EHR variables, among patients with medical oncology encounters at a tertiary academic oncology and a community oncology practice.

RESULTS

Among 12,350 patients, 5,870 (47.5%) completed PRO assessments. Compared with EHR- and PRO-only algorithms, the EHR + PRO model improved predictive performance in both tertiary oncology (EHR + PRO v EHR v PRO: area under the curve [AUC] 0.86 [0.85-0.87] v 0.82 [0.81-0.83] v 0.74 [0.74-0.74]) and community oncology (area under the curve 0.89 [0.88-0.90] v 0.86 [0.85-0.88] v 0.77 [0.76-0.79]) practices.

CONCLUSION

Routinely collected PROs contain added prognostic information not captured by an EHR-based ML mortality risk algorithm. Augmenting an EHR-based algorithm with PROs resulted in a more accurate and clinically relevant model, which can facilitate earlier and targeted supportive care for patients with cancer.

A multi-method evaluation of tobacco treatment trial recruitment messages for current smokers recently diagnosed with cancer: A pilot factorial randomized control trial (Preprint)

Background

A cancer diagnosis can catalyze motivation to quit smoking. Tobacco treatment trials offer cessation resources but have low accrual rates. Digital outreach may improve accrual, but knowledge of how best to recruit smokers with recent diagnoses is limited.

Objective

This study aims to identify the message frames that were most effective in promoting intent to talk to a physician about participating in a tobacco treatment trial for smokers recently diagnosed with cancer.

Methods

From February to April 2019, current smokers diagnosed within the past 24 months were recruited from a national web-based panel for a multimethod pilot randomized trial (N=99). Participants were randomized to a 2×3 plus control factorial design that tested 3 unique message frames: proximal versus distal threats of smoking, costs of continued smoking versus benefits of quitting, and gains of participating versus losses of not participating in a tobacco treatment trial. The primary outcome was intent to talk to a physician about participating in a tobacco treatment trial. In phase 1, the main effect within each message factor level was examined using ANOVA and compared with the control condition. Other message evaluation and effectiveness measures were collected and explored in a multivariable model predicting intent to talk to a physician. In phase 2, open-text evaluations of the messages were analyzed using natural language processing software (Leximancer) to generate a thematic concept map and Linguistic Inquiry Word Count to identify and compare the prevalence of linguistic markers among message factors.

Results

Of the 99 participants, 76 (77%) completed the intervention. Participants who received the cost of continued smoking frame were significantly more likely to intend to talk to their physician about participating in a tobacco treatment trial than those who received the benefits of the quitting frame (mean costs 5.13, SD 1.70 vs mean benefits 4.23, SD 1.86; P=.04). Participants who received the proximal risks of continued smoking frame were significantly more likely to seek more information about participating (mean distal 4.83, SD 1.61 vs mean proximal 5.55, SD 1.15; P=.04), and those who received the losses of not participating frame reported significantly improved perceptions of smoking cessation research (mean gain 3.98, SD 0.83 vs mean loss 4.38, SD 0.78; P=.01). Male participants (P=.006) and those with greater message relevancy (P=.001) were significantly more likely to intend to talk to their physician. Participants’ perceptions of their smoking habits, as well as their motivation to quit smoking, were prevalent themes in the open-text data. Differences in the percentages of affective words across message frames were identified.

Conclusions

Multimethod approaches are needed to develop evidence-based recruitment messages for patients recently diagnosed with cancer. Future tobacco treatment trials should evaluate the effectiveness of different message frames on smoker enrollment rates.

Trial Registration

Clinicaltrials.gov NCT05471284; https://clinicaltrials.gov/ct2/show/NCT05471284