An Acoustical and Lexical Machine-Learning Pipeline to Identify Connectional Silences.

Context: Developing scalable methods for conversation analytics is essential for health care communication science and quality improvement. Purpose: To assess the feasibility of automating the identification of a conversational feature, Connectional Silence, which is associated with important patient outcomes. Methods: Using audio recordings from the Palliative Care Communication Research Initiative cohort study, we develop and test an automated measurement pipeline comprising three machine-learning (ML) tools-a random forest algorithm and a custom convolutional neural network that operate in parallel on audio recordings, and subsequently a natural language processing algorithm that uses brief excerpts of automated speech-to-text transcripts. Results: Our ML pipeline identified Connectional Silence with an overall sensitivity of 84% and specificity of 92%. For Emotional and Invitational subtypes, we observed sensitivities of 68% and 67%, and specificities of 95% and 97%, respectively. Conclusion: These findings support the capacity for coordinated and complementary ML methods to fully automate the identification of Connectional Silence in natural hospital-based clinical conversations.

A Formative Mixed-Methods Study of Emotional Responsiveness in Telepalliative Care.

Background: It is unknown whether telemedicine-delivered palliative care (tele-PC) supports emotionally responsive patient-clinician interactions. Objectives: We conducted a mixed-methods formative study at two academic medical centers in rural U.S. states to explore the acceptability, feasibility, and emotional responsiveness of tele-PC. Design: We assessed clinicians' emotional responsiveness through questionnaires, qualitative interviews, and video coding. Results: We completed 11 tele-PC consultations. Mean age was 71 years, 30% did not complete high school, 55% experienced at least moderate financial insecurity, and 2/3 rated their overall health poorly. All patients rated tele-PC as equal to, or better than, in-person PC at providing emotional support. There was a tendency toward higher positive and lower negative emotions following the consultation. Video coding identified 114 instances of patients expressing emotions, and clinicians detected and responded to 98% of these events. Conclusion: Tele-PC appears to support emotionally responsive patient-clinician interactions. A mixed-methods approach to evaluating tele-PC yields useful, complementary insights.

Evidence for the early emergence of piperaquine-resistant Plasmodium falciparum malaria and modeling strategies to mitigate resistance.

Multidrug-resistant Plasmodium falciparum parasites have emerged in Cambodia and neighboring countries in Southeast Asia, compromising the efficacy of first-line antimalarial combinations. Dihydroartemisinin + piperaquine (PPQ) treatment failure rates have risen to as high as 50% in some areas in this region. For PPQ, resistance is driven primarily by a series of mutant alleles of the P. falciparum chloroquine resistance transporter (PfCRT). PPQ resistance was reported in China three decades earlier, but the molecular driver remained unknown. Herein, we identify a PPQ-resistant pfcrt allele (China C) from Yunnan Province, China, whose genotypic lineage is distinct from the PPQ-resistant pfcrt alleles currently observed in Cambodia. Combining gene editing and competitive growth assays, we report that PfCRT China C confers moderate PPQ resistance while re-sensitizing parasites to chloroquine (CQ) and incurring a fitness cost that manifests as a reduced rate of parasite growth. PPQ transport assays using purified PfCRT isoforms, combined with molecular dynamics simulations, highlight differences in drug transport kinetics and in this transporter's central cavity conformation between China C and the current Southeast Asian PPQ-resistant isoforms. We also report a novel computational model that incorporates empirically determined fitness landscapes at varying drug concentrations, combined with antimalarial susceptibility profiles, mutation rates, and drug pharmacokinetics. Our simulations with PPQ-resistant or -sensitive parasite lines predict that a three-day regimen of PPQ combined with CQ can effectively clear infections and prevent the evolution of PfCRT variants. This work suggests that including CQ in combination therapies could be effective in suppressing the evolution of PfCRT-mediated multidrug resistance in regions where PPQ has lost efficacy.

Epidemiology of Connectional Silence in specialist serious illness conversations.

CONTEXT: Human connection can reduce suffering and facilitate meaningful decision-making amid the often terrifying experience of hospitalization for advanced cancer. Some conversational pauses indicate human connection, but we know little about their prevalence, distribution or association with outcomes. PURPOSE: To describe the epidemiology of Connectional Silence during serious illness conversations in advanced cancer. METHODS: We audio-recorded 226 inpatient palliative care consultations at two academic centers. We identified pauses lasting 2+ seconds and distinguished Connectional Silences from other pauses, sub-categorized as either Invitational (ICS) or Emotional (ECS). We identified treatment decisional status pre-consultation from medical records and post-consultation via clinicians. Patients self-reported quality-of-life before and one day after consultation. RESULTS: Among all 6769 two-second silences, we observed 328 (4.8%) ECS and 240 (3.5%) ICS. ECS prevalence was associated with decisions favoring fewer disease-focused treatments (ORadj: 2.12; 95% CI: 1.12, 4.06). Earlier conversational ECS was associated with improved quality-of-life (p = 0.01). ICS prevalence was associated with clinicians' prognosis expectations. CONCLUSIONS: Connectional Silences during specialist serious illness conversations are associated with decision-making and improved patient quality-of-life. Further work is necessary to evaluate potential causal relationships. PRACTICE IMPLICATIONS: Pauses offer important opportunities to advance the science of human connection in serious illness decision-making.

Conversational stories & self organizing maps: Innovations for the scalable study of uncertainty in healthcare communication.

BACKGROUND: Understanding uncertainty in participatory decision-making requires scientific attention to interaction between what actually happens when patients, families and clinicians engage one another in conversation and the multi-level contexts in which these occur. Achieving this understanding will require conceptually grounded and scalable methods for use in large samples of people representing diversity in cultures, speaking and decision-making norms, and clinical situations. DISCUSSION: Here, we focus on serious illness and describe Conversational Stories as a scalable and conceptually grounded framework for characterizing uncertainty expression in these clinical contexts. Using actual conversations from a large direct-observation cohort study, we demonstrate how natural language processing and unsupervised machine learning methods can reveal underlying types of uncertainty stories in serious illness conversations. CONCLUSIONS: Conversational Storytelling offers a meaningful analytic framework for scalable computational methods to study uncertainty in healthcare conversations.

A general model of conversational dynamics and an example application in serious illness communication.

Conversation has been a primary means for the exchange of information since ancient times. Understanding patterns of information flow in conversations is a critical step in assessing and improving communication quality. In this paper, we describe COnversational DYnamics Model (CODYM) analysis, a novel approach for studying patterns of information flow in conversations. CODYMs are Markov Models that capture sequential dependencies in the lengths of speaker turns. The proposed method is automated and scalable, and preserves the privacy of the conversational participants. The primary function of CODYM analysis is to quantify and visualize patterns of information flow, concisely summarized over sequential turns from one or more conversations. Our approach is general and complements existing methods, providing a new tool for use in the analysis of any type of conversation. As an important first application, we demonstrate the model on transcribed conversations between palliative care clinicians and seriously ill patients. These conversations are dynamic and complex, taking place amidst heavy emotions, and include difficult topics such as end-of-life preferences and patient values. We use CODYMs to identify normative patterns of information flow in serious illness conversations, show how these normative patterns change over the course of the conversations, and show how they differ in conversations where the patient does or doesn't audibly express anger or fear. Potential applications of CODYMs range from assessment and training of effective healthcare communication to comparing conversational dynamics across languages, cultures, and contexts with the prospect of identifying universal similarities and unique "fingerprints" of information flow.

Epidemiology of Fear, Sadness, and Anger Expression in Palliative Care Conversations.

CONTEXT: Advancing the science of serious illness communication requires methods for measuring characteristics of conversations in large studies. Understanding which characteristics predict clinically important outcomes can help prioritize attention to scalable measure development. OBJECTIVES: To understand whether audibly recognizable expressions of distressing emotion during palliative care serious illness conversations are associated with ratings of patient experience or six-month enrollment in hospice. METHODS: We audiorecorded initial palliative care consultations involving 231 hospitalized people with advanced cancer at two large academic medical centers. We coded conversations for expressions of fear, anger, and sadness. We examined the distribution of these expressions and their association with pre/post ratings of feeling heard and understood and six-month hospice enrollment after the consultation. RESULTS: Nearly six in 10 conversations included at least one audible expression of distressing emotion (59%; 137 of 231). Among conversations with such an expression, fear was the most prevalent (72%; 98 of 137) followed by sadness (50%; 69 of 137) and anger (45%; 62 of 137). Anger expression was associated with more disease-focused end-of-life treatment preferences, pre/post consultation improvement in feeling heard and understood and lower six-month hospice enrollment. Fear was strongly associated with preconsultation patient ratings of shorter survival expectations. Sadness did not exhibit strong association with patient descriptors or outcomes. CONCLUSION: Fear, anger, and sadness are commonly expressed in hospital-based palliative care consultations with people who have advanced cancer. Anger is an epidemiologically useful predictor of important clinical outcomes.

A Tandem Evolutionary Algorithm for Identifying Causal Rules from Complex Data.

We propose a new evolutionary approach for discovering causal rules in complex classification problems from batch data. Key aspects include (a) the use of a hypergeometric probability mass function as a principled statistic for assessing fitness that quantifies the probability that the observed association between a given clause and target class is due to chance, taking into account the size of the dataset, the amount of missing data, and the distribution of outcome categories, (b) tandem age-layered evolutionary algorithms for evolving parsimonious archives of conjunctive clauses, and disjunctions of these conjunctions, each of which have probabilistically significant associations with outcome classes, and (c) separate archive bins for clauses of different orders, with dynamically adjusted order-specific thresholds. The method is validated on majority-on and multiplexer benchmark problems exhibiting various combinations of heterogeneity, epistasis, overlap, noise in class associations, missing data, extraneous features, and imbalanced classes. We also validate on a more realistic synthetic genome dataset with heterogeneity, epistasis, extraneous features, and noise. In all synthetic epistatic benchmarks, we consistently recover the true causal rule sets used to generate the data. Finally, we discuss an application to a complex real-world survey dataset designed to inform possible ecohealth interventions for Chagas disease.

Story Arcs in Serious Illness: Natural Language Processing features of Palliative Care Conversations.

OBJECTIVE: Serious illness conversations are complex clinical narratives that remain poorly understood. Natural Language Processing (NLP) offers new approaches for identifying hidden patterns within the lexicon of stories that may reveal insights about the taxonomy of serious illness conversations. METHODS: We analyzed verbatim transcripts from 354 consultations involving 231 patients and 45 palliative care clinicians from the Palliative Care Communication Research Initiative. We stratified each conversation into deciles of "narrative time" based on word counts. We used standard NLP analyses to examine the frequency and distribution of words and phrases indicating temporal reference, illness terminology, sentiment and modal verbs (indicating possibility/desirability). RESULTS: Temporal references shifted steadily from talking about the past to talking about the future over deciles of narrative time. Conversations progressed incrementally from "sadder" to "happier" lexicon; reduction in illness terminology accounted substantially for this pattern. We observed the following sequence in peak frequency over narrative time: symptom terms, treatment terms, prognosis terms and modal verbs indicating possibility. CONCLUSIONS: NLP methods can identify narrative arcs in serious illness conversations. PRACTICE IMPLICATIONS: Fully automating NLP methods will allow for efficient, large scale and real time measurement of serious illness conversations for research, education and system re-design.

Identifying Connectional Silence in Palliative Care Consultations: A Tandem Machine-Learning and Human Coding Method.

Background: Systematic measurement of conversational features in the natural clinical setting is essential to better understand, disseminate, and incentivize high quality serious illness communication. Advances in machine-learning (ML) classification of human speech offer exceptional opportunity to complement human coding (HC) methods for measurement in large scale studies. Objectives: To test the reliability, efficiency, and sensitivity of a tandem ML-HC method for identifying one feature of clinical importance in serious illness conversations: Connectional Silence. Design: This was a cross-sectional analysis of 354 audio-recorded inpatient palliative care consultations from the Palliative Care Communication Research Initiative multisite cohort study. Setting/Subjects: Hospitalized people with advanced cancer. Measurements: We created 1000 brief audio "clips" of randomly selected moments predicted by a screening ML algorithm to be two-second or longer pauses in conversation. Each clip included 10 seconds of speaking before and 5 seconds after each pause. Two HCs independently evaluated each clip for Connectional Silence as operationalized from conceptual taxonomies of silence in serious illness conversations. HCs also evaluated 100 minutes from 10 additional conversations having unique speakers to identify how frequently the ML screening algorithm missed episodes of Connectional Silence. Results:Connectional Silences were rare (5.5%) among all two-second or longer pauses in palliative care conversations. Tandem ML-HC demonstrated strong reliability (kappa 0.62; 95% confidence interval: 0.47-0.76). HC alone required 61% more time than the Tandem ML-HC method. No Connectional Silences were missed by the ML screening algorithm. Conclusions: Tandem ML-HC methods are reliable, efficient, and sensitive for identifying Connectional Silence in serious illness conversations.

Automated Detection of Conversational Pauses from Audio Recordings of Serious Illness Conversations in Natural Hospital Settings.

OBJECTIVE: Automating conversation analysis in the natural clinical setting is essential to scale serious illness communication research to samples that are large enough for traditional epidemiological studies. Our objective is to automate the identification of pauses in conversations because these are important linguistic targets for evaluating dynamics of speaker involvement and turn-taking, listening and human connection, or distraction and disengagement. DESIGN: We used 354 audio recordings of serious illness conversations from the multisite Palliative Care Communication Research Initiative cohort study. SETTING/SUBJECTS: Hospitalized people with advanced cancer seen by the palliative care team. MEASUREMENTS: We developed a Random Forest machine learning (ML) algorithm to detect Conversational Pauses of two seconds or longer. We triple-coded 261 minutes of audio with human coders to establish a gold standard for evaluating ML performance characteristics. RESULTS: ML automatically identified Conversational Pauses with a sensitivity of 90.5 and a specificity of 94.5. CONCLUSIONS: ML is a valid method for automatically identifying Conversational Pauses in the natural acoustic setting of inpatient serious illness conversations.

Dynamic changes in diffusion measures improve sensitivity in identifying patients with mild traumatic brain injury.

The goal of this study was to investigate patterns of axonal injury in the first week after mild traumatic brain injury (mTBI). We performed a prospective cohort study of 20 patients presenting to the emergency department with mTBI, using 3.0T diffusion tensor MRI immediately after injury and again at 1 week post-injury. Corresponding data were acquired from 16 controls over a similar time interval. Fractional anisotropy (FA) and other diffusion measures were calculated from 11 a priori selected axon tracts at each time-point, and the change across time in each region was quantified for each subject. Clinical outcomes were determined by standardized neurocognitive assessment. We found that mTBI subjects were significantly more likely to have changes in FA in those 11 regions of interest across the one week time period, compared to control subjects whose FA measurements were stable across time. Longitudinal imaging was more sensitive to these subtle changes in white matter integrity than cross-sectional assessments at either of two time points, alone. Analyzing the sources of variance in our control population, we show that this increased sensitivity is likely due to the smaller within-subject variability obtained by longitudinal analysis with each subject as their own control. This is in contrast to the larger between-subject variability obtained by cross-sectional analysis of each individual subject to normalized data from a control group. We also demonstrated that inclusion of all a priori ROIs in an analytic model as opposed to measuring individual ROIs improves detection of white matter changes by overcoming issues of injury heterogeneity. Finally, we employed genetic programming (a bio-inspired computational method for model estimation) to demonstrate that longitudinal changes in FA have utility in predicting the symptomatology of patients with mTBI. We conclude concussive brain injury caused acute, measurable changes in the FA of white matter tracts consistent with evolving axonal injury and/or edema, which may contribute to post-concussive symptoms.

Tunably Rugged Landscapes With Known Maximum and Minimum.

We propose NM landscapes as a new class of tunably rugged benchmark problems. NM landscapes are well defined on alphabets of any arity, including both discrete and real-valued alphabets, include epistasis in a natural and transparent manner, are proven to have known value and location of the global maximum and, with some additional constraints, are proven to also have a known global minimum. Empirical studies are used to illustrate that, when coefficients are selected from a recommended distribution, the ruggedness of NM landscapes is smoothly tunable and correlates with several measures of search difficulty. We discuss why these properties make NM landscapes preferable to both NK landscapes and Walsh polynomials as benchmark landscape models with tunable epistasis.

Competition along trajectories governs adaptation rates towards antimicrobial resistance.

The increasing availability of genotype-phenotype maps for different combinations of mutations has empowered evolutionary biologists with the tools to interrogate the predictability of adaptive evolution, especially in the context of the evolution of antimicrobial resistance. Large microbial populations are known to generate competing beneficial mutations, but determining how these mutations contribute to the adaptive trajectories that are most likely to be followed remains a challenge. Despite a recognition that there may also be competition between successive alleles on the same trajectory, prior studies have not fully considered how this impacts adaptation rates along, or likelihood of following, individual trajectories. Here, we develop a metric that quantifies the competition between successive alleles along adaptive trajectories and show how this competition largely governs the rate of evolution in simulations on empirical fitness landscapes for proteins involved in drug resistance in two species of malaria (Plasmodium falciparum and P. vivax). Our findings reveal that a trajectory with a larger-than-average initial fitness increase may have smaller fitness increases in later steps, which slows adaptation. In some circumstances, these trajectories may be outcompeted by alleles on faster alternative trajectories that are being explored simultaneously. The ability to predict adaptation rates along accessible trajectories has implications for efforts to manage antimicrobial resistance in real-world settings and for the broader intellectual pursuit of predictive evolution in complex adaptive fitness landscapes for a variety of application domains.

Evolving modular genetic regulatory networks with a recursive, top-down approach.

Being able to design genetic regulatory networks (GRNs) to achieve a desired cellular function is one of the main goals of synthetic biology. However, determining minimal GRNs that produce desired time-series behaviors is non-trivial. In this paper, we propose a 'top-down' approach to evolving small GRNs and then use these to recursively boot-strap the identification of larger, more complex, modular GRNs. We start with relatively dense GRNs and then use differential evolution (DE) to evolve interaction coefficients. When the target dynamical behavior is found embedded in a dense GRN, we narrow the focus of the search and begin aggressively pruning out excess interactions at the end of each generation. We first show that the method can quickly rediscover known small GRNs for a toggle switch and an oscillatory circuit. Next we include these GRNs as non-evolvable subnetworks in the subsequent evolution of more complex, modular GRNs. Successful solutions found in canonical DE where we truncated small interactions to zero, with or without an interaction penalty term, invariably contained many excess interactions. In contrast, by incorporating aggressive pruning and the penalty term, the DE was able to find minimal or nearly minimal GRNs in all test problems.

Exploratory Analysis in Time-Varying Data Sets: a Healthcare Network Application.

We introduce a new method for exploratory analysis of large data sets with time-varying features, where the aim is to automatically discover novel relationships between features (over some time period) that are predictive of any of a number of time-varying outcomes (over some other time period). Using a genetic algorithm, we co-evolve (i) a subset of predictive features, (ii) which attribute will be predicted (iii) the time period over which to assess the predictive features, and (iv) the time period over which to assess the predicted attribute. After validating the method on 15 synthetic test problems, we used the approach for exploratory analysis of a large healthcare network data set. We discovered a strong association, with 100% sensitivity, between hospital participation in multi-institutional quality improvement collaboratives during or before 2002, and changes in the risk-adjusted rates of mortality and morbidity observed after a 1-2 year lag. The proposed approach is a potentially powerful and general tool for exploratory analysis of a wide range of time-series data sets.

Team Learning for Healthcare Quality Improvement.

In organized healthcare quality improvement collaboratives (QICs), teams of practitioners from different hospitals exchange information on clinical practices with the aim of improving health outcomes at their own institutions. However, what works in one hospital may not work in others with different local contexts because of nonlinear interactions among various demographics, treatments, and practices. In previous studies of collaborations where the goal is a collective problem solving, teams of diverse individuals have been shown to outperform teams of similar individuals. However, when the purpose of collaboration is knowledge diffusion in complex environments, it is not clear whether team diversity will help or hinder effective learning. In this paper, we first use an agent-based model of QICs to show that teams comprising similar individuals outperform those with more diverse individuals under nearly all conditions, and that this advantage increases with the complexity of the landscape and level of noise in assessing performance. Examination of data from a network of real hospitals provides encouraging evidence of a high degree of similarity in clinical practices, especially within teams of hospitals engaging in QIC teams. However, our model also suggests that groups of similar hospitals could benefit from larger teams and more open sharing of details on clinical outcomes than is currently the norm. To facilitate this, we propose a secure virtual collaboration system that would allow hospitals to efficiently identify potentially better practices in use at other institutions similar to theirs without any institutions having to sacrifice the privacy of their own data. Our results may also have implications for other types of data-driven diffusive learning such as in personalized medicine and evolutionary search in noisy, complex combinatorial optimization problems.

Searching the clinical fitness landscape.

Widespread unexplained variations in clinical practices and patient outcomes suggest major opportunities for improving the quality and safety of medical care. However, there is little consensus regarding how to best identify and disseminate healthcare improvements and a dearth of theory to guide the debate. Many consider multicenter randomized controlled trials to be the gold standard of evidence-based medicine, although results are often inconclusive or may not be generally applicable due to differences in the contexts within which care is provided. Increasingly, others advocate the use "quality improvement collaboratives", in which multi-institutional teams share information to identify potentially better practices that are subsequently evaluated in the local contexts of specific institutions, but there is concern that such collaborative learning approaches lack the statistical rigor of randomized trials. Using an agent-based model, we show how and why a collaborative learning approach almost invariably leads to greater improvements in expected patient outcomes than more traditional approaches in searching simulated clinical fitness landscapes. This is due to a combination of greater statistical power and more context-dependent evaluation of treatments, especially in complex terrains where some combinations of practices may interact in affecting outcomes. The results of our simulations are consistent with observed limitations of randomized controlled trials and provide important insights into probable reasons for effectiveness of quality improvement collaboratives in the complex socio-technical environments of healthcare institutions. Our approach illustrates how modeling the evolution of medical practice as search on a clinical fitness landscape can aid in identifying and understanding strategies for improving the quality and safety of medical care.