Closed-Loop Uncertainty: The Evaluation and Calibration of Uncertainty for Human-Machine Teams under Data Drift.

Though an accurate measurement of entropy, or more generally uncertainty, is critical to the success of human-machine teams, the evaluation of the accuracy of such metrics as a probability of machine correctness is often aggregated and not assessed as an iterative control process. The entropy of the decisions made by human-machine teams may not be accurately measured under cold start or at times of data drift unless disagreements between the human and machine are immediately fed back to the classifier iteratively. In this study, we present a stochastic framework by which an uncertainty model may be evaluated iteratively as a probability of machine correctness. We target a novel problem, referred to as the threshold selection problem, which involves a user subjectively selecting the point at which a signal transitions to a low state. This problem is designed to be simple and replicable for human-machine experimentation while exhibiting properties of more complex applications. Finally, we explore the potential of incorporating feedback of machine correctness into a baseline naïve Bayes uncertainty model with a novel reinforcement learning approach. The approach refines a baseline uncertainty model by incorporating machine correctness at every iteration. Experiments are conducted over a large number of realizations to properly evaluate uncertainty at each iteration of the human-machine team. Results show that our novel approach, called closed-loop uncertainty, outperforms the baseline in every case, yielding about 45% improvement on average.

Barriers and enablers to telehealth use by physiotherapists during the COVID-19 pandemic.

OBJECTIVES: To investigate the perceived barriers and enablers experienced by physiotherapists whilst delivering community and outpatient services during the COVID-19 pandemic. METHODS: Qualitative study undertaken at a University-affiliated hospital in Melbourne, Australia. Physiotherapists working in the outpatient setting participated in a semi-structured interview. Interviews were analysed using Framework Analysis, with themes mapped to the Theoretical Domains Framework (TDF). RESULTS: From 19 interviews, we identified major themes and mapped these to the TDF domains: 1. The pandemic rapidly closed the knowledge-practice gap; 2. Adaptation of existing skills and integration of new skills were required; 3. Supportive senior leadership helped the transition; 4. Capabilities and confidence improved with time; 5. Environmental factors were crucial to success or failure of telehealth; 6. Access to and delivery of care improved for some; 7. Identification of appropriate patients and future hybrid models of care; 8. Changes in work practices, role certainty and identity; 9. Development of educational resources consolidated knowledge; 10. Socialisation of telehealth and optimism for the future. Within each domain, key barriers and enablers were also identified. CONCLUSIONS: The findings of this study suggest that the delivery of community and outpatient physiotherapy via telehealth during the COVID-19 pandemic was an initial challenge. Growing knowledge, confidence and ability to problem solve barriers enabled physiotherapists to move along the continuum of 'fear' to 'triumph' as the pandemic progressed. These results can be used by clinicians, managers, and academics to guide future workforce planning, hospital environmental design, and service delivery. CLINICAL MESSAGE.

Using Human-Centered Design to Improve a Surgery Resident Well-Being Program.

INTRODUCTION: Surgery resident mental health, burnout, and overall well-being are constantly scrutinized, and improving surgery resident well-being programs continuously requires refinement. We sought to evaluate the effectiveness of human-centered design (HCD) sprints to enhance our surgery resident well-being program. METHODS: An HCD sprint was conducted with 34 surgery residents in a single session using seven separate domains, including Mental Health/Reflection and Therapy; Mentoring or Faculty Engagement; Physical Well-being; Retreats; Scheduled Breaks or Free Time; Social Connection; and Well-being Lectures, Emails, or Curriculum. Responses were characterized as: "How might we", Suggestions, Useful, and Not Useful. RESULTS: Well-being Lectures, Emails, or Curriculum were overwhelmingly viewed, as Not Useful (77%), as was Mental Health/Reflection and Therapy (42%). Scheduled Breaks or Free Time was viewed as the most Useful (42%). This category also had the most suggestions and "How might we" ideas for improvement (41%). Lastly, Suggestions and "How might we" ideas were also common for improving Mentoring or Faculty Engagement (31% and 29%, respectively). These results were incorporated into multiple strategies to improve surgery resident well-being and also shared in a Department of Surgery Grand Rounds. CONCLUSIONS: Surgery resident well-being and a targeted approach by a well-being program are critical to a residency program, particularly with the arduous nature of surgical training during the pandemic resulting in periods of prolonged social isolation. HCD sprints are an effective means to refine a surgery resident well-being program and to involve the residents themselves in that process.

Primary Immune Deficiency: Patients' Preferences for Replacement Immunoglobulin Therapy.

Purpose: Immunoglobulin (Ig) replacement therapy is an important life-saving treatment modality for patients with primary antibody immune deficiency disorders (PAD). IVIG and SCIg are suitable alternatives to treat patients with PAD but vary in key ways. Existing evidence on patient preferences for Ig treatments given the complexities associated with IVIG and SCIg treatment is limited and fails to account for variations in preferences across patients. For this reason, we sought to evaluate PAD patient preferences for features of IVIG and SCIg across different patient characteristics. Materials and Methods: 119 PAD patients completed a discrete-choice experiment (DCE) survey. The DCE asked respondents to make choices between carefully constructed treatment alternatives described in terms of generic treatment features. Choices from the DCE were analyzed to determine the relative influence of attribute changes on treatment preferences. We used subgroup analysis to evaluate systematic variations in preferences by patients' age, gender, time since diagnosis, and treatment experience. Results: Patients were primarily concerned about the duration of treatment side effects, but preferences were heterogeneous. This was particularly true around administration features. Time since diagnosis was associated with an increase in patients' concerns with the number of needles required per infusion. Also, patients appear to prefer the kind of therapy they are currently using which could be the result of properly aligned patient preferences or evidence of patient adaptive behavior. Conclusions: Heterogeneity in preferences for Ig replacement treatments suggests that a formal shared decision making process could have an important role in improving patient care.

Higher Radiation Dose to the Immune Cells Correlates with Worse Tumor Control and Overall Survival in Patients with Stage III NSCLC: A Secondary Analysis of RTOG0617.

Background: We hypothesized that the Effective radiation Dose to the Immune Cells (EDIC) in circulating blood is a significant factor for the treatment outcome in patients with locally advanced non-small-cell lung cancer (NSCLC). Methods: This is a secondary study of a phase III trial, NRG/RTOG 0617, in patients with stage III NSCLC treated with radiation-based treatment. The EDIC was computed as equivalent uniform dose to the entire blood based on radiation doses to all blood-containing organs, with consideration of blood flow and fractionation effect. The primary endpoint was overall survival (OS), and the secondary endpoints were progression-free survival (PFS) and local progression-free survival (LPFS). The EDIC-survival relationship was analyzed with consideration of clinical significant factors. Results: A total of 456 patients were eligible. The median EDIC values were 5.6 Gy (range, 2.1-12.2 Gy) and 6.3 Gy (2.1-11.6 Gy) for the low- and high-dose groups, respectively. The EDIC was significantly associated with OS (hazard ratio [HR] = 1.12, p = 0.005) and LPFS (HR = 1.09, p = 0.02) but PFS (HR = 1.05, p = 0.17) after adjustment for tumor dose, gross tumor volume and other factors. OS decreased with an increasing EDIC in a non-linear pattern: the two-year OS decreased first with a slope of 8%/Gy when the EDIC < 6 Gy, remained relatively unchanged when the EDIC was 6-8 Gy, and followed by a further reduction with a slope of 12%/Gy when the EDIC > 8 Gy. Conclusions: The EDIC is a significant independent risk factor for poor OS and LPFS in RTOG 0617 patients with stage III NSCLC, suggesting that radiation dose to circulating immune cells is critical for tumor control. Organ at risk for the immune system should be considered during RT plan.

The value of clinical breast examination in a breast cancer surveillance program for women with germline BRCA1 or BRCA2 mutations.

OBJECTIVE: To assess the sensitivity and specificity of clinical breast examination for detecting breast cancer in asymptomatic women with predisposing germline mutations enrolled in a cancer risk management program that includes radiologic screening. DESIGN, SETTING: Retrospective, longitudinal cohort study of women with BRCA1/2 mutations who attended the Breast and Ovarian Cancer Risk Management Clinic at the Peter MacCallum Cancer Centre, a tertiary referral centre in Melbourne, during 1 September 2001 - 31 December 2019. PARTICIPANTS: Consecutive women with BRCA1/2 mutations who did not have personal histories of cancer and had not undergone bilateral risk-reducing mastectomy, and who had visited the clinic at least twice during the study period. Participants had generally undergone breast examination at 6- or 12-month intervals, and annual breast imaging (mammography; and magnetic resonance imaging [MRI] for women aged 50 years or younger). MAIN OUTCOME MEASURES: Sensitivity (proportion of all biopsy-confirmed breast cancers detected by breast examination alone) and specificity of breast examination for detecting breast cancer. RESULTS: Of 414 eligible women (mean age, 35.5 years; SD, 11.2 years), 35 were diagnosed with breast cancer during 1761 woman-years of follow-up. Only two were diagnosed based on breast examination alone (ie, without radiologic evidence), neither of whom was undergoing MRI screening. The sensitivity of breast examination was 6% (95% CI, 1-19%), the specificity 97% (95% CI, 95-98%); the positive predictive value was 14% (95% CI, 2-43%), the negative predictive value 92% (95% CI, 89-94%). CONCLUSION: Clinical breast examination did not increase the number of breast cancers detected in MRI-screened women with BRCA1/2 mutations. Removing breast examination from surveillance programs that include MRI may be reasonable for these women.

Social Restrictions versus Testing Campaigns in the COVID-19 Crisis: A Predictive Model Based on the Spanish Case.

The global COVID-19 spread has forced countries to implement non-pharmacological interventions (NPI) (i.e., mobility restrictions and testing campaigns) to preserve health systems. Spain is one of the most severely impacted countries, both clinically and economically. In an effort to support policy decision-making, we aimed to assess the impacts of different NPI on COVID-19 epidemiology, healthcare costs and Gross Domestic Product (GDP). A modified Susceptible-Exposed-Infectious-Removed epidemiological model was created to simulate the pandemic evolution. Its output was used to populate an economic model to quantify healthcare costs and GDP variation through a regression model which correlates NPI and GDP change from 42 countries. Thirteen scenarios combining different NPI were consecutively simulated in the epidemiological and economic models. Both increased testing and stringency could reduce cases, hospitalizations and deaths. While policies based on increased testing rates lead to higher healthcare costs, increased stringency is correlated with greater GDP declines, with differences of up to 4.4% points. Increased test sensitivity may lead to a reduction of cases, hospitalizations and deaths and to the implementation of pooling techniques that can increase throughput testing capacity. Alternative strategies to control COVID-19 spread entail differing economic outcomes. Decision-makers may utilize this tool to identify the most suitable strategy considering epidemiological and economic outcomes.

Reducing disparities in health care quality: the role of health IT in underresourced settings.

Health information technology (health IT) and the Internet have tremendous potential to transform health care. The implications of the adoption of health IT within the context of underresourced health care settings have not been adequately evaluated. For this reason, the Agency for Healthcare Research and Quality sponsored a 2-day invitational expert meeting in October 2009. The theme of the meeting was "Reducing Disparities in Healthcare Quality in Under Resourced Settings Using HIT and Other Quality Improvement Strategies." In preparation for the meeting, five background manuscripts were commissioned, presented, and discussed. Attendees also participated in one of three breakout sessions. Several overarching themes, key recommendations, and research topics emerged across five general categories of (a) the health care delivery setting, (b) research and evaluation methodologies, (c) patients and target populations, (d) technology applications and platforms, and (e) providers and clinicians. This article outlines the process, findings, and key recommendations of this expert workshop.