scholarly journals Patient similarity analytics for explainable clinical risk prediction

2021 ◽  
Vol 21 (1) ◽  
Author(s):  
Hao Sen Andrew Fang ◽  
Ngiap Chuan Tan ◽  
Wei Ying Tan ◽  
Ronald Wihal Oei ◽  
Mong Li Lee ◽  
...  
Keyword(s):  
Decision Making ◽  
Logistic Regression ◽  
Clinical Practice ◽  
Risk Prediction ◽  
Clinical Decision ◽  
Support Vector ◽  
Prototype System ◽  
Care Clinic ◽  
Clinical Risk ◽  
Similarity Model

Abstract Background Clinical risk prediction models (CRPMs) use patient characteristics to estimate the probability of having or developing a particular disease and/or outcome. While CRPMs are gaining in popularity, they have yet to be widely adopted in clinical practice. The lack of explainability and interpretability has limited their utility. Explainability is the extent of which a model’s prediction process can be described. Interpretability is the degree to which a user can understand the predictions made by a model. Methods The study aimed to demonstrate utility of patient similarity analytics in developing an explainable and interpretable CRPM. Data was extracted from the electronic medical records of patients with type-2 diabetes mellitus, hypertension and dyslipidaemia in a Singapore public primary care clinic. We used modified K-nearest neighbour which incorporated expert input, to develop a patient similarity model on this real-world training dataset (n = 7,041) and validated it on a testing dataset (n = 3,018). The results were compared using logistic regression, random forest (RF) and support vector machine (SVM) models from the same dataset. The patient similarity model was then implemented in a prototype system to demonstrate the identification, explainability and interpretability of similar patients and the prediction process. Results The patient similarity model (AUROC = 0.718) was comparable to the logistic regression (AUROC = 0.695), RF (AUROC = 0.764) and SVM models (AUROC = 0.766). We packaged the patient similarity model in a prototype web application. A proof of concept demonstrated how the application provided both quantitative and qualitative information, in the form of patient narratives. This information was used to better inform and influence clinical decision-making, such as getting a patient to agree to start insulin therapy. Conclusions Patient similarity analytics is a feasible approach to develop an explainable and interpretable CRPM. While the approach is generalizable, it can be used to develop locally relevant information, based on the database it searches. Ultimately, such an approach can generate a more informative CRPMs which can be deployed as part of clinical decision support tools to better facilitate shared decision-making in clinical practice.

scholarly journals Patient Similarity Analytics for Explainable Clinical Risk Prediction

2021 ◽  
Author(s):  
Hao Sen Andrew Fang ◽  
Ngiap Chuan Tan ◽  
Wei Ying Tan ◽  
Ronald Wihal Oei ◽  
Mong Li Lee ◽  
...  
Keyword(s):  
Decision Making ◽  
Support Vector Machine ◽  
Logistic Regression ◽  
Clinical Practice ◽  
Random Forest ◽  
Risk Prediction ◽  
Clinical Decision ◽  
Support Vector ◽  
Clinical Risk ◽  
Similarity Model

Abstract Background: A Clinical Risk Prediction Model (CRPM) uses patient characteristics to estimate the probability about having or developing a particular disease and/or outcome. While CRPMs are gaining in popularity, they have yet to be adopted routinely in clinical practice. The lack of explainability and interpretability has limited its utility. Explainability is the extent of which a model’s prediction process can be described. Interpretability is the degree to which a user can understand the predictions made by a model.Methods: The study aimed to demonstrate utility of patient similarity analytics in developing an explainable and interpretable CPRM. Data was extracted from the electronic medical records of patients with type-2 diabetes mellitus, hypertension and dyslipidaemia in a Singapore public primary care clinic. We used various techniques, including patient similarity analytics, to develop various models on this real-world training dataset (n=7,041) and validated each of them on the same test dataset (n=3,018). The results were compared using logistic regression, random forest and support vector machine models from the same dataset. The CRPM was then implemented in a prototype system to demonstrate the identification, explainability and interpretability of similar patients and the prediction process.Results: The patient similarity model (AUROC=0.718) was comparable to the logistic regression (AUROC=0.695), random forest (AUROC=0.764) and support vector machine models (AUROC=0.766). We incorporated the patient similarity model in a prototype web application. A case study demonstrated how the application was provided both quantitative and qualitative information, in the form of patient narratives. This information was used to better inform and influence clinical decision-making, such as getting a patient to agree to start insulin therapy.Conclusions: A patient similarity approach is feasible to develop an explainable and interpretable CRPM. It is a general approach which can be used to develop locally relevant information, based on the database it searches. Ultimately, such an approach can generate a more informative CRPMs which can be deployed as part of clinical decision support tools to better facilitate shared decision-making in clinical practice.

scholarly journals The use of multiple-criteria decision-making theory to measure students’ perceptions of high-fidelity simulation

2017 ◽  
Vol 3 (3) ◽  
pp. 88-93 ◽  
Author(s):  
Maureen Anne Jersby ◽  
Paul Van-Schaik ◽  
Stephen Green ◽  
Lili Nacheva-Skopalik
Keyword(s):  
Decision Making ◽  
Clinical Practice ◽  
Repeated Measures ◽  
Clinical Decision Making ◽  
Learning Experience ◽  
Clinical Decision ◽  
High Fidelity ◽  
High Fidelity Simulation ◽  
Initial Exposure ◽  
Repeated Measures Design

BackgroundHigh-Fidelity Simulation (HFS) has great potential to improve decision-making in clinical practice. Previous studies have found HFS promotes self-confidence, but its effectiveness in clinical practice has not been established. The aim of this research is to establish if HFS facilitates learning that informs decision-making skills in clinical practice using MultipleCriteria DecisionMaking Theory (MCDMT).MethodsThe sample was 2nd year undergraduate pre-registration adult nursing students.MCDMT was used to measure the students’ experience of HFS and how it developed their clinical decision-making skills. MCDMT requires characteristic measurements which for the learning experience were based on five factors that underpin successful learning, and for clinical decision-making, an analytical framework was used. The study used a repeated-measures design to take two measurements: the first one after the first simulation experience and the second one after clinical placement. Baseline measurements were obtained from academics. Data were analysed using the MCDMT tool.ResultsAfter their initial exposure to simulation learning, students reported that HFS provides a high-quality learning experience (87%) and supports all aspects of clinical decision-making (85%). Following clinical practice, the level of support for clinical decision-making remained at 85%, suggesting that students believe HFS promotes transferability of knowledge to the practice setting.ConclusionOverall, students report a high level of support for learning and developing clinical decision-making skills from HFS. However, there are no comparative data available from classroom teaching of similar content so it cannot be established if these results are due to HFS alone.

Between a logic of disruption and a logic of continuation: Negotiating the legitimacy of algorithms used in automated clinical decision-making

2021 ◽  
pp. 136345932199674
Author(s):  
Rikke Torenholt ◽  
Henriette Langstrup
Keyword(s):  
Decision Making ◽  
Clinical Practice ◽  
Patient Reported Outcomes ◽  
Patient Involvement ◽  
Clinical Decision Making ◽  
Diagnostic Procedures ◽  
Clinical Decision ◽  
Patient Reported ◽  
Pro Tools

In both popular and academic discussions of the use of algorithms in clinical practice, narratives often draw on the decisive potentialities of algorithms and come with the belief that algorithms will substantially transform healthcare. We suggest that this approach is associated with a logic of disruption. However, we argue that in clinical practice alongside this logic, another and less recognised logic exists, namely that of continuation: here the use of algorithms constitutes part of an established practice. Applying these logics as our analytical framing, we set out to explore how algorithms for clinical decision-making are enacted by political stakeholders, healthcare professionals, and patients, and in doing so, study how the legitimacy of delegating to an algorithm is negotiated and obtained. Empirically we draw on ethnographic fieldwork carried out in relation to attempts in Denmark to develop and implement Patient Reported Outcomes (PRO) tools – involving algorithmic sorting – in clinical practice. We follow the work within two disease areas: heart rehabilitation and breast cancer follow-up care. We show how at the political level, algorithms constitute tools for disrupting inefficient work and unsystematic patient involvement, whereas closer to the clinical practice, algorithms constitute a continuation of standardised and evidence-based diagnostic procedures and a continuation of the physicians’ expertise and authority. We argue that the co-existence of the two logics have implications as both provide a push towards the use of algorithms and how a logic of continuation may divert attention away from new issues introduced with automated digital decision-support systems.

scholarly journals Awareness, Knowledge, and Utility of RCT Data vs RWE: Results From a Survey of US Cardiologists: Real-world Evidence in Clinical Decision Making

2020 ◽  
Vol 14 ◽  
pp. 117954682095341 ◽  
Author(s):  
Todd C Villines ◽  
Mark J Cziraky ◽  
Alpesh N Amin
Keyword(s):  
Decision Making ◽  
Clinical Practice ◽  
Real World ◽  
Randomized Clinical Trials ◽  
Clinical Decision Making ◽  
Clinical Decision ◽  
The Body ◽  
Representative Sampling ◽  
Real World Evidence ◽  
Wide Range

Real-world evidence (RWE) provides a potential rich source of additional information to the body of data available from randomized clinical trials (RCTs), but there is a need to understand the strengths and limitations of RWE before it can be applied to clinical practice. To gain insight into current thinking in clinical decision making and utility of different data sources, a representative sampling of US cardiologists selected from the current, active Fellows of the American College of Cardiology (ACC) were surveyed to evaluate their perceptions of findings from RCTs and RWE studies and their application in clinical practice. The survey was conducted online via the ACC web portal between 12 July and 11 August 2017. Of the 548 active ACC Fellows invited as panel members, 173 completed the survey (32% response), most of whom were board certified in general cardiology (n = 119, 69%) or interventional cardiology (n = 40, 23%). The survey results indicated a wide range of familiarity with and utilization of RWE amongst cardiologists. Most cardiologists were familiar with RWE and considered RWE in clinical practice at least some of the time. However, a significant minority of survey respondents had rarely or never applied RWE learnings in their clinical practice, and many did not feel confident in the results of RWE other than registry data. These survey findings suggest that additional education on how to assess and interpret RWE could help physicians to integrate data and learnings from RCTs and RWE to best guide clinical decision making.

Implementing machine learning algorithms for suicide risk prediction in clinical practice: A focus group study

2021 ◽  
Author(s):  
Kate Bentley ◽  
Kelly Zuromski ◽  
Rebecca Fortgang ◽  
Emily Madsen ◽  
Daniel Kessler ◽  
...  
Keyword(s):  
Machine Learning ◽  
Risk Assessment ◽  
Clinical Practice ◽  
Risk Prediction ◽  
Suicide Risk ◽  
Prediction Models ◽  
Clinical Decision ◽  
Machine Learning Algorithms ◽  
Suicide Risk Assessment ◽  
Focus Group Study

Background: Interest in developing machine learning algorithms that use electronic health record data to predict patients’ risk of suicidal behavior has recently proliferated. Whether and how such models might be implemented and useful in clinical practice, however, remains unknown. In order to ultimately make automated suicide risk prediction algorithms useful in practice, and thus better prevent patient suicides, it is critical to partner with key stakeholders (including the frontline providers who will be using such tools) at each stage of the implementation process.Objective: The aim of this focus group study was to inform ongoing and future efforts to deploy suicide risk prediction models in clinical practice. The specific goals were to better understand hospital providers’ current practices for assessing and managing suicide risk; determine providers’ perspectives on using automated suicide risk prediction algorithms; and identify barriers, facilitators, recommendations, and factors to consider for initiatives in this area. Methods: We conducted 10 two-hour focus groups with a total of 40 providers from psychiatry, internal medicine and primary care, emergency medicine, and obstetrics and gynecology departments within an urban academic medical center. Audio recordings of open-ended group discussions were transcribed and coded for relevant and recurrent themes by two independent study staff members. All coded text was reviewed and discrepancies resolved in consensus meetings with doctoral-level staff. Results: Though most providers reported using standardized suicide risk assessment tools in their clinical practices, existing tools were commonly described as unhelpful and providers indicated dissatisfaction with current suicide risk assessment methods. Overall, providers’ general attitudes toward the practical use of automated suicide risk prediction models and corresponding clinical decision support tools were positive. Providers were especially interested in the potential to identify high-risk patients who might be missed by traditional screening methods. Some expressed skepticism about the potential usefulness of these models in routine care; specific barriers included concerns about liability, alert fatigue, and increased demand on the healthcare system. Key facilitators included presenting specific patient-level features contributing to risk scores, emphasizing changes in risk over time, and developing systematic clinical workflows and provider trainings. Participants also recommended considering risk-prediction windows, timing of alerts, who will have access to model predictions, and variability across treatment settings.Conclusions: Providers were dissatisfied with current suicide risk assessment methods and open to the use of a machine learning-based risk prediction system to inform clinical decision-making. They also raised multiple concerns about potential barriers to the usefulness of this approach and suggested several possible facilitators. Future efforts in this area will benefit from incorporating systematic qualitative feedback from providers, patients, administrators, and payers on the use of new methods in routine care, especially given the complex, sensitive, and unfortunately still stigmatized nature of suicide risk.

scholarly journals P114: Barriers and facilitators affecting implementation of a decision aid for the diagnosis of acute aortic syndrome: a qualitative study

CJEM ◽  
2020 ◽  
Vol 22 (S1) ◽  
pp. S106-S106
Author(s):  
C. Dmitriew ◽  
R. Ohle
Keyword(s):  
Decision Making ◽  
Clinical Practice ◽  
Practice Guidelines ◽  
Resource Use ◽  
Decision Aid ◽  
Clinical Decision Making ◽  
Clinical Decision ◽  
Barriers And Facilitators ◽  
Acute Aortic Syndrome ◽  
D Dimer

Introduction: Acute aortic syndrome (AAS) is an uncommon, life-threatening emergency that is frequently misdiagnosed. The Canadian clinical practice guidelines for the diagnosis of AAS were developed in order to reduce the frequency of misdiagnoses and number of diagnostic tests. As part of the guidelines, a clinical decision aid was developed in order to facilitate clinician decision-making based on practice recommendations. The objective of this study was to identify barriers and facilitators among physicians to implementation of the decision aid. Methods: We conducted semi-structured interviews with emergency room physicians working at 5 sites distributed between urban academic and rural settings. We used purposive sampling, contacting ED physicians until data saturation was reached. Interview questions were designed to understand potential barriers and facilitators affecting the probability of decision aid uptake and accurate application of the tool. Two independent raters coded interview transcripts using an integrative approach to theme identification, combining an inductive approach to identification of themes within an organizing framework (Theoretical Domains Framework), discrepancies in coding were resolved through discussion until consensus was reached. Results: A majority of interviewees anticipated that the decision aid would support clinical decision making and risk stratification while reducing resource use and missed diagnoses. Facilitators identified included validation and publication of the guidelines as well as adoption by peers. Barriers to implementation and application of the tool included the fact that the use of D-dimer and knowledge of the rationale for its use in the investigation of AAS were not widespread. Furthermore, scoring components were, at times, out of alignment with clinician practices and understanding of risk factors. The complexity of the decision aid was also identified as a potential barrier to accurate use. Conclusion: Physicians were amenable to using the AAS decision aid to support clinical decision-making and to reduce resource use, particularly within rural contexts. Key barriers identified included the complexity of scoring and inclusion criteria, and the variable acceptance of D-dimer among clinicians. These barriers should be addressed prior to implementation of the decision aid during validation studies of the clinical practice guidelines.

Interrogating metastatic prostate cancer treatment switch decisions.

2016 ◽  
Vol 34 (2_suppl) ◽  
pp. 296-296
Author(s):  
David Lorente ◽  
Praful Ravi ◽  
Niven Mehra ◽  
Carmel Jo Pezaro ◽  
Aurelius Gabriel Omlin ◽  
...  
Keyword(s):  
Decision Making ◽  
Clinical Practice ◽  
Bone Scan ◽  
Clinical Decision Making ◽  
Clinical Decision ◽  
Response To Treatment ◽  
Clinical Progression ◽  
Online Questionnaire ◽  
Main Challenge ◽  
Access To Technology

296 Background: Increased availability of treatment options in CRPC requires improved biomarkers to optimize decision making for therapeutic sequencing. Despite evidence for the value of CTCs in assessing prognosis and response to treatment, their use in clinical practice is not widely implemented. Clinicians rely on PCWG2 criteria based on PSA, clinical and radiological criteria although these are only inconsistently used in clinical practice. We evaluated the trends for clinical decision-making by physicians treating CRPC. Methods: An online questionnaire was distributed to physicians treating PC from the UK, Switzerland and Australia. Questions on clinical practice, familiarity with progression criteria, use of CTCs and clinical-decision making were formulated. Results: 111 participants replied. Most (84.7%) were oncologists treating ≥ 50 patients per year (65.3%). Although only 39.6% usually used PCWG2 in clinical practice, 74.5% considered PSA, bone scans and CT to be useful for monitoring disease. 55.6% considered PSA to be an important biomarker. A minority were able to identify PSA (41.4%) and bone scan (39.4%) progression criteria by PCWG2. On average, more physicians discontinued cabazitaxel (28%) than docetaxel (10.4%) before cycle 4. Similar number of cycles were given to bone only disease compared to RECIST evaluable patients. Clinical progression was most important for switching treatment for most physicians (90.5%), followed by RECIST (71.6%), bone scan (47.7%), CTC (23.2%) and PSA (21.1%). The main challenge associated with the use of CTCs was the access to technology (84.7%). Most respondents (92%) would not stop therapy with rising PSA but falling CTC counts; most (88.8%) would not stop with declining PSA but rising CTCs. Although 50% acknowledged the prognostic value of CTCs, only 33% would use them to guide decision-making. Conclusions: A significant number of physicians discontinue treatment before 12 weeks. Most physicians rely on clinical progression for decision-making. Knowledge of PCWG2 response and progression criteria is generally suboptimal. Greater physician awareness, access to technology and further evidence and will be required for the implementation of CTCs as a routine biomarker in CRPC.

Understanding critical care nurses’ autonomy in Jordan

2017 ◽  
Vol 30 (4) ◽  
pp. 432-442 ◽  
Author(s):  
Mahmoud Maharmeh
Keyword(s):  
Decision Making ◽  
Critical Care ◽  
Clinical Practice ◽  
Clinical Decision Making ◽  
Clinical Decision ◽  
Critical Care Nurses ◽  
Cross Sectional Survey ◽  
Eligibility Criteria ◽  
Cross Sectional ◽  
Content Type

Purpose The aim of this study was to describe Jordanian critical care nurses’ experiences of autonomy in their clinical practice. Design/methodology/approach A descriptive correlational design was applied using a self-reported cross-sectional survey. A total of 110 registered nurses who met the eligibility criteria participated in this study. The data were collected by a structured questionnaire. Findings A majority of critical care nurses were autonomous in their decision-making and participation in decisions to take action in their clinical settings. Also, they were independent to develop their own knowledge. The study identified that their autonomy in action and acquired knowledge were influenced by a number of factors such as gender and area of practice. Practical implications Nurse’s autonomy could be increased if nurses are made aware of the current level of autonomy and explore new ways to increase empowerment. This could be offered through classroom lectures that concentrate on the concept of autonomy and its implication in practice. Nurses should demonstrate autonomous nursing care at the same time in the clinical practice. This could be done through collaboration between educators and clinical practice to help merge theory to practice. Originality/value Critical care nurses were more autonomous in action and knowledge base. This may negatively affect the quality of patient care and nurses’ job satisfaction. Therefore, improving nurses’ clinical decision-making autonomy could be done by the support of both hospital administrators and nurses themselves.

Sign in / Sign up

Export Citation Format

Share Document