A community-of-practice-based evaluation methodology for knowledge intensive computational methods and its application to multimorbidity decision support.

OBJECTIVE: The study has dual objectives. Our first objective (1) is to develop a community-of-practice-based evaluation methodology for knowledge-intensive computational methods. We target a whitebox analysis of the computational methods to gain insight on their functional features and inner workings. In more detail, we aim to answer evaluation questions on (i) support offered by computational methods for functional features within the application domain; and (ii) in-depth characterizations of the underlying computational processes, models, data and knowledge of the computational methods. Our second objective (2) involves applying the evaluation methodology to answer questions (i) and (ii) for knowledge-intensive clinical decision support (CDS) methods, which operationalize clinical knowledge as computer interpretable guidelines (CIG); we focus on multimorbidity CIG-based clinical decision support (MGCDS) methods that target multimorbidity treatment plans. MATERIALS AND METHODS: Our methodology directly involves the research community of practice in (a) identifying functional features within the application domain; (b) defining exemplar case studies covering these features; and (c) solving the case studies using their developed computational methods-research groups detail their solutions and functional feature support in solution reports. Next, the study authors (d) perform a qualitative analysis of the solution reports, identifying and characterizing common themes (or dimensions) among the computational methods. This methodology is well suited to perform whitebox analysis, as it directly involves the respective developers in studying inner workings and feature support of computational methods. Moreover, the established evaluation parameters (e.g., features, case studies, themes) constitute a re-usable benchmark framework, which can be used to evaluate new computational methods as they are developed. We applied our community-of-practice-based evaluation methodology on MGCDS methods. RESULTS: Six research groups submitted comprehensive solution reports for the exemplar case studies. Solutions for two of these case studies were reported by all groups. We identified four evaluation dimensions: detection of adverse interactions, management strategy representation, implementation paradigms, and human-in-the-loop support. Based on our whitebox analysis, we present answers to the evaluation questions (i) and (ii) for MGCDS methods. DISCUSSION: The proposed evaluation methodology includes features of illuminative and comparison-based approaches; focusing on understanding rather than judging/scoring or identifying gaps in current methods. It involves answering evaluation questions with direct involvement of the research community of practice, who participate in setting up evaluation parameters and solving exemplar case studies. Our methodology was successfully applied to evaluate six MGCDS knowledge-intensive computational methods. We established that, while the evaluated methods provide a multifaceted set of solutions with different benefits and drawbacks, no single MGCDS method currently provides a comprehensive solution for MGCDS. CONCLUSION: We posit that our evaluation methodology, applied here to gain new insights into MGCDS, can be used to assess other types of knowledge-intensive computational methods and answer other types of evaluation questions. Our case studies can be accessed at our GitHub repository (https://github.com/william-vw/MGCDS).

Towards a goal-oriented methodology for clinical-guideline-based management recommendations for patients with multimorbidity: GoCom and its preliminary evaluation.

Patients with chronic multimorbidity are becoming more common as life expectancy increases, making it necessary for physicians to develop complex management plans. We are looking at the patient management process as a goal-attainment problem. Hence, our aim is to develop a goal-oriented methodology for providing decision support for managing patients with multimorbidity continuously, as the patient's health state is progressing and new goals arise (e.g., treat ulcer, prevent osteoporosis). Our methodology allows us to detect and mitigate inconsistencies among guideline recommendations stemming from multiple clinical guidelines, while consulting medical ontologies and terminologies and relying on patient information standards. This methodology and its implementation as a decision-support system, called GoCom, starts with computer-interpretable clinical guidelines (CIGs) for single problems that are formalized using the PROforma CIG language. We previously published the architecture of the system as well as a CIG elicitation guide for enriching PROforma tasks with properties referring to vocabulary codes of goals and physiological effects of management plans. In this paper, we provide a formalization of the conceptual model of GoCom that generates, for each morbidity of the patient, a patient-specific goal tree that results from the PROforma engine's enactment of the CIG with the patient's data. We also present the "Controller" algorithm that drives the GoCom system. Given a new problem that a patient develops, the Controller detects inconsistencies among goals pertaining to different comorbid problems and consults the CIGs to generate alternative non-conflicted and goal-oriented management plans that address the multiple goals simultaneously. In this stage of our research, the inconsistencies that can be detected are of two types - starting vs. stopping medications that belong to the same medication class hierarchy, and detecting opposing physiological effect goals that are specified in concurrent CIGs (e.g., decreased blood pressure vs. increased blood pressure). However, the design of GoCom is modular and generic and allows the future introduction of additional interaction detection and mitigation strategies. Moreover, GoCom generates explanations of the alternative non-conflicted management plans, based on recommendations stemming from the clinical guidelines and reasoning patterns. GoCom's functionality was evaluated using three cases of multimorbidity interactions that were checked by our three clinicians. Usefulness was evaluated with two studies. The first evaluation was a pilot study with ten 6th year medical students and the second evaluation was done with 27 6th medical students and interns. The participants solved complex realistic cases of multimorbidity patients: with and without decision-support, two cases in the first evaluation and 6 cases in the second evaluation. Use of GoCom increased completeness of the patient management plans produced by the medical students from 0.44 to 0.71 (P-value of 0.0005) in the first evaluation, and from 0.31 to 0.78 (P-value < 0.0001) in the second evaluation. Correctness in the first evaluation was very high with (0.98) or without the system (0.91), with non-significant difference (P-value ≥ 0.17). In the second evaluation, use of GoCom increased correctness from 0.68 to 0.83 (P-value of 0.001). In addition, GoCom's explanation and visualization were perceived as useful by the vast majority of participants. While GoCom's detection of goal interactions is currently limited to detection of starting vs. stopping the same medication or medication subclasses and detecting conflicting physiological effects of concurrent medications, the evaluation demonstrated potential of the system for improving clinical decision-making for multimorbidity patients.

Use of ontology structure and Bayesian models to aid the crowdsourcing of ICD-11 sanctioning rules.

The International Classification of Diseases (ICD) is the de facto standard international classification for mortality reporting and for many epidemiological, clinical, and financial use cases. The next version of ICD, ICD-11, will be submitted for approval by the World Health Assembly in 2018. Unlike previous versions of ICD, where coders mostly select single codes from pre-enumerated disease and disorder codes, ICD-11 coding will allow extensive use of multiple codes to give more detailed disease descriptions. For example, "severe malignant neoplasms of left breast" may be coded using the combination of a "stem code" (e.g., code for malignant neoplasms of breast) with a variety of "extension codes" (e.g., codes for laterality and severity). The use of multiple codes (a process called post-coordination), while avoiding the pitfall of having to pre-enumerate vast number of possible disease and qualifier combinations, risks the creation of meaningless expressions that combine stem codes with inappropriate qualifiers. To prevent that from happening, "sanctioning rules" that define legal combinations are necessary. In this work, we developed a crowdsourcing method for obtaining sanctioning rules for the post-coordination of concepts in ICD-11. Our method utilized the hierarchical structures in the domain to improve the accuracy of the sanctioning rules and to lower the crowdsourcing cost. We used Bayesian networks to model crowd workers' skills, the accuracy of their responses, and our confidence in the acquired sanctioning rules. We applied reinforcement learning to develop an agent that constantly adjusted the confidence cutoffs during the crowdsourcing process to maximize the overall quality of sanctioning rules under a fixed budget. Finally, we performed formative evaluations using a skin-disease branch of the draft ICD-11 and demonstrated that the crowd-sourced sanctioning rules replicated those defined by an expert dermatologist with high precision and recall. This work demonstrated that a crowdsourcing approach could offer a reasonably efficient method for generating a first draft of sanctioning rules that subject matter experts could verify and edit, thus relieving them of the tedium and cost of formulating the initial set of rules.

Summarizing and visualizing structural changes during the evolution of biomedical ontologies using a Diff Abstraction Network.

Biomedical ontologies are a critical component in biomedical research and practice. As an ontology evolves, its structure and content change in response to additions, deletions and updates. When editing a biomedical ontology, small local updates may affect large portions of the ontology, leading to unintended and potentially erroneous changes. Such unwanted side effects often go unnoticed since biomedical ontologies are large and complex knowledge structures. Abstraction networks, which provide compact summaries of an ontology's content and structure, have been used to uncover structural irregularities, inconsistencies and errors in ontologies. In this paper, we introduce Diff Abstraction Networks ("Diff AbNs"), compact networks that summarize and visualize global structural changes due to ontology editing operations that result in a new ontology release. A Diff AbN can be used to support curators in identifying unintended and unwanted ontology changes. The derivation of two Diff AbNs, the Diff Area Taxonomy and the Diff Partial-area Taxonomy, is explained and Diff Partial-area Taxonomies are derived and analyzed for the Ontology of Clinical Research, Sleep Domain Ontology, and eagle-i Research Resource Ontology. Diff Taxonomy usage for identifying unintended erroneous consequences of quality assurance and ontology merging are demonstrated.

The Ontology of Clinical Research (OCRe): an informatics foundation for the science of clinical research.

To date, the scientific process for generating, interpreting, and applying knowledge has received less informatics attention than operational processes for conducting clinical studies. The activities of these scientific processes - the science of clinical research - are centered on the study protocol, which is the abstract representation of the scientific design of a clinical study. The Ontology of Clinical Research (OCRe) is an OWL 2 model of the entities and relationships of study design protocols for the purpose of computationally supporting the design and analysis of human studies. OCRe's modeling is independent of any specific study design or clinical domain. It includes a study design typology and a specialized module called ERGO Annotation for capturing the meaning of eligibility criteria. In this paper, we describe the key informatics use cases of each phase of a study's scientific lifecycle, present OCRe and the principles behind its modeling, and describe applications of OCRe and associated technologies to a range of clinical research use cases. OCRe captures the central semantics that underlies the scientific processes of clinical research and can serve as an informatics foundation for supporting the entire range of knowledge activities that constitute the science of clinical research.

Linking Health Terminologies: A Unified Approach to the WHO Family of International Classifications.

This paper presents an effort by the World Health Organization (WHO) to integrate the reference classifications of the Family of International Classifications (ICD, ICF, and ICHI) into a unified digital framework. The integration was accomplished via an expanded Content Model and a single Foundation that hosts all entities from these classifications, allowing the traditional use cases of individual classifications to be retained while enhancing their combined use. The harmonized WHO-FIC Content Model and the unified Foundation has streamlined the content management, enhanced the web-based tool functionalities, and provided opportunities for linkage with external terminologies and ontologies. This integration promises reduced maintenance cost, seamless joint application, complete representation of health-related concepts while enabling better interoperability with other informatics infrastructures.

Harmonization of ICF Body Structures and ICD-11 Anatomic Detail: One foundation for multiple classifications.

The Family of International Classifications of the World Health Organization (WHO-FIC) currently includes three reference classifications, namely International Classification of Diseases (ICD), International Classification of Functioning, Disability, and Health (ICF), and International Classification of Health Interventions (ICHI). Recently, the three classifications have been incorporated into a single WHO-FIC Foundation that serves as the repository of all concepts in the classifications. Each classification serves a specific classification need. However, they share some common concepts that are present, in different forms, in two or all of them. For the WHO-FIC Foundation to be a logically consistent repository without duplicates, these common concepts must be reconciled. One important set of shared concepts is the representation of human anatomy entities, which are not always modeled in the same way and with the same level of detail. To understand the relationships among the three anatomical representations, an effort is needed to compare them, identifying common areas, gaps, and compatible and incompatible modeling. The work presented here contributes to this effort, focusing on the anatomy representations in ICF and ICD-11. For this aim, three experts were asked to identify, for each entity in the ICF Body Structures, one or more entities in the ICD-11 Anatomic Detail that could be considered identical, broader or narrower. To do this, they used a specifically developed web application, which also automatically identified the most obvious equivalences. A total of 631 maps were independently identified by the three mappers for 218 ICF Body Structures, with an interobserver agreement of 93.5%. Together with 113 maps identified by the software, they were then consolidated into 434 relations. The results highlight some differences between the two classifications: in general, ICF is less detailed than ICD-11; ICF favors lumping of structures; in very few cases, the two classifications follow different anatomic models. For these issues, solutions have to be found that are compliant with the WHO approach to classification modeling and maintenance.

Structural Patterns in Chronic Disease Clinical Practice Guidelines Formalized for Clinical Decision Support.

Making recommendations from clinical practice guidelines (CPGs) computable for clinical decision support (CDS) has typically been a laborious and costly process. Identifying domain-specific regularities helps clinicians and knowledge engineers conceptualize, extract, and encode evidence-based recommendations. Based on our work to provide complex CDS in the management of multiple chronic diseases, we propose nine chronic disease CPG structural patterns, discuss considerations in representing the necessary knowledge, and illustrate them with the solutions that our CDS system provides.

A practical method for transforming free-text eligibility criteria into computable criteria.

Formalizing eligibility criteria in a computer-interpretable language would facilitate eligibility determination for study subjects and the identification of studies on similar patient populations. Because such formalization is extremely labor intensive, we transform the problem from one of fully capturing the semantics of criteria directly in a formal expression language to one of annotating free-text criteria in a format called ERGO annotation. The annotation can be done manually, or it can be partially automated using natural-language processing techniques. We evaluated our approach in three ways. First, we assessed the extent to which ERGO annotations capture the semantics of 1000 eligibility criteria randomly drawn from ClinicalTrials.gov. Second, we demonstrated the practicality of the annotation process in a feasibility study. Finally, we demonstrate the computability of ERGO annotation by using it to (1) structure a library of eligibility criteria, (2) search for studies enrolling specified study populations, and (3) screen patients for potential eligibility for a study. We therefore demonstrate a new and practical method for incrementally capturing the semantics of free-text eligibility criteria into computable form.

Towards a framework for comparing functionalities of multimorbidity clinical decision support: A literature-based feature set and benchmark cases.

Multimorbidity, the coexistence of two or more health conditions, has become more prevalent as mortality rates in many countries have declined and their populations have aged. Multimorbidity presents significant difficulties for Clinical Decision Support Systems (CDSS), particularly in cases where recommendations from relevant clinical guidelines offer conflicting advice. A number of research groups are developing computer-interpretable guideline (CIG) modeling formalisms that integrate recommendations from multiple Clinical Practice Guidelines (CPGs) for knowledge-based multimorbidity decision support. In this paper we describe work towards the development of a framework for comparing the different approaches to multimorbidity CIG-based clinical decision support (MGCDS). We present (1) a set of features for MGCDS, which were derived using a literature review and evaluated by physicians using a survey, and (2) a set of benchmarking case studies, which illustrate the clinical application of these features. This work represents the first necessary step in a broader research program aimed at the development of a benchmark framework that allows for standardized and comparable MGCDS evaluations, which will facilitate the assessment of functionalities of MGCDS, as well as highlight important gaps in the state-of-the-art. We also outline our future work on developing the framework, specifically, (3) a standard for reporting MGCDS solutions for the benchmark case studies, and (4) criteria for evaluating these MGCDS solutions. We plan to conduct a large-scale comparison study of existing MGCDS based on the comparative framework.

Formal representation of eligibility criteria: a literature review.

Standards-based, computable knowledge representations for eligibility criteria are increasingly needed to provide computer-based decision support for automated research participant screening, clinical evidence application, and clinical research knowledge management. We surveyed the literature and identified five aspects of eligibility criteria knowledge representation that contribute to the various research and clinical applications: the intended use of computable eligibility criteria, the classification of eligibility criteria, the expression language for representing eligibility rules, the encoding of eligibility concepts, and the modeling of patient data. We consider three of these aspects (expression language, codification of eligibility concepts, and patient data modeling) to be essential constructs of a formal knowledge representation for eligibility criteria. The requirements for each of the three knowledge constructs vary for different use cases, which therefore should inform the development and choice of the constructs toward cost-effective knowledge representation efforts. We discuss the implications of our findings for standardization efforts toward knowledge representation for sharable and computable eligibility criteria.

Evaluation of the acceptability and usability of a decision support system to encourage safe and effective use of opioid therapy for chronic, noncancer pain by primary care providers.

OBJECTIVE: To develop and evaluate a clinical decision support system (CDSS) named Assessment and Treatment in Healthcare: Evidenced-Based Automation (ATHENA)-Opioid Therapy, which encourages safe and effective use of opioid therapy for chronic, noncancer pain. DESIGN: CDSS development and iterative evaluation using the analysis, design, development, implementation, and evaluation process including simulation-based and in-clinic assessments of usability for providers followed by targeted system revisions. RESULTS: Volunteers provided detailed feedback to guide improvements in the graphical user interface, and content and design changes to increase clinical usefulness, understandability, clinical workflow fit, and ease of completing guideline recommended practices. Revisions based on feedback increased CDSS usability ratings over time. Practice concerns outside the scope of the CDSS were also identified. CONCLUSIONS: Usability testing optimized the CDSS to better address barriers such as lack of provider education, confusion in dosing calculations and titration schedules, access to relevant patient information, provider discontinuity, documentation, and access to validated assessment tools. It also highlighted barriers to good clinical practice that are difficult to address with CDSS technology in its current conceptualization. For example, clinicians indicated that constraints on time and competing priorities in primary care, discomfort in patient-provider communications, and lack of evidence to guide opioid prescribing decisions impeded their ability to provide effective, guideline-adherent pain management. Iterative testing was essential for designing a highly usable and acceptable CDSS; however, identified barriers may limit the impact of the ATHENA-Opioid Therapy system and other CDSS on clinical practices and outcomes unless CDSS are paired with parallel initiatives to address these issues.

Toward a Harmonized WHO Family of International Classifications Content Model.

An overarching WHO-FIC Content Model will allow uniform modeling of classifications in the WHO Family of International Classifications (WHO-FIC) and promote their joint use. We provide an initial conceptualization of such a model.

Understanding detection performance in public health surveillance: modeling aberrancy-detection algorithms.

OBJECTIVE: Statistical aberrancy-detection algorithms play a central role in automated public health systems, analyzing large volumes of clinical and administrative data in real-time with the goal of detecting disease outbreaks rapidly and accurately. Not all algorithms perform equally well in terms of sensitivity, specificity, and timeliness in detecting disease outbreaks and the evidence describing the relative performance of different methods is fragmented and mainly qualitative. DESIGN: We developed and evaluated a unified model of aberrancy-detection algorithms and a software infrastructure that uses this model to conduct studies to evaluate detection performance. We used a task-analytic methodology to identify the common features and meaningful distinctions among different algorithms and to provide an extensible framework for gathering evidence about the relative performance of these algorithms using a number of evaluation metrics. We implemented our model as part of a modular software infrastructure (Biological Space-Time Outbreak Reasoning Module, or BioSTORM) that allows configuration, deployment, and evaluation of aberrancy-detection algorithms in a systematic manner. MEASUREMENT: We assessed the ability of our model to encode the commonly used EARS algorithms and the ability of the BioSTORM software to reproduce an existing evaluation study of these algorithms. RESULTS: Using our unified model of aberrancy-detection algorithms, we successfully encoded the EARS algorithms, deployed these algorithms using BioSTORM, and were able to reproduce and extend previously published evaluation results. CONCLUSION: The validated model of aberrancy-detection algorithms and its software implementation will enable principled comparison of algorithms, synthesis of results from evaluation studies, and identification of surveillance algorithms for use in specific public health settings.

Goal-driven management of interacting clinical guidelines for multimorbidity patients.

Computer-interpretable guidelines (CIGs) are based on clinical practice guidelines, which typically address a single morbidity. However, most of the aging population suffers from multiple morbidities. Currently, there is no demonstrated effective mechanism that integrates recommendations from multiple CIGs. We are developing a goal-based method that utilizes knowledge of drugs' physiological effects and therapeutic usage to combine knowledge from CIGs. It incrementally detects interactions and plans non-contradicting therapies. Our algorithm uses patterns to check consistency and respond to events, including data enquiries, diagnoses, adverse events, recommended medications, tests, and goals. Our method utilizes existing standards and CIG tools, including the Fast Healthcare Interoperability Resources (FHIR) patient data model, SNOMED-CT, and the PROforma CIG formalism with its Alium knowledge-engineering environment and PROforma enactment engine. We demonstrate our approach using a case study involving two clinical guidelines with templates for responding to a new goal and to a medication request that causes an inconsistency which can be automatically detected and resolved based on the knowledge of the two CIGs.

Selecting Test Cases from the Electronic Health Record for Software Testing of Knowledge-Based Clinical Decision Support Systems.

Software testing of knowledge-based clinical decision support systems is challenging, labor intensive, and expensive; yet, testing is necessary since clinical applications have heightened consequences. Thoughtful test case selection improves testing coverage while minimizing testing burden. ATHENA-CDS is a knowledge-based system that provides guideline-based recommendations for chronic medical conditions. Using the ATHENA-CDS diabetes knowledgebase, we demonstrate a generalizable approach for selecting test cases using rules/ filters to create a set of paths that mimics the system's logic. Test cases are allocated to paths using a proportion heuristic. Using data from the electronic health record, we found 1,086 cases with glycemic control above target goals. We created a total of 48 filters and 50 unique system paths, which were used to allocate 200 test cases. We show that our method generates a comprehensive set of test cases that provides adequate coverage for the testing of a knowledge-based CDS.

The SAGE Guideline Model: achievements and overview.

The SAGE (Standards-Based Active Guideline Environment) project was formed to create a methodology and infrastructure required to demonstrate integration of decision-support technology for guideline-based care in commercial clinical information systems. This paper describes the development and innovative features of the SAGE Guideline Model and reports our experience encoding four guidelines. Innovations include methods for integrating guideline-based decision support with clinical workflow and employment of enterprise order sets. Using SAGE, a clinician informatician can encode computable guideline content as recommendation sets using only standard terminologies and standards-based patient information models. The SAGE Model supports encoding large portions of guideline knowledge as re-usable declarative evidence statements and supports querying external knowledge sources.

Creating interoperable guidelines: requirements of vocabulary standards in immunization decision support.

Interoperable support of electronic health records and clinical decision support technology are central to the vision of sustainable information infrastructure. Efforts to implement interoperable clinical guidelines for immunization practice have been sparse. We used the SAGE knowledge workbench to develop a knowledge base to provide immunization decision support in primary care. We translated the written clinical guideline into a structured decision logic format. The semantic content to completely capture CDC clinical decision logic required 197 separate concepts but was completely captured with SNOMED CT and LOINC. Although 88% of concepts employed precoordinated codes, 6% of guideline concepts required expanded vocabulary services employing Boolean logical definition using two or more SNOMED concepts. Postcoordination requirements were modest, representing just 6% of guideline semantic concepts. We conclude that creation of interoperable knowledge bases employing clinical vocabulary standards is achievable and realistic. Employment of information model (HL7 RIM) and vocabulary (SNOMED CT, LOINC) standards is a necessary and feasible requirement to achieve interoperability in clinical decision support.

High-Risk Drug-Drug Interactions Between Clinical Practice Guidelines for Management of Chronic Conditions.

Clinicians and clinical decision-support systems often follow pharmacotherapy recommendations for patients based on clinical practice guidelines (CPGs). In multimorbid patients, these recommendations can potentially have clinically significant drug-drug interactions (DDIs). In this study, we describe and validate a method for programmatically detecting DDIs among CPG recommendations. The system extracts pharmacotherapy intervention recommendations from narrative CPGs, normalizes the terms, creates a mapping of drugs and drug classes, and then identifies occurrences of DDIs between CPG pairs. We used this system to analyze 75 CPGs written by authoring entities in the United States that discuss outpatient management of common chronic diseases. Using a reference list of high-risk DDIs, we identified 2198 of these DDIs in 638 CPG pairs (46 unique CPGs). Only 9 high-risk DDIs were discussed by both CPGs in a pairing. In 69 of the pairings, neither CPG had a pharmacologic reference or a warning of the possibility of a DDI.