Pragmatic Considerations on Clinical Decision Support from the 2019 Literature.

OBJECTIVES: To summarize significant research contributions published in 2019 in the field of computerized clinical decision support and select the best papers for the Decision Support section of the International Medical Informatics Association (IMIA) Yearbook. METHODS: Two bibliographic databases were searched for papers referring to clinical decision support systems (CDSSs) and computerized provider order entry (CPOE) systems. From search results, section editors established a list of candidate best papers, which were then peer-reviewed by external reviewers. The IMIA Yearbook editorial committee finally selected the best papers on the basis of all reviews including the section editors' evaluation. RESULTS: A total of 1,378 articles were retrieved. Fifteen best paper candidates were selected, the reviews of which resulted in the selection of three best papers. One paper reports on a guideline modeling approach based on clinical decision trees, both clinically interpretable and suitable for implementation in CDSSs. In another paper, authors promote the use of extended Timed Transition Diagrams in CDSSs to formalize consistently recurrent medical processes for chronic diseases management. The third paper proposes a conceptual framework and a grid for assessing the performance of predictive tools based on the critical appraisal of published evidence. CONCLUSIONS: As showed by the number and the variety of works related to decision support, research in the field is very active. This year's selection highlighted pragmatic works that promote transparency and trust required by decision support tools.

Contributions from the 2017 Literature on Clinical Decision Support.

OBJECTIVES: To summarize recent research and select the best papers published in 2017 in the field of computerized clinical decision support for the Decision Support section of the International Medical Informatics Association (IMIA) yearbook. METHODS: A literature review was performed by searching two bibliographic databases for papers referring to clinical decision support systems (CDSSs). The aim was to identify a list of candidate best papers from the retrieved bibliographic records, which were then peer-reviewed by external reviewers. A consensus meeting of the IMIA editorial team finally selected the best papers on the basis of all reviews and the section editors' evaluation. RESULTS: Among the 1,194 retrieved papers, the entire review process resulted in the selection of four best papers. The first paper studies the impact of recency and of longitudinal extent of electronic health record (EHR) datasets used to train a data-driven predictive model of inpatient admission orders. The second paper presents a decision support tool for surgical team selection, relying on the history of surgical team members and the specific characteristics of the patient. The third paper compares three commercial drug-drug interaction knowledge bases, particularly against a reference list of highly-significant known interactions. The fourth paper focuses on supporting the diagnosis of postoperative delirium using an adaptation of the "anchor and learn" framework, which was applied in unstructured texts contained in EHRs. CONCLUSIONS: The conducted review illustrated also this year that research in the field of CDSS is very active. Of note is the increase in publications concerning data-driven CDSSs, as revealed by the review process and also reflected by the four papers that have been selected. This trend is in line with the current attention that "Big Data" and data-driven artificial intelligence have gained in the domain of health and CDSSs in particular.

Contributions from the 2016 Literature on Clinical Decision Support.

Objectives: To summarize recent research and select the best papers published in 2016 in the field of computerized clinical decision support for the Decision Support section of the IMIA yearbook. Methods: A literature review was performed by searching two bibliographic databases for papers related to clinical decision support systems (CDSSs). The aim was to identify a list of candidate best papers from the retrieved papers that were then peer-reviewed by external reviewers. A consensus meeting of the IMIA editorial team finally selected the best papers on the basis of all reviews and section editor evaluation. Results: Among the 1,145 retrieved papers, the entire review process resulted in the selection of four best papers. The first paper describes machine learning models used to predict breast cancer multidisciplinary team decisions and compares them with two predictors based on guideline knowledge. The second paper introduces a linked-data approach for publication, discovery, and interoperability of CDSSs. The third paper assessed the variation in high-priority drug-drug interaction (DDI) alerts across 14 Electronic Health Record systems, operating in different institutions in the US. The fourth paper proposes a generic framework for modeling multiple concurrent guidelines and detecting their recommendation interactions using semantic web technologies. Conclusions: The process of identifying and selecting best papers in the domain of CDSSs demonstrated that the research in this field is very active concerning diverse dimensions, such as the types of CDSSs, e.g. guideline-based, machine-learning-based, knowledge-fusion-based, etc., and addresses challenging areas, such as the concurrent application of multiple guidelines for comorbid patients, the resolution of interoperability issues, and the evaluation of CDSSs. Nevertheless, this process also showed that CDSSs are not yet fully part of the digitalized healthcare ecosystem. Many challenges remain to be faced with regard to the evidence of their output, the dissemination of their technologies, as well as their adoption for better and safer healthcare delivery.

Computerized Clinical Decision Support: Contributions from 2015.

OBJECTIVE: To summarize recent research and select the best papers published in 2015 in the field of computerized clinical decision support for the Decision Support section of the IMIA yearbook. METHOD: A literature review was performed by searching two bibliographic databases for papers related to clinical decision support systems (CDSSs) and computerized provider order entry (CPOE) systems. The aim was to identify a list of candidate best papers from the retrieved papers that were then peer-reviewed by external reviewers. A consensus meeting between the two section editors and the IMIA editorial team was finally conducted to conclude in the best paper selection. RESULTS: Among the 974 retrieved papers, the entire review process resulted in the selection of four best papers. One paper reports on a CDSS routinely applied in pediatrics for more than 10 years, relying on adaptations of the Arden Syntax. Another paper assessed the acceptability and feasibility of an important CPOE evaluation tool in hospitals outside the US where it was developed. The third paper is a systematic, qualitative review, concerning usability flaws of medication-related alerting functions, providing an important evidence-based, methodological contribution in the domain of CDSS design and development in general. Lastly, the fourth paper describes a study quantifying the effect of a complex, continuous-care, guideline-based CDSS on the correctness and completeness of clinicians' decisions. CONCLUSIONS: While there are notable examples of routinely used decision support systems, this 2015 review on CDSSs and CPOE systems still shows that, despite methodological contributions, theoretical frameworks, and prototype developments, these technologies are not yet widely spread (at least with their full functionalities) in routine clinical practice. Further research, testing, evaluation, and training are still needed for these tools to be adopted in clinical practice and, ultimately, illustrate the benefits that they promise.

Computerized Clinical Decision Support: Contributions from 2014.

OBJECTIVE: To summarize recent research and propose a selection of best papers published in 2014 in the field of computerized clinical decision support for the Decision Support section of the IMIA yearbook. METHOD: A literature review was performed by searching two bibliographic databases for papers related to clinical decision support systems (CDSSs) and computerized provider order entry systems in order to select a list of candidate best papers to be then peer-reviewed by external reviewers. A consensus meeting between the two section editors and the editorial team was finally organized to conclude on the selection of best papers. RESULTS: Among the 1,254 returned papers published in 2014, the full review process selected four best papers. The first one is an experimental contribution to a better understanding of unintended uses of CDSSs. The second paper describes the effective use of previously collected data to tailor and adapt a CDSS. The third paper presents an innovative application that uses pharmacogenomic information to support personalized medicine. The fourth paper reports on the long-term effect of the routine use of a CDSS for antibiotic therapy. CONCLUSIONS: As health information technologies spread more and more meaningfully, CDSSs are improving to answer users' needs more accurately. The exploitation of previously collected data and the use of genomic data for decision support has started to materialize. However, more work is still needed to address issues related to the correct usage of such technologies, and to assess their effective impact in the long term.

Toward a formalization of the process to select IMIA Yearbook best papers.

BACKGROUND: Each year, the International Medical Informatics Association Yearbook recognizes significant scientific papers, labelled as "best papers", published the previous year in the subfields of biomedical informatics that correspond to the different section topics of the journal. For each section, about fifteen pre-selected "candidate" best papers are externally peer-reviewed to select the actual best papers. Although based on the available literature, little is known about the pre-selection process. OBJECTIVE: To move toward an explicit formalization of the candidate best papers selection process to reduce variability in the literature search across sections and over years. METHODS: A methodological framework is proposed to build for each section topic specific queries tailored to PubMed and Web of Science citation databases. The two sets of returned papers are merged and reviewed by two independent section editors and citations are tagged as "discarded", "pending", and "kept". A protocolized consolidation step is then jointly conducted to resolve conflicts. A bibliographic software tool, BibReview, was developed to support the whole process. RESULTS: The proposed search strategy was fully applied to the Decision Support section of the 2013 edition of the Yearbook. For this section, 1124 references were returned (689 PubMed-specific, 254 WoS-specific, 181 common to both databases) among which the 15 candidate best papers were selected. CONCLUSIONS: The search strategy for determining candidate best papers for an IMIA Yearbook's section is now explicitly specified and allows for reproducibility. However, some aspects of the whole process remain reviewer-dependent, mostly because there is no characterization of a "best paper".

A 2014 medical informatics perspective on clinical decision support systems: do we hit the ceiling of effectiveness?

OBJECTIVE: To summarize recent research and propose a selection of best papers published in 2013 in the field of computer-based decision support in health care. METHOD: Two literature reviews were performed by the two section editors from bibliographic databases with a focus on clinical decision support systems (CDSSs) and computer provider order entry in order to select a list of candidate best papers to be peer-reviewed by external reviewers. RESULTS: The full review process highlighted three papers, illustrating current trends in the domain of clinical decision support. The first trend is the development of theoretical approaches for CDSSs, and is exemplified by a paper proposing the integration of family histories and pedigrees in a CDSS. The second trend is illustrated by well-designed CDSSs, showing good theoretical performances and acceptance, while failing to show a clinical impact. An example is given with a paper reporting on scorecards aiming to reduce adverse drug events. The third trend is represented by research works that try to understand the limits of CDSS use, for instance by analyzing interactions between general practitioners, patients, and a CDSS. CONCLUSIONS: CDSSs can achieve good theoretical results in terms of sensibility and specificity, as well as a good acceptance, but evaluations often fail to demonstrate a clinical impact. Future research is needed to better understand the causes of this observation and imagine new effective solutions for CDSS implementation.

Which breast cancer decisions remain non-compliant with guidelines despite the use of computerised decision support?

BACKGROUND: Despite multidisciplinary tumour boards (MTBs), non-compliance with clinical practice guidelines is still observed for breast cancer patients. Computerised clinical decision support systems (CDSSs) may improve the implementation of guidelines, but cases of non-compliance persist. METHODS: OncoDoc2, a guideline-based decision support system, has been routinely used to remind MTB physicians of patient-specific recommended care plans. Non-compliant MTB decisions were analysed using a multivariate adjusted logistic regression model. RESULTS: Between 2007 and 2009, 1624 decisions for invasive breast cancers with a global non-compliance rate of 8.3% were analysed. Patient factors associated with non-compliance were age>80 years (odds ratio (OR): 7.7; 95% confidence interval (CI): 3.7-15.7) in pre-surgical decisions; microinvasive tumour (OR: 5.2; 95% CI: 1.5-17.5), surgical discovery of microinvasion in addition to a unique invasive tumour (OR: 4.2; 95% CI: 1.4-12.5), and prior neoadjuvant treatment (OR: 4.2; 95% CI: 1.1-15.1) in decisions with recommendation of re-excision; age<35 years (OR: 4.7; 95% CI: 1.9-11.4), positive hormonal receptors with human epidermal growth factor receptor 2 overexpression (OR: 15.7; 95% CI: 3.1-78.7), and the absence of prior axillary surgery (OR: 17.2; 95% CI: 5.1-58.1) in adjuvant decisions. CONCLUSION: Residual non-compliance despite the use of OncoDoc2 illustrates the need to question the clinical profiles where evidence is missing. These findings challenge the weaknesses of guideline content rather than the use of CDSSs.

A medical informatics perspective on clinical decision support systems. Findings from the yearbook 2013 section on decision support.

OBJECTIVE: To summarize excellent research and to select best papers published in 2012 in the field of computer-based decision support in healthcare. METHODS: A bibliographic search focused on clinical decision support systems (CDSSs) and computer provider order entry was performed, followed by a double-blind literature review. RESULTS: The review process yielded six papers, illustrating various aspects of clinical decision support. The first paper is a systematic review of CDSS intervention trials in real settings, and considers different types of possible outcomes. It emphasizes the heterogeneity of studies and confirms that CDSSs can improve process measures but that evidence lacks for other types of outcomes, especially clinical or economic. Four other papers tackle the safety of drug prescribing and show that CDSSs can be efficient in reducing prescription errors. The sixth paper exemplifies the growing role of ontological resources which can be used for several applications including decision support. CONCLUSIONS: CDSS research has to be continuously developed and assessed. The wide variety of systems and of interventions limits the understanding of factors of success of CDSS implementations. A standardization in the characterization of CDSSs and of intervention trial reporting will help to overcome this obstacle.

Why GPs do not follow computerized guidelines: an attempt of explanation involving usability with ASTI guiding mode.

Clinical decision support systems (CDSSs) have the potential to increase guideline adherence, but factors of success are not well understood. ASTI-GM is an on demand guideline-based CDSS where the user interactively characterizes her patient by browsing the system knowledge base to obtain the recommended treatment. We conducted a web-based evaluation of ASTI-GM as a before-after study to assess whether the system improves general practitioners' (GPs) performance and how they would use it. Five clinical cases had to be solved, as usual in the before phase, and using ASTI-GM in the after phase. On a 2-month period, 266 GPs participated and 1,981 prescription orders were collected. The overall guideline adherence rate increased from 27.2% to 64.3%. Only 56.4% of ASTI-GM uses corresponded to a "good use" of the system. Adherence increased from 28.5% to 86.1% in the sub-group of "good uses", whereas it only increased from 28.1% to 36.6% in the complementary sub-group. Reasons for non "good uses" of CDSSs should be investigated since they impede their potential impact.

Role of physicians' reactance in e-iatrogenesis: a case study with ASTI guiding mode on the management of hypertension.

Clinical decision support systems (CDSSs) have the potential to increase guideline adherence, but factors of success are not yet understood. ASTI guiding mode (ASTI-GM) is an on-demand guideline-based CDSS where the user navigates in a knowledge base to get the best treatment for a given patient. We conducted a web-based evaluation of ASTI-GM, carried out as a before-after study, where general practitioners (GPs) were asked to solve 5 clinical cases, first without ASTI-GM, then using the system. Of the 136 GPs that resolved the case on the management of hypertension, compliance with best practices increased from 69.1% to 80.9% with ASTI-GM. When the navigation matched the set of patient parameters described in the clinical case, the increase was even higher and reached 92.9%. E-iatrogenesis has been measured at 19.1%, with 5.1% of commission errors, 8.1% of negative reactance, and 5.9% of neutral reactance. Role of physicians' reactance in noncompliance with guideline-based CDSSs should be further investigated.

Consequences of the verification of completeness in clinical practice guideline modeling: a theoretical and empirical study with hypertension.

Building clinical decision support systems requires a formalization of clinical practice guidelines (CPGs) including the verification of completeness to ensure all medically relevant situations are addressed. Recommendations that rely on completed knowledge cannot be but expert-based. Using French hypertension management guidelines, we characterized the status of a patient profile as evidence-based (EB), consensus-based (CB), or expert-based (XB). The distribution of these status on the formal patient profiles of ASTIGM knowledge base showed that 12.6% (0.5% EB and 12.1% CB) lead to explicit CPG recommendations. The same analysis on a sample of 435 actual patients medical records showed that 55% were covered by CPGs. The characterization of guideline-based CDSSs should be based on empirical data estimated from the target population of CPGs.

[PneumoDoc: a computer-based decision-making system for drug-related pulmonary disease]. / Formalisation de la démarche diagnostique des pneumopathies médicamenteuses: le système PneumoDoc.

Establishing the diagnosis of drug-related pulmonary disease (DRPD) remains a difficult task because of the large number of drug-related toxic situations and the variety of clinical presentations. PneumoDoc is a computer-based support system designed to facilitate the diagnosis of lung disease using chronological, clinical, imaging, and cytological (alveolar lavage) input. These intrinsic items are crosschecked against extrinsic items reported in the literature (Pneumotox). Data input is in the form of yes-no questions. The final output displays the characteristic features of the observed clinical situation and calculates the probability of DRPD defined in five categories: incompatible, doubtful, compatible, suggestive, and highly suggestive. Use of multiple drugs, interaction with cardiopulmonary disease, and the absence of reported cases are limitations of the system.

Supporting multidisciplinary staff meetings for guideline-based breast cancer management: a study with OncoDoc2.

In order to reduce practice variations and offer cancer patients the best treatments according to reference guidelines, therapeutic decisions have to be taken, in France, by "multidisciplinary staff meetings'' (MSMs) as patient-specific care plans which are then implemented by cancer specialists. OncoDoc2 is a CDSS implementing CancerEst guidelines, a "local reference guideline'', on breast cancer management. The system has been assessed in a pragmatic before/after study. The intervention consisted in the routine use of OncoDoc2 during MSMs of Tenon hospital. The MSM decision compliance rate with the reference guideline was significantly higher in the after period, increasing from 79% to 93%. MSM decision analysis showed that missing steps in treatment plans were the main cause of noncompliance during the before period. This cause was drastically reduced in the after period.

Design factors for success or failure of guideline-based decision support systems: an hypothesis involving case complexity.

Computer-based decision support systems (CDSSs) are currently mostly reminder systems. However, the effectiveness of such systems to modify physician behavior is not always observed. We assume that this approach is appropriate when physicians think they know how to prescribe and consider they don't need to be helped, i.e. for simple clinical cases. On the opposite, on-demand approaches allowing for flexibility in the interpretation of patient conditions are more appropriate for more complex cases, e.g. in chronic disease management. ASTI is a CDSS operating in two modes, a critiquing mode working as a reminder-based system and a user-initiated guiding mode. Using a clinical case complexity score, a pre/post-intervention experiment with 10 GPs and 15 cases of hypertensive patients has been performed. Preliminary results tend to indicate that reminder-based interaction is appropriate for simple cases and that physicians are willing to use on-demand systems as clinical situations become complex, making both modes complementary.

Guideline-based modeling of therapeutic strategies in the special case of chronic diseases.

Despite the availability of evidence-based clinical practice guidelines in most countries, patients with chronic diseases are still generally inadequately managed. One difficulty lies in the optimal synchronization of a patient with the guideline therapeutic strategy, especially when the history of her past treatments does not follow the recommended sequence of therapies. We propose a formal model to represent guideline-based therapeutic strategies as a two-level decision tree. The clinical level is used to identify a patient-specific clinical situation, on the basis of key elements of clinical examination (complication of hypertension, associated diseases). The therapeutic level is derived from the formalization of guideline-based strategies first represented as bidimensional matrices structured in lines of therapy and levels of therapeutic intention. A revised version based on the ordering of levels of therapeutic combination is then developed. The aim is to dynamically provide the best next step of treatment from the patient's therapeutic-history-based customization of the general therapeutic sequence established in the guideline for the corresponding clinical situation. A preliminary in vitro evaluation of the system on actual clinical cases showed a positive impact on physician compliance with a significant increase from 16 to 57%.

Reminder-based or on-demand decision support systems: a preliminary study in primary care with the management of hypertension.

ASTI is a guideline-based decision support system for therapeutic prescribing in primary care with two modes of interaction. The "critic mode" operates as a reminder system to detect non guideline-compliant physician drug orders, whereas the "guided mode" operates on demand and provides physician guidance to help her establishing best recommended drug prescriptions for the management of hypertension. A preliminary evaluation study was conducted with 10 GPs to test the complementary nature of both modes of decision support. Results tend to validate our assumption that reminder-based interaction is appropriate for simple cases and that physicians are willing to use on-demand systems as clinical situations become more complex.

Automatic generation of a metamodel from an existing knowledge base to assist the development of a new analogous knowledge base.

Knowledge acquisition is a key step in the development of knowledge-based systems and methods have been proposed to help elicitating a domain-specific task model from a generic task model. We explored how an existing validated knowledge base (KB) represented by a decision tree could be automatically processed to infer a higher level domain-specific task model. On-codoc is a guideline-based decision support system applied to breast cancer therapy. Assuming task identity and ontological proximity between breast and lung cancer domains, the generalization of the breast can-cer KB should allow to build a metamodel to serve as a guide for the elaboration of a new specific KB on lung cancer. Two types of parametrized generalization methods based on tree structure simplification and ontological abstraction were used. We defined a similarity distance and a generalization coefficient to select the best metamodel identified as the closest to the original decision tree of the most generalized metamodels.

A before-after study using OncoDoc, a guideline-based decision support-system on breast cancer management: impact upon physician prescribing behaviour.

Guideline-based decision support systems have been developed to influence the prescribing behaviour of clinicians, but they have not yet shown to increase physician compliance with best practices in routine. OncoDoc is a non-automated system that allows flexibility in guideline interpretation to obtain best patient-specific recommendations at the point of care. OncoDoc is applied to breast cancer management. We have experimented the system at the Institut Gustave Roussy with a before-after study in which treatment decisions for breast cancer patients were measured before and after using the system in order to evaluate its impact upon physicians' prescribing behaviour. After 4 months, 127 decisions were recorded. Physicians compliance with OncoDoc was significantly improved (p < 10(-4) ) to reach 85.03% after using the system. Comparison of initial and final decisions showed that physicians modified their prescription in 31% of the cases. Clinical trial accrual rate increased of 50%, though not statistically significant because estimated on small figures.

ASTI: a guideline-based drug-ordering system for primary care.

Existing computer-based ordering systems for physicians provide effective drug-centered checks but offer little assistance for optimizing the overall patient-centered treatment strategy. Evidence-based clinical practice guidelines have been developed to disseminate state-of-the-art information concerning treatment strategy but these guidelines are poorly used in routine practice. The ASTI project aims to design a guideline-based ordering system to enable general practitioners to avoid prescription errors and to improve compliance with best therapeutic practices. The " critic mode " operates as a background process and corrects the physician's prescription on the basis of automatically triggered elementary rules that account for isolated guideline recommendations. The " guided mode " directs the physician to the best treatment by browsing a comprehensive guideline knowledge base represented as a decision tree. A first prototype, applied to hypertension, is currently under development.