Effect of clinical decision support for severe hypercholesterolemia on low-density lipoprotein cholesterol levels.

Severe hypercholesterolemia/possible familial hypercholesterolemia (FH) is relatively common but underdiagnosed and undertreated. We investigated whether implementing clinical decision support (CDS) was associated with lower low-density lipoprotein cholesterol (LDL-C) in patients with severe hypercholesterolemia/possible FH (LDL-C ≥ 190 mg/dL). As part of a pre-post implementation study, a CDS alert was deployed in the electronic health record (EHR) in a large health system comprising 3 main sites, 16 hospitals and 53 clinics. Data were collected for 3 months before ('silent mode') and after ('active mode') its implementation. Clinicians were only able to view the alert in the EHR during active mode. We matched individuals 1:1 in both modes, based on age, sex, and baseline lipid lowering therapy (LLT). The primary outcome was difference in LDL-C between the two groups and the secondary outcome was initiation/intensification of LLT after alert trigger. We identified 800 matched patients in each mode (mean ± SD age 56.1 ± 11.8 y vs. 55.9 ± 11.8 y; 36.0% male in both groups; mean ± SD initial LDL-C 211.3 ± 27.4 mg/dL vs. 209.8 ± 23.9 mg/dL; 11.2% on LLT at baseline in each group). LDL-C levels were 6.6 mg/dL lower (95% CI, -10.7 to -2.5; P = 0.002) in active vs. silent mode. The odds of high-intensity statin use (OR, 1.78; 95% CI, 1.41-2.23; P < 0.001) and LLT initiation/intensification (OR, 1.30, 95% CI, 1.06-1.58, P = 0.01) were higher in active vs. silent mode. Implementation of a CDS was associated with lowering of LDL-C levels in patients with severe hypercholesterolemia/possible FH, likely due to higher rates of clinician led LLT initiation/intensification.

Clinical Implementation of an AI Early Warning System Algorithm: Lessons Learned.

The Deterioration Index (DI) is an automatic early warning system that utilizes a machine learning algorithm integrated into the electronic health record and was implemented to improve risk stratification of inpatients. Our pilot implementation showed superior diagnostic accuracy than standard care. A score >60 had a specificity of 88.5% and a sensitivity of 59.8% (PPV 0.1758, NPP 0.9817). However, acceptance in the clinical workflow was divided; nurses preferred standard care, while providers found it helpful.

Decision Support System Detecting Patients at Risk of Prolonged QT and Associated Mortality.

Prolonged QT interval is an independent risk factor for all-cause mortality. However, evaluation of mortality associated to the implementation of a clinical decision support system to increase awareness and provide management recommendations has been challenging. Here we present our attempt to develop a model using only electronic data and different control groups.

Contrastive Transfer Learning for Prediction of Adverse Events in Hospitalized Patients.

OBJECTIVE: Deterioration index (DI) is a computer-generated score at a specific frequency that represents the overall condition of hospitalized patients using a variety of clinical, laboratory and physiologic data. In this paper, a contrastive transfer learning method is proposed and validated for early prediction of adverse events in hospitalized patients using DI scores. METHODS AND PROCEDURES: An unsupervised contrastive learning (CL) model with a classifier is proposed to predict adverse outcome using a single temporal variable (DI scores). The model is pretrained on an unsupervised fashion with large-scale time series data and fine-tuned with retrospective DI score data. RESULTS: The performance of this model is compared with supervised deep learning models for time series classification. Results show that unsupervised contrastive transfer learning with a classifier outperforms supervised deep learning solutions. Pretraining of the proposed CL model with large-scale time series data and fine-tuning that with DI scores can enhance prediction accuracy. CONCLUSION: A relationship exists between longitudinal DI scores of a patient and the corresponding outcome. DI scores and contrastive transfer learning can be used to predict and prevent adverse outcomes in hospitalized patients. CLINICAL IMPACT: This paper successfully developed an unsupervised contrastive transfer learning algorithm for prediction of adverse events in hospitalized patients. The proposed model can be deployed in hospitals as an early warning system for preemptive intervention in hospitalized patients, which can mitigate the likelihood of adverse outcomes.

Clinician Perspectives on Clinical Decision Support for Familial Hypercholesterolemia.

Familial Hypercholesterolemia (FH) is underdiagnosed in the United States. Clinical decision support (CDS) could increase FH detection once implemented in clinical workflows. We deployed CDS for FH at an academic medical center and sought clinician insights using an implementation survey. In November 2020, the FH CDS was deployed in the electronic health record at all Mayo Clinic sites in two formats: a best practice advisory (BPA) and an in-basket alert. Over three months, 104 clinicians participated in the survey (response rate 11.1%). Most clinicians (81%) agreed that CDS implementation was a good option for identifying FH patients; 78% recognized the importance of implementing the tool in practice, and 72% agreed it would improve early diagnosis of FH. In comparing the two alert formats, clinicians found the in-basket alert more acceptable (p = 0.036) and more feasible (p = 0.042) than the BPA. Overall, clinicians favored implementing the FH CDS in clinical practice and provided feedback that led to iterative refinement of the tool. Such a tool can potentially increase FH detection and optimize patient management.

Returning integrated genomic risk and clinical recommendations: The eMERGE study.

PURPOSE: Assessing the risk of common, complex diseases requires consideration of clinical risk factors as well as monogenic and polygenic risks, which in turn may be reflected in family history. Returning risks to individuals and providers may influence preventive care or use of prophylactic therapies for those individuals at high genetic risk. METHODS: To enable integrated genetic risk assessment, the eMERGE (electronic MEdical Records and GEnomics) network is enrolling 25,000 diverse individuals in a prospective cohort study across 10 sites. The network developed methods to return cross-ancestry polygenic risk scores, monogenic risks, family history, and clinical risk assessments via a genome-informed risk assessment (GIRA) report and will assess uptake of care recommendations after return of results. RESULTS: GIRAs include summary care recommendations for 11 conditions, education pages, and clinical laboratory reports. The return of high-risk GIRA to individuals and providers includes guidelines for care and lifestyle recommendations. Assembling the GIRA required infrastructure and workflows for ingesting and presenting content from multiple sources. Recruitment began in February 2022. CONCLUSION: Return of a novel report for communicating monogenic, polygenic, and family history-based risk factors will inform the benefits of integrated genetic risk assessment for routine health care.

Novel machine learning model to improve performance of an early warning system in hospitalized patients: a retrospective multisite cross-validation study.

Background: Threshold-based early warning systems (EWS) are used to predict adverse events (Aes). Machine learning (ML) algorithms that incorporate all EWS scores prior to an event may perform better in hospitalized patients. Methods: The deterioration index (DI) is a proprietary EWS. A threshold of DI >60 is used to predict a composite AE: all-cause mortality, cardiac arrest, transfer to intensive care, and evaluation by the rapid response team in practice. The DI scores were collected for adult patients (≥18 y-o) hospitalized on medical or surgical services during 8-23-2021 to 3-31-2022 from four different Mayo Clinic sites in the United States. A novel ML model was developed and trained on a retrospective cohort of hospital encounters. DI scores were represented in a high-dimensional space using random convolution kernels to facilitate training of a classifier and the area under the receiver operator characteristics curve (AUC) was calculated. Multiple time intervals prior to an AE were analyzed. A leave-one-out cross-validation protocol was used to evaluate performance across separate clinic sites. Findings: Three different classifiers were trained on 59,617 encounter-derived DI scores in high-dimensional feature space and the AUCs were compared to two threshold models. All three tested classifiers improved the AUC over the threshold approaches from 0.56 and 0.57 to 0.76, 0.85 and 0.94. Time interval analysis of the top performing classifier showed best accuracy in the hour before an event occurred (AUC 0.91), but prediction held up even in the 12 h before an AE (AUC 0.80 at minus 12 h, 0.81 at minus 9 h, 0.85 at minus 6 h, and 0.88 at minus 3 h before an AE). Multisite cross-validation using leave-one-out approach on data from four different clinical sites showed broad generalization performance of the top performing ML model with AUC of 0.91, 0.91, 0.95, and 0.91. Interpretation: A novel ML model that incorporates all the longitudinal DI scores prior to an AE in a hospitalized patient performs better at outcome prediction than the currently used threshold model. The use of clinical data, a generalized ML technique, and successful multisite cross-validation demonstrate the feasibility of our model in clinical implementation. Funding: No funding to report.

Implementation of preemptive DNA sequence-based pharmacogenomics testing across a large academic medical center: The Mayo-Baylor RIGHT 10K Study.

PURPOSE: The Mayo-Baylor RIGHT 10K Study enabled preemptive, sequence-based pharmacogenomics (PGx)-driven drug prescribing practices in routine clinical care within a large cohort. We also generated the tools and resources necessary for clinical PGx implementation and identified challenges that need to be overcome. Furthermore, we measured the frequency of both common genetic variation for which clinical guidelines already exist and rare variation that could be detected by DNA sequencing, rather than genotyping. METHODS: Targeted oligonucleotide-capture sequencing of 77 pharmacogenes was performed using DNA from 10,077 consented Mayo Clinic Biobank volunteers. The resulting predicted drug response-related phenotypes for 13 genes, including CYP2D6 and HLA, affecting 21 drug-gene pairs, were deposited preemptively in the Mayo electronic health record. RESULTS: For the 13 pharmacogenes of interest, the genomes of 79% of participants carried clinically actionable variants in 3 or more genes, and DNA sequencing identified an average of 3.3 additional conservatively predicted deleterious variants that would not have been evident using genotyping. CONCLUSION: Implementation of preemptive rather than reactive and sequence-based rather than genotype-based PGx prescribing revealed nearly universal patient applicability and required integrated institution-wide resources to fully realize individualized drug therapy and to show more efficient use of health care resources.

Pharmacogenomics testing in patients with liver transplant and potential impact on prospective management.

Aim: Pharmacogenomics (PGx) tests are performed on whole-blood or saliva specimens. In patients with a transplanted liver, PGx results may be discordant with hepatic drug metabolizing enzyme activity. We evaluate the incidence and impact of PGx testing in liver transplant recipients, detail potential errors and describe clinical decision support (CDS) solution implemented. Materials & methods: A retrospective cohort study of liver transplant recipients at Mayo Clinic who underwent PGx testing between 1 January 1996 and 7 October 2019 were characterized. Impact of a CDS solution was evaluated. Results: There were 129 PGx tests in 117 patients. PGx testing incidence increased before (per year incidence rate ratio = 1.45, 95% CI: 1.20-1.74, p < 0.001) and after transplant (incidence rate ratio = 1.48, 95% CI: 1.27-1.72, p < 0.001). Three erroneous PGx tests were avoided 6 months following CDS implementation. Conclusion: Incidence of PGx testing in liver transplant recipients is increasing, leading to erroneous therapeutic decisions. CDS interventions and education are needed to prevent errors.

A novel method for causal structure discovery from EHR data and its application to type-2 diabetes mellitus.

Modern AI-based clinical decision support models owe their success in part to the very large number of predictors they use. Safe and robust decision support, especially for intervention planning, requires causal, not associative, relationships. Traditional methods of causal discovery, clinical trials and extracting biochemical pathways, are resource intensive and may not scale up to the number and complexity of relationships sufficient for precision treatment planning. Computational causal structure discovery (CSD) from electronic health records (EHR) data can represent a solution, however, current CSD methods fall short on EHR data. This paper presents a CSD method tailored to the EHR data. The application of the proposed methodology was demonstrated on type-2 diabetes mellitus. A large EHR dataset from Mayo Clinic was used as development cohort, and another large dataset from an independent health system, M Health Fairview, as external validation cohort. The proposed method achieved very high recall (.95) and substantially higher precision than the general-purpose methods (.84 versus .29, and .55). The causal relationships extracted from the development and external validation cohorts had a high (81%) overlap. Due to the adaptations to EHR data, the proposed method is more suitable for use in clinical decision support than the general-purpose methods.

Association of BMI, comorbidities and all-cause mortality by using a baseline mortality risk model.

OBJECTIVE: The association of body mass index (BMI) and all-cause mortality is controversial, frequently referred to as a paradox. Whether the cause is metabolic factors or statistical biases is still controversial. We assessed the association of BMI and all-cause mortality considering a wide range of comorbidities and baseline mortality risk. METHODS: Retrospective cohort study of Olmsted County residents with at least one BMI measurement between 2000-2005, clinical data in the electronic health record and minimum 8 year follow-up or death within this time. The cohort was categorized based on baseline mortality risk: Low, Medium, Medium-high, High and Very-high. All-cause mortality was assessed for BMI intervals of 5 and 0.5 Kg/m2. RESULTS: Of 39,739 subjects (average age 52.6, range 18-89; 38.1% male) 11.86% died during 8-year follow-up. The 8-year all-cause mortality risk had a "U" shape with a flat nadir in all the risk groups. Extreme BMI showed higher risk (BMI <15 = 36.4%, 15 to <20 = 15.4% and ≥45 = 13.7%), while intermediate BMI categories showed a plateau between 10.6 and 12.5%. The increased risk attributed to baseline risk and comorbidities was more obvious than the risk based on BMI increase within the same risk groups. CONCLUSIONS: There is a complex association between BMI and all-cause mortality when evaluated including comorbidities and baseline mortality risk. In general, comorbidities are better predictors of mortality risk except at extreme BMIs. In patients with no or few comorbidities, BMI seems to better define mortality risk. Aggressive management of comorbidities may provide better survival outcome for patients with body mass between normal and moderate obesity.

Preferences for Updates on General Research Results: A Survey of Participants in Genomic Research from Two Institutions.

There is a need for multimodal strategies to keep research participants informed about study results. Our aim was to characterize preferences of genomic research participants from two institutions along four dimensions of general research result updates: content, timing, mechanism, and frequency. METHODS: We conducted a web-based cross-sectional survey that was administered from 25 June 2018 to 5 December 2018. RESULTS: 397 participants completed the survey, most of whom (96%) expressed a desire to receive research updates. Preferences with high endorsement included: update content (brief descriptions of major findings, descriptions of purpose and goals, and educational material); update timing (when the research is completed, when findings are reviewed, when findings are published, and when the study status changes); update mechanism (email with updates, and email newsletter); and update frequency (every three months). Hierarchical cluster analyses based on the four update preferences identified four profiles of participants with similar preference patterns. Very few participants in the largest profile were comfortable with budgeting less money for research activities so that researchers have money to set up services to send research result updates to study participants. CONCLUSION: Future studies may benefit from exploring preferences for research result updates, as we have in our study. In addition, this work provides evidence of a need for funders to incentivize researchers to communicate results to participants.

A Computational Method for Learning Disease Trajectories From Partially Observable EHR Data.

Diseases can show different courses of progression even when patients share the same risk factors. Recent studies have revealed that the use of trajectories, the order in which diseases manifest throughout life, can be predictive of the course of progression. In this study, we propose a novel computational method for learning disease trajectories from EHR data. The proposed method consists of three parts: first, we propose an algorithm for extracting trajectories from EHR data; second, three criteria for filtering trajectories; and third, a likelihood function for assessing the risk of developing a set of outcomes given a trajectory set. We applied our methods to extract a set of disease trajectories from Mayo Clinic EHR data and evaluated it internally based on log-likelihood, which can be interpreted as the trajectories' ability to explain the observed (partial) disease progressions. We then externally evaluated the trajectories on EHR data from an independent health system, M Health Fairview. The proposed algorithm extracted a comprehensive set of disease trajectories that can explain the observed outcomes substantially better than competing methods and the proposed filtering criteria selected a small subset of disease trajectories that are highly interpretable and suffered only a minimal (relative 5%) loss of the ability to explain disease progression in both the internal and external validation.

Genomic considerations for FHIR®; eMERGE implementation lessons.

Structured representation of clinical genetic results is necessary for advancing precision medicine. The Electronic Medical Records and Genomics (eMERGE) Network's Phase III program initially used a commercially developed XML message format for standardized and structured representation of genetic results for electronic health record (EHR) integration. In a desire to move towards a standard representation, the network created a new standardized format based upon Health Level Seven Fast Healthcare Interoperability Resources (HL7® FHIR®), to represent clinical genomics results. These new standards improve the utility of HL7® FHIR® as an international healthcare interoperability standard for management of genetic data from patients. This work advances the establishment of standards that are being designed for broad adoption in the current health information technology landscape.

Pharmacogenomics education and perceptions: is there a gap between internal medicine resident and attending physicians?

Aim: To determine if differences in self-reported pharmacogenomics knowledge, skills and perceptions exist between internal medicine residents and attending physicians. Materials & methods: Forty-six internal medicine residents and 54 attending physicians completed surveys. Thirteen participated in focus groups to explore themes emerging from the surveys. Results: Resident physicians reported a greater amount of pharmacogenomics training compared with attending physicians (48 vs 13%, p < 0.00012). No differences were found in self-reported knowledge, skills and perceptions. Conclusion: Both groups expressed pharmacogenomics was relevant to their current clinical practice; they should be able to provide information to patients and use to guide prescribing, but lacked sufficient education to be able to do so effectively. Practical approaches are needed to teach pharmacogenomics concepts and address point of care gaps.

An Implementation Science Framework to Develop a Clinical Decision Support Tool for Familial Hypercholesterolemia.

Electronic health record (EHR)-based clinical decision support (CDS) can address the low awareness and undertreatment of familial hypercholesterolemia (FH), a disorder associated with a markedly increased risk of coronary heart disease. We aimed to incorporate provider perspectives into the development and implementation of a CDS tool for FH. An implementation science framework and a user-centered design process were used to create a CDS tool for FH. Primary care physicians and specialist physicians participated in qualitative interviews, usability testing and an implementation survey. The CDS was configured in two formats-a best practice alert (BPA) and an in-basket message and subsequently deployed in the EHR in silent mode. The key themes that emerged from the analysis of interview transcripts included understanding and awareness of FH, clinical workflow, physician preferences and value of CDS tools, perspectives on patient needs and values and dissemination and implementation. Recommendations related to usability included preferred CDS format and placement, content, timing and frequency, and level of alert urgency/prioritization. In response to the survey, 84.6% of physicians agreed that the CDS would improve early FH diagnosis and 92.3% agreed that it would help them identify and manage FH patients. Physician feedback led to iterative CDS refinement. In summary, we developed a CDS tool for FH using an implementation science framework and physician feedback. Initial deployment revealed a significant burden of FH and the potential for the CDS tool to have a large impact.

Development and Validation of a Risk Stratification Model Using Disease Severity Hierarchy for Mortality or Major Cardiovascular Event.

Importance: Clinical domain knowledge about diseases and their comorbidities, severity, treatment pathways, and outcomes can facilitate diagnosis, enhance preventive strategies, and help create smart evidence-based practice guidelines. Objective: To introduce a new representation of patient data called disease severity hierarchy that leverages domain knowledge in a nested fashion to create subpopulations that share increasing amounts of clinical details suitable for risk prediction. Design, Setting, and Participants: This retrospective cohort study included 51 969 patients aged 45 to 85 years, with 10 674 patients who received primary care at the Mayo Clinic between January 2004 and December 2015 in the training cohort and 41 295 patients who received primary care at Fairview Health Services from January 2010 to December 2017 in the validation cohort. Data were analyzed from May 2018 to December 2019. Main Outcomes and Measures: Several binary classification measures, including the area under the receiver operating characteristic curve (AUC), Gini score, sensitivity, and positive predictive value, were used to evaluate models predicting all-cause mortality and major cardiovascular events at ages 60, 65, 75, and 80 years. Results: The mean (SD) age and proportions of women and white individuals were 59.4 (10.8) years, 6324 (59.3%) and 9804 (91.9%), respectively, in the training cohort and 57.4 (7.9) years, 21 975 (53.1%), and 37 653 (91.2%), respectively, in the validation cohort. During follow-up, 945 patients (8.9%) in the training cohort died, while 787 (7.4%) had major cardiovascular events. Models using the new representation achieved AUCs for predicting death in the training cohort at ages 60, 65, 75, and 80 years of 0.96 (95% CI, 0.94-0.97), 0.96 (95% CI, 0.95-0.98), 0.97 (95% CI, 0.96-0.98), and 0.98 (95% CI, 0.98-0.99), respectively, while standard methods achieved modest AUCs of 0.67 (95% CI, 0.55-0.80), 0.66 (95% CI, 0.56-0.79), 0.64 (95% CI, 0.57-0.71), and 0.63 (95% CI, 0.54-0.70), respectively. Conclusions and Relevance: In this study, the proposed patient data representation accurately predicted the age at which a patient was at risk of dying or developing major cardiovascular events substantially better than standard methods. The representation uses known relationships contained in electronic health records to capture disease severity in a natural and clinically meaningful way. Furthermore, it is expressive and interpretable. This novel patient representation can help to support critical decision-making, develop smart guidelines, and enhance health care and disease management by helping to identify patients with high risk.

Challenges in returning results in a genomic medicine implementation study: the Return of Actionable Variants Empirical (RAVE) study.

To inform the process of returning results in genome sequencing studies, we conducted a quantitative and qualitative assessment of challenges encountered during the Return of Actionable Variants Empiric (RAVE) study conducted at Mayo Clinic. Participants (n = 2535, mean age 63 ± 7, 57% female) were sequenced for 68 clinically actionable genes and 14 single nucleotide variants. Of 122 actionable results detected, 118 were returnable; results were returned by a genetic counselor-86 in-person and 12 by phone. Challenges in returning actionable results were encountered in a significant proportion (38%) of the cohort and were related to sequencing and participant contact. Sequencing related challenges (n = 14), affecting 13 participants, included reports revised based on clinical presentation (n = 3); reports requiring corrections (n = 2); mosaicism requiring alternative DNA samples for confirmation (n = 3); and variant re-interpretation due to updated informatics pipelines (n = 6). Participant contact related challenges (n = 44), affecting 38 participants, included nonresponders (n = 20), decedents (n = 1), and previously known results (n = 23). These results should be helpful to investigators preparing for return of results in large-scale genomic sequencing projects.

Predicting diabetes clinical outcomes using longitudinal risk factor trajectories.

BACKGROUND: The ubiquity of electronic health records (EHR) offers an opportunity to observe trajectories of laboratory results and vital signs over long periods of time. This study assessed the value of risk factor trajectories available in the electronic health record to predict incident type 2 diabetes. STUDY DESIGN AND METHODS: Analysis was based on a large 13-year retrospective cohort of 71,545 adult, non-diabetic patients with baseline in 2005 and median follow-up time of 8 years. The trajectories of fasting plasma glucose, lipids, BMI and blood pressure were computed over three time frames (2000-2001, 2002-2003, 2004) before baseline. A novel method, Cumulative Exposure (CE), was developed and evaluated using Cox proportional hazards regression to assess risk of incident type 2 diabetes. We used the Framingham Diabetes Risk Scoring (FDRS) Model as control. RESULTS: The new model outperformed the FDRS Model (.802 vs .660; p-values <2e-16). Cumulative exposure measured over different periods showed that even short episodes of hyperglycemia increase the risk of developing diabetes. Returning to normoglycemia moderates the risk, but does not fully eliminate it. The longer an individual maintains glycemic control after a hyperglycemic episode, the lower the subsequent risk of diabetes. CONCLUSION: Incorporating risk factor trajectories substantially increases the ability of clinical decision support risk models to predict onset of type 2 diabetes and provides information about how risk changes over time.

Deploying Clinical Decision Support for Familial Hypercholesterolemia.

Objective: Familial hypercholesterolemia (FH), a prevalent genomic disorder that increases risk of coronary heart disease, remains significantly underdiagnosed. Clinical decision support (CDS) tools have the potential to increase FH detection. We describe our experience in the development and implementation of a genomic CDS for FH at a large academic medical center. Methods: CDS development and implementation were conducted in four phases: (1) development and validation of an algorithm to identify "possible FH"; (2) obtaining approvals from institutional committees to develop the CDS; (3) development of the initial prototype; and (4) use of an implementation science framework to evaluate the CDS. Results: The timeline for this work was approximately 4 years; algorithm development and validation occurred from August 2018 to February 2020. During this 4-year period, we engaged with 15 stakeholder groups to build and integrate the CDS, including health care providers who gave feedback at each stage of development. During CDS implementation six main challenges were identified: (1) need for multiple institutional committee approvals; (2) need to align the CDS with institutional knowledge resources; (3) need to adapt the CDS to differing workflows; (4) lack of institutional guidelines for CDS implementation; (5) transition to a new institutional electronic health record (EHR) system; and (6) limitations of the EHR related to genomic medicine. Conclusion: We identified multiple challenges in different domains while developing CDS for FH and integrating it with the EHR. The lessons learned herein may be helpful in streamlining the development and deployment of CDS to facilitate genomic medicine implementation.