Measuring the Impact of Primary Care Team Composition on Patient Activation Utilizing Electronic Health Record Big Data Analytics

Purpose

Team-based care has been linked to key outcomes associated with the Quadruple Aim and a key driver of high-value patient-centered care. Use of the electronic health record (EHR) and machine learning have significant potential to overcome previous barriers to studying the impact of teams, including delays in accessing data to improve teamwork and optimize patient outcomes.

Methods

This study utilized a large EHR dataset (n=316,542) from an urban health system to explore the relationship between team composition and patient activation, a key driver of patient engagement. Teams were operationalized using consensus definitions of teamwork from the literature. Patient activation was measured using the Patient Activation Measure (PAM). Results from multilevel regression analyses were compared to machine learning analyses using multinomial logistic regression to calculate propensity scores for the effect of team composition on PAM scores. Under the machine learning approach, a causal inference model with generalized overlap weighting was used to calculate the average treatment effect of teamwork.

Results

Seventeen different team types were observed in the data from the analyzed sample (n=12,448). Team sizes ranged from 2 to 5 members. After controlling for confounding variables in both analyses, more diverse, multidisciplinary teams (team size of 4 or more) were observed to have improved patient activation scores.

Conclusions

This is the first study to explore the relationship between team composition and patient activation using the EHR and big data analytics. Implications for further research using EHR data and machine learning to study teams and other patient-centered care are promising and could be used to advance team science.

Development and Validation of the Pharmacological Statin-Associated Muscle Symptoms Risk Stratification Score Using Electronic Health Record Data

Statin-associated muscle symptoms (SAMS) can lead to statin nonadherence. This paper aims to develop a pharmacological SAMS risk stratification (PSAMS-RS) score using a previously developed PSAMS phenotyping algorithm that distinguishes objective vs. nocebo SAMS using electronic health record (EHR) data. Using our PSAMS phenotyping algorithm, SAMS cases and controls were identified from Minnesota Fairview EHR, with the statin user cohort divided into derivation (January 1, 2010, to December 31, 2018) and validation (January 1, 2019, to December 31, 2020) cohorts. A Least Absolute Shrinkage and Selection Operator regression model was applied to identify significant features for PSAMS. PSAMS-RS scores were calculated and the clinical utility of stratifying PSAMS risk was assessed by comparing hazard ratios (HRs) between fourth vs. first score quartiles. PSAMS cases were identified in 1.9% (310/16,128) of the derivation and 1.5% (64/4,182) of the validation cohorts. Sixteen out of 38 clinical features were determined to be significant predictors for PSAMS risk. Patients within the fourth quartile of the PSAMS scores had an over sevenfold (HR: 7.1, 95% confidence interval (CI): 4.03-12.45, derivation cohort) or sixfold (HR: 6.1, 95% CI: 2.15-17.45, validation cohort) higher hazard of developing PSAMS vs. those in their respective first quartile. The PSAMS-RS score is a simple tool to stratify patients' risk of developing PSAMS after statin initiation which could inform clinician-guided pre-emptive measures to prevent PSAMS-related statin nonadherence.

Decision rules for personalized statin treatment prescriptions over multi-objectives

In our previous study, we demonstrated the feasibility of producing a proactive statin prescription strategy - a personalized statin treatment plan (PSTP) - using neural networks with big data. However, its non-transparency limited result interpretations and clinical usability. To improve the transparency of our previous approach with minimal compromise to the maximal statin treatment benefit-to-risk ratio, this study proposed a five-step pipeline approach called the decision rules for statin treatment (DRST). Steps 1-3 of our proposed pipeline improved our previous PSTP model in optimizing individual benefit-to-risk ratio; Step 4 used a decision tree model (DRST) to provide straightforward rules in the initial statin treatment plan; Step 5 aimed to evaluate the efficacy of these decision rules by conducting a clinical trial simulation. We included 107,739 de-identified patient data from Optum Labs Database Warehouse in this study. The final decision rules were compact and efficient, resulting from a decision tree with only a maximum depth of 3 and 11 nodes. The DRST identified three factors that are easily obtainable at the point of care: age, low-density lipoprotein cholesterol (LDL-C) level, and age-adjusted Charlson score. Moreover, it also identified six subpopulations that can benefit most from these decision rules. In our clinical trial simulations, DRST was found to improve statin benefit in LDL-C reduction by 4.15 percentage points (pp) and reduce risks of statin-associated symptoms (SAS) and statin discontinuation by 11.71 and 3.96 pp, respectively, when compared to the standard of care. Moreover, these DRST results were only less than 0.6 pp suboptimal to PSTP, demonstrating that building DRST that provide transparency with minimal compromise to the maximal benefit-to-risk ratio of statin treatments is feasible.

Development and application of pharmacological statin-associated muscle symptoms phenotyping algorithms using structured and unstructured electronic health records data

Importance

Statins are widely prescribed cholesterol-lowering medications in the United States, but their clinical benefits can be diminished by statin-associated muscle symptoms (SAMS), leading to discontinuation.

Objectives

In this study, we aimed to develop and validate a pharmacological SAMS clinical phenotyping algorithm using electronic health records (EHRs) data from Minnesota Fairview.

Materials and Methods

We retrieved structured and unstructured EHR data of statin users and manually ascertained a gold standard set of SAMS cases and controls using the published SAMS-Clinical Index tool from clinical notes in 200 patients. We developed machine learning algorithms and rule-based algorithms that incorporated various criteria, including ICD codes, statin allergy, creatine kinase elevation, and keyword mentions in clinical notes. We applied the best-performing algorithm to the statin cohort to identify SAMS.

Results

We identified 16 889 patients who started statins in the Fairview EHR system from 2010 to 2020. The combined rule-based (CRB) algorithm, which utilized both clinical notes and structured data criteria, achieved similar performance compared to machine learning algorithms with a precision of 0.85, recall of 0.71, and F1 score of 0.77 against the gold standard set. Applying the CRB algorithm to the statin cohort, we identified the pharmacological SAMS prevalence to be 1.9% and selective risk factors which included female gender, coronary artery disease, hypothyroidism, and use of immunosuppressants or fibrates.

Discussion and Conclusion

Our study developed and validated a simple pharmacological SAMS phenotyping algorithm that can be used to create SAMS case/control cohort to enable further analysis which can lead to the development of a SAMS risk prediction model.

Development and validation of the pharmacological statin-associated muscle symptoms risk stratification (PSAMS-RS) score using real-world electronic health record data

Introduction

Statin-associated muscle symptoms (SAMS) contribute to the nonadherence to statin therapy. In a previous study, we successfully developed a pharmacological SAMS (PSAMS) phenotyping algorithm that distinguishes objective versus nocebo SAMS using structured and unstructured electronic health records (EHRs) data. Our aim in this paper was to develop a pharmacological SAMS risk stratification (PSAMS-RS) score using these same EHR data.

Method

Using our PSAMS phenotyping algorithm, SAMS cases and controls were identified using University of Minnesota (UMN) Fairview EHR data. The statin user cohort was temporally divided into derivation (1/1/2010 to 12/31/2018) and validation (1/1/2019 to 12/31/2020) cohorts. First, from a feature set of 38 variables, a Least Absolute Shrinkage and Selection Operator (LASSO) regression model was fitted to identify important features for PSAMS cases and their coefficients. A PSAMS-RS score was calculated by multiplying these coefficients by 100 and then adding together for individual integer scores. The clinical utility of PSAMS-RS in stratifying PSAMS risk was assessed by comparing the hazard ratio (HR) between 4th vs 1st score quartile.

Results

PSAMS cases were identified in 1.9% (310/16128) of the derivation and 1.5% (64/4182) of the validation cohort. After fitting LASSO regression, 16 out of 38 clinical features were determined to be significant predictors for PSAMS risk. These factors are male gender, chronic pulmonary disease, neurological disease, tobacco use, renal disease, alcohol use, ACE inhibitors, polypharmacy, cerebrovascular disease, hypothyroidism, lymphoma, peripheral vascular disease, coronary artery disease and concurrent uses of fibrates, beta blockers or ezetimibe. After adjusting for statin intensity, patients in the PSAMS score 4th quartile had an over seven-fold (derivation) (HR, 7.1; 95% CI, 4.03-12.45) and six-fold (validation) (HR, 6.1; 95% CI, 2.15-17.45) higher hazard of developing PSAMS versus those in 1st score quartile.

Conclusion

The PSAMS-RS score can be a simple tool to stratify patients' risk of developing PSAMS after statin initiation which can facilitate clinician-guided preemptive measures that may prevent potential PSAMS-related statin non-adherence.

Development and Application of Pharmacological Statin-Associated Muscle Symptoms Phenotyping Algorithms Using Structured and Unstructured Electronic Health Records Data

Background

Statins are widely prescribed cholesterol-lowering medications in the US, but their clinical benefits can be diminished by statin-associated muscle symptoms (SAMS), leading to discontinuation. In this study, we aimed to develop and validate a pharmacological SAMS clinical phenotyping algorithm using electronic health records (EHRs) data from Minnesota Fairview.

Methods

We retrieved structured and unstructured EHR data of statin users and manually ascertained a gold standard set of SAMS cases and controls using the SAMS-CI tool from clinical notes in 200 patients. We developed machine learning algorithms and rule-based algorithms that incorporated various criteria, including ICD codes, statin allergy, creatine kinase elevation, and keyword mentions in clinical notes. We applied the best performing algorithm to the statin cohort to identify SAMS.

Results

We identified 16,889 patients who started statins in the Fairview EHR system from 2010-2020. The combined rule-based (CRB) algorithm, which utilized both clinical notes and structured data criteria, achieved similar performance compared to machine learning algorithms with a precision of 0.85, recall of 0.71, and F1 score of 0.77 against the gold standard set. Applying the CRB algorithm to the statin cohort, we identified the pharmacological SAMS prevalence to be 1.9% and selective risk factors which included female gender, coronary artery disease, hypothyroidism, use of immunosuppressants or fibrates.

Conclusion

Our study developed and validated a simple pharmacological SAMS phenotyping algorithm that can be used to create SAMS case/control cohort for further analysis such as developing SAMS risk prediction model.

Using artificial intelligence to improve pain assessment and pain management: a scoping review

CONTEXT

Over 20% of US adults report they experience pain on most days or every day. Uncontrolled pain has led to increased healthcare utilization, hospitalization, emergency visits, and financial burden. Recognizing, assessing, understanding, and treating pain using artificial intelligence (AI) approaches may improve patient outcomes and healthcare resource utilization. A comprehensive synthesis of the current use and outcomes of AI-based interventions focused on pain assessment and management will guide the development of future research.

OBJECTIVES

This review aims to investigate the state of the research on AI-based interventions designed to improve pain assessment and management for adult patients. We also ascertain the actual outcomes of Al-based interventions for adult patients.

METHODS

The electronic databases searched include Web of Science, CINAHL, PsycINFO, Cochrane CENTRAL, Scopus, IEEE Xplore, and ACM Digital Library. The search initially identified 6946 studies. After screening, 30 studies met the inclusion criteria. The Critical Appraisals Skills Programme was used to assess study quality.

RESULTS

This review provides evidence that machine learning, data mining, and natural language processing were used to improve efficient pain recognition and pain assessment, analyze self-reported pain data, predict pain, and help clinicians and patients to manage chronic pain more effectively.

CONCLUSIONS

Findings from this review suggest that using AI-based interventions has a positive effect on pain recognition, pain prediction, and pain self-management; however, most reports are only pilot studies. More pilot studies with physiological pain measures are required before these approaches are ready for large clinical trial.

Comparison of the Warfarin Dosing and Outcomes in Hmong Versus East Asians Patients: Real-World Data From an Integrated Healthcare System

Background

Previous research predicted that Hmong, an understudied East Asian subpopulation, might require significantly lower warfarin doses than East Asian patients partially due to their unique genetic and clinical factors. However, such findings have not been corroborated using real-world data.

Methods

This was a retrospective cohort study of Hmong and East Asian patients receiving warfarin. Warfarin stable doses (WSD) and time to the composite outcome, including international normalized ratio (INR) greater than four incidences or major bleeding within six months of warfarin initiation, were compared.

Results

This cohort study included 55 Hmong and 100 East Asian patients. Compared to East Asian patients, Hmong had a lower mean WSD (14.5 vs. 20.4 mg/week, p<0.05). In addition, Hmong had a 3.1-fold (95% CI: 1.1-9.3, p<0.05) higher hazard of the composite outcome.

Conclusion

Using real-world data, significant differences in warfarin dosing and hazard for the composite outcome of INR>4 and major bleeding were observed between Hmong and East Asian patients. These observations further underscore the importance of recognizing subpopulation-based differences in warfarin dosing and outcomes.

Clinical Data Cohort Quality Improvement: The Case of the Medication Data in The University of Minnesota's Clinical Data Repository

Clinical and translational research centers (CTRCs) have emerged as key centers for electronic medical record related research through integrated data repositories (IDRs) and the 'secondary use' of clinical data. Researchers accessing and pre-processing ever increasing amounts of electronic medical records for data mining tasks have a growing need for best practice approaches for clinical data quality assessment and improvement. This project focused on a large data extract for 7 statin medication prescriptions for patients with cardiovascular disease. After the initial data extraction, we proceeded to analyze the data for completeness, correctness, currency, and percentage populated using established data quality frameworks. Assessment of the said data was performed through medication possession ratios, medication discontinuation reasons, and drug dosages. When we compared distributions of data elements such as drug dosage before and after changes were introduced by our pre-processing protocols, only a minimal noticeable difference was found as the clinical data cohort quality assessment and pre-processing were completed without substantially altering the original data structure. Our study demonstrated practical steps for clinical data cohort quality improvement using medication data and illustrates a best practice approach in clinical data cohort quality improvement for any data mining tasks.

Development, implementation and participant evaluation of combining text messaging and peer group support in a weight management programme for African-American women

Background

Development of highly accessible interventions that are effective in reducing body weight, preventing weight gain, and maintaining weight loss is urgently needed to solve the current obesity epidemic, especially among African-American women.

Aims

The purpose of this paper is to describe the development, implementation, and participant evaluation processes of a combined text messaging and peer support group programme to enhance weight management skills among African-American women.

Methods

The programme's conceptual framework and operational model were developed to enhance the research design and protocol to support the study rationale and to lay a solid theoretical base for programme implementation. The programme curriculum and schedule were established and embedded into the programme protocol.

Results

The 16-week text messaging and peer support group intervention was implemented from September 2014 to March 2015. In total, 2089 messages were sent using an online text messaging application. Eight support group sessions were held in the participant's community centre or community church bi-weekly for approximately one hour.

Conclusions

This paper provides a blueprint of the methodological aspects and insights from participants' evaluation of a combined weight management intervention that can be used or adapted by public health nurses and other community health professionals in their work to develop weight management skills among African-American women.

A Population-Based Study of Simvastatin Drug-Drug Interactions in Cardiovascular Disease Patients

Simvastatin is a commonly used medication for lipid management and cardiovascular disease, however, the risk of adverse events (AEs) with its use increases via drug-drug interaction (DDI) exposures. Patients were extracted if initially diagnosed with cardiovascular disease and newly initiated simvastatin therapy. The cohort was divided into a DDI-exposed group and a non-DDI exposed group. The DDI-exposed group was further divided into gemfibrozil, clarithromycin, and erythromycin exposure groups. The outcome was defined as a composite of predefined AEs. Our results show that the simvastatin-DDI group had a higher illness burden with longer simvastatin exposure time and more medical care follow-up compared with the simvastatin-non-DDI exposed group. AEs occurred more frequently in subjects exposed to interacting drugs with a higher risk for clarithromycin and erythromycin exposed subjects than for gemfibrozil subjects.

Discovering Associations Among Older Adults' Characteristics and Planned Nursing Interventions Using Electronic Health Record Data

BACKGROUND AND PURPOSE

Little is known about how nursing assessments of strengths and signs/symptoms inform intervention planning in assisted living communities. The purpose of this study was to discover associations among older adults' characteristics and their planned nursing interventions.

METHODS

This study employed a data-driven method, latent class analysis, using existing electronic health record data from a senior living community in the Midwest. A convenience sample comprised de-identified data of well-being assessments and care plans for 243 residents. Latent class analysis, descriptive, and inferential statistics were used to group the sample, summarize strengths and problems attributes, nursing interventions, and Knowledge, Behavior, and Status scores, and detect differences.

RESULTS

Three groups presented based on patterns of strengths and signs/symptoms combined with problem concepts: Living Well (n = 95) had more strengths and fewer signs/symptoms; Lower Strengths (n = 99) had fewer strengths and more signs/symptoms; and Resilient Survivors (n = 49) had more strengths and more signs/symptoms. Some associations were found among group characteristics and planned interventions. Living Well had the lowest average number of planned interventions per resident (Mean = 2.7; standard deviation [SD] = 1.7) followed by Lower Strengths (Mean = 3.8; SD = 2.6) and Resilient Survivors (Mean = 4.1; SD = 3.4).

IMPLICATIONS FOR PRACTICE

This study offers new knowledge in the use of a strengths-based ontology to facilitate a nursing discourse that leverages use of older adults' strengths to address their problems and support their living a healthier life. It also offers the potential to complement the problem-based infrastructure in clinical practice and documentation.

Big Data Cohort Extraction for Personalized Statin Treatment and Machine Learning

The creation of big clinical data cohorts for machine learning and data analysis require a number of steps from the beginning to successful completion. Similar to data set preprocessing in other fields, there is an initial need to complete data quality evaluation; however, with large heterogeneous clinical data sets, it is important to standardize the data in order to facilitate dimensionality reduction. This is particularly important for clinical data sets including medications as a core data component due to the complexity of coded medication data. Data integration at the individual subject level is essential with medication-related machine learning applications since it can be difficult to accurately identify drug exposures, therapeutic effects, and adverse drug events without having high-quality data integration of insurance, medication, and medical data. Successful data integration and standardization efforts can substantially improve the ability to identify and replicate personalized treatment pathways to optimize drug therapy.

Exploring Older Adults' Strengths, Problems, and Wellbeing Using De-identified Electronic Health Record Data

As new data sources including individuals' strengths emerge in electronic health records, such data provide whole-person oriented information to generate integrated knowledge for person-centered practice. The purpose of this study is to describe older adults' strengths and problems within a wellbeing context documented by the Omaha System. The Wellbeing Model is employed as a conceptual framework for wellbeing and is operationalized by the Omaha System Problem Classification Scheme. This study has a retrospective, descriptive design using de-identified EHR data of wellbeing assessments including problems, strengths, and signs/symptoms for a convenience sample of 440 assisted-living residents in a Midwest metropolitan area. Descriptive statistics and data visualization were used to summarize and display strength and signs/symptom attributes within wellbeing contexts. The study reveals cutting-edge knowledge regarding older adults' strengths and wellbeing, and creates a platform for further research use of a strength-based ontology in clinical practice and electronic system of documentation.

Personalized Anticoagulation: Optimizing Warfarin Management Using Genetics and Simulated Clinical Trials

BACKGROUND

Clinical trials testing pharmacogenomic-guided warfarin dosing for patients with atrial fibrillation have demonstrated conflicting results. Non-vitamin K antagonist oral anticoagulants are expensive and contraindicated for several conditions. A strategy optimizing anticoagulant selection remains an unmet clinical need.

METHODS AND RESULTS

Characteristics from 14 206 patients with atrial fibrillation were integrated into a validated warfarin clinical trial simulation framework using iterative Bayesian network modeling and a pharmacokinetic-pharmacodynamic model. Individual dose-response for patients was simulated for 5 warfarin protocols-a fixed-dose protocol, a clinically guided protocol, and 3 increasingly complex pharmacogenomic-guided protocols. For each protocol, a complexity score was calculated using the variables predicting warfarin dose and the number of predefined international normalized ratio (INR) thresholds for each adjusted dose. Study outcomes included optimal time in therapeutic range ≥65% and clinical events. A combination of age and genotype identified different optimal protocols for various subpopulations. A fixed-dose protocol provided well-controlled INR only in normal responders ≥65, whereas for normal responders <65 years old, a clinically guided protocol was necessary to achieve well-controlled INR. Sensitive responders ≥65 and <65 and highly sensitive responders ≥65 years old required pharmacogenomic-guided protocols to achieve well-controlled INR. However, highly sensitive responders <65 years old did not achieve well-controlled INR and had higher associated clinical events rates than other subpopulations.

CONCLUSIONS

Under the assumptions of this simulation, patients with atrial fibrillation can be triaged to an optimal warfarin therapy protocol by age and genotype. Clinicians should consider alternative anticoagulation therapy for patients with suboptimal outcomes under any warfarin protocol.

A strengths-based data capture model: mining data-driven and person-centered health assets

With health care policy directives advancing value-based care, risk assessments and management have permeated health care discourse. The conventional problem-based infrastructure defines what data are employed to build this discourse and how it unfolds. Such a health care model tends to bias data for risk assessment and risk management toward problems and does not capture data about health assets or strengths. The purpose of this article is to explore and illustrate the incorporation of a strengths-based data capture model into risk assessment and management by harnessing data-driven and person-centered health assets using the Omaha System. This strengths-based data capture model encourages and enables use of whole-person data including strengths at the individual level and, in aggregate, at the population level. When aggregated, such data may be used for the development of strengths-based population health metrics that will promote evaluation of data-driven and person-centered care, outcomes, and value.

Using simulation and optimization approach to improve outcome through warfarin precision treatment

We apply a treatment simulation and optimization approach to develop decision support guidance for warfarin precision treatment plans. Simulation include the use of ∼1,500,000 clinical avatars (simulated patients) generated by an integrated data-driven and domain-knowledge based Bayesian Network Modeling approach. Subsequently, we simulate 30-day individual patient response to warfarin treatment of five clinical and genetic treatment plans followed by both individual and subpopulation based optimization. Sub-population optimization (compared to individual optimization) provides a cost effective and realistic means of implementation of a precision-driven treatment plan in practical settings. In this project, we use the property of minimal entropy to minimize overall adverse risks for the largest possible patient sub-populations and we temper the results by considering both transparency and ease of implementation. Finally, we discuss the improved outcome of the precision treatment plan based on the sub-population optimized decision support rules.

Personalized long-term prediction of cognitive function: Using sequential assessments to improve model performance

Prediction of onset and progression of cognitive decline and dementia is important both for understanding the underlying disease processes and for planning health care for populations at risk. Predictors identified in research studies are typically accessed at one point in time. In this manuscript, we argue that an accurate model for predicting cognitive status over relatively long periods requires inclusion of time-varying components that are sequentially assessed at multiple time points (e.g., in multiple follow-up visits). We developed a pilot model to test the feasibility of using either estimated or observed risk factors to predict cognitive status. We developed two models, the first using a sequential estimation of risk factors originally obtained from 8 years prior, then improved by optimization. This model can predict how cognition will change over relatively long time periods. The second model uses observed rather than estimated time-varying risk factors and, as expected, results in better prediction. This model can predict when newly observed data are acquired in a follow-up visit. Performances of both models that are evaluated in10-fold cross-validation and various patient subgroups show supporting evidence for these pilot models. Each model consists of multiple base prediction units (BPUs), which were trained using the same set of data. The difference in usage and function between the two models is the source of input data: either estimated or observed data. In the next step of model refinement, we plan to integrate the two types of data together to flexibly predict dementia status and changes over time, when some time-varying predictors are measured only once and others are measured repeatedly. Computationally, both data provide upper and lower bounds for predictive performance.

Peer Group and Text Message-Based Weight-Loss and Management Intervention for African American Women

About 80% of African American (AA) women are overweight or obese. Accessible and effective weight management programs targeting weight loss, weight maintenance and the prevention of weight regain are needed to improve health of AA women. A feasibility study was conducted to examine the feasibility, acceptability, and potential efficacy of a 16-week intervention protocol for weight loss and management that combined daily text messages and biweekly peer group sessions. Modest but statistically significant reductions were detected in weight and body mass index from baseline to 16 weeks. At baseline, 36% of participants were in action and maintenance stages in measures of the stages of change for weight loss and management; this percent increased to 82% at 16 weeks. Findings of this feasibility study provide preliminary evidence of an educational intervention that could motivate women and lead to successful behavior change, and successful weight loss and management for AA women.

Big Data Cohort Extraction to Facilitate Machine Learning to Improve Statin Treatment

Health care Big Data studies hold substantial promise for improving clinical practice. Among analytic tools, machine learning (ML) is an important approach that has been widely used by many industries for data-driven decision support. In Big Data, thousands of variables and millions of patient records are commonly encountered, but most data elements cannot be directly used to support decision making. Although many feature-selection tools can help identify relevant data, these tools are typically insufficient to determine a patient data cohort to support learning. Therefore, domain experts with nursing or clinic knowledge play critical roles in determining value criteria or the type of variables that should be included in the patient cohort to maximize project success. We demonstrate this process by extracting a patient cohort (37,506 individuals) to support our ML work (i.e., the production of a proactive strategy to prevent statin adverse events) from 130 million de-identified lives in the OptumLabs™ Data Warehouse.

A systems approach to designing effective clinical trials using simulations

BACKGROUND

Pharmacogenetics in warfarin clinical trials have failed to show a significant benefit in comparison with standard clinical therapy. This study demonstrates a computational framework to systematically evaluate preclinical trial design of target population, pharmacogenetic algorithms, and dosing protocols to optimize primary outcomes.

METHODS AND RESULTS

We programmatically created an end-to-end framework that systematically evaluates warfarin clinical trial designs. The framework includes options to create a patient population, multiple dosing strategies including genetic-based and nongenetic clinical-based, multiple-dose adjustment protocols, pharmacokinetic/pharmacodynamics modeling and international normalization ratio prediction, and various types of outcome measures. We validated the framework by conducting 1000 simulations of the applying pharmacogenetic algorithms to individualize dosing of warfarin (CoumaGen) clinical trial primary end points. The simulation predicted a mean time in therapeutic range of 70.6% and 72.2% (P=0.47) in the standard and pharmacogenetic arms, respectively. Then, we evaluated another dosing protocol under the same original conditions and found a significant difference in the time in therapeutic range between the pharmacogenetic and standard arm (78.8% versus 73.8%; P=0.0065), respectively.

CONCLUSIONS

We demonstrate that this simulation framework is useful in the preclinical assessment phase to study and evaluate design options and provide evidence to optimize the clinical trial for patient efficacy and reduced risk.

Individualized patient-centered lifestyle recommendations: an expert system for communicating patient specific cardiovascular risk information and prioritizing lifestyle options

We propose a proof-of-concept machine-learning expert system that learned knowledge of lifestyle and the associated 10-year cardiovascular disease (CVD) risks from individual-level data (i.e., Atherosclerosis Risk in Communities Study, ARIC). The expert system prioritizes lifestyle options and identifies the one that maximally reduce an individual's 10-year CVD risk by (1) using the knowledge learned from the ARIC data and (2) communicating for patient-specific cardiovascular risk information and personal limitations and preferences (as defined by variables used in this study). As a result, the optimal lifestyle is not only prioritized based on an individual's characteristics but is also relevant to personal circumstances. We also explored probable uses and tested the system in several examples using real-world scenarios and patient preferences. For example, the system identifies the most effective lifestyle activities as the starting point for an individual's behavior change, shows different levels of BMI changes and the associated CVD risk reductions to encourage weight loss, identifies whether weight loss or smoking cessation is the most urgent change for a diabetes patient, etc. Answers to the questions noted above vary based on an individual's characteristics. Our validation results from clinical trial simulations, which compared original with the optimal lifestyle using an independent dataset, show that the optimal individualized patient-centered lifestyle consistently reduced 10-year CVD risks.

A simulation platform to examine heterogeneity influence on treatment

Although a protocol aims to guide treatment management and optimize overall outcomes, the benefits and harms for each individual vary due to heterogeneity. Some protocols integrate clinical and genetic variation to provide treatment recommendation; it is not clear whether such integration is sufficient. If not, treatment outcomes may be sub-optimal for certain patient sub-populations. Unfortunately, running a clinical trial to examine such outcome responses is cost prohibitive and requires a significant amount of time to conduct the study. We propose a simulation approach to discover this knowledge from electronic medical records; a rapid method to reach this goal. We use the well-known drug warfarin as an example to examine whether patient characteristics, including race and the genes CYP2C9 and VKORC1, have been fully integrated into dosing protocols. The two genes mentioned above have been shown to be important in patient response to warfarin.

The optimal diagnostic decision sequence

We describe a data mining model for constructing an optimal diagnostic sequence that assists cost-effective sequential decisions. We use heuristic search, i.e., hill climbing and genetic algorithms (GAs), and the evaluation function of cost-based Mean Accuracy Gain (cMAG), which is provided by SVM classifiers, to find this optimal sequence. GA can find a good sequence because of the ability to escape from local optima.

Application of artificial neural network-based survival analysis on two breast cancer datasets

This paper applies artificial neural networks (ANNs) to the survival analysis problem. Because ANNs can easily consider variable interactions and create a non-linear prediction model, they offer more flexible prediction of survival time than traditional methods. This study compares ANN results on two different breast cancer datasets, both of which use nuclear morphometric features. The results show that ANNs can successfully predict recurrence probability and separate patients with good (more than five years) and bad (less than five years) prognoses. Results are not as clear when the separation is done within subgroups such as lymph node positive or negative.

A data mining technique for risk-stratification diagnosis

We describe a data mining model for sequential diagnosis, called the Optimal Decision Path Finder (ODPF), which is built based on the idea of risk stratification. A filter was used to stratify patients depending on ease of diagnosis, and a series of patient-specific classifiers was built to diagnose with confidence while reducing exam cost. Results show that applying stratification to data mining can improve the diagnostic performance and reduce waste of medical resource. This resulting model can assist the physician in triage decisions.

Another look at differences in the susceptibility of Escherichia coli and Klebsiella pneumoniae to cephalothin and cefazolin

The significance of in vitro susceptibility tests on Enterobacteriaceae to cephalothin and cefazolin has not been exactly defined in the guidelines of the National Committee for Clinical Laboratory Standards. In the hope of clarifying this confusion, we provide additional information from an ancillary study of the Taiwan Surveillance of Antimicrobial Resistance 1998 (TSAR I). There were 505 Escherichia coli and 227 Klebsiella pneumoniae isolates susceptible to cephalothin, reported by 42 participating hospitals. The susceptibility of these isolates were re-tested at the Microbial Infections Reference Laboratory using cefazolin, with the result that 72% of the 252 cephalothin-resistant E. coli isolates and 24% of the 41 cephalothin-resistant K. pneumoniae isolates were found to be susceptible to cefazolin. We further surveyed the availability of cephalothin and cefazolin in Pharmacy Departments; all of the TSAR I hospitals had cefazolin available in their pharmacies. The resistance rate of E. coli was significantly lower for 12 hospitals that had cefazolin in both pharmacy and laboratory compared with 11 hospitals that had cefazolin available in pharmacy but cephalothin in laboratory. In addition, for all the hospitals that had cephalothin available for clinical use, the resistance rate was twice as low in two hospitals reporting cefazolin susceptibility as in the seven hospitals reporting cephalothin susceptibility. Our findings suggest that inappropriate selection of cephalothin and cefazolin for susceptibility testing contribute to inaccurate indications of in vivo activity for first generation cephalosporins in the treatment of E. coli infections.

Vertebrate Sprouty genes are induced by FGF signaling and can cause chondrodysplasia when overexpressed

The Drosophila sprouty gene encodes an antagonist of FGF and EGF signaling whose expression is induced by the signaling pathways that it inhibits. Here we describe a family of vertebrate Sprouty homologs and demonstrate that the regulatory relationship with FGF pathways has been conserved. In both mouse and chick embryos, Sprouty genes are expressed in intimate association with FGF signaling centers. Gain- and loss-of-function experiments demonstrate that FGF signaling induces Sprouty gene expression in various tissues. Sprouty overexpression obtained by infecting the prospective wing territory of the chick embryo with a retrovirus containing a mouse Sprouty gene causes a reduction in limb bud outgrowth and other effects consistent with reduced FGF signaling from the apical ectodermal ridge. At later stages of development in the infected limbs there was a dramatic reduction in skeletal element length due to an inhibition of chondrocyte differentiation. The results provide evidence that vertebrate Sprouty proteins function as FGF-induced feedback inhibitors, and suggest a possible role for Sprouty genes in the pathogenesis of specific human chondrodysplasias caused by activating mutations in Fgfr3.

Expression of N-methyl-D-aspartate receptor subunit mRNAs in the rat pheochromocytoma cell line PC12

The expression of the N-methyl-D-aspartate (NMDA) receptor subunit mRNAs NMDAR1 and NMDAR2A-D was characterized in undifferentiated and nerve growth factor (NGF)-differentiated PC12 cells using Northern blotting, RNase protection assays (RPA) and polymerase chain reaction (PCR). PC12 cells expressed predominately the splice variant NMDAR1-4a and smaller amounts of NMDAR1-1a, NMDAR1-2a and NMDAR1-3a. No splice isoforms containing exon 5 were detected. The NMDAR2C subunit was detected in PC12 cells by Northern blotting and trace amounts of NMDAR2A, B and D were detected by PCR. PC12 cells may be a useful model system for the study of the transcriptional and post-transcriptional regulation of expression of the NMDA receptor subunit genes, including the alternative splicing of NMDAR1 pre-mRNAs.

Developmental and regional expression pattern of a novel NMDA receptor-like subunit (NMDAR-L) in the rodent brain

A novel NMDA receptor-like (NMDAR-L) cDNA was isolated that contained an open reading frame coding for a predicted polypeptide of 1115 amino acids that shares approximately 27% identity with NMDA receptor subunits. In situ hybridization experiments indicated that NMDAR-L mRNA was expressed in the developing rodent CNS. On postnatal day 1 (P1), NMDAR-L mRNA expression was pronounced in the entorhinal cortex, the subiculum and the thalamus, in layer V of the developing neocortex, in the superior and inferior colliculi, and various regions of the hindbrain, excluding the cerebellum. On P5, NMDAR-L mRNA was expressed in layer V of the neocortex, in the entorhinal cortex, in the subiculum, and in the thalamus. On P14, NMDAR-L mRNA was expressed in layers II-VI of the neocortex, in the entorhinal and piriform cortex, in the subiculum and CA1 field, and in the nucleus of the lateral olfactory tract. In the adult brain, NMDAR-L mRNA was detected predominately in the nucleus of the lateral olfactory tract. Injection of NMDAR-L cRNA into Xenopus oocytes did not lead to the expression of homomeric glutamate-activated channels. However, coinjection of the triple combination of NMDAR-L with NMDAR1 and NMDAR2B cRNAs led to a striking decrease in the current magnitude compared to currents obtained after coexpression of the double combination of NMDAR1 with NMDAR2B. While the function of NMDAR-L remains to be established, its developmental and regional expression pattern suggests that NMDAR-L may influence axonal outgrowth and synaptogenesis during brain development.

Protein binding of brequinar in the plasma of healthy donors and cancer patients and analysis of the relationship between protein binding and pharmacokinetics in cancer patients

The protein binding of weakly acidic and basic drugs has been shown to be altered in cancer patients. Brequinar is a weakly acidic, low-clearance, and highly protein-bound (> 98% bound) antitumor agent. The pharmacokinetic parameters of brequinar are subject to large interpatient variability. This large interpatient variability may be related to brequinar's plasma protein-binding capacity (assuming no change in the intrinsic clearance of the unbound drug). The objectives of this study, therefore, were (a) to characterize brequinar's protein binding in the plasma of healthy donors and cancer patients and (b) to examine the relationships between brequinar's plasma protein binding and its pharmacokinetics in patients. Brequinar protein binding was determined in human serum albumin (HSA) solution, drug-free donor plasma, and brequinar-free, predose plasma samples obtained from a phase I cancer trial. Pharmacokinetic results from this study were used to examine relationships between plasma protein binding and drug disposition. In HSA solution and healthy donor plasma, brequinar's protein binding as determined using spiked samples was concentration-dependent. The unbound brequinar fraction increased by a factor of 3 (from 0.3% to 0.9% free) in 4% HSA solution and by a factor of 4 (from 0.4% to 1.6% free) in donor plasma as the brequinar concentrations increased from 0.1 to 2.3 mM in the HSA solution and from 0.076 to 1.5 mM in the donor plasma. Analysis of brequinar binding characteristics using the binding ratio and Rosenthal binding plots showed that albumin was the primary protein for brequinar binding in human plasma. The addition of various concentrations of alpha 1-acid glycoprotein to 4% HSA solution did not affect the protein binding of brequinar to HSA. The protein binding determined in the plasma of cancer patients was not quantitatively different, except for variability, from that observed in the plasma of healthy donors. Examination of relationships between the unbound brequinar fraction and pharmacokinetics suggested that plasma protein binding was not a major determinant of brequinar disposition in cancer patients.