Determinants of appropriate antibiotic and NSAID prescribing in unscheduled outpatient settings in the veterans health administration.

BACKGROUND: Despite efforts to enhance the quality of medication prescribing in outpatient settings, potentially inappropriate prescribing remains common, particularly in unscheduled settings where patients can present with infectious and pain-related complaints. Two of the most commonly prescribed medication classes in outpatient settings with frequent rates of potentially inappropriate prescribing include antibiotics and nonsteroidal anti-inflammatory drugs (NSAIDs). In the setting of persistent inappropriate prescribing, we sought to understand a diverse set of perspectives on the determinants of inappropriate prescribing of antibiotics and NSAIDs in the Veterans Health Administration. METHODS: We conducted a qualitative study guided by the Consolidated Framework for Implementation Research and Theory of Planned Behavior. Semi-structured interviews were conducted with clinicians, stakeholders, and Veterans from March 1, 2021 through December 31, 2021 within the Veteran Affairs Health System in unscheduled outpatient settings at the Tennessee Valley Healthcare System. Stakeholders included clinical operations leadership and methodological experts. Audio-recorded interviews were transcribed and de-identified. Data coding and analysis were conducted by experienced qualitative methodologists adhering to the Consolidated Criteria for Reporting Qualitative Studies guidelines. Analysis was conducted using an iterative inductive/deductive process. RESULTS: We conducted semi-structured interviews with 66 participants: clinicians (N = 25), stakeholders (N = 24), and Veterans (N = 17). We identified six themes contributing to potentially inappropriate prescribing of antibiotics and NSAIDs: 1) Perceived versus actual Veterans expectations about prescribing; 2) the influence of a time-pressured clinical environment on prescribing stewardship; 3) Limited clinician knowledge, awareness, and willingness to use evidence-based care; 4) Prescriber uncertainties about the Veteran condition at the time of the clinical encounter; 5) Limited communication; and 6) Technology barriers of the electronic health record and patient portal. CONCLUSIONS: The diverse perspectives on prescribing underscore the need for interventions that recognize the detrimental impact of high workload on prescribing stewardship and the need to design interventions with the end-user in mind. This study revealed actionable themes that could be addressed to improve guideline concordant prescribing to enhance the quality of prescribing and to reduce patient harm.

Heart failure subtype after acute kidney injury.

INTRODUCTION: Acute kidney injury (AKI) is associated with increased risk of heart failure (HF). Determining the type of HF experienced by AKI survivors (heart failure with preserved or reduced ejection fraction, HFpEF or HFrEF) could suggest potential mechanisms underlying the association and opportunities for improving post-AKI care. METHODS: In this retrospective study of adults within the Vanderbilt University health system with a diagnosis of HF, we tested whether AKI events in the two years preceding incident HF associated more with HFpEF or HFrEF while controlling for known predictors. HF outcomes were defined by administrative codes and classified as HFpEF or HFrEF by echocardiogram data. We used multivariable logistic regression models to estimate the effects of AKI on the odds of incident HFpEF versus HFrEF. RESULTS: AKI (all stages) trended towards a preferential association with HFpEF in adjusted analyses (adjusted OR 0.80, 95% CI 0.63 - 1.01). Stage 1 AKI was associated with higher odds of HFpEF that was statistically significant (adjusted OR 0.62, 95% CI 0.43 - 0.88), whereas stages 2-3 AKI showed a trend toward HFrEF that did not reach statistical significance (adjusted OR 1.11, 95% CI 0.76 - 1.63). CONCLUSIONS: AKI as a binary outcome trended towards a preferential association with HFpEF. Stage 1 AKI was associated with higher odds of HFpEF, whereas stage 2-3 trended towards an association with HFrEF that did not meet statistical significance. Different mechanisms may predominate in incident HF following mild versus more severe AKI. Close follow-up with particular attention to volume status and cardiac function after discharge is warranted after even mild AKI.

Use of noisy labels as weak learners to identify incompletely ascertainable outcomes: A Feasibility study with opioid-induced respiratory depression.

Objective: Assigning outcome labels to large observational data sets in a timely and accurate manner, particularly when outcomes are rare or not directly ascertainable, remains a significant challenge within biomedical informatics. We examined whether noisy labels generated from subject matter experts' heuristics using heterogenous data types within a data programming paradigm could provide outcomes labels to a large, observational data set. We chose the clinical condition of opioid-induced respiratory depression for our use case because it is rare, has no administrative codes to easily identify the condition, and typically requires at least some unstructured text to ascertain its presence. Materials and methods: Using de-identified electronic health records of 52,861 post-operative encounters, we applied a data programming paradigm (implemented in the Snorkel software) for the development of a machine learning classifier for opioid-induced respiratory depression. Our approach included subject matter experts creating 14 labeling functions that served as noisy labels for developing a probabilistic Generative model. We used probabilistic labels from the Generative model as outcome labels for training a Discriminative model on the source data. We evaluated performance of the Discriminative model with a hold-out test set of 599 independently-reviewed patient records. Results: The final Discriminative classification model achieved an accuracy of 0.977, an F1 score of 0.417, a sensitivity of 1.0, and an AUC of 0.988 in the hold-out test set with a prevalence of 0.83% (5/599). Discussion: All of the confirmed Cases were identified by the classifier. For rare outcomes, this finding is encouraging because it reduces the number of manual reviews needed by excluding visits/patients with low probabilities. Conclusion: Application of a data programming paradigm with expert-informed labeling functions might have utility for phenotyping clinical phenomena that are not easily ascertainable from highly-structured data.

Similar Risk of Kidney Failure among Patients with Blinding Diseases Who Receive Ranibizumab, Aflibercept, and Bevacizumab: An Observational Health Data Sciences and Informatics Network Study.

PURPOSE: To characterize the incidence of kidney failure associated with intravitreal anti-VEGF exposure; and compare the risk of kidney failure in patients treated with ranibizumab, aflibercept, or bevacizumab. DESIGN: Retrospective cohort study across 12 databases in the Observational Health Data Sciences and Informatics (OHDSI) network. SUBJECTS: Subjects aged ≥ 18 years with ≥ 3 monthly intravitreal anti-VEGF medications for a blinding disease (diabetic retinopathy, diabetic macular edema, exudative age-related macular degeneration, or retinal vein occlusion). METHODS: The standardized incidence proportions and rates of kidney failure while on treatment with anti-VEGF were calculated. For each comparison (e.g., aflibercept versus ranibizumab), patients from each group were matched 1:1 using propensity scores. Cox proportional hazards models were used to estimate the risk of kidney failure while on treatment. A random effects meta-analysis was performed to combine each database's hazard ratio (HR) estimate into a single network-wide estimate. MAIN OUTCOME MEASURES: Incidence of kidney failure while on anti-VEGF treatment, and time from cohort entry to kidney failure. RESULTS: Of the 6.1 million patients with blinding diseases, 37 189 who received ranibizumab, 39 447 aflibercept, and 163 611 bevacizumab were included; the total treatment exposure time was 161 724 person-years. The average standardized incidence proportion of kidney failure was 678 per 100 000 persons (range, 0-2389), and incidence rate 742 per 100 000 person-years (range, 0-2661). The meta-analysis HR of kidney failure comparing aflibercept with ranibizumab was 1.01 (95% confidence interval [CI], 0.70-1.47; P = 0.45), ranibizumab with bevacizumab 0.95 (95% CI, 0.68-1.32; P = 0.62), and aflibercept with bevacizumab 0.95 (95% CI, 0.65-1.39; P = 0.60). CONCLUSIONS: There was no substantially different relative risk of kidney failure between those who received ranibizumab, bevacizumab, or aflibercept. Practicing ophthalmologists and nephrologists should be aware of the risk of kidney failure among patients receiving intravitreal anti-VEGF medications and that there is little empirical evidence to preferentially choose among the specific intravitreal anti-VEGF agents. FINANCIAL DISCLOSURES: Proprietary or commercial disclosure may be found in the Footnotes and Disclosures at the end of this article.

Use of Noisy Labels as Weak Learners to Identify Incompletely Ascertainable Outcomes: A Feasibility Study with Opioid-Induced Respiratory Depression.

Objective: Assigning outcome labels to large observational data sets in a timely and accurate manner, particularly when outcomes are rare or not directly ascertainable, remains a significant challenge within biomedical informatics. We examined whether noisy labels generated from subject matter experts' heuristics using heterogenous data types within a data programming paradigm could provide outcomes labels to a large, observational data set. We chose the clinical condition of opioid-induced respiratory depression for our use case because it is rare, has no administrative codes to easily identify the condition, and typically requires at least some unstructured text to ascertain its presence. Materials and Methods: Using de-identified electronic health records of 52,861 post-operative encounters, we applied a data programming paradigm (implemented in the Snorkel software) for the development of a machine learning classifier for opioid-induced respiratory depression. Our approach included subject matter experts creating 14 labeling functions that served as noisy labels for developing a probabilistic Generative model. We used probabilistic labels from the Generative model as outcome labels for training a Discriminative model on the source data. We evaluated performance of the Discriminative model with a hold-out test set of 599 independently-reviewed patient records. Results: The final Discriminative classification model achieved an accuracy of 0.977, an F1 score of 0.417, a sensitivity of 1.0, and an AUC of 0.988 in the hold-out test set with a prevalence of 0.83% (5/599). Discussion: All of the confirmed Cases were identified by the classifier. For rare outcomes, this finding is encouraging because it reduces the number of manual reviews needed by excluding visits/patients with low probabilities. Conclusion: Application of a data programming paradigm with expert-informed labeling functions might have utility for phenotyping clinical phenomena that are not easily ascertainable from highly-structured data.

Sustainable deployment of clinical prediction tools-a 360° approach to model maintenance.

BACKGROUND: As the enthusiasm for integrating artificial intelligence (AI) into clinical care grows, so has our understanding of the challenges associated with deploying impactful and sustainable clinical AI models. Complex dataset shifts resulting from evolving clinical environments strain the longevity of AI models as predictive accuracy and associated utility deteriorate over time. OBJECTIVE: Responsible practice thus necessitates the lifecycle of AI models be extended to include ongoing monitoring and maintenance strategies within health system algorithmovigilance programs. We describe a framework encompassing a 360° continuum of preventive, preemptive, responsive, and reactive approaches to address model monitoring and maintenance from critically different angles. DISCUSSION: We describe the complementary advantages and limitations of these four approaches and highlight the importance of such a coordinated strategy to help ensure the promise of clinical AI is not short-lived.

Comparative Effectiveness of Second-line Antihyperglycemic Agents for Cardiovascular Outcomes: A Large-scale, Multinational, Federated Analysis of the LEGEND-T2DM Study.

Background: SGLT2 inhibitors (SGLT2is) and GLP-1 receptor agonists (GLP1-RAs) reduce major adverse cardiovascular events (MACE) in patients with type 2 diabetes mellitus (T2DM). However, their effectiveness relative to each other and other second-line antihyperglycemic agents is unknown, without any major ongoing head-to-head trials. Methods: Across the LEGEND-T2DM network, we included ten federated international data sources, spanning 1992-2021. We identified 1,492,855 patients with T2DM and established cardiovascular disease (CVD) on metformin monotherapy who initiated one of four second-line agents (SGLT2is, GLP1-RAs, dipeptidyl peptidase 4 inhibitor [DPP4is], sulfonylureas [SUs]). We used large-scale propensity score models to conduct an active comparator, target trial emulation for pairwise comparisons. After evaluating empirical equipoise and population generalizability, we fit on-treatment Cox proportional hazard models for 3-point MACE (myocardial infarction, stroke, death) and 4-point MACE (3-point MACE + heart failure hospitalization) risk, and combined hazard ratio (HR) estimates in a random-effects meta-analysis. Findings: Across cohorts, 16·4%, 8·3%, 27·7%, and 47·6% of individuals with T2DM initiated SGLT2is, GLP1-RAs, DPP4is, and SUs, respectively. Over 5·2 million patient-years of follow-up and 489 million patient-days of time at-risk, there were 25,982 3-point MACE and 41,447 4-point MACE events. SGLT2is and GLP1-RAs were associated with a lower risk for 3-point MACE compared with DPP4is (HR 0·89 [95% CI, 0·79-1·00] and 0·83 [0·70-0·98]), and SUs (HR 0·76 [0·65-0·89] and 0·71 [0·59-0·86]). DPP4is were associated with a lower 3-point MACE risk versus SUs (HR 0·87 [0·79-0·95]). The pattern was consistent for 4-point MACE for the comparisons above. There were no significant differences between SGLT2is and GLP1-RAs for 3-point or 4-point MACE (HR 1·06 [0·96-1·17] and 1·05 [0·97-1·13]). Interpretation: In patients with T2DM and established CVD, we found comparable cardiovascular risk reduction with SGLT2is and GLP1-RAs, with both agents more effective than DPP4is, which in turn were more effective than SUs. These findings suggest that the use of GLP1-RAs and SGLT2is should be prioritized as second-line agents in those with established CVD. Funding: National Institutes of Health, United States Department of Veterans Affairs.

Toward Real-World Reproducibility: Verifying Value Sets for Clinical Research.

Standardized operational definitions are an important tool to improve reproducibility of research using secondary real-world healthcare data. This approach was leveraged for studies evaluating the effectiveness of AZD7442 as COVID-19 pre-exposure prophylaxis across multiple healthcare systems. Value sets were defined, grouped, and mapped. Results of this exercise were reviewed and recorded. Value sets were updated to reflect findings.

Factors Shaping the Implementation of Strategies to Prevent Acute Kidney Injury: A Qualitative Study.

Reducing the prevalence of acute kidney injury (AKI) is an important patient safety objective set forth by the National Quality Forum. Despite international guidelines to prevent AKI, there continues to be an inconsistent uptake of these interventions by cardiac teams across practice settings. The IMPROVE-AKI study was designed to test the effectiveness and implementation of AKI preventive strategies delivered through team-based coaching activities. Qualitative methods were used to identify factors that shaped sites' implementation of AKI prevention strategies. Semi-structured interviews were conducted with staff in a range of roles within the cardiac catheterization laboratories, including nurses, laboratory managers, and interventional cardiologists (N = 50) at multiple time points over the course of the study. Interview transcripts were qualitatively coded, and aggregated code reports were reviewed to construct main themes through memoing. In this paper, we report insights from semi-structured interviews regarding workflow, organizational culture, and leadership factors that impacted implementation of AKI prevention strategies.

Understanding Prescribing Practices and Patient Experiences with Renin Angiotensin System Inhibitors Use in Chronic Kidney Disease: A Qualitative Study.

INTRODUCTION: Angiotensin-converting enzyme inhibitors (ACEIs) and angiotensin receptor blockers (ARBs) improve outcomes but are underutilized in patients with chronic kidney disease (CKD). Little is known about reasons for discontinuation and lack of reinitiating these medications. We aimed to explore clinicians' and patients' experiences and perceptions of ACEI/ARB use in CKD. METHODS: A multi-profession sample of health care clinicians and patients with documented ACEI/ARB-associated side effects in the past 6 months. Participants were recruited from 2 Veterans Affairs healthcare systems in Texas and Tennessee. A total of 15 clinicians and 10 patients completed interviews. We used inductive and deductive qualitative data analysis approaches to identify themes related to clinician and patient experiences with ACEI/ARB. Thematic analysis focused on prescribing decisions and practices, clinical guidelines, and perception of side effects. Data were analyzed as they amassed, and recruitment was stopped at the point of thematic saturation. RESULTS: Clinicians prescribe ACEI/ARB for blood pressure control and kidney protection and underscored the importance of these medications in patients with diabetes. While clinicians described providing comprehensive patient education about ACEI/ARB in CKD, patient interviews revealed significant knowledge gaps about CKD and ACEI/ARB use. Many patients were unaware of their CKD status, and some did not know why they were prescribed ACEI/ARB. Clinicians' drug management strategies varied widely, as did their understanding of prescribing guidelines. They identified structural and patient-level barriers to prescribing and many endorsed the development of a decision support tool to facilitate ACEI/ARB prescribing and management. DISCUSSION/CONCLUSION: Our qualitative study of clinicians and providers identified key target areas for improvement to increase ACEI/ARB utilization in patients with CKD with the goal to improve long-term outcomes in high-risk patients. These findings will also inform the development of a decision support tool to assist with prescribing ACEI/ARBs for patients with CKD.

Scalable Incident Detection via Natural Language Processing and Probabilistic Language Models.

Post marketing safety surveillance depends in part on the ability to detect concerning clinical events at scale. Spontaneous reporting might be an effective component of safety surveillance, but it requires awareness and understanding among healthcare professionals to achieve its potential. Reliance on readily available structured data such as diagnostic codes risk under-coding and imprecision. Clinical textual data might bridge these gaps, and natural language processing (NLP) has been shown to aid in scalable phenotyping across healthcare records in multiple clinical domains. In this study, we developed and validated a novel incident phenotyping approach using unstructured clinical textual data agnostic to Electronic Health Record (EHR) and note type. It's based on a published, validated approach (PheRe) used to ascertain social determinants of health and suicidality across entire healthcare records. To demonstrate generalizability, we validated this approach on two separate phenotypes that share common challenges with respect to accurate ascertainment: 1) suicide attempt; 2) sleep-related behaviors. With samples of 89,428 records and 35,863 records for suicide attempt and sleep-related behaviors, respectively, we conducted silver standard (diagnostic coding) and gold standard (manual chart review) validation. We showed Area Under the Precision-Recall Curve of ∼ 0.77 (95% CI 0.75-0.78) for suicide attempt and AUPR ∼ 0.31 (95% CI 0.28-0.34) for sleep-related behaviors. We also evaluated performance by coded race and demonstrated differences in performance by race were dissimilar across phenotypes and require algorithmovigilance and debiasing prior to implementation.

Genetic Inhibition of APOL1 Pore-Forming Function Prevents APOL1-Mediated Kidney Disease.

SIGNIFICANCE STATEMENT: African Americans are at increased risk of CKD in part due to high-risk (HR) variants in the apolipoprotein L1 ( APOL1 ) gene, termed G1/G2. A different APOL1 variant, p.N264K , reduced the risk of CKD and ESKD among carriers of APOL1 HR variants to levels comparable with individuals with APOL1 low-risk variants in an analysis of 121,492 participants of African ancestry from the Million Veteran Program (MVP). Functional genetic studies in cell models showed that APOL1 p.N264K blocked APOL1 pore-forming function and ion channel conduction and reduced toxicity of APOL1 HR mutations. Pharmacologic inhibitors that mimic this mutation blocking APOL1 -mediated pore formation may be able to prevent and/or treat APOL1 -associated kidney disease. BACKGROUND: African Americans are at increased risk for nondiabetic CKD in part due to HR variants in the APOL1 gene. METHODS: We tested whether a different APOL1 variant, p.N264K , modified the association between APOL1 HR genotypes (two copies of G1/G2) and CKD in a cross-sectional analysis of 121,492 participants of African ancestry from the MVP. We replicated our findings in the Vanderbilt University Biobank ( n =14,386) and National Institutes of Health All of Us ( n =14,704). Primary outcome was CKD and secondary outcome was ESKD among nondiabetic patients. Primary analysis compared APOL1 HR genotypes with and without p.N264K . Secondary analyses included APOL1 low-risk genotypes and tested for interaction. In MVP, we performed sequential logistic regression models adjusting for demographics, comorbidities, medications, and ten principal components of ancestry. Functional genomic studies expressed APOL1 HR variants with and without APOL1 p.N264K in cell models. RESULTS: In the MVP cohort, 15,604 (12.8%) had two APOL1 HR variants, of which 582 (0.5%) also had APOL1 p.N264K . In MVP, 18,831 (15%) had CKD, 4177 (3%) had ESKD, and 34% had diabetes. MVP APOL1 HR, without p.N264K , was associated with increased odds of CKD (odds ratio [OR], 1.72; 95% confidence interval [CI], 1.60 to 1.85) and ESKD (OR, 3.94; 95% CI, 3.52 to 4.41). In MVP, APOL1 p.N264K mitigated the renal risk of APOL1 HR, in CKD (OR, 0.43; 95% CI, 0.28 to 0.65) and ESKD (OR, 0.19; CI 0.07 to 0.51). In the replication cohorts meta-analysis, APOL1 p.N264K mitigated the renal risk of APOL1 HR in CKD (OR, 0.40; 95% CI, 0.18 to 0.92) and ESKD (OR, 0.19; 95% CI, 0.05 to 0.79). In the mechanistic studies, APOL1 p.N264K blocked APOL1 pore-forming function and ion channel conduction and reduced toxicity of APOL1 HR variants. CONCLUSIONS: APOL1 p.N264K is associated with reduced risk of CKD and ESKD among carriers of APOL1 HR to levels comparable with individuals with APOL1 low-risk genotypes.

Inpatient nurses' preferences and decisions with risk information visualization.

OBJECTIVE: We examined the influence of 4 different risk information formats on inpatient nurses' preferences and decisions with an acute clinical deterioration decision-support system. MATERIALS AND METHODS: We conducted a comparative usability evaluation in which participants provided responses to multiple user interface options in a simulated setting. We collected qualitative data using think aloud methods. We collected quantitative data by asking participants which action they would perform after each time point in 3 different patient scenarios. RESULTS: More participants (n = 6) preferred the probability format over relative risk ratios (n = 2), absolute differences (n = 2), and number of persons out of 100 (n = 0). Participants liked average lines, having a trend graph to supplement the risk estimate, and consistent colors between trend graphs and possible actions. Participants did not like too much text information or the presence of confidence intervals. From a decision-making perspective, use of the probability format was associated with greater concordance in actions taken by participants compared to the other 3 risk information formats. DISCUSSION: By focusing on nurses' preferences and decisions with several risk information display formats and collecting both qualitative and quantitative data, we have provided meaningful insights for the design of clinical decision-support systems containing complex quantitative information. CONCLUSION: This study adds to our knowledge of presenting risk information to nurses within clinical decision-support systems. We encourage those developing risk-based systems for inpatient nurses to consider expressing risk in a probability format and include a graph (with average line) to display the patient's recent trends.

Multinational patterns of second line antihyperglycaemic drug initiation across cardiovascular risk groups: federated pharmacoepidemiological evaluation in LEGEND-T2DM.

Objective: To assess the uptake of second line antihyperglycaemic drugs among patients with type 2 diabetes mellitus who are receiving metformin. Design: Federated pharmacoepidemiological evaluation in LEGEND-T2DM. Setting: 10 US and seven non-US electronic health record and administrative claims databases in the Observational Health Data Sciences and Informatics network in eight countries from 2011 to the end of 2021. Participants: 4.8 million patients (≥18 years) across US and non-US based databases with type 2 diabetes mellitus who had received metformin monotherapy and had initiated second line treatments. Exposure: The exposure used to evaluate each database was calendar year trends, with the years in the study that were specific to each cohort. Main outcomes measures: The outcome was the incidence of second line antihyperglycaemic drug use (ie, glucagon-like peptide-1 receptor agonists, sodium-glucose cotransporter-2 inhibitors, dipeptidyl peptidase-4 inhibitors, and sulfonylureas) among individuals who were already receiving treatment with metformin. The relative drug class level uptake across cardiovascular risk groups was also evaluated. Results: 4.6 million patients were identified in US databases, 61 382 from Spain, 32 442 from Germany, 25 173 from the UK, 13 270 from France, 5580 from Scotland, 4614 from Hong Kong, and 2322 from Australia. During 2011-21, the combined proportional initiation of the cardioprotective antihyperglycaemic drugs (glucagon-like peptide-1 receptor agonists and sodium-glucose cotransporter-2 inhibitors) increased across all data sources, with the combined initiation of these drugs as second line drugs in 2021 ranging from 35.2% to 68.2% in the US databases, 15.4% in France, 34.7% in Spain, 50.1% in Germany, and 54.8% in Scotland. From 2016 to 2021, in some US and non-US databases, uptake of glucagon-like peptide-1 receptor agonists and sodium-glucose cotransporter-2 inhibitors increased more significantly among populations with no cardiovascular disease compared with patients with established cardiovascular disease. No data source provided evidence of a greater increase in the uptake of these two drug classes in populations with cardiovascular disease compared with no cardiovascular disease. Conclusions: Despite the increase in overall uptake of cardioprotective antihyperglycaemic drugs as second line treatments for type 2 diabetes mellitus, their uptake was lower in patients with cardiovascular disease than in people with no cardiovascular disease over the past decade. A strategy is needed to ensure that medication use is concordant with guideline recommendations to improve outcomes of patients with type 2 diabetes mellitus.

Using intervention mapping to design and implement a multicomponent intervention to improve antibiotic and NSAID prescribing.

Successfully changing prescribing behavior to reduce inappropriate antibiotic and nonsteroidal anti-inflammatory drug (NSAID) prescriptions often requires combining components into a multicomponent intervention. However, multicomponent interventions often fail because of development and implementation complexity. To increase the likelihood of successfully changing prescribing behavior, we applied a systematic process to design and implement a multicomponent intervention. We used Intervention Mapping to create a roadmap for a multicomponent intervention in unscheduled outpatient care settings in the Veterans Health Administration. Intervention Mapping is a systematic process consisting of six steps that we grouped into three phases: (i) understand behavioral determinants and barriers to implementation, (ii) develop the intervention, and (iii) define evaluation plan and implementation strategies. A targeted literature review, combined with 25 prescriber and 25 stakeholder interviews, helped identify key behavioral determinants to inappropriate prescribing (e.g. perceived social pressure from patients to prescribe). We targeted three desired prescriber behaviors: (i) review guideline-concordant prescribing and patient outcomes, (ii) manage diagnostic and treatment uncertainty, and (iii) educate patients and caregivers. The intervention consisted of components for academic detailing, prescribing feedback, and alternative prescription order sets. Implementation strategies consisted of preparing clinical champions, conducting readiness assessments, and incentivizing use of the intervention. We chose a mixed-method study design with a commonly used evaluation framework to assess effectiveness and implementation outcomes in a subsequent trial. This study furthers knowledge about causes of inappropriate antibiotic and NSAID prescribing and demonstrates how theoretical, empirical, and practical information can be systematically applied to develop a multicomponent intervention to help address these causes.

Reducing adverse drug events from antibiotics and nonsteroidal anti-inflammatory drugs (NSAIDs) is a patient safety priority. Successfully changing prescribing behavior to reduce inappropriate prescriptions can require combining intervention components, each with different mechanisms for behavior change, into a multicomponent intervention. However, multicomponent interventions often fail because of development and implementation complexity. To increase the chance of successfully changing antibiotic and NSAID prescribing, the objective this study was to apply a systematic process to design and implement a multicomponent intervention. Three desired prescriber behaviors were targeted: (i) review guideline-concordant prescribing and patient outcomes, (ii) manage diagnostic and treatment uncertainty, and (iii) educate patients and caregivers. The designed intervention consisted of components for prescribing feedback, academic detailing, and alternative prescription order sets. Strategies to improve use of the intervention consisted of preparing clinical champions, conducting readiness assessments prior to study onset, and incentivizing use of the intervention. We chose a mixed-method study design with a commonly used evaluation framework to assess effectiveness and implementation outcomes of the multicomponent intervention in a subsequent trial.

A framework for understanding label leakage in machine learning for health care.

INTRODUCTION: The pitfalls of label leakage, contamination of model input features with outcome information, are well established. Unfortunately, avoiding label leakage in clinical prediction models requires more nuance than the common advice of applying "no time machine rule." FRAMEWORK: We provide a framework for contemplating whether and when model features pose leakage concerns by considering the cadence, perspective, and applicability of predictions. To ground these concepts, we use real-world clinical models to highlight examples of appropriate and inappropriate label leakage in practice. RECOMMENDATIONS: Finally, we provide recommendations to support clinical and technical stakeholders as they evaluate the leakage tradeoffs associated with model design, development, and implementation decisions. By providing common language and dimensions to consider when designing models, we hope the clinical prediction community will be better prepared to develop statistically valid and clinically useful machine learning models.

Machine learning to develop a predictive model of pressure injury in persons with spinal cord injury.

STUDY DESIGN: A 5-year longitudinal, retrospective, cohort study. OBJECTIVES: Develop a prediction model based on electronic health record (EHR) data to identify veterans with spinal cord injury/diseases (SCI/D) at highest risk for new pressure injuries (PIs). SETTING: Structured (coded) and text EHR data, for veterans with SCI/D treated in a VHA SCI/D Center between October 1, 2008, and September 30, 2013. METHODS: A total of 4709 veterans were available for analysis after randomly selecting 175 to act as a validation (gold standard) sample. Machine learning models were created using ten-fold cross validation and three techniques: (1) two-step logistic regression; (2) regression model employing adaptive LASSO; (3) and gradient boosting. Models based on each method were compared using area under the receiver-operating curve (AUC) analysis. RESULTS: The AUC value for the gradient boosting model was 0.62 (95% CI = 0.54-0.70), for the logistic regression model it was 0.67 (95% CI = 0.59-0.75), and for the adaptive LASSO model it was 0.72 (95% CI = 0.65-80). Based on these results, the adaptive LASSO model was chosen for interpretation. The strongest predictors of new PI cases were having fewer total days in the hospital in the year before the annual exam, higher vs. lower weight and most severe vs. less severe grade of injury based on the American Spinal Cord Injury Association (ASIA) Impairment Scale. CONCLUSIONS: While the analyses resulted in a potentially useful predictive model, clinical implications were limited because modifiable risk factors were absent in the models.

Novel integration of governmental data sources using machine learning to identify super-utilization among U.S. counties.

Background: Super-utilizers consume the greatest share of resource intensive healthcare (RIHC) and reducing their utilization remains a crucial challenge to healthcare systems in the United States (U.S.). The objective of this study was to predict RIHC among U.S. counties, using routinely collected data from the U.S. government, including information on consumer spending, offering an alternative method for identifying super-utilization among population units rather than individuals. Methods: Cross-sectional data from 5 governmental sources in 2017 were used in a machine learning pipeline, where target-prediction features were selected and used in 4 distinct algorithms. Outcome metrics of RIHC utilization came from the American Hospital Association and included yearly: (1) emergency rooms visit, (2) inpatient days, and (3) hospital expenditures. Target-prediction features included: 149 demographic characteristics from the U.S. Census Bureau, 151 adult and child health characteristics from the Centers for Disease Control and Prevention, 151 community characteristics from the American Community Survey, and 571 consumer expenditures from the Bureau of Labor Statistics. SHAP analysis identified important target-prediction features for 3 RIHC outcome metrics. Results: 2475 counties with emergency rooms and 2491 counties with hospitals were included. The median yearly emergency room visits per capita was 0.450 [IQR:0.318, 0.618], the median inpatient days per capita was 0.368 [IQR: 0.176, 0.826], and the median hospital expenditures per capita was $2104 [IQR: $1299.93, 3362.97]. The coefficient of determination (R2), calculated on the test set, ranged between 0.267 and 0.447. Demographic and community characteristics were among the important predictors for all 3 RIHC outcome metrics. Conclusions: Integrating diverse population characteristics from numerous governmental sources, we predicted 3-outcome metrics of RIHC among U.S. counties with good performance, offering a novel and actionable tool for identifying super-utilizer segments in the population. Wider integration of routinely collected data can be used to develop alternative methods for predicting RIHC among population units.