Search

BackgroundThe impact variant-specific immune evasion and waning protection have on declining coronavirus disease 2019 (COVID-19) vaccine effectiveness (VE) remains unclear. Using whole-genome sequencing (WGS), we examined the contribution these factors had on the decline that followed the introduction of the Delta variant. Furthermore, we evaluated calendar-period–based classification as a WGS alternative.MethodsWe conducted a test-negative case-control study among people tested for severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) between 1 April and 24 August 2021. Variants were classified using WGS and calendar period.ResultsWe included 2029 cases (positive, sequenced samples) and 343 727 controls (negative tests). VE 14–89 days after second dose was significantly higher against Alpha (84.4%; 95% confidence interval [CI], 75.6%–90.0%) than Delta infection (68.9%; 95% CI, 58.0%–77.1%). The odds of Delta infection were significantly higher 90–149 than 14–89 days after second dose (P value = .003). Calendar-period–classified VE estimates approximated WGS-classified estimates; however, calendar-period–based classification was subject to misclassification (35% Alpha, 4% Delta).ConclusionsBoth waning protection and variant-specific immune evasion contributed to the lower effectiveness. While calendar-period–classified VE estimates mirrored WGS-classified estimates, our analysis highlights the need for WGS when variants are cocirculating and misclassification is likely.

ObjectiveWe aimed to discover computationally-derived phenotypes of opioid-related patient presentations to the emergency department (ED) via clinical notes and structured electronic health record (EHR) data.MethodsThis was a retrospective study of ED visits from 2013-2020 across ten sites within a regional healthcare network. We derived phenotypes from visits for patients 18 years of age with at least one prior or current documentation of an opioid-related diagnosis. Natural language processing was used to extract clinical entities from notes, which were combined with structured data within the EHR to create a set of features. We performed Latent Dirichlet allocation to identify topics within these features. Groups of patient presentations with similar attributes were identified by cluster analysis.ResultsIn total 82,577 ED visits met inclusion criteria. The 30 topics discovered ranged from those related to substance use disorder, chronic conditions, mental health, and medical management. Clustering on these topics identified nine unique cohorts with one-year survivals ranging from 84.2-96.8%, rates of one-year ED returns from 9-34%, rates of one-year opioid event 10-17%, rates of medications for opioid use disorder from 17-43%, and a median Carlson comorbidity index of 2-8. Two cohorts of phenotypes were identified related to chronic substance use disorder, or acute overdose.ConclusionsOur results indicate distinct phenotypic clusters with varying patient-oriented outcomes which provide future targets better allocation of resources and therapeutics. This highlights the heterogeneity of the overall population, and the need to develop targeted interventions for each population.

Artificial intelligence (AI) can detect left ventricular systolic dysfunction (LVSD) from electrocardiograms (ECGs). Wearable devices could allow for broad AI-based screening but frequently obtain noisy ECGs. We report a novel strategy that automates the detection of hidden cardiovascular diseases, such as LVSD, adapted for noisy single-lead ECGs obtained on wearable and portable devices. Overall, 385,601 ECGs were used for development of a standard and noise-adapted model. For the noise-adapted model, ECGs were augmented during training with random gaussian noise within four distinct frequency ranges, each emulating real-world noise sources. Both models performed comparably on clean ECGs with an AUROC of 0.90. The noise-adapted model performed significantly better on the same test set augmented with four distinct real-world noise recordings at multiple signal-to-noise ratios (SNRs), including noise isolated from a portable device ECG. The standard and noise-adapted models had an AUROC of 0.72 and 0.87, respectively when evaluated on ECGs augmented with portable ECG device noise at an SNR of 0.5. This approach represents a novel strategy for the development of wearable adapted tools from clinical ECG repositories.

Background and AimsEarly diagnosis of aortic stenosis (AS) is critical to prevent morbidity and mortality but requires skilled examination with Doppler imaging. This study reports the development and validation of a novel deep learning model that relies on 2-dimensional parasternal long axis (PLAX) videos from transthoracic echocardiography (TTE) without Doppler imaging to identify severe AS, suitable for point-of-care ultrasonography.MethodsIn a training set of 5,257 studies (17,570 videos) from 2016-2020 (Yale-New Haven Hospital [YNHH], Connecticut), an ensemble of 3-dimensional convolutional neural networks was developed to detect severe AS, leveraging self-supervised contrastive pretraining for label-efficient model development. This deep learning model was validated in a temporally distinct set of 2,040 consecutive studies from 2021 from YNHH as well as two geographically distinct cohorts of 5,572 and 865 studies, from California and other hospitals in New England, respectively.ResultsThe deep learning model achieved an AUROC of 0.978 (95% CI: 0.966, 0.988) for detecting severe AS with 95.4% specificity and 90% sensitivity in the temporally distinct test set, maintaining its diagnostic performance in both geographically distinct cohorts (AUROC 0.972 [95% CI: 0.969, 0.975] in California and 0.915 [95% CI: 0.896, 0.933] in New England, respectively). The model was interpretable with saliency maps identifying the aortic valve as the predictive region. Among non-severe AS cases, predicted probabilities were associated with worse quantitative metrics of AS suggesting association with various stages of AS severity.ConclusionsThis study developed and externally validated an automated approach for severe AS detection using single-view 2D echocardiography, with implications for point-of-care screening.STRUCTURED GRAPHICAL ABSTRACTKey QuestionIs it feasible to automatically screen for the presence of severe aortic stenosis (AS) using single-view echocardiographic videos without the use of Doppler imaging?Key FindingUsing self-supervised pretraining and ensemble learning, we trained a deep learning model to detect severe AS using single-view echocardiography without Doppler imaging. The model maintained its high performance in multiple geographically and temporally distinct cohorts.Take-home MessageWe present an automated method to detect severe AS using single-view TTE videos, with implications for point-of-care ultrasound screening as part of routine clinic visits and in limited resource settings by individuals with minimal training.An automated deep learning approach for severe aortic stenosis detection from single-view echocardiography evaluated across geographically and temporally distinct cohorts.

BackgroundThe decline in COVID-19 mRNA vaccine effectiveness (VE) is well established, however the impact of variant-specific immune evasion and waning protection remains unclear. Here, we use whole-genome-sequencing (WGS) to tease apart the contribution of these factors on the decline observed following the introduction of the Delta variant. Further, we evaluate the utility of calendar-period-based variant classification as an alternative to WGS.MethodsWe conducted a test-negative-case-control study among people who received SARS-CoV-2 RT-PCR testing in the Yale New Haven Health System between April 1 and August 24, 2021. Variant classification was performed using WGS and secondarily by calendar-period. We estimated VE as one minus the ratio comparing the odds of infection among vaccinated and unvaccinated people.ResultsOverall, 2,029 cases (RT-PCR positive, sequenced samples) and 343,985 controls (negative RT-PCRs) were included. VE 14-89 days after 2nd dose was significantly higher against WGS-classified Alpha infection (84.4%, 95% confidence interval: 75.6-90.0%) than Delta infection (68.9%, CI: 58.0-77.1%, p-value: 0.013). The odds of WGS-classified Delta infection were significantly higher 90-149 than 14-89 days after 2nd dose (Odds ratio: 1.6, CI: 1.2-2.3). While estimates of VE against calendar-period-classified infections approximated estimates against WGS-classified infections, calendar-period-based classification was subject to outcome misclassification (35% during Alpha period, 4% during Delta period).ConclusionsThese findings suggest that both waning protection and variant-specific immune evasion contributed to the lower effectiveness. While estimates of VE against calendar-period-classified infections mirrored that against WGS-classified infections, our analysis highlights the need for WGS when variants are co-circulating and misclassification is likely.Summary of main pointsUsing whole genome sequencing, we provide direct evidence of waning vaccine effectiveness and variant-specific immune evasion during the Delta wave. Effectiveness estimates against calendar-period-classified infections approximated estimates against WGS-classified infections, however, calendar-period classification was associated with a variant misclassification.

COVID-19 manifests with a wide spectrum of clinical phenotypes that are characterized by exaggerated and misdirected host immune responses1-6. Although pathological innate immune activation is well-documented in severe disease1, the effect of autoantibodies on disease progression is less well-defined. Here we use a high-throughput autoantibody discovery technique known as rapid extracellular antigen profiling7 to screen a cohort of 194 individuals infected with SARS-CoV-2, comprising 172 patients with COVID-19 and 22 healthcare workers with mild disease or asymptomatic infection, for autoantibodies against 2,770 extracellular and secreted proteins (members of the exoproteome). We found that patients with COVID-19 exhibit marked increases in autoantibody reactivities as compared to uninfected individuals, and show a high prevalence of autoantibodies against immunomodulatory proteins (including cytokines, chemokines, complement components and cell-surface proteins). We established that these autoantibodies perturb immune function and impair virological control by inhibiting immunoreceptor signalling and by altering peripheral immune cell composition, and found that mouse surrogates of these autoantibodies increase disease severity in a mouse model of SARS-CoV-2 infection. Our analysis of autoantibodies against tissue-associated antigens revealed associations with specific clinical characteristics. Our findings suggest a pathological role for exoproteome-directed autoantibodies in COVID-19, with diverse effects on immune functionality and associations with clinical outcomes.

The emergence of severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) variants continues to shape the coronavirus disease 2019 (Covid-19) pandemic. The detection and rapid spread of the SARS-CoV-2 ‘ Omicron’ variant (lineage B.1.1.529) in Botswana and South Africa became a global concern because it contained 15 mutations in the spike protein immunogenic receptor binding domain and was less neutralized by sera derived from vaccinees compared to the previously dominant Delta variant. To investigate if Omicron is more likely than Delta to cause infections in vaccinated persons, we analyzed 37,877 nasal swab PCR tests conducted from 12-26 December 2021 and calculated the test positivity rates for each variant by vaccination status. We found that the positivity rate among unvaccinated persons was higher for Delta (5.2%) than Omicron (4.5%). We found similar results in persons who received a single vaccine dose. Conversely, our results show that Omicron had higher positivity rates than Delta among those who received two doses within five months (Omicron = 4.7% vs. Delta = 2.6%), two doses more than five months ago (4.2% vs. 2.9%), and three vaccine doses (2.2% vs. 0.9%). Our estimates of Omicron positivity rates in persons receiving one or two vaccine doses were not significantly lower than unvaccinated persons but were 49.7% lower after three doses. In comparison, the reduction in Delta positivity rates from unvaccinated to 2 vaccine doses was 45.6-49.6% and to 3 vaccine doses was 83.2%. Despite the higher positivity rates for Omicron in vaccinated persons, we still found that 91.2% of the Omicron infections in our study occurred in persons who were eligible for 1 or more vaccine doses at the time of PCR testing. In conclusion, escape from vaccine-induced immunity likely contributed to the rapid rise in Omicron infections.

Background: Intraoperative data may improve models predicting postoperative events. We evaluated the effect of incorporating intraoperative variables to the existing preoperative model on the predictive performance of the model for coronary artery bypass graft. Methods: We analyzed 378 572 isolated coronary artery bypass graft cases performed across 1083 centers, using the national Society of Thoracic Surgeons Adult Cardiac Surgery Database between 2014 and 2016. Outcomes were operative mortality, 5 postoperative complications, and composite representation of all events. We fitted models by logistic regression or extreme gradient boosting (XGBoost). For each modeling approach, we used preoperative only, intraoperative only, or pre+intraoperative variables. We developed 84 models with unique combinations of the 3 variable sets, 2 variable selection methods, 2 modeling approaches, and 7 outcomes. Each model was tested in 20 iterations of 70:30 stratified random splitting into development/testing samples. Model performances were evaluated on the testing dataset using the C statistic, area under the precision-recall curve, and calibration metrics, including the Brier score. Results: The mean patient age was 65.3 years, and 24.7% were women. Operative mortality, excluding intraoperative death, occurred in 1.9%. In all outcomes, models that considered pre+intraoperative variables demonstrated significantly improved Brier score and area under the precision-recall curve compared with models considering pre or intraoperative variables alone. XGBoost without external variable selection had the best C statistics, Brier score, and area under the precision-recall curve values in 4 of the 7 outcomes (mortality, renal failure, prolonged ventilation, and composite) compared with logistic regression models with or without variable selection. Based on the calibration plots, risk restratification for mortality showed that the logistic regression model underestimated the risk in 11 114 patients (9.8%) and overestimated in 12 005 patients (10.6%). In contrast, the XGBoost model underestimated the risk in 7218 patients (6.4%) and overestimated in 0 patients (0%). Conclusions: In isolated coronary artery bypass graft, adding intraoperative variables to preoperative variables resulted in improved predictions of all 7 outcomes. Risk models based on XGBoost may provide a better prediction of adverse events to guide clinical care.

COVID-19 manifests with a wide spectrum of clinical phenotypes that are characterized by exaggerated and misdirected host immune responses

Background The electronic health record (EHR) holds the prospect of providing more complete and timely access to clinical information for biomedical research, quality assessments, and quality improvement compared to other data sources, such as administrative claims. In this study, we sought to assess the completeness and timeliness of structured diagnoses in the EHR compared to computed diagnoses for hypertension (HTN), hyperlipidemia (HLD), and diabetes mellitus (DM). Methods We determined the amount of time for a structured diagnosis to be recorded in the EHR from when an equivalent diagnosis could be computed from other structured data elements, such as vital signs and laboratory results. We used EHR data for encounters from January 1, 2012 through February 10, 2019 from an academic health system. Diagnoses for HTN, HLD, and DM were computed for patients with at least two observations above threshold separated by at least 30 days, where the thresholds were outpatient blood pressure of ≥ 140/90 mmHg, any low-density lipoprotein ≥ 130 mg/dl, or any hemoglobin A1c ≥ 6.5%, respectively. The primary measure was the length of time between the computed diagnosis and the time at which a structured diagnosis could be identified within the EHR history or problem list. Results We found that 39.8% of those with HTN, 21.6% with HLD, and 5.2% with DM did not receive a corresponding structured diagnosis recorded in the EHR. For those who received a structured diagnosis, a mean of 389, 198, and 166 days elapsed before the patient had the corresponding diagnosis of HTN, HLD, or DM, respectively, recorded in the EHR. Conclusions We found a marked temporal delay between when a diagnosis can be computed or inferred and when an equivalent structured diagnosis is recorded within the EHR. These findings demonstrate the continued need for additional study of the EHR to avoid bias when using observational data and reinforce the need for computational approaches to identify clinical phenotypes.

Rationale and Objective Acute kidney injury (AKI) is common in patients with COVID-19 and associated with poor outcomes. Urinary biomarkers have been associated with adverse kidney outcomes in other settings and may provide additional prognostic information in patients with COVID-19. We investigated the association between urinary biomarkers with adverse kidney outcomes among patients hospitalized with COVID-19. Study Design Prospective cohort study. Setting and Participants Patients hospitalized with COVID-19 (n=153) at 2 academic medical centers between April and June 2020. Exposures 19 urinary biomarkers of injury, inflammation, and repair. Outcomes Composite of KDIGO stage 3 AKI, requirement for dialysis, or death within 60 days of hospital admission. We also compared various kidney biomarker levels in the setting of COVID-19 versus other common AKI settings. Analytic Approach Time-varying Cox proportional hazards regression to associate biomarker level with composite outcome. Results Out of 153 patients, 24 (15.7%) experienced the primary outcome. Two-fold higher levels of neutrophil gelatinase-associated lipocalin (NGAL) (HR: 1.34; 95% CI: 1.14-1.57), monocyte chemoattractant protein (MCP-1) (HR: 1.42; 95% CI: 1.09-1.84), and kidney injury molecule-1 (KIM-1) (HR: 2.03; 95% CI: 1.38-2.99) were associated with highest risk of sustaining primary composite outcome. Higher epidermal growth factor (EGF) levels were associated with a lower risk of the primary outcome (HR 0.61; 95% CI: 0.47-0.79). Individual biomarkers provided moderate discrimination and biomarker combinations improved discrimination for the primary outcome. The degree of kidney injury by biomarker level in COVID-19 was comparable to other settings of clinical AKI. There was evidence of subclinical AKI in COVID-19 patients based on elevated injury biomarker level in patients without clinical AKI defined by serum creatinine. Limitations Small sample size with low number of composite outcome events. Conclusion Urinary biomarkers are associated with adverse kidney outcomes in patients hospitalized with COVID-19 and may provide valuable information to monitor kidney disease progression and recovery., Graphical abstract

Electronic health records (EHR) provide opportunities to leverage vast arrays of data to help prevent adverse events, improve patient outcomes, and reduce hospital costs. This paper develops a postoperative complications prediction system by extracting data from the EHR and creating features. The analytic engine then provides model accuracy, calibration, feature ranking, and personalized feature responses. This allows clinicians to interpret the likelihood of an adverse event occurring, general causes for these events, and the contributing factors for each specific patient. The patient cohort considered was 5214 patients in Yale-New Haven Hospital undergoing major cardiovascular procedures. Cohort-specific models predicted the likelihood of postoperative respiratory failure and infection, and achieved an area under the receiver operating characteristic curve of 0.81 for respiratory failure and 0.83 for infection.

This comparative effectiveness research study assesses whether using present on admission codes and single, rather than grouped, diagnostic codes can enhance Centers for Medicare & Medicaid (CMS) models to predict payment for hospitalization for acute myocardial infarction, heart failure, and pneumonia., Key Points Question Does leveraging present on admission codes and using single, rather than grouped, diagnostic codes enhance risk models for acute myocardial infarction, heart failure, and pneumonia payment measures? Findings In this comparative effectiveness research study of risk models on 1 667 983 patients with 1 943 049 Medicare fee-for-service hospitalizations, use of present on admission codes and single diagnosis codes and separation of index admission codes from codes in the previous year improved models predicting payment that were compared with models based on Centers for Medicare & Medicaid Services grouped codes. The patient-level pseudo R2 improved from 0.077 to 0.129 for acute myocardial infarction, from 0.042 to 0.129 for heart failure, and from 0.114 to 0.237 for pneumonia. Meaning Changing candidate variables from the current standard improved models predicting payments, which has implications for research, benchmarking, public reporting, and calculations for population-based programs., Importance Predicting payments for particular conditions or populations is essential for research, benchmarking, public reporting, and calculations for population-based programs. Centers for Medicare & Medicaid Services (CMS) models often group codes into disease categories, but using single, rather than grouped, diagnostic codes and leveraging present on admission (POA) codes may enhance these models. Objective To determine whether changes to the candidate variables in CMS models would improve risk models predicting patient total payment within 30 days of hospitalization for acute myocardial infarction (AMI), heart failure (HF), and pneumonia. Design, Setting, and Participants This comparative effectiveness research study used data from Medicare fee-for-service hospitalizations for AMI, HF, and pneumonia at acute care hospitals from July 1, 2013, through September 30, 2015. Payments across multiple care settings, services, and supplies were included and adjusted for geographic and policy variations, corrected for inflation, and winsorized. The same data source was used but varied for the candidate variables and their selection, and the method used by CMS for public reporting that used grouped codes was compared with variations that used POA codes and single diagnostic codes. Combinations of use of POA codes, separation of index admission diagnoses from those in the previous 12 months, and use of individual International Classification of Diseases, Ninth Revision, Clinical Modification codes instead of grouped diagnostic categories were tested. Data analysis was performed from December 4, 2017, to June 10, 2019. Main Outcomes and Measures The models’ goodness of fit was compared using root mean square error (RMSE) and the McFadden pseudo R2. Results Among the 1 943 049 total hospitalizations of the study participants, 343 116 admissions were for AMI (52.5% male; 37.4% aged ≤74 years), 677 044 for HF (45.5% male; 25.9% aged ≤74 years), and 922 889 for pneumonia (46.4% male; 28.2% aged ≤74 years). The mean (SD) 30-day payment was $23 103 ($18 221) for AMI, $16 365 ($12 527) for HF, and $17 097 ($12 087) for pneumonia. Each incremental model change improved the pseudo R2 and RMSE. Incorporating all 3 changes improved the pseudo R2 of the patient-level models from 0.077 to 0.129 for AMI, from 0.042 to 0.129 for HF, and from 0.114 to 0.237 for pneumonia. Parallel improvements in RMSE were found for all 3 conditions. Conclusions and Relevance Leveraging POA codes, separating index from previous diagnoses, and using single diagnostic codes improved payment models. Better models can potentially improve research, benchmarking, public reporting, and calculations for population-based programs.

BACKGROUND Health care data are increasing in volume and complexity. Storing and analyzing these data to implement precision medicine initiatives and data-driven research has exceeded the capabilities of traditional computer systems. Modern big data platforms must be adapted to the specific demands of health care and designed for scalability and growth. OBJECTIVE The objectives of our study were to (1) demonstrate the implementation of a data science platform built on open source technology within a large, academic health care system and (2) describe 2 computational health care applications built on such a platform. METHODS We deployed a data science platform based on several open source technologies to support real-time, big data workloads. We developed data-acquisition workflows for Apache Storm and NiFi in Java and Python to capture patient monitoring and laboratory data for downstream analytics. RESULTS Emerging data management approaches, along with open source technologies such as Hadoop, can be used to create integrated data lakes to store large, real-time datasets. This infrastructure also provides a robust analytics platform where health care and biomedical research data can be analyzed in near real time for precision medicine and computational health care use cases. CONCLUSIONS The implementation and use of integrated data science platforms offer organizations the opportunity to combine traditional datasets, including data from the electronic health record, with emerging big data sources, such as continuous patient monitoring and real-time laboratory results. These platforms can enable cost-effective and scalable analytics for the information that will be key to the delivery of precision medicine initiatives. Organizations that can take advantage of the technical advances found in data science platforms will have the opportunity to provide comprehensive access to health care data for computational health care and precision medicine research.

Background: Health care data are increasing in volume and complexity. Storing and analyzing these data to implement precision medicine initiatives and data-driven research has exceeded the capabilities of traditional computer systems. Modern big data platforms must be adapted to the specific demands of health care and designed for scalability and growth. Objective: The objectives of our study were to (1) demonstrate the implementation of a data science platform built on open source technology within a large, academic health care system and (2) describe 2 computational health care applications built on such a platform. Methods: We deployed a data science platform based on several open source technologies to support real-time, big data workloads. We developed data-acquisition workflows for Apache Storm and NiFi in Java and Python to capture patient monitoring and laboratory data for downstream analytics. Results: Emerging data management approaches, along with open source technologies such as Hadoop, can be used to create integrated data lakes to store large, real-time datasets. This infrastructure also provides a robust analytics platform where health care and biomedical research data can be analyzed in near real time for precision medicine and computational health care use cases. Conclusions: The implementation and use of integrated data science platforms offer organizations the opportunity to combine traditional datasets, including data from the electronic health record, with emerging big data sources, such as continuous patient monitoring and real-time laboratory results. These platforms can enable cost-effective and scalable analytics for the information that will be key to the delivery of precision medicine initiatives. Organizations that can take advantage of the technical advances found in data science platforms will have the opportunity to provide comprehensive access to health care data for computational health care and precision medicine research.

PURPOSE: To estimate medical device utilization needed to detect safety differences among implantable cardioverter defibrillators (ICDs) generator models and compare these estimates to utilization in practice. METHODS: We conducted repeated sample size estimates to calculate the medical device utilization needed, systematically varying device‐specific safety event rate ratios and significance levels while maintaining 80% power, testing 3 average adverse event rates (3.9, 6.1, and 12.6 events per 100 person‐years) estimated from the American College of Cardiology’s 2006 to 2010 National Cardiovascular Data Registry of ICDs. We then compared with actual medical device utilization. RESULTS: At significance level 0.05 and 80% power, 34% or fewer ICD models accrued sufficient utilization in practice to detect safety differences for rate ratios

Background SPRINT (Systolic Blood Pressure Intervention Trial) and the ACCORD (Action to Control Cardiovascular Risk in Diabetes) blood pressure trial used similar interventions but produced discordant results. We investigated whether differences in systolic blood pressure ( SBP ) response contributed to the discordant trial results. Methods and Results We evaluated the distributions of SBP response during the first year for the intensive and standard treatment groups of SPRINT and ACCORD using growth mixture models. We assessed whether significant differences existed between trials in the distributions of SBP achieved at 1 year and the treatment‐independent relationships of achieved SBP with risks of primary outcomes defined in each trial, heart failure, stroke, and all‐cause death. We examined whether visit‐to‐visit variability was associated with heterogeneous treatment effects. Among the included 9027 SPRINT and 4575 ACCORD participants, the difference in mean SBP achieved between treatment groups was 15.7 mm Hg in SPRINT and 14.2 mm Hg in ACCORD , but SPRINT had significantly less between‐group overlap in the achieved SBP (standard deviations of intensive and standard groups, respectively: 6.7 and 5.9 mm Hg in SPRINT versus 8.8 and 8.2 mm Hg in ACCORD ; P SBP and outcomes was consistent across trials except for stroke and all‐cause death. Higher visit‐to‐visit variability was more common in SPRINT but without treatment‐effect heterogeneity. Conclusions SPRINT and ACCORD had different degrees of separation in achieved SBP between treatment groups, even as they had similar mean differences. The greater between‐group overlap of achieved SBP may have contributed to the discordant trial results.

Randomized trials of hypertension have seldom examined heterogeneity in response to treatments over time and the implications for cardiovascular outcomes. Understanding this heterogeneity, however, is a necessary step toward personalizing antihypertensive therapy. We applied trajectory-based modeling to data on 39 763 study participants of the ALLHAT (Antihypertensive and Lipid-Lowering Treatment to Prevent Heart Attack Trial) to identify distinct patterns of systolic blood pressure (SBP) response to randomized medications during the first 6 months of the trial. Two trajectory patterns were identified: immediate responders (85.5%), on average, had a decreasing SBP, whereas nonimmediate responders (14.5%), on average, had an initially increasing SBP followed by a decrease. Compared with those randomized to chlorthalidone, participants randomized to amlodipine (odds ratio, 1.20; 95% confidence interval [CI], 1.10–1.31), lisinopril (odds ratio, 1.88; 95% CI, 1.73–2.03), and doxazosin (odds ratio, 1.65; 95% CI, 1.52–1.78) had higher adjusted odds ratios associated with being a nonimmediate responder (versus immediate responder). After multivariable adjustment, nonimmediate responders had a higher hazard ratio of stroke (hazard ratio, 1.49; 95% CI, 1.21–1.84), combined cardiovascular disease (hazard ratio, 1.21; 95% CI, 1.11–1.31), and heart failure (hazard ratio, 1.48; 95% CI, 1.24–1.78) during follow-up between 6 months and 2 years. The SBP response trajectories provided superior discrimination for predicting downstream adverse cardiovascular events than classification based on difference in SBP between the first 2 measurements, SBP at 6 months, and average SBP during the first 6 months. Our findings demonstrate heterogeneity in response to antihypertensive therapies and show that chlorthalidone is associated with more favorable initial response than the other medications.

Identifying temporal variation in hospitalization rates may provide insights about disease patterns and thereby inform research, policy, and clinical care. However, the majority of medical conditions have not been studied for their potential seasonal variation. The objective of this study was to apply a data-driven approach to characterize temporal variation in condition-specific hospitalizations. Using a dataset of 34 million inpatient discharges gathered from hospitals in New York State from 2008–2011, we grouped all discharges into 263 clinical conditions based on the principal discharge diagnosis using Clinical Classification Software in order to mitigate the limitation that administrative claims data reflect clinical conditions to varying specificity. After applying Seasonal-Trend Decomposition by LOESS, we estimated the periodicity of the seasonal component using spectral analysis and applied harmonic regression to calculate the amplitude and phase of the condition’s seasonal utilization pattern. We also introduced four new indices of temporal variation: mean oscillation width, seasonal coefficient, trend coefficient, and linearity of the trend. Finally, K-means clustering was used to group conditions across these four indices to identify common temporal variation patterns. Of all 263 clinical conditions considered, 164 demonstrated statistically significant seasonality. Notably, we identified conditions for which seasonal variation has not been previously described such as ovarian cancer, tuberculosis, and schizophrenia. Clustering analysis yielded three distinct groups of conditions based on multiple measures of seasonal variation. Our study was limited to New York State and results may not directly apply to other regions with distinct climates and health burden. A substantial proportion of medical conditions, larger than previously described, exhibit seasonal variation in hospital utilization. Moreover, the application of clustering tools yields groups of clinically heterogeneous conditions with similar seasonal phenotypes. Further investigation is necessary to uncover common etiologies underlying these shared seasonal phenotypes.

This comparative effectiveness study examines current US Centers for Medicare & Medicaid Services 30-day mortality risk models vs novel risk models for acute myocardial infarction, heart failure, and pneumonia hospitalizations., Key Points Question Could present on admission indicators and ungrouped diagnostic codes enhance risk models for acute myocardial infarction, heart failure, and pneumonia mortality measures and improve discrimination of hospital-level performance? Findings In this comparative effectiveness study including all Medicare fee-for-service beneficiaries hospitalized for acute myocardial infarction, heart failure, or pneumonia at acute care hospitals, incorporating present on admission coding and ungrouped historical and index admission International Classification of Diseases, Ninth Revision, Clinical Modification codes was associated with greater discrimination in patient-level and hospital-level 30-day mortality risk models. Meaning Changes incurring no additional cost could enhance the risk adjustment for mortality and increase discrimination of hospital-level performance., Importance Risk adjustment models using claims-based data are central in evaluating health care performance. Although US Centers for Medicare & Medicaid Services (CMS) models apply well-vetted statistical approaches, recent changes in the International Classification of Diseases, Ninth Revision, Clinical Modification (ICD-9-CM) coding system and advances in computational capabilities may provide an opportunity for enhancement. Objective To examine whether changes using already available data would enhance risk models and yield greater discrimination in hospital-level performance measures. Design, Setting, and Participants This comparative effectiveness study used ICD-9-CM codes from all Medicare fee-for-service beneficiary claims for hospitalizations for acute myocardial infarction (AMI), heart failure (HF), or pneumonia among patients 65 years and older from July 1, 2013, through September 30, 2015. Changes to current CMS mortality risk models were applied incrementally to patient-level models, and the best model was tested on hospital performance measures to model 30-day mortality. Analyses were conducted from April 19, 2018, to September 19, 2018. Main Outcomes and Measures The main outcome was all-cause death within 30 days of hospitalization for AMI, HF, or pneumonia, examined using 3 changes to current CMS mortality risk models: (1) incorporating present on admission coding to better exclude potential complications of care, (2) separating index admission diagnoses from those of the 12-month history, and (3) using ungrouped ICD-9-CM codes. Results There were 361 175 hospital admissions (mean [SD] age, 78.6 [8.4] years; 189 225 [52.4%] men) for AMI, 716 790 hospital admissions (mean [SD] age, 81.1 [8.4] years; 326 825 [45.6%] men) for HF, and 988 225 hospital admissions (mean [SD] age, 80.7 [8.6] years; 460 761 [46.6%] men) for pneumonia during the study; mean 30-day mortality rates were 13.8% for AMI, 12.1% for HF, and 16.1% for pneumonia. Each change to the models was associated with incremental gains in C statistics. The best model, incorporating all changes, was associated with significantly improved patient-level C statistics, from 0.720 to 0.826 for AMI, 0.685 to 0.776 for HF, and 0.715 to 0.804 for pneumonia. Compared with current CMS models, the best model produced wider predicted probabilities with better calibration and Brier scores. Hospital risk-standardized mortality rates had wider distributions, with more hospitals identified as good or bad performance outliers. Conclusions and Relevance Incorporating present on admission coding and using ungrouped index and historical ICD-9-CM codes were associated with improved patient-level and hospital-level risk models for mortality compared with the current CMS models for all 3 conditions.

Background: Prior randomized trials of hypertension have rarely examined patient heterogeneity in response to treatments and the implications for outcomes. Methods: We applied growth mixture modeling to identify distinct SBP trajectory classes within the first 6 months of ALLHAT. We assessed the

Throughout this Phase I project, we have integrated a suite of nonlinear signal processing algorithms derived from diffusion geometry into an existing proprietary Hyperspectral processing toolbox. These methods enable the organization and comparison of spatio-spectral features of hyperspectral images acquired under different conditions, for target detection, change and anomaly assessment. The main ingredients in our approach involve a high level "geometrization" of spatio spectral signatures. We developed an approach to simultaneously segment a scene in terms of similarities of spatio spectral signatures at different inference as well as a partition of the feature space of spectra and morphology into groups of features related to the various locations on the scene. We refer to this approach in which we interrogate and organize both the pixels and their responses as the questionnaire organization paradigm. This spectral segmentation methodology is critical for change detection as it enables to isolate changes by comparing their relation to their spatio-spectral folders. The folder identity provides invariant features for change detection.

We apply a unique micro-optoelectromechanical tuned light source and new algorithms to the hyper-spectral microscopic analysis of human colon biopsies. The tuned light prototype (Plain Sight Systems Inc.) transmits any combination of light frequencies, range 440nm 700nm, trans-illuminating H and E stained tissue sections of normal (N), benign adenoma (B) and malignant carcinoma (M) colon biopsies, through a Nikon Biophot microscope. Hyper-spectral photomicrographs, randomly collected 400X magnication, are obtained with a CCD camera (Sensovation) from 59 different patient biopsies (20 N, 19 B, 20 M) mounted as a microarray on a single glass slide. The spectra of each pixel are normalized and analyzed to discriminate among tissue features: gland nuclei, gland cytoplasm and lamina propria/lumens. Spectral features permit the automatic extraction of 3298 nuclei with classification as N, B or M. When nuclei are extracted from each of the 59 biopsies the average classification among N, B and M nuclei is 97.1%; classification of the biopsies, based on the average nuclei classification, is 100%. However, when the nuclei are extracted from a subset of biopsies, and the prediction is made on nuclei in the remaining biopsies, there is a marked decrement in performance to 60% across the 3 classes. Similarly the biopsy classification drops to 54%. In spite of these classification differences, which we believe are due to instrument and biopsy normalization issues, hyper-spectral analysis has the potential to achieve diagnostic efficiency needed for objective microscopic diagnosis.