Wyndy L. Wiitala

Follow-up of Glycemic Control and Cardiovascular Outcomes in Type 2 Diabetes

BACKGROUND The Veterans Affairs Diabetes Trial previously showed that intensive glucose lowering, as compared with standard therapy, did not significantly reduce the rate of major cardiovascular events among 1791 military veterans (median follow-up, 5.6 years). We report the extended follow-up of the study participants. METHODS After the conclusion of the clinical trial, we followed participants, using central databases to identify procedures, hospitalizations, and deaths (complete cohort, with follow-up data for 92.4% of participants). Most participants agreed to additional data collection by means of annual surveys and periodic chart reviews (survey cohort, with 77.7% follow-up). The primary outcome was the time to the first major cardiovascular event (heart attack, stroke, new or worsening congestive heart failure, amputation for ischemic gangrene, or cardiovascular-related death). Secondary outcomes were cardiovascular mortality and all-cause mortality. RESULTS The difference in glycated hemoglobin levels between the intensive-therapy group and the standard-therapy group averaged 1.5 percentage points during the trial (median level, 6.9% vs. 8.4%) and declined to 0.2 to 0.3 percentage points by 3 years after the trial ended. Over a median follow-up of 9.8 years, the intensive-therapy group had a significantly lower risk of the primary outcome than did the standard-therapy group (hazard ratio, 0.83; 95% confidence interval [CI], 0.70 to 0.99; P=0.04), with an absolute reduction in risk of 8.6 major cardiovascular events per 1000 person-years, but did not have reduced cardiovascular mortality (hazard ratio, 0.88; 95% CI, 0.64 to 1.20; P=0.42). No reduction in total mortality was evident (hazard ratio in the intensive-therapy group, 1.05; 95% CI, 0.89 to 1.25; P=0.54; median follow-up, 11.8 years). CONCLUSIONS After nearly 10 years of follow-up, patients with type 2 diabetes who had been randomly assigned to intensive glucose control for 5.6 years had 8.6 fewer major cardiovascular events per 1000 person-years than those assigned to standard therapy, but no improvement was seen in the rate of overall survival. (Funded by the VA Cooperative Studies Program and others; VADT ClinicalTrials.gov number, NCT00032487.).

Intensive Care Unit Admitting Patterns in the Veterans Affairs Health Care System

BACKGROUND Critical care resource use accounts for almost 1% of US gross domestic product and varies widely among hospitals. However, we know little about the initial decision to admit a patient to the intensive care unit (ICU). METHODS To describe hospital ICU admitting patterns for medical patients after accounting for severity of illness on admission, we performed a retrospective cohort study of the first nonsurgical admission of 289,310 patients admitted from the emergency department or the outpatient clinic to 118 Veterans Affairs acute care hospitals between July 1, 2009, and June 30, 2010. Severity (30-day predicted mortality rate) was measured using a modified Veterans Affairs ICU score based on laboratory data and comorbidities around admission. The main outcome measure was direct admission to an ICU. RESULTS Of the 31,555 patients (10.9%) directly admitted to the ICU, 53.2% had 30-day predicted mortality at admission of 2% or less. The rate of ICU admission for this low-risk group varied from 1.2% to 38.9%. For high-risk patients (predicted mortality >30%), ICU admission rates also varied widely. For a 1-SD increase in predicted mortality, the adjusted odds of ICU admission varied substantially across hospitals (odds ratio = 0.85-2.22). As a result, 66.1% of hospitals were in different quartiles of ICU use for low- vs high-risk patients (weighted κ = 0.50). CONCLUSIONS The proportion of low- and high-risk patients admitted to the ICU, variation in ICU admitting patterns among hospitals, and the sensitivity of hospital rankings to patient risk all likely reflect a lack of consensus about which patients most benefit from ICU admission.

Improved cardiovascular risk prediction using nonparametric regression and electronic health record data

Background:Use of the electronic health record (EHR) is expected to increase rapidly in the near future, yet little research exists on whether analyzing internal EHR data using flexible, adaptive statistical methods could improve clinical risk prediction. Extensive implementation of EHR in the Veterans Health Administration provides an opportunity for exploration. Objectives:To compare the performance of various approaches for predicting risk of cerebrovascular and cardiovascular (CCV) death, using traditional risk predictors versus more comprehensive EHR data. Research Design:Retrospective cohort study. We identified all Veterans Health Administration patients without recent CCV events treated at 12 facilities from 2003 to 2007, and predicted risk using the Framingham risk score, logistic regression, generalized additive modeling, and gradient tree boosting. Measures:The outcome was CCV-related death within 5 years. We assessed each method’s predictive performance with the area under the receiver operating characteristic curve (AUC), the Hosmer-Lemeshow goodness-of-fit test, plots of estimated risk, and reclassification tables, using cross-validation to penalize overfitting. Results:Regression methods outperformed the Framingham risk score, even with the same predictors (AUC increased from 71% to 73% and calibration also improved). Even better performance was attained in models using additional EHR-derived predictor variables (AUC increased to 78% and net reclassification improvement was as large as 0.29). Nonparametric regression further improved calibration and discrimination compared with logistic regression. Conclusions:Despite the EHR lacking some risk factors and its imperfect data quality, health care systems may be able to substantially improve risk prediction for their patients by using internally developed EHR-derived models and flexible statistical methodology.

Despite variation in volume, Veterans Affairs hospitals show consistent outcomes among patients with non-postoperative mechanical ventilation

Objective:To assess the relationship between volume of nonoperative mechanically ventilated patients receiving care in a specific Veterans Health Administration hospital and their mortality. Design:Retrospective cohort study. Setting:One-hundred nineteen Veterans Health Administration medical centers. Patients:We identified 5,131 hospitalizations involving mechanically ventilated patients in an intensive care unit during 2009, who did not receive surgery. Interventions:None. Measurements and Main Results:We extracted demographic and clinical data from the VA Inpatient Evaluation Center. For each hospital, we defined volume as the total number of nonsurgical admissions receiving mechanical ventilation in an intensive care unit during 2009. We examined the hospital contribution to 30-day mortality using multilevel logistic regression models with a random intercept for each hospital. We quantified the extent of interhospital variation in 30-day mortality using the intraclass correlation coefficient and median odds ratio. We used generalized estimating equations to examine the relationship between volume and 30-day mortality and risk-adjusted all models using a patient-level prognostic score derived from clinical data representing the risk of death conditional on treatment at a high-volume hospital. Mean age for the sample was 65 (SD 11) yrs, 97% were men, and 60% were white. The median VA hospital cared for 40 (interquartile range 19–62) mechanically ventilated patients in 2009. Crude 30-day mortality for these patients was 36.9%. After reliability and risk adjustment to the median patient, adjusted hospital-level mortality varied from 33.5% to 40.6%. The intraclass correlation coefficient for the hospital-level variation was 0.6% (95% confidence interval 0.1, 3.4%), with a median odds ratio of 1.15 (95% confidence interval 1.06, 1.38). The relationship between hospital volume of mechanically ventilated and 30-day mortality was not statistically significant: each 50-patient increase in volume was associated with a nonsignificant 2% decrease in the odds of death within 30 days (odds ratio 0.98, 95% confidence interval 0.87–1.10). Conclusions:Veterans Health Administration hospitals caring for lower volumes of mechanically ventilated patients do not have worse mortality. Mechanisms underlying this finding are unclear, but, if elucidated, may offer other integrated health systems ways to overcome the disadvantages of small-volume centers in achieving good outcomes.

Temporal Changes in the Influence of Hospitals and Regional Healthcare Networks on Severe Sepsis Mortality.

Objectives:There is systematic variation between hospitals in their care of severe sepsis, but little information on whether this variation impacts sepsis-related mortality, or how hospitals’ and health-systems’ impacts have changed over time. We examined whether hospital and regional organization of severe sepsis care is associated with meaningful differences in 30-day mortality in a large integrated health care system, and the extent to which those effects are stable over time. Design:In this retrospective cohort study, we used risk- and reliability-adjusted hierarchical logistic regression to estimate hospital- and region-level random effects after controlling for severity of illness using a rich mix of administrative and clinical laboratory data. Setting:One hundred fourteen U.S. Department of Veterans Affairs hospitals in 21 geographic regions. Patients:Forty-three thousand seven hundred thirty-three patients with severe sepsis in 2012, compared to 33,095 such patients in 2008. Interventions:None. Measurements and Main Results:The median hospital in the worst quintile of performers had a risk-adjusted 30-day mortality of 16.7% (95% CI, 13.5%, 20.5%) in 2012 compared with the best quintile, which had a risk-adjusted mortality of 12.8% (95% CI, 10.7%, 15.3%). Hospitals and regions explained a statistically and clinically significant proportion of the variation in patient outcomes. Thirty-day mortality after severe sepsis declined from 18.3% in 2008 to 14.7% in 2012 despite very similar severity of illness between years. The proportion of the variance in sepsis-related mortality explained by hospitals and regions was stable between 2008 and 2012. Conclusions:In this large integrated healthcare system, there is clinically significant variation in sepsis-related mortality associated with hospitals and regions. The proportion of variance explained by hospitals and regions has been stable over time, although sepsis-related mortality has declined.

Corticosteroid use and complications in a US inflammatory bowel disease cohort

Background and Aims Corticosteroids are effective for the short-term treatment of inflammatory bowel disease (IBD). Long-term use, however, is associated with significant adverse effects. To define the: (1) frequency and duration of corticosteroid use, (2) frequency of escalation to corticosteroid-sparing therapy, (3) rate of complications related to corticosteroid use, (4) rate of appropriate bone density measurements (dual energy X-ray absorptiometry [DEXA] scans), and (5) factors associated with escalation and DEXA scans. Methods Retrospective review of Veterans Health Administration (VHA) data from 2002–2010. Results Of the 30,456 Veterans with IBD, 32% required at least one course of corticosteroids during the study time period, and 17% of the steroid users had a prolonged course. Among these patients, only 26.2% underwent escalation of therapy. Patients visiting a gastroenterology (GI) physician were significantly more likely to receive corticosteroid-sparing medications. Factors associated with corticosteroid-sparing medications included younger age (OR = 0.96 per year,95%CI:0.95, 0.97), male gender (OR = 2.00,95%CI:1.16,3.46), GI visit during the corticosteroid evaluation period (OR = 8.01,95%CI:5.85,10.95) and the use of continuous corticosteroids vs. intermittent corticosteroids (OR = 2.28,95%CI:1.33,3.90). Rates of complications per 1000 person-years after IBD diagnosis were higher among corticosteroid users (venous thromboembolism [VTE] 9.0%; fragility fracture 2.6%; Infections 54.3) than non-corticosteroid users (VTE 4.9%; fragility fracture 1.9%; Infections 26.9). DEXA scan utilization rates among corticosteroid users were only 7.8%. Conclusions Prolonged corticosteroid therapy for the treatment of IBD is common and is associated with significant harm to patients. Patients with prolonged use of corticosteroids for IBD should be referred to gastroenterology early and universal efforts to improve the delivery of high quality care should be undertaken.

Age Disparities in the Use of Steroid-sparing Therapy for Inflammatory Bowel Disease.

Background:Corticosteroids are effective rescue therapies for patients with inflammatory bowel disease (IBD), but have significant side effects, which may be amplified in the growing population of elderly patients with IBD. We aimed to compare the use of steroids and steroid-sparing therapies (immunomodulators and biologics) and rates of complications among elderly (≥65) and younger patients in a national cohort of veterans with IBD. Methods:We used national Veterans Health Administrative data to conduct a retrospective study of veterans with IBD between 2002 and 2010. Medications and the incidence of complications were obtained from the Veterans Health Administrative Decision Support Systems. Multivariate logistic regression accounting for facility-level clustering was used to identify predictors of use of steroid-sparing medications. Results:We identified 30,456 veterans with IBD. Of these, 94% were men and 40% were more than 65, and 32% were given steroids. Elderly veterans were less likely to receive steroids (23.8% versus 38.3%, P < 0.001) and were less likely to be prescribed steroid-sparing medications (25.5% versus 46.9%, respectively, P < 0.001). In multivariate analysis controlling for sex, age <65 (odds ratio, 2.19; 95% CI, 1.54–3.11) and gastroenterology care (odds ratio, 8.42; 95% CI, 6.18–11.47) were associated with initiation of steroid-sparing medications. After starting steroids, fracture rates increased in the elderly patients with IBD, whereas increases in venous thromboembolism and infections after starting steroids affected both age groups. Conclusions:Elderly veterans are less likely to receive steroids and steroid-sparing medications than younger veterans; elderly patients exposed to steroids were more likely to have fractures than the younger population.

Predicting Hospitalization and Outpatient Corticosteroid Use in Inflammatory Bowel Disease Patients Using Machine Learning

Background Inflammatory bowel disease (IBD) is a chronic disease characterized by unpredictable episodes of flares and periods of remission. Tools that accurately predict disease course would substantially aid therapeutic decision-making. This study aims to construct a model that accurately predicts the combined end point of outpatient corticosteroid use and hospitalizations as a surrogate for IBD flare. Methods Predictors evaluated included age, sex, race, use of corticosteroid-sparing immunosuppressive medications (immunomodulators and/or anti-TNF), longitudinal laboratory data, and number of previous IBD-related hospitalizations and outpatient corticosteroid prescriptions. We constructed models using logistic regression and machine learning methods (random forest [RF]) to predict the combined end point of hospitalization and/or corticosteroid use for IBD within 6 months. Results We identified 20,368 Veterans Health Administration patients with the first (index) IBD diagnosis between 2002 and 2009. Area under the receiver operating characteristic curve (AuROC) for the baseline logistic regression model was 0.68 (95% confidence interval [CI], 0.67-0.68). AuROC for the RF longitudinal model was 0.85 (95% CI, 0.84-0.85). AuROC for the RF longitudinal model using previous hospitalization or steroid use was 0.87 (95% CI, 0.87-0.88). The 5 leading independent risk factors for future hospitalization or steroid use were age, mean serum albumin, immunosuppressive medication use, and mean and highest platelet counts. Previous hospitalization and corticosteroid use were highly predictive when included in specified models. Conclusions A novel machine learning model substantially improved our ability to predict IBD-related hospitalization and outpatient steroid use. This model could be used at point of care to distinguish patients at high and low risk for disease flare, allowing individualized therapeutic management.

Corrected ROC analysis for misclassified binary outcomes

Creating accurate risk prediction models from Big Data resources such as Electronic Health Records (EHRs) is a critical step toward achieving precision medicine. A major challenge in developing these tools is accounting for imperfect aspects of EHR data, particularly the potential for misclassified outcomes. Misclassification, the swapping of case and control outcome labels, is well known to bias effect size estimates for regression prediction models. In this paper, we study the effect of misclassification on accuracy assessment for risk prediction models and find that it leads to bias in the area under the curve (AUC) metric from standard ROC analysis. The extent of the bias is determined by the false positive and false negative misclassification rates as well as disease prevalence. Notably, we show that simply correcting for misclassification while building the prediction model is not sufficient to remove the bias in AUC. We therefore introduce an intuitive misclassification-adjusted ROC procedure that accounts for uncertainty in observed outcomes and produces bias-corrected estimates of the true AUC. The method requires that misclassification rates are either known or can be estimated, quantities typically required for the modeling step. The computational simplicity of our method is a key advantage, making it ideal for efficiently comparing multiple prediction models on very large datasets. Finally, we apply the correction method to a hospitalization prediction model from a cohort of over 1 million patients from the Veterans Health Administrations EHR. Implementations of the ROC correction are provided for Stata and R. Published 2017. This article is a U.S. Government work and is in the public domain in the USA.

Correlates and consequences of central sleep apnea in a national sample of US veterans

The prevalence and consequences of central sleep apnea (CSA) in adults are not well described. By utilizing the large Veterans Health Administration (VHA) national administrative databases, we sought to determine the incidence, clinical correlates, and impact of CSA on healthcare utilization in Veterans. Analysis of a retrospective cohort of patients with sleep disorders was performed from outpatient visits and inpatient admissions from fiscal years 2006 through 2012. The CSA group, defined by International Classification of Diseases-9, was compared with a comparison group. The number of newly diagnosed CSA cases increased fivefold during this timeframe; however, the prevalence was highly variable depending on the VHA site. The important predictors of CSA were male gender (odds ratio [OR] = 2.31, 95% confidence interval [CI]: 1.94-2.76, p < 0.0001), heart failure (HF) (OR = 1.78, 95% CI: 1.64-1.92, p < 0.0001), atrial fibrillation (OR = 1.83, 95% CI: 1.69-2.00, p < 0.0001), pulmonary hypertension (OR = 1.38, 95% CI:1.19-1.59, p < 0.0001), stroke (OR = 1.65, 95% CI: 1.50-1.82, p < 0.0001), and chronic prescription opioid use (OR = 1.99, 95% CI: 1.87-2.13, p < 0.0001). Veterans with CSA were at an increased risk for hospital admissions related to cardiovascular disorders compared with the comparison group (incidence rate ratio [IRR] = 1.50, 95% CI: 1.16-1.95, p = 0.002). Additionally, the effect of prior HF on future admissions was greater in the CSA group (IRR: 4.78, 95% CI: 3.87-5.91, p < 0.0001) compared with the comparison group (IRR = 3.32, 95% CI: 3.18-3.47, p < 0.0001). Thus, CSA in veterans is associated with cardiovascular disorders, chronic prescription opioid use, and increased admissions related to the comorbid cardiovascular disorders. Furthermore, there is a need for standardization of diagnostics methods across the VHA to accurately diagnose CSA in high-risk populations.