Centralized Interactive Phenomics Resource: an integrated online phenomics knowledgebase for health data users

OBJECTIVE

Development of clinical phenotypes from electronic health records (EHRs) can be resource intensive. Several phenotype libraries have been created to facilitate reuse of definitions. However, these platforms vary in target audience and utility. We describe the development of the Centralized Interactive Phenomics Resource (CIPHER) knowledgebase, a comprehensive public-facing phenotype library, which aims to facilitate clinical and health services research.

MATERIALS AND METHODS

The platform was designed to collect and catalog EHR-based computable phenotype algorithms from any healthcare system, scale metadata management, facilitate phenotype discovery, and allow for integration of tools and user workflows. Phenomics experts were engaged in the development and testing of the site.

RESULTS

The knowledgebase stores phenotype metadata using the CIPHER standard, and definitions are accessible through complex searching. Phenotypes are contributed to the knowledgebase via webform, allowing metadata validation. Data visualization tools linking to the knowledgebase enhance user interaction with content and accelerate phenotype development.

DISCUSSION

The CIPHER knowledgebase was developed in the largest healthcare system in the United States and piloted with external partners. The design of the CIPHER website supports a variety of front-end tools and features to facilitate phenotype development and reuse. Health data users are encouraged to contribute their algorithms to the knowledgebase for wider dissemination to the research community, and to use the platform as a springboard for phenotyping.

CONCLUSION

CIPHER is a public resource for all health data users available at https://phenomics.va.ornl.gov/ which facilitates phenotype reuse, development, and dissemination of phenotyping knowledge.

New statin use, mortality, and first cardiovascular events in older US Veterans by frailty status

BACKGROUND

Statins are part of long-term medical regimens for many older adults. Whether frailty modifies the protective relationship between statins, mortality, and major adverse cardiovascular events (MACE) is unknown.

METHODS

This was a retrospective study of US Veterans ≥65, without CVD or prior statin use seen in 2002-2012, followed through 2017. A 31-item frailty index was used. The co-primary endpoint was all-cause mortality or MACE (MI, stroke/TIA, revascularization, or cardiovascular death). Cox proportional hazards models were developed to evaluate the association of statin use with outcomes; propensity score overlap weighting accounted for confounding by indication.

RESULTS

We identified 710,313 Veterans (mean age (SD) 75.3(6.5), 98% male, 89% white); 86,327 (12.1%) were frail. Over mean follow-up of 8 (5) years, there were 48.6 and 72.6 deaths per 1000 person-years (PY) among non-frail statin-users vs nonusers (weighted Incidence Rate Difference (wIRD)/1000 person years (PY), -24.0[95% CI, -24.5 to -23.6]), and 90.4 and 130.4 deaths per 1000PY among frail statin-users vs nonusers (wIRD/1000PY, -40.0[95% CI, -41.8 to -38.2]). There were 51.7 and 60.8 MACE per 1000PY among non-frail statin-users vs nonusers (wIRD/1000PY, -9.1[95% CI, -9.7 to -8.5]), and 88.2 and 102.0 MACE per 1000PY among frail statin-users vs nonusers (wIRD/1000PY, -13.8[95% CI, -16.2 to -11.4]). There were no significant interactions by frailty for statin users vs non-users by either mortality or MACE outcomes, p-interaction 0.770 and 0.319, respectively. Statin use was associated with lower risk of all-cause mortality (HR, 0.61 (0.60-0.61)) and MACE (HR 0.86 (0.85-0.87)).

CONCLUSIONS

New statin use is associated with a lower risk of mortality and MACE, independent of frailty. These findings should be confirmed in a randomized clinical trial.

Incidence and prognostic significance of newly-diagnosed atrial fibrillation among older U.S. veterans hospitalized with COVID-19

Most prior studies on the prognostic significance of newly-diagnosed atrial fibrillation (AF) in COVID-19 did not differentiate newly-diagnosed AF from pre-existing AF. To determine the association between newly-diagnosed AF and in-hospital and 30-day mortality among regular users of Veterans Health Administration using data linked to Medicare. We identified Veterans aged ≥ 65 years who were hospitalized for ≥ 24 h with COVID-19 from 06/01/2020 to 1/31/2022 and had ≥ 2 primary care visits within 24 months prior to the index hospitalization. We performed multivariable logistic regression analyses to estimate adjusted risks, risk differences (RD), and odds ratios (OR) for the association between newly-diagnosed AF and the mortality outcomes adjusting for patient demographics, baseline comorbidities, and presence of acute organ dysfunction on admission. Of 23,299 patients in the study cohort, 5.3% had newly-diagnosed AF, and 29.2% had pre-existing AF. In newly-diagnosed AF adjusted in-hospital and 30-day mortality were 16.5% and 22.7%, respectively. Newly-diagnosed AF was associated with increased mortality compared to pre-existing AF (in-hospital: OR 2.02, 95% confidence interval [CI] 1.72-2.37; RD 7.58%, 95% CI 5.54-9.62) (30-day: OR 1.86; 95% CI 1.60-2.16; RD 9.04%, 95% CI 6.61-11.5) or no AF (in-hospital: OR 2.24, 95% CI 1.93-2.60; RD 8.40%, 95% CI 6.44-10.4) (30-day: 2.07, 95% CI 1.80-2.37; RD 10.2%, 95% CI 7.89-12.6). There was a smaller association between pre-existing AF and the mortality outcomes. Newly-diagnosed AF is an important prognostic marker for patients hospitalized with COVID-19. Whether prevention or treatment of AF improves clinical outcomes in these patients remains unknown.

Statins, Mortality, and Major Adverse Cardiovascular Events Among US Veterans With Chronic Kidney Disease

Importance

There are limited data for the utility of statins for primary prevention of atherosclerotic cardiovascular disease (ASCVD) and death in adults with chronic kidney disease (CKD).

Objective

To evaluate the association of statin use with all-cause mortality and major adverse cardiovascular events (MACE) among US veterans older than 65 years with CKD stages 3 to 4.

Design, Setting, and Participants

This cohort study used a target trial emulation design for statin initiation among veterans with moderate CKD (stages 3 or 4) using nested trials with a propensity weighting approach. Linked Veterans Affairs (VA) Healthcare System, Medicare, and Medicaid data were used. This study considered veterans newly diagnosed with moderate CKD between 2005 and 2015 in the VA, with follow-up through December 31, 2017. Veterans were older than 65 years, within 5 years of CKD diagnosis, had no prior ASCVD or statin use, and had at least 1 clinical visit in the year prior to trial baseline. Eligibility criteria were assessed for each nested trial, and Cox proportional hazards models with bootstrapping were run. Analysis was conducted from July 2021 to October 2023.

Exposure

Statin initiation vs none.

Main Outcomes and Measures

Primary outcome was all-cause mortality; secondary outcome was time to first MACE (myocardial infarction, transient ischemic attack, stroke, revascularization, or mortality).

Results

Included in the analysis were 14 828 veterans. Mean (SD) age at CKD diagnosis was 76.9 (8.2) years, 14 616 (99%) were men, 10 539 (72%) White, and 2568 (17%) Black. After expanding to person-trials and assessing eligibility at each baseline, there were 151 243 person-trials (14 685 individuals) of nonstatin initiators and 2924 person-trials (2924 individuals) of statin initiators included. Propensity score adjustment via overlap weighting with nonparametric bootstrapping resulted in covariate balance, with mean (SD) follow-up of 3.6 (2.7) years. The hazard ratio for all-cause mortality was 0.91 (95% CI, 0.85-0.97) comparing statin initiators to noninitiators. The hazard ratio for MACE was 0.96 (95% CI, 0.91-1.02). Results remained consistent in prespecified subgroup analyses.

Conclusions and Relevance

In this target trial emulation of statin initiation in US veterans older than 65 years with CKD stages 3 to 4 and no prior ASCVD, statin initiation was significantly associated with a lower risk of all-cause mortality but not MACE. Results should be confirmed in a randomized clinical trial.

Serum Cholesterol and Impact of Age on Coronary Heart Disease Death in More Than 4 Million Veterans

Background The lipid hypothesis postulates that lower blood cholesterol is associated with reduced coronary heart disease (CHD) risk, which has been challenged by reports of a U-shaped relation between cholesterol and death in recent studies. We sought to examine whether the U-shaped relationship is true and to assess the impact of age on this association. Method and Results We conducted a prospective cohort study of 4 467 942 veterans aged >18 years, with baseline outpatient visits from 2002 to 2007 and follow-up to December 30, 2018, in the Veterans Health Administration electronic health record system. We observed a J-shaped relation between total cholesterol (TC) and CHD mortality after a comprehensive adjustment of confounding factors: flat for TC <180 mg/dL, and greater risk was present at higher cholesterol levels. Compared with veterans with TC between 180 and 199 mg/dL, the multiadjusted hazard ratios (HRs) for CHD death were 1.03 (95% CI, 1.02-1.04), 1.07 (95% CI, 1.06-1.09), 1.15 (95% CI, 1.13-1.18), 1.25 (95% CI, 1.22-1.28), and 1.45 (95% CI, 1.42-1.49) times greater among veterans with TC (mg/dL) of 200 to 219, 220 to 239, 140 to 259, 260 to 279 and ≥280, respectively. Similar J-shaped TC-CHD mortality patterns were observed among veterans with and without statin use at or before baseline. Conclusions The cholesterol paradox, for example, higher CHD death in patients with a low cholesterol level, was a reflection of reverse causality, especially among older participants. Our results support the lipid hypothesis that lower blood cholesterol is associated with reduced CHD. Furthermore, the hypothesis remained true when TC was low due to use of statins or other lipid-lowering medication.

Predictive value of ASCVD risk score for mortality and major adverse cardiovascular events in the year following a COVID-19 infection among US Veterans

BACKGROUND

Morbidity and mortality following COVID-19 infection may be influenced by baseline atherosclerotic cardiovascular disease (ASCVD) risk, yet limited data are available to identify those at highest risk. We examined the association between baseline ASCVD risk with mortality and major adverse cardiovascular events (MACE) in the year following COVID-19 infection.

METHODS

We evaluated a nationwide retrospective cohort of US Veterans free of ASCVD who were tested for COVID-19. The primary outcome was absolute risk of all-cause mortality in the year following a COVID-19 test among those hospitalized vs. not stratified by baseline VA-ASCVD risk scores. Secondarily, risk of MACE was examined.

RESULTS

There were 393,683 Veterans tested for COVID-19 and 72,840 tested positive. Mean age was 57 years, 86% were male, and 68% were white. Within 30 days following infection, hospitalized Veterans with VA-ASCVD scores >20% had an absolute risk of death of 24.6% vs. 9.7% (P ≤0.0001) for those who tested positive and negative for COVID-19 respectively. In the year following infection, risk of mortality attenuated with no difference in risk after 60 days. The absolute risk of MACE was similar for Veterans who tested positive or negative for COVID-19.

CONCLUSIONS

Veterans without clinical ASCVD experienced an increased absolute risk of death within 30 days of a COVID-19 infection compared to Veterans with the same VA-ASCVD risk score who tested negative, but this risk attenuated after 60 days. Whether cardiovascular preventive medications can lower the risk of mortality and MACE in the acute period following COVID-19 infection should be evaluated.

Cardiovascular Disease Risk Assessment Using Traditional Risk Factors and Polygenic Risk Scores in the Million Veteran Program

Importance

Primary prevention of atherosclerotic cardiovascular disease (ASCVD) relies on risk stratification. Genome-wide polygenic risk scores (PRSs) are proposed to improve ASCVD risk estimation.

Objective

To determine whether genome-wide PRSs for coronary artery disease (CAD) and acute ischemic stroke improve ASCVD risk estimation with traditional clinical risk factors in an ancestrally diverse midlife population.

Design, Setting, and Participants

This was a prognostic analysis of incident events in a retrospectively defined longitudinal cohort conducted from January 1, 2011, to December 31, 2018. Included in the study were adults free of ASCVD and statin naive at baseline from the Million Veteran Program (MVP), a mega biobank with genetic, survey, and electronic health record data from a large US health care system. Data were analyzed from March 15, 2021, to January 5, 2023.

Exposures

PRSs for CAD and ischemic stroke derived from cohorts of largely European descent and risk factors, including age, sex, systolic blood pressure, total cholesterol, high-density lipoprotein (HDL) cholesterol, smoking, and diabetes status.

Main Outcomes and Measures

Incident nonfatal myocardial infarction (MI), ischemic stroke, ASCVD death, and composite ASCVD events.

Results

A total of 79 151 participants (mean [SD] age, 57.8 [13.7] years; 68 503 male [86.5%]) were included in the study. The cohort included participants from the following harmonized genetic ancestry and race and ethnicity categories: 18 505 non-Hispanic Black (23.4%), 6785 Hispanic (8.6%), and 53 861 non-Hispanic White (68.0%) with a median (5th-95th percentile) follow-up of 4.3 (0.7-6.9) years. From 2011 to 2018, 3186 MIs (4.0%), 1933 ischemic strokes (2.4%), 867 ASCVD deaths (1.1%), and 5485 composite ASCVD events (6.9%) were observed. CAD PRS was associated with incident MI in non-Hispanic Black (hazard ratio [HR], 1.10; 95% CI, 1.02-1.19), Hispanic (HR, 1.26; 95% CI, 1.09-1.46), and non-Hispanic White (HR, 1.23; 95% CI, 1.18-1.29) participants. Stroke PRS was associated with incident stroke in non-Hispanic White participants (HR, 1.15; 95% CI, 1.08-1.21). A combined CAD plus stroke PRS was associated with ASCVD deaths among non-Hispanic Black (HR, 1.19; 95% CI, 1.03-1.17) and non-Hispanic (HR, 1.11; 95% CI, 1.03-1.21) participants. The combined PRS was also associated with composite ASCVD across all ancestry groups but greater among non-Hispanic White (HR, 1.20; 95% CI, 1.16-1.24) than non-Hispanic Black (HR, 1.11; 95% CI, 1.05-1.17) and Hispanic (HR, 1.12; 95% CI, 1.00-1.25) participants. Net reclassification improvement from adding PRS to a traditional risk model was modest for the intermediate risk group for composite CVD among men (5-year risk >3.75%, 0.38%; 95% CI, 0.07%-0.68%), among women, (6.79%; 95% CI, 3.01%-10.58%), for age older than 55 years (0.25%; 95% CI, 0.03%-0.47%), and for ages 40 to 55 years (1.61%; 95% CI, -0.07% to 3.30%).

Conclusions and Relevance

Study results suggest that PRSs derived predominantly in European samples were statistically significantly associated with ASCVD in the multiancestry midlife and older-age MVP cohort. Overall, modest improvement in discrimination metrics were observed with addition of PRSs to traditional risk factors with greater magnitude in women and younger age groups.

Mild Neurocognitive Disorder, Social Engagement, and Falls Among Older Primary Care Patients

OBJECTIVES

(1) To estimate the association between social engagement (SE) and falls; (2) To examine the relation between mild neurocognitive disorder (MNCD) and falls by different levels of SE.

DESIGN

We performed a secondary data analysis using prospective cohort study design.

SETTING

Primary care.

PARTICIPANTS

A total of 425 older adult primary care patients at risk for mobility decline (N=425). As previously reported, at baseline, 42% of participants exhibit MNCD.

MAIN OUTCOME MEASURES

The outcome variable was the number of falls during 2 years of follow-up. Exposure variables at baseline included (1) MNCD identified using a cut-off of 1.5 SD below the age-adjusted mean on at least 2 measures within a cognitive performance battery and (2) SE, which was assessed using the social component of the Late-Life Function and Disability Instrument. High SE was defined as having a score ≥ median value (≥49 out of 100). All models were adjusted for age, sex, education, marital status, comorbidities, and pain status.

RESULTS

Over 2 years of follow-up, 48% of participants fell at least once. MNCD was associated with a higher rate of falls, adjusting for the covariates (Incidence Rate Ratio=1.6, 95% confidence interval: 1.1-2.3). There was no significant association between MNCD and the rate of falls among people with high SE. In participants with low SE (having a score less than 49.5 out 100), MNCD was associated with a higher rate of falls as compared with participants with no neurocognitive disorder (No-NCD).

CONCLUSIONS

Among participants with low SE, MNCD was associated with a higher rate of falls, but not among participants with high SE. The findings suggest that high SE may be protective against falls among older primary care patients with MNCD.

Framework of the Centralized Interactive Phenomics Resource (CIPHER) standard for electronic health data-based phenomics knowledgebase

The development of phenotypes using electronic health records is a resource-intensive process. Therefore, the cataloging of phenotype algorithm metadata for reuse is critical to accelerate clinical research. The Department of Veterans Affairs (VA) has developed a standard for phenotype metadata collection which is currently used in the VA phenomics knowledgebase library, CIPHER (Centralized Interactive Phenomics Resource), to capture over 5000 phenotypes. The CIPHER standard improves upon existing phenotype library metadata collection by capturing the context of algorithm development, phenotyping method used, and approach to validation. While the standard was iteratively developed with VA phenomics experts, it is applicable to the capture of phenotypes across healthcare systems. We describe the framework of the CIPHER standard for phenotype metadata collection, the rationale for its development, and its current application to the largest healthcare system in the United States.

Abstract P271: Spatiotemporal Trends of Cardiovascular Incidence in U.S. Veterans From 2003-2018

Introduction: Cardiovascular disease (CVD) is the leading cause of mortality and disability in the United States. Spatiotemporal modeling of disease incidence may guide prevention efforts toward regions most at risk in the future.

Methods: We examined spatiotemporal trends in composite CVD events, defined as myocardial infarction, ischemic stroke, atrial fibrillation, heart failure, coronary artery disease, or cardiovascular death, using electronic health records from 9,718,107 U.S. Veterans who used VA healthcare facilities from 2003-2018. Age-standardized annual incidence by state was modeled using the Bernardinelli model, a Bayesian Poisson regression with linear temporal trends for the US and individual states.

Results: There were 626,271 CVD events over 16 years of follow up and annual incidence fell gradually from 11.5% in 2003 to 6.6% in 2018. Incidence among all US veterans decreased an estimated 3.1% [95% CI: -3.3%, -2.9%] per year (RR year = 0.969). State-level incidences also decreased monotonically over the follow-up period (Figure panel A ), ranging from an absolute decrease 5.0% (RI) to 1.7% (OR). Estimated incidence for the five states most above (RI, FL, NC, MT, LA) and most below (OR, NV, CT, OK, MI) the national trend are plotted against the US average in Figure panel B . These outlying states differed significantly from the overall trend (posterior probabilities ≥ 99.7%), and the magnitude of trend was loosely correlated with initial incidence rate (in 2003). Notably, four states crossed the national average over the study period: CT and OK started below average in 2003 but ended above in 2018, whereas RI and NC started with high incidence and ended with below average incidence in 2018.

Conclusions: Among US veterans, there was significant geographic variation in the rate of decline of incident CVD from 2003-2018. Future research will investigate the mechanisms underpinning the observed trends, particularly prevalence and control of risk factors.

Long-Term Outcomes and Duration of Dual Antiplatelet Therapy After Coronary Intervention With Second-Generation Drug-Eluting Stents: The Veterans Affairs Extended DAPT Study

Background Recent guidelines on dual antiplatelet therapy (DAPT) duration after percutaneous coronary intervention (PCI) balance the subsequent risks of major bleeding with ischemic events. Although generally favoring shorter DAPT duration with second-generation drug-eluting stents, the effects on long-term outcomes in the wider population are uncertain. Methods and Results We tracked all patients having PCI with second-generation drug-eluting stents in the Veterans Affairs Healthcare System between 2006 and 2016 for death, myocardial infarction, stroke, and major bleeding up to 13 years. We compared these outcomes with 4 DAPT durations of 1 to 5, 6 to 9, 10 to 12, and 13 to 18 months after the index PCI using hazard ratios (HRs) and 95% CIs from Cox proportional hazards models adjusted by inverse probability weighting. A total of 40 882 subjects with PCI were followed up for a median of 4.3 (25%-75%: 2.4-6.5) years. DAPT discontinuation was rare early after PCI (5.8% at 1-5 months and 6.3% at 6-9 months) but increased (19% and 44%) >9 months. The risk of cardiovascular and noncardiovascular death was higher (HR, 2.03-3.41) with DAPT discontinuation <9 months, likely reflecting premature cessation from factors related to early death. DAPT discontinuation after 9 months following PCI was associated with lower risks of death (HR, 0.93 [95% CI, 0.88-0.99]), cardiac death (HR, 0.79 [95% CI, 0.70-0.90]), myocardial infarction (HR, 0.75 [95% CI, 0.69-0.82]), and major bleeding (HR, 0.82 [95% CI, 0.74-0.91]). Results were similar with an index PCI for an acute coronary syndrome. Conclusions Stopping DAPT after 9 months is associated with lower long-term risks of adverse ischemic and bleeding events and supports recent guidelines of shorter duration DAPT after PCI with second-generation drug-eluting stents.

Comparative effectiveness of third doses of mRNA-based COVID-19 vaccines in US veterans

Vaccination against SARS-CoV-2 has been effective in reducing the burden of severe disease and death from COVID-19. Third doses of mRNA-based vaccines have provided a way to address waning immunity and broaden protection against emerging SARS-CoV-2 variants. However, their comparative effectiveness for a range of COVID-19 outcomes across diverse populations is unknown. We emulated a target trial using electronic health records of US veterans who received a third dose of either BNT162b2 or mRNA-1273 vaccines between 20 October 2021 and 8 February 2022, during a period that included Delta- and Omicron-variant waves. Eligible veterans had previously completed an mRNA vaccine primary series. We matched recipients of each vaccine in a 1:1 ratio according to recorded risk factors. Each vaccine group included 65,196 persons. The excess number of events over 16 weeks per 10,000 persons for BNT162b2 compared with mRNA-1273 was 45.4 (95% CI: 19.4, 84.7) for documented infection, 3.7 (2.2, 14.1) for symptomatic COVID-19, 10.6 (5.1, 19.7) for COVID-19 hospitalization, 2.0 (-3.1, 6.3) for COVID-19 intensive care unit admission and 0.2 (-2.2, 4.0) for COVID-19 death. After emulating a second target trial of veterans who received a third dose between 1 January and 1 March 2022, during a period restricted to Omicron-variant predominance, the excess number of events over 9 weeks per 10,000 persons for BNT162b2 compared with mRNA-1273 was 63.2 (95% CI: 15.2, 100.7) for documented infection. The 16-week risks of COVID-19 outcomes were low after a third dose of mRNA-1273 or BNT162b2, although risks were lower with mRNA-1273 than with BNT162b2, particularly for documented infection.

Mild-to-Moderate Kidney Dysfunction and Cardiovascular Disease: Observational and Mendelian Randomization Analyses

BACKGROUND

End-stage renal disease is associated with a high risk of cardiovascular events. It is unknown, however, whether mild-to-moderate kidney dysfunction is causally related to coronary heart disease (CHD) and stroke.

METHODS

Observational analyses were conducted using individual-level data from 4 population data sources (Emerging Risk Factors Collaboration, EPIC-CVD [European Prospective Investigation into Cancer and Nutrition-Cardiovascular Disease Study], Million Veteran Program, and UK Biobank), comprising 648 135 participants with no history of cardiovascular disease or diabetes at baseline, yielding 42 858 and 15 693 incident CHD and stroke events, respectively, during 6.8 million person-years of follow-up. Using a genetic risk score of 218 variants for estimated glomerular filtration rate (eGFR), we conducted Mendelian randomization analyses involving 413 718 participants (25 917 CHD and 8622 strokes) in EPIC-CVD, Million Veteran Program, and UK Biobank.

RESULTS

There were U-shaped observational associations of creatinine-based eGFR with CHD and stroke, with higher risk in participants with eGFR values <60 or >105 mL·min-1·1.73 m-2, compared with those with eGFR between 60 and 105 mL·min-1·1.73 m-2. Mendelian randomization analyses for CHD showed an association among participants with eGFR <60 mL·min-1·1.73 m-2, with a 14% (95% CI, 3%-27%) higher CHD risk per 5 mL·min-1·1.73 m-2 lower genetically predicted eGFR, but not for those with eGFR >105 mL·min-1·1.73 m-2. Results were not materially different after adjustment for factors associated with the eGFR genetic risk score, such as lipoprotein(a), triglycerides, hemoglobin A1c, and blood pressure. Mendelian randomization results for stroke were nonsignificant but broadly similar to those for CHD.

CONCLUSIONS

In people without manifest cardiovascular disease or diabetes, mild-to-moderate kidney dysfunction is causally related to risk of CHD, highlighting the potential value of preventive approaches that preserve and modulate kidney function.

Visualizing novel connections and genetic similarities across diseases using a network-medicine based approach

Understanding the genetic relationships between human disorders could lead to better treatment and prevention strategies, especially for individuals with multiple comorbidities. A common resource for studying genetic-disease relationships is the GWAS Catalog, a large and well curated repository of SNP-trait associations from various studies and populations. Some of these populations are contained within mega-biobanks such as the Million Veteran Program (MVP), which has enabled the genetic classification of several diseases in a large well-characterized and heterogeneous population. Here we aim to provide a network of the genetic relationships among diseases and to demonstrate the utility of quantifying the extent to which a given resource such as MVP has contributed to the discovery of such relations. We use a network-based approach to evaluate shared variants among thousands of traits in the GWAS Catalog repository. Our results indicate many more novel disease relationships that did not exist in early studies and demonstrate that the network can reveal clusters of diseases mechanistically related. Finally, we show novel disease connections that emerge when MVP data is included, highlighting methodology that can be used to indicate the contributions of a given biobank.

Scalable relevance ranking algorithm via semantic similarity assessment improves efficiency of medical chart review

OBJECTIVE

Accurately assigning phenotype information to individual patients via computational phenotyping using Electronic Health Records (EHRs) has been seen as the first step towards enabling EHRs for precision medicine research. Chart review labels annotated by clinical experts, also known as "gold standard" labels, are essential for the development and validation of computational phenotyping algorithms. However, given the complexity of EHR systems, the process of chart review is both labor intensive and time consuming. We propose a fully automated algorithm, referred to as pGUESS, to rank EHR notes according to their relevance to a given phenotype. By identifying the most relevant notes, pGUESS can greatly improve the efficiency and accuracy of chart reviews.

METHOD

pGUESS uses prior guided semantic similarity to measure the informativeness of a clinical note to a given phenotype. We first select candidate clinical concepts from a pool of comprehensive medical concepts using public knowledge sources and then derive the semantic embedding vector (SEV) for a reference article (SEV_ref) and each note (SEV_note). The algorithm scores the relevance of a note as the cosine similarity between SEV_note and SEV_ref.

RESULTS

The algorithm was validated against four sets of 200 notes that were manually annotated by clinical experts to assess their informativeness to one of three disease phenotypes. pGUESS algorithm substantially outperforms existing unsupervised approaches for classifying the relevance status with respect to both accuracy and scalability across phenotypes. Averaging over the three phenotypes, the rank correlation between the algorithm ranking and gold standard label was 0.64 for pGUESS, but only 0.47 and 0.35 for the next two best performing algorithms. pGUESS is also much more computationally scalable compared to existing algorithms.

CONCLUSION

pGUESS algorithm can substantially reduce the burden of chart review and holds potential in improving the efficiency and accuracy of human annotation.

Comparative Safety of BNT162b2 and mRNA-1273 Vaccines in a Nationwide Cohort of US Veterans

Importance

The risk of adverse events has been found to be low for participants receiving the BNT162b2 (Pfizer-BioNTech) and mRNA-1273 (Moderna Inc) vaccines in randomized trials. However, a head-to-head comparison of their safety for a broader range of potential adverse events over longer follow-up and in larger and more diverse populations is lacking, to our knowledge.

Objective

To compare the head-to-head safety in terms of risk of adverse events of the BNT162b2 and mRNA-1273 vaccines in the national health care databases of the US Department of Veterans Affairs, the largest integrated health care system in the US.

Design, Setting, and Participants

In this cohort study, the electronic health records of US veterans who received a first dose of the BNT162b2 or mRNA-1273 vaccine between January 4 and September 20, 2021, were used. Recipients of each vaccine were matched in a 1:1 ratio according to their risk factors.

Exposures

Vaccination with either the BNT162b2 vaccine, with a second dose scheduled 21 days later, or the mRNA-1273 vaccine, with a second dose scheduled 28 days later.

Main Outcomes and Measures

A large panel of potential adverse events was evaluated; the panel included neurologic events, hematologic events, hemorrhagic stroke, ischemic stroke, myocardial infarction, other thromboembolic events, myocarditis or pericarditis, arrhythmia, kidney injury, appendicitis, autoimmune events, herpes zoster or simplex, arthritis or arthropathy, and pneumonia. Risks over 38 weeks were estimated using the Kaplan-Meier estimator.

Results

Among 433 672 persons included in the matched vaccine groups, the median age was 69 years (IQR, 60-74 years), 93% of individuals were male, and 20% were Black. Estimated 38-week risks of adverse events were generally low after administration of either the BNT162b2 or the mRNA-1273 vaccine. Compared with the mRNA-1273 group, the BNT162b2 group had an excess per 10 000 persons of 10.9 events (95% CI, 1.9-17.4 events) of ischemic stroke, 14.8 events (95% CI, 7.9-21.8 events) of myocardial infarction, 11.3 events (95% CI, 3.4-17.7 events) of other thromboembolic events, and 17.1 events (95% CI, 8.8-30.2 events) of kidney injury. Estimates were largely similar among subgroups defined by age (<40, 40-69, and ≥70 years) and race (Black, White), but there were higher magnitudes of risk differences of ischemic stroke among older persons and White persons, kidney injury among older persons, and other thromboembolic events among Black persons. Small-magnitude differences between the 2 vaccines were seen within 42 days of the first dose, and few differences were seen within 14 days of the first dose.

Conclusions and Relevance

The findings of this cohort study suggest that there were few differences in risk of adverse events within 14 days of the first dose of either the BNT162b2 or the mRNA-1273 vaccine and small-magnitude differences within 42 days of the first dose. The 38-week risks of adverse events were low in both vaccine groups, although risks were lower for recipients of the mRNA-1273 vaccine than for recipients of the BNT162b2 vaccine. Although the primary analysis was designed to detect safety events unrelated to SARS-CoV-2 infection, the possibility that these differences may partially be explained by a lower effectiveness of the BNT162b2 vaccine in preventing the sequelae of SARS-CoV-2 infection compared with the mRNA-1273 vaccine could not be ruled out. These findings may help inform decision-making in future vaccination campaigns.

Impact of Coronavirus Disease 2019 (COVID-19) Severity on Long-term Events in United States Veterans Using the Veterans Affairs Severity Index for COVID-19 (VASIC)

In this retrospective cohort study of 94 595 severe acute respiratory syndrome coronavirus 2-positive cases, we developed and validated an algorithm to assess the association between coronavirus disease 2019 (COVID-19) severity and long-term complications (stroke, myocardial infarction, pulmonary embolism/deep vein thrombosis, heart failure, and mortality). COVID-19 severity was associated with a greater risk of experiencing a long-term complication 31-120 days postinfection. Most incident events occurred 31-60 days postinfection and diminished after day 91, except heart failure for severe patients and death for moderate patients, which peaked on days 91-120. Understanding the differential impact of COVID-19 severity on long-term events provides insight into possible intervention modalities and critical prevention strategies.

30-day Mortality Following COVID-19 and Influenza Hospitalization Among US Veterans Aged 65 and Older

Background

COVID‐19 and influenza are important sources of morbidity and mortality among older adults. Understanding how outcomes differ for older adults hospitalized with either infection is important for improving care. We compared outcomes from infection with COVID‐19 and influenza among hospitalized older adults.

Methods

We conducted a retrospective study of 30‐day mortality among veterans aged 65+ hospitalized with COVID‐19 from March 1, 2020–December 31, 2020 or with influenza A/B from September 1, 2017 to August 31, 2019 in Veterans Affairs Health Care System (VAHCS). COVID‐19 infection was determined by a positive PCR test and influenza by tests used in the VA system. Frailty was defined by the claims‐based Veterans Affairs Frailty Index (VA‐FI). Logistic regressions of mortality on frailty, age, and infection were adjusted for multiple confounders.

Results

A total of 15,474 veterans were admitted with COVID‐19 and 7867 with influenza. Mean (SD) ages were 76.1 (7.8) and 75.8 (8.3) years, 97.7% and 97.4% were male, and 66.9% and 76.4% were white in the COVID‐19 and influenza cohorts respectively. Crude 30‐day mortality (95% CI) was 18.9% (18.3%–19.5%) for COVID‐19 and 4.3% (3.8%–4.7%) for influenza.

Combining cohorts, the odds ratio for 30‐day mortality from COVID‐19 (versus influenza) was 6.61 (5.74–7.65). There was a statistically significant interaction between infection with COVID‐19 and frailty, but there was no significant interaction between COVID‐19 and age.

Separating cohorts, greater 30‐day mortality was significantly associated with older age (p: COVID‐19: <0.001, Influenza: <0.001) and for frail compared with robust individuals (p for trend: COVID‐19: <0.001, Influenza: <0.001).

Conclusion

Mortality from COVID‐19 exceeded that from influenza among hospitalized older adults. However, odds of mortality were higher at every level of frailty among those admitted with influenza compared to COVID‐19. Prevention will remain key to reducing mortality from viral illnesses among older adults.

Prediction of Cardiovascular and All-Cause Mortality After Myocardial Infarction in US Veterans

Risk prediction models for cardiovascular disease (CVD) death developed from patients without vascular disease may not be suitable for myocardial infarction (MI) survivors. Prediction of mortality risk after MI may help to guide secondary prevention. Using national electronic record data from the Veterans Health Administration 2002 to 2012, we developed risk prediction models for CVD death and all-cause death based on 5-year follow-up data of 100,601 survivors of MI using Cox proportional hazards models. Model performance was evaluated using a cross-validation approach. During follow-up, there were 31,622 deaths and 12,901 CVD deaths. In men, older age, current smoking, atrial fibrillation, heart failure, peripheral artery disease, and lower body mass index were associated with greater risk of death from CVD or all-causes, and statin treatment, hypertension medication, estimated glomerular filtration rate level, and high body mass index were significantly associated with reduced risk of fatal outcomes. Similar associations and slightly different predictors were observed in women. The estimated Harrell's C-statistics of the final model versus the cross-validation estimates were 0.77 versus 0.77 in men and 0.81 versus 0.77 in women for CVD death. Similarly, the C-statistics were 0.75 versus 0.75 in men, 0.78 versus 0.75 in women for all-cause mortality. The predicted risk of death was well calibrated compared with the observed risk. In conclusion, we developed and internally validated risk prediction models of 5-year risk for CVD and all-cause death for outpatient survivors of MI. Traditional risk factors, co-morbidities, and lack of blood pressure or lipid treatment were all associated with greater risk of CVD and all-cause mortality.

Frailty and cardiovascular mortality in more than 3 million US Veterans

AIMS

Frailty is associated with an increased risk of all-cause mortality and cardiovascular (CV) events. Limited data exist from the modern era of CV prevention on the relationship between frailty and CV mortality. We hypothesized that frailty is associated with an increased risk of CV mortality.

METHODS AND RESULTS

All US Veterans aged ≥65 years who were regular users of Veteran Affairs care from 2002 to 2017 were included. Frailty was defined using a 31-item previously validated frailty index, ranging from 0 to 1. The primary outcome was CV mortality with secondary analyses examining the relationship between frailty and CV events (myocardial infarction, stroke, revascularization). Survival analysis models were adjusted for age, sex, ethnicity, geographic region, smoking, hyperlipidaemia, statin use, and blood pressure medication use. There were 3 068 439 US Veterans included in the analysis. Mean age was 74.1 ± 5.8 years in 2002, 76.0 ± 8.3 years in 2014, 98% male, and 87.5% White. In 2002, the median (interquartile range) frailty score was 0.16 (0.10-0.23). This increased and stabilized to 0.19 (0.10-0.32) for 2006-14. The presence of frailty was associated with an increased risk of CV mortality at every stage of frailty. Frailty was associated with an increased risk of myocardial infarction and stroke, but not revascularization.

CONCLUSION

In this population, both the presence and severity of frailty are tightly correlated with CV death, independent of underlying CV disease. This study is the largest and most contemporary evaluation of the relationship between frailty and CV mortality to date. Further work is needed to understand how this risk can be diminished.

KEY QUESTION

Can an electronic frailty index identify adults aged 65 and older who are at risk of CV mortality and major CV events?

KEY FINDING

Among 3 068 439 US Veterans aged 65 and older, frailty was associated with an increased risk of CV mortality at every level of frailty. Frailty was also associated with an increased risk of myocardial infarction and stroke, but not revascularization.

TAKE HOME MESSAGE

Both the presence and severity of frailty are associated with CV mortality and major CV events, independent of underlying CV disease.

Social Characteristics, Health, and Mortality Among Male Centenarians Using Veterans Affairs (VA) Health Care

We studied male centenarian Veterans using VA health care to understand the impact of social characteristics on their annual mortality rate, adjusting for prevalent health conditions. This longitudinal study used VA Electronic Health Record data from 1997 to 2012 (n = 1,858). Covariates included age, race, marital status, and periods of military service. The mean age was 100.4 ± 1.4 years, 76% were white, and 49% were married. The average annual mortality rate was 32 per 100 person-years. The annual mortality rate was stable and not affected by race but did vary by marital status. Divorced or separated centenarians had a 21% higher rate of death than married centenarians. A diagnosis of dementia or of congestive heart failure each increased the mortality risk by 37%. Providers should consider prevalent health conditions, as well as marital status, in managing care of centenarian Veterans.

Comparative Effectiveness of BNT162b2 and mRNA-1273 Vaccines in U.S. Veterans

BACKGROUND

The messenger RNA (mRNA)-based vaccines BNT162b2 and mRNA-1273 are more than 90% effective against coronavirus disease 2019 (Covid-19). However, their comparative effectiveness for a range of outcomes across diverse populations is unknown.

METHODS

We emulated a target trial using the electronic health records of U.S. veterans who received a first dose of the BNT162b2 or mRNA-1273 vaccine between January 4 and May 14, 2021, during a period marked by predominance of the SARS-CoV-2 B.1.1.7 (alpha) variant. We matched recipients of each vaccine in a 1:1 ratio according to their risk factors. Outcomes included documented severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) infection, symptomatic Covid-19, hospitalization for Covid-19, admission to an intensive care unit (ICU) for Covid-19, and death from Covid-19. We estimated risks using the Kaplan-Meier estimator. To assess the influence of the B.1.617.2 (delta) variant, we emulated a second target trial that involved veterans vaccinated between July 1 and September 20, 2021.

RESULTS

Each vaccine group included 219,842 persons. Over 24 weeks of follow-up in a period marked by alpha-variant predominance, the estimated risk of documented infection was 5.75 events per 1000 persons (95% confidence interval [CI], 5.39 to 6.23) in the BNT162b2 group and 4.52 events per 1000 persons (95% CI, 4.17 to 4.84) in the mRNA-1273 group. The excess number of events per 1000 persons for BNT162b2 as compared with mRNA-1273 was 1.23 (95% CI, 0.72 to 1.81) for documented infection, 0.44 (95% CI, 0.25 to 0.70) for symptomatic Covid-19, 0.55 (95% CI, 0.36 to 0.83) for hospitalization for Covid-19, 0.10 (95% CI, 0.00 to 0.26) for ICU admission for Covid-19, and 0.02 (95% CI, -0.06 to 0.12) for death from Covid-19. The corresponding excess risk (BNT162b2 vs. mRNA-1273) of documented infection over 12 weeks of follow-up in a period marked by delta-variant predominance was 6.54 events per 1000 persons (95% CI, -2.58 to 11.82).

CONCLUSIONS

The 24-week risk of Covid-19 outcomes was low after vaccination with mRNA-1273 or BNT162b2, although risks were lower with mRNA-1273 than with BNT162b2. This pattern was consistent across periods marked by alpha- and delta-variant predominance. (Funded by the Department of Veterans Affairs and others.).

Pharmacoepidemiology, Machine Learning, and COVID-19: An Intent-to-Treat Analysis of Hydroxychloroquine, With or Without Azithromycin, and COVID-19 Outcomes Among Hospitalized US Veterans

Hydroxychloroquine (HCQ) was proposed as an early therapy for coronavirus disease 2019 (COVID-19) after in vitro studies indicated possible benefit. Previous in vivo observational studies have presented conflicting results, though recent randomized clinical trials have reported no benefit from HCQ amongst hospitalized COVID-19 patients. We examined the effects of HCQ alone, and in combination with azithromycin, in a hospitalized COVID-19 positive, United States (US) Veteran population using a propensity score adjusted survival analysis with imputation of missing data. From March 1, 2020 through April 30, 2020, 64,055 US Veterans were tested for COVID-19 based on Veteran Affairs Healthcare Administration electronic health record data. Of the 7,193 positive cases, 2,809 were hospitalized, and 657 individuals were prescribed HCQ within the first 48-hours of hospitalization for the treatment of COVID-19. There was no apparent benefit associated with HCQ receipt, alone or in combination with azithromycin, and an increased risk of intubation when used in combination with azithromycin [Hazard Ratio (95% Confidence Interval): 1.55 (1.07, 2.24)]. In conclusion, we assessed the effectiveness of HCQ with or without azithromycin in treating patients hospitalized with COVID-19 using a national sample of the US Veteran population. Using rigorous study design and analytic methods to reduce confounding and bias, we found no evidence of a survival benefit from the administration of HCQ.

Trajectories of Frailty in the 5 Years Prior to Death Among U.S. Veterans Born 1927-1934

BACKGROUND

Electronic frailty indices (eFIs) are increasingly used to identify patients at risk for morbidity and mortality. Whether eFIs capture the spectrum of frailty change, including decline, stability, and improvement, is unknown.

METHODS

In a nationwide retrospective birth cohort of U.S. Veterans, a validated eFI, including 31 health deficits, was calculated annually using medical record and insurance claims data (2002-2012). K-means clustering was used to assign patients into frailty trajectories measured 5 years prior to death.

RESULTS

There were 214 250 veterans born between 1927 and 1934 (mean [SD] age at death = 79.4 [2.8] years, 99.2% male, 90.3% White) with an annual eFI in the 5 years before death. Nine frailty trajectories were identified. Those starting at nonfrail or prefrail had 2 stable trajectories (nonfrail to prefrail, n = 29 786 and stable prefrail, n = 28 499) and 2 rapidly increasing trajectories (prefrail to moderately frail, n = 28 244 and prefrail to severely frail, n = 22 596). Those who were mildly frail at baseline included 1 gradually increasing trajectory (mildly to moderately frail, n = 33 806) and 1 rapidly increasing trajectory (mildly to severely frail, n = 15 253). Trajectories that started at moderately or severely frail included 2 gradually increasing trajectories (moderately to severely frail, n = 27 662 and progressing severely frail, n = 14 478) and 1 recovering trajectory (moderately frail to mildly frail, n = 13 926).

CONCLUSIONS

Nine frailty trajectories, including 1 recovering trajectory, were identified in this cohort of older U.S. Veterans. Future work is needed to understand whether prevention and treatment strategies can improve frailty trajectories and contribute to compression of morbidity toward the end of life.

Early Convalescent Plasma Therapy and Mortality Among US Veterans Hospitalized With Nonsevere COVID-19: An Observational Analysis Emulating a Target Trial

BACKGROUND

Early convalescent plasma transfusion may reduce mortality in patients with nonsevere coronavirus disease 2019 (COVID-19).

METHODS

This study emulates a (hypothetical) target trial using observational data from a cohort of US veterans admitted to a Department of Veterans Affairs (VA) facility between 1 May and 17 November 2020 with nonsevere COVID-19. The intervention was convalescent plasma initiated within 2 days of eligibility. Thirty-day mortality was compared using cumulative incidence curves, risk differences, and hazard ratios estimated from pooled logistic models with inverse probability weighting to adjust for confounding.

RESULTS

Of 11 269 eligible person-trials contributed by 4755 patients, 402 trials were assigned to the convalescent plasma group. Forty and 671 deaths occurred within the plasma and nonplasma groups, respectively. The estimated 30-day mortality risk was 6.5% (95% confidence interval [CI], 4.0%-9.7%) in the plasma group and 6.2% (95% CI, 5.6%-7.0%) in the nonplasma group. The associated risk difference was 0.30% (95% CI, -2.30% to 3.60%) and the hazard ratio was 1.04 (95% CI, .64-1.62).

CONCLUSIONS

Our target trial emulation estimated no meaningful differences in 30-day mortality between nonsevere COVID-19 patients treated and untreated with convalescent plasma. Clinical Trials Registration. NCT04545047.

Risk factors and prediction models for incident heart failure with reduced and preserved ejection fraction

Aims

This study aims to develop the first race‐specific and sex‐specific risk prediction models for heart failure with preserved (HFpEF) and reduced ejection fraction (HFrEF).

Methods and results

We created a cohort of 1.8 million individuals who had an outpatient clinic visit between 2002 and 2007 within the Veterans Affairs (VA) Healthcare System and obtained information on HFpEF, HFrEF, and several risk factors from electronic health records (EHR). Variables were selected for the risk prediction models in a ‘derivation cohort’ that consisted of individuals with baseline date in 2002, 2003, or 2004 using a forward stepwise selection based on a change in C‐index threshold. Discrimination and calibration were assessed in the remaining participants (internal ‘validation cohort’). A total of 66 831 individuals developed HFpEF, and 92 233 developed HFrEF (52 679 and 71 463 in the derivation cohort) over a median of 11.1 years of follow‐up. The HFpEF risk prediction model included age, diabetes, BMI, COPD, previous MI, antihypertensive treatment, SBP, smoking status, atrial fibrillation, and estimated glomerular filtration rate (eGFR), while the HFrEF model additionally included previous CAD. For the HFpEF model, C‐indices were 0.74 (SE = 0.002) for white men, 0.76 (0.005) for black men, 0.79 (0.015) for white women, and 0.77 (0.026) for black women, compared with 0.72 (0.002), 0.72 (0.004), 0.77 (0.017), and 0.75 (0.028), respectively, for the HFrEF model. These risk prediction models were generally well calibrated in each race‐specific and sex‐specific stratum of the validation cohort.

Conclusions

Our race‐specific and sex‐specific risk prediction models, which used easily obtainable clinical variables, can be a useful tool to implement preventive strategies or subtype‐specific prevention trials in the nine million users of the VA healthcare system and the general population after external validation.

A randomised trial of a web-based physical activity self-management intervention in COPD

Improving exercise capacity is a primary objective in COPD. Declines in exercise capacity result in reduced physical activity and health-related quality of life (HRQoL). Self-management interventions can teach patients skills and behaviours to manage their disease. Technology-mediated interventions have the potential to provide easily accessible support for disease self-management. We evaluated the effectiveness of a web-based self-management intervention, focused on physical activity promotion, on exercise capacity in COPD. This 6-month randomised controlled trial (NCT02099799) enrolled 153 persons with COPD at two US sites (VABoston, n=108; VABirmingham, n=45). Participants were allocated (1:1) to the web-based self-management intervention (physical activity promotion through personalised, progressive step-count goals, feedback, online COPD-related education and social support via an online community) or usual care. The primary outcome was exercise capacity (6-min walk distance (6 MWD)). Secondary outcomes included physical activity (daily steps per day), HRQoL (St. George's Respiratory Questionnaire Total Score), dyspnoea, COPD-related knowledge and social support. Change in step-count goals reflected intervention engagement. Participants' mean age was 69 (sd=7), and mean forced expiratory volume in 1 s % predicted was 61% (sd=21%). Change in 6MWD did not differ between groups. Intervention participants improved their mean daily step counts by 1312 more than those in the usual care group (p<0.001). Groups did not differ on other secondary outcomes. VABirmingham participants were significantly more engaged with the intervention, although site did not modify the effect of the intervention on 6MWD or secondary outcomes. The intervention did not improve exercise capacity but improved physical activity at 6 months. Additional intervention modifications are needed to optimise its COPD self-management capabilities.

A web-based self-management intervention improved physical activity but not exercise capacity. There is a need to develop and study accessible self-management interventions for COPD.https://bit.ly/3iT1yvU

Phenome-wide association of 1809 phenotypes and COVID-19 disease progression in the Veterans Health Administration Million Veteran Program

BACKGROUND

The risk factors associated with the stages of Coronavirus Disease-2019 (COVID-19) disease progression are not well known. We aim to identify risk factors specific to each state of COVID-19 progression from SARS-CoV-2 infection through death.

METHODS AND RESULTS

We included 648,202 participants from the Veteran Affairs Million Veteran Program (2011-). We identified characteristics and 1,809 ICD code-based phenotypes from the electronic health record. We used logistic regression to examine the association of age, sex, body mass index (BMI), race, and prevalent phenotypes to the stages of COVID-19 disease progression: infection, hospitalization, intensive care unit (ICU) admission, and 30-day mortality (separate models for each). Models were adjusted for age, sex, race, ethnicity, number of visit months and ICD codes, state infection rate and controlled for multiple testing using false discovery rate (≤0.1). As of August 10, 2020, 5,929 individuals were SARS-CoV-2 positive and among those, 1,463 (25%) were hospitalized, 579 (10%) were in ICU, and 398 (7%) died. We observed a lower risk in women vs. men for ICU and mortality (Odds Ratio (95% CI): 0.48 (0.30-0.76) and 0.59 (0.31-1.15), respectively) and a higher risk in Black vs. Other race patients for hospitalization and ICU (OR (95%CI): 1.53 (1.32-1.77) and 1.63 (1.32-2.02), respectively). We observed an increased risk of all COVID-19 disease states with older age and BMI ≥35 vs. 20-24 kg/m2. Renal failure, respiratory failure, morbid obesity, acid-base balance disorder, white blood cell diseases, hydronephrosis and bacterial infections were associated with an increased risk of ICU admissions; sepsis, chronic skin ulcers, acid-base balance disorder and acidosis were associated with mortality.

CONCLUSIONS

Older age, higher BMI, males and patients with a history of respiratory, kidney, bacterial or metabolic comorbidities experienced greater COVID-19 severity. Future studies to investigate the underlying mechanisms associated with these phenotype clusters and COVID-19 are warranted.

Egg consumption, overall diet quality, and risk of type 2 diabetes and coronary heart disease: A pooling project of US prospective cohorts

BACKGROUND AND AIMS

Data on the relation of egg consumption with risk of type 2 diabetes (T2D) and coronary heart disease (CHD) are limited and inconsistent. Few studies have controlled for overall dietary patterns in egg-T2D or egg-CHD analyses, and it is unclear whether any observed elevated risks of T2D and CHD with frequent egg consumption is real or due to confounding by dietary habits. We tested the hypothesis that frequent egg consumption is associated with a higher risk of T2D and CHD risk after adjustment for overall dietary patterns among adults.

DESIGN

We used prospective cohort design to complete time-to-event analyses.

METHODS

We pooled de novo, harmonized, individual-level analyses from nine US cohorts (n = 103,811). Cox regression was used to estimate hazard ratios separately in each cohort adjusting for age, ethnicity, body mass index (BMI), exercise, smoking, alcohol intake, and dietary patterns. We pooled cohort-specific results using an inverse-variance weighted method to estimate summary relative risks.

RESULTS

Median age ranged from 25 to 72 years. Median egg consumption was 1 egg per week in most of the cohorts. While egg consumption up to one per week was not associated with T2D risk, consumption of ≥2 eggs per week was associated with elevated risk [27% elevated risk of T2D comparing 7+ eggs/week with none (95% CI: 16%-37%)]. There was little evidence for heterogeneity across cohorts and we observed similar conclusions when stratified by BMI. Overall, egg consumption was not associated with the risk of CHD. However, in a sensitivity analysis, there was a 30% higher risk of CHD (95% CI: 3%-56%) restricted to older adults consuming 5-6 eggs/week.

CONCLUSIONS

Our data showed an elevated risk of T2D with egg consumption of ≥2 eggs per week but not with <2 eggs/week. While there was no overall association of egg consumption with CHD risk, the elevated CHD observed with consumption of 5-6 eggs/week in older cohorts merits further investigation.

Actionable druggable genome-wide Mendelian randomization identifies repurposing opportunities for COVID-19

Drug repurposing provides a rapid approach to meet the urgent need for therapeutics to address COVID-19. To identify therapeutic targets relevant to COVID-19, we conducted Mendelian randomization analyses, deriving genetic instruments based on transcriptomic and proteomic data for 1,263 actionable proteins that are targeted by approved drugs or in clinical phase of drug development. Using summary statistics from the Host Genetics Initiative and the Million Veteran Program, we studied 7,554 patients hospitalized with COVID-19 and >1 million controls. We found significant Mendelian randomization results for three proteins (ACE2, P = 1.6 × 10-6; IFNAR2, P = 9.8 × 10-11 and IL-10RB, P = 2.3 × 10-14) using cis-expression quantitative trait loci genetic instruments that also had strong evidence for colocalization with COVID-19 hospitalization. To disentangle the shared expression quantitative trait loci signal for IL10RB and IFNAR2, we conducted phenome-wide association scans and pathway enrichment analysis, which suggested that IFNAR2 is more likely to play a role in COVID-19 hospitalization. Our findings prioritize trials of drugs targeting IFNAR2 and ACE2 for early management of COVID-19.

Chocolate consumption and risk of coronary artery disease: the Million Veteran Program

Background

Although previous studies have suggested cocoa products may promote cardiovascular health in the general population, no public data are available from patients receiving care in a national integrated health care system.

Objectives

We tested the hypothesis that regular chocolate consumption is associated with a lower risk of coronary artery disease (CAD) events among participants of the Million Veteran Program (MVP). Secondary analysis examined if the main hypothesis was observed among participants with type 2 diabetes.

Methods

We analyzed data from MVP participants who completed the food frequency section of the MVP Lifestyle Survey and were free of CAD at the time of survey completion. CAD events during follow-up (International Statistical Classification of Diseases Ninth Revision codes 410-411 and 413-414, and Tenth Revision codes I20-I25 except I25.2) were assessed using electronic health records. We fitted a Cox proportional hazard model to estimate the RR of CAD.

Results

Of 188,447 MVP enrollees with survey data, mean ± SD age was 64 ± 12.0 y and 90% were men. For regular chocolate (28.3 g/serving) consumption of <1 serving/mo, 1-3 servings/mo, 1 serving/wk, 2-4 servings/wk, and ≥5 servings/wk, crude incidence rates (per 1000 person-years) for fatal and nonfatal CAD events or coronary procedures were 20.2, 17.5, 16.7, 17.1, and 16.9, respectively, during a mean follow-up of 3.2 y. After adjusting for age, sex, race, and lifestyle factors, the corresponding HRs (95% CIs) were 1.00 (ref), 0.92 (0.87, 0.96), 0.88 (0.83, 0.93), 0.89 (0.84, 0.95), and 0.89 (0.84, 0.96), respectively (P for linear trend < 0.0001). In a secondary analysis of 47,265 diabetics, we did not observe a decreasing trend in CAD mortality among those who consumed ≥1 serving of chocolate a month compared with those who consumed <1 serving/mo.

Conclusions

Regular chocolate consumption was associated with a lower risk of CAD among veterans, but was not associated with cardiovascular disease risk in veterans with type 2 diabetes.

Revisiting methods for modeling longitudinal and survival data: Framingham Heart Study

Background

Statistical methods for modeling longitudinal and time-to-event data has received much attention in medical research and is becoming increasingly useful. In clinical studies, such as cancer and AIDS, longitudinal biomarkers are used to monitor disease progression and to predict survival. These longitudinal measures are often missing at failure times and may be prone to measurement errors. More importantly, time-dependent survival models that include the raw longitudinal measurements may lead to biased results. In previous studies these two types of data are frequently analyzed separately where a mixed effects model is used for the longitudinal data and a survival model is applied to the event outcome.

Methods

In this paper we compare joint maximum likelihood methods, a two-step approach and a time dependent covariate method that link longitudinal data to survival data with emphasis on using longitudinal measures to predict survival. We apply a Bayesian semi-parametric joint method and maximum likelihood joint method that maximizes the joint likelihood of the time-to-event and longitudinal measures. We also implement the Two-Step approach, which estimates random effects separately, and a classic Time Dependent Covariate Model. We use simulation studies to assess bias, accuracy, and coverage probabilities for the estimates of the link parameter that connects the longitudinal measures to survival times.

Results

Simulation results demonstrate that the Two-Step approach performed best at estimating the link parameter when variability in the longitudinal measure is low but is somewhat biased downwards when the variability is high. Bayesian semi-parametric and maximum likelihood joint methods yield higher link parameter estimates with low and high variability in the longitudinal measure. The Time Dependent Covariate method resulted in consistent underestimation of the link parameter. We illustrate these methods using data from the Framingham Heart Study in which lipid measurements and Myocardial Infarction data were collected over a period of 26 years.

Conclusions

Traditional methods for modeling longitudinal and survival data, such as the time dependent covariate method, that use the observed longitudinal data, tend to provide downwardly biased estimates. The two-step approach and joint models provide better estimates, although a comparison of these methods may depend on the underlying residual variance.

Supplementary Information

The online version contains supplementary material available at 10.1186/s12874-021-01207-y.

Development and validation of a 30-day mortality index based on pre-existing medical administrative data from 13,323 COVID-19 patients: The Veterans Health Administration COVID-19 (VACO) Index

BACKGROUND

Available COVID-19 mortality indices are limited to acute inpatient data. Using nationwide medical administrative data available prior to SARS-CoV-2 infection from the US Veterans Health Administration (VA), we developed the VA COVID-19 (VACO) 30-day mortality index and validated the index in two independent, prospective samples.

METHODS AND FINDINGS

We reviewed SARS-CoV-2 testing results within the VA between February 8 and August 18, 2020. The sample was split into a development cohort (test positive between March 2 and April 15, 2020), an early validation cohort (test positive between April 16 and May 18, 2020), and a late validation cohort (test positive between May 19 and July 19, 2020). Our logistic regression model in the development cohort considered demographics (age, sex, race/ethnicity), and pre-existing medical conditions and the Charlson Comorbidity Index (CCI) derived from ICD-10 diagnosis codes. Weights were fixed to create the VACO Index that was then validated by comparing area under receiver operating characteristic curves (AUC) in the early and late validation cohorts and among important validation cohort subgroups defined by sex, race/ethnicity, and geographic region. We also evaluated calibration curves and the range of predictions generated within age categories. 13,323 individuals tested positive for SARS-CoV-2 (median age: 63 years; 91% male; 42% non-Hispanic Black). We observed 480/3,681 (13%) deaths in development, 253/2,151 (12%) deaths in the early validation cohort, and 403/7,491 (5%) deaths in the late validation cohort. Age, multimorbidity described with CCI, and a history of myocardial infarction or peripheral vascular disease were independently associated with mortality-no other individual comorbid diagnosis provided additional information. The VACO Index discriminated mortality in development (AUC = 0.79, 95% CI: 0.77-0.81), and in early (AUC = 0.81 95% CI: 0.78-0.83) and late (AUC = 0.84, 95% CI: 0.78-0.86) validation. The VACO Index allows personalized estimates of 30-day mortality after COVID-19 infection. For example, among those aged 60-64 years, overall mortality was estimated at 9% (95% CI: 6-11%). The Index further discriminated risk in this age stratum from 4% (95% CI: 3-7%) to 21% (95% CI: 12-31%), depending on sex and comorbid disease.

CONCLUSION

Prior to infection, demographics and comorbid conditions can discriminate COVID-19 mortality risk overall and within age strata. The VACO Index reproducibly identified individuals at substantial risk of COVID-19 mortality who might consider continuing social distancing, despite relaxed state and local guidelines.

Egg consumption and risk of coronary artery disease in the Million Veteran Program

BACKGROUND & AIMS

Limited and inconsistent data are available on the relation between egg consumption and risk of myocardial infarction (MI) and it is unclear if adiposity or type 2 diabetes modifies egg-MI relation. We tested the primary hypothesis that egg consumption is positively associated with incidence of MI among veterans. In secondary analyses, we examined potential effect modification of egg-MI relation by adiposity and type 2 diabetes.

METHODS

We analyzed data collected on 188,267 US veterans who were enrolled in the Million Veteran Program (MVP) from 2011 to 2018. Information on egg consumption was obtained via self-administered food frequency questionnaire and we used electronic health records to identify incident MI.

RESULTS

The mean age was 64.4 (SD = 12.0) years and 9.9% of the population were female. We ascertained 10,260 new cases of non-fatal MI during an average follow up of 3.24 years (range: 0.002 to 7.49 y). Hazard ratio (95% CI) for non-fatal MI were 1.00 (ref), 0.93 (0.85-0.1.02), 0.96 (0.87-1.05), 0.98 (0.89-1.07), 1.08 (0.98-1.19), 1.11 (1.00-1.24), and 1.13 (1.00-1.28) for egg consumption of <1/month, 1-3/month, 1/week, 2-4/week, 5-6/week, 1/d, and 2+/d, respectively, controlling for age, sex, race, body mass index, smoking, exercise, alcohol intake, and overall dietary pattern (p non-linear trend 0.019). In secondary analyses, we observed similar results with a composite endpoint including fatal MI, coronary angioplasty and revascularization.

CONCLUSIONS

Our data showed no association of infrequent consumption of eggs with non-fatal MI but a slightly elevated risk with intake of 1 or more eggs per day among US veterans.

Optimizing Atherosclerotic Cardiovascular Disease Risk Estimation for Veterans With Diabetes Mellitus

BACKGROUND

Estimated 10-year atherosclerotic cardiovascular disease (ASCVD) risk in diabetes mellitus patients is used to guide primary prevention, but the performance of risk estimators (2013 Pooled Cohort Equations [PCE] and Risk Equations for Complications of Diabetes [RECODe]) varies across populations. Data from electronic health records could be used to improve risk estimation for a health system's patients. We aimed to evaluate risk equations for initial ASCVD events in US veterans with diabetes mellitus and improve model performance in this population.

METHODS AND RESULTS

We studied 183 096 adults with diabetes mellitus and without prior ASCVD who received care in the Veterans Affairs Healthcare System (VA) from 2002 to 2016 with mean follow-up of 4.6 years. We evaluated model discrimination, using Harrell's C statistic, and calibration, using the reclassification χ2 test, of the PCE and RECODe equations to predict fatal or nonfatal myocardial infarction or stroke and cardiovascular mortality. We then tested whether model performance was affected by deriving VA-specific β-coefficients. Discrimination of ASCVD events by the PCE was improved by deriving VA-specific β-coefficients (C statistic increased from 0.560 to 0.597) and improved further by including measures of glycemia, renal function, and diabetes mellitus treatment (C statistic, 0.632). Discrimination by the RECODe equations was improved by substituting VA-specific coefficients (C statistic increased from 0.604 to 0.621). Absolute risk estimation by PCE and RECODe equations also improved with VA-specific coefficients; the calibration P increased from <0.001 to 0.08 for PCE and from <0.001 to 0.005 for RECODe, where higher P indicates better calibration. Approximately two-thirds of veterans would meet a guideline indication for high-intensity statin therapy based on the PCE versus only 10% to 15% using VA-fitted models.

CONCLUSIONS

Existing ASCVD risk equations overestimate risk in veterans with diabetes mellitus, potentially impacting guideline-indicated statin therapy. Prediction model performance can be improved for a health system's patients using readily available electronic health record data.

Association of Statin Use With All-Cause and Cardiovascular Mortality in US Veterans 75 Years and Older

IMPORTANCE

Data are limited regarding statin therapy for primary prevention of atherosclerotic cardiovascular disease (ASCVD) in adults 75 years and older.

OBJECTIVE

To evaluate the role of statin use for mortality and primary prevention of ASCVD in veterans 75 years and older.

DESIGN, SETTING, AND PARTICIPANTS

Retrospective cohort study that used Veterans Health Administration (VHA) data on adults 75 years and older, free of ASCVD, and with a clinical visit in 2002-2012. Follow-up continued through December 31, 2016. All data were linked to Medicare and Medicaid claims and pharmaceutical data. A new-user design was used, excluding those with any prior statin use. Cox proportional hazards models were fit to evaluate the association of statin use with outcomes. Analyses were conducted using propensity score overlap weighting to balance baseline characteristics.

EXPOSURES

Any new statin prescription.

MAIN OUTCOMES AND MEASURES

The primary outcomes were all-cause and cardiovascular mortality. Secondary outcomes included a composite of ASCVD events (myocardial infarction, ischemic stroke, and revascularization with coronary artery bypass graft surgery or percutaneous coronary intervention).

RESULTS

Of 326 981 eligible veterans (mean [SD] age, 81.1 [4.1] years; 97% men; 91% white), 57 178 (17.5%) newly initiated statins during the study period. During a mean follow-up of 6.8 (SD, 3.9) years, a total 206 902 deaths occurred including 53 296 cardiovascular deaths, with 78.7 and 98.2 total deaths/1000 person-years among statin users and nonusers, respectively (weighted incidence rate difference [IRD]/1000 person-years, -19.5 [95% CI, -20.4 to -18.5]). There were 22.6 and 25.7 cardiovascular deaths per 1000 person-years among statin users and nonusers, respectively (weighted IRD/1000 person-years, -3.1 [95 CI, -3.6 to -2.6]). For the composite ASCVD outcome there were 123 379 events, with 66.3 and 70.4 events/1000 person-years among statin users and nonusers, respectively (weighted IRD/1000 person-years, -4.1 [95% CI, -5.1 to -3.0]). After propensity score overlap weighting was applied, the hazard ratio was 0.75 (95% CI, 0.74-0.76) for all-cause mortality, 0.80 (95% CI, 0.78-0.81) for cardiovascular mortality, and 0.92 (95% CI, 0.91-0.94) for a composite of ASCVD events when comparing statin users with nonusers.

CONCLUSIONS AND RELEVANCE

Among US veterans 75 years and older and free of ASCVD at baseline, new statin use was significantly associated with a lower risk of all-cause and cardiovascular mortality. Further research, including from randomized clinical trials, is needed to more definitively determine the role of statin therapy in older adults for primary prevention of ASCVD.

Estimation of Atherosclerotic Cardiovascular Disease Risk Among Patients in the Veterans Affairs Health Care System

IMPORTANCE

Current guidelines recommend statin therapy for millions of US residents for the primary prevention of atherosclerotic cardiovascular disease (ASCVD). It is unclear whether traditional prediction models that do not account for current widespread statin use are sufficient for risk assessment.

OBJECTIVES

To examine the performance of the Pooled Cohort Equations (PCE) for 5-year ASCVD risk estimation in a contemporary cohort and to test the hypothesis that inclusion of statin therapy improves model performance.

DESIGN, SETTING, AND PARTICIPANTS

This cohort study included adult patients in the Veterans Affairs health care system without baseline ASCVD. Using national electronic health record data, 3 Cox proportional hazards models were developed to estimate 5-year ASCVD risk, as follows: the variables and published β coefficients from the PCE (model 1), the PCE variables with cohort-derived β coefficients (model 2), and model 2 plus baseline statin use (model 3). Data were collected from January 2002 to December 2012 and analyzed from June 2016 to March 2020.

EXPOSURES

Traditional ASCVD risk factors from the PCE plus baseline statin use.

MAIN OUTCOMES AND MEASURES

Incident ASCVD and ASCVD mortality.

RESULTS

Of 1 672 336 patients in the cohort (mean [SD] baseline age 58.0 [13.8] years, 1 575 163 [94.2%] men, 1 383 993 [82.8%] white), 312 155 (18.7%) were receiving statin therapy at baseline. During 5 years of follow-up, 66 605 (4.0%) experienced an ASCVD event, and 31 878 (1.9%) experienced ASCVD death. Compared with the original PCE, the cohort-derived model did not improve model discrimination in any of the 4 age-sex strata but did improve model calibration. The PCE overestimated ASCVD risk compared with the cohort-derived model; 211 237 of 1 136 161 white men (18.6%), 29 634 of 218 463 black men (13.6%), 1741 of 44 399 white women (3.9%), and 836 of 16 034 black women (5.2%) would be potentially eligible for statin therapy under the PCE but not the cohort-derived model. When added to the cohort-derived model, baseline statin therapy was associated with a 7% (95% CI, 5%-9%) lower relative risk of ASCVD and a 25% (95% CI, 23%-28%) lower relative risk for ASCVD death.

CONCLUSIONS AND RELEVANCE

In this study, lower than expected rates of incident ASCVD events in a contemporary national cohort were observed. The PCE overestimated ASCVD risk, and more than 15% of patients would be potentially eligible for statin therapy based on the PCE but not on a cohort-derived model. In the statin era, health care professionals and systems should base ASCVD risk assessment on models calibrated to their patient populations.

Standardized Architecture for a Mega-Biobank Phenomic Library: The Million Veteran Program (MVP)

Electronic health records (EHRs) provide a wealth of data for phenotype development in population health studies, and researchers invest considerable time to curate data elements and validate disease definitions. The ability to reproduce well-defined phenotypes increases data quality, comparability of results and expedites research. In this paper, we present a standardized approach to organize and capture phenotype definitions, resulting in the creation of an open, online repository of phenotypes. This resource captures phenotype development, provenance and process from the Million Veteran Program, a national mega-biobank embedded in the Veterans Health Administration (VHA). To ensure that the repository is searchable, extendable, and sustainable, it is necessary to develop both a proper digital catalog architecture and underlying metadata infrastructure to enable effective management of the data fields required to define each phenotype. Our methods provide a resource for VHA investigators and a roadmap for researchers interested in standardizing their phenotype definitions to increase portability.

Diabetes Mellitus-Related All-Cause and Cardiovascular Mortality in a National Cohort of Adults

Background Diabetes mellitus is a risk factor for cardiovascular disease ( CVD ) and has been associated with 2- to 4-fold higher mortality. Diabetes mellitus-related mortality has not been reassessed in individuals receiving routine care in the United States in the contemporary era of CVD risk reduction. Methods and Results We retrospectively studied 963 648 adults receiving care in the US Veterans Affairs Healthcare System from 2002 to 2014; mean follow-up was 8 years. We estimated associations of diabetes mellitus status and hemoglobin A1c (HbA1c) with all-cause and CVD mortality using covariate-adjusted incidence rates and multivariable Cox proportional hazards regression. Of participants, 34% had diabetes mellitus. Compared with nondiabetic individuals, patients with diabetes mellitus had 7.0 (95% CI , 6.7-7.4) and 3.5 (95% CI, 3.3-3.7) deaths/1000-person-years higher all-cause and CVD mortality, respectively. The age-, sex-, race-, and ethnicity-adjusted hazard ratio for diabetes mellitus-related mortality was 1.29 (95% CI, 1.28-1.31), and declined with adjustment for CVD risk factors (hazard ratio, 1.18 [95% CI, 1.16-1.19]) and glycemia (hazard ratio, 1.03 [95% CI, 1.02-1.05]). Among individuals with diabetes mellitus, CVD mortality increased as HbA1c exceeded 7% (hazard ratios, 1.11 [95% CI, 1.08-1.14], 1.25 [95% CI, 1.22-1.29], and 1.52 [95% CI, 1.48-1.56] for HbA1c 7%-7.9%, 8%-8.9%, and ≥9%, respectively, relative to HbA1c 6%-6.9%). HbA1c 6% to 6.9% was associated with the lowest mortality risk irrespective of CVD history or age. Conclusions Diabetes mellitus remains significantly associated with all-cause and CVD mortality, although diabetes mellitus-related excess mortality is lower in the contemporary era than previously. We observed a gradient of mortality risk with increasing HbA1c >6% to 6.9%, suggesting HbA1c remains an informative predictor of outcomes even if causality cannot be inferred.

Association of pulse rate with outcomes in heart failure with reduced ejection fraction: a retrospective cohort study

BACKGROUND

In a real-world setting, the effect of pulse rate measured at the time of diagnosis and serially during follow-up and management, on outcomes in heart failure with reduced ejection fraction (HFrEF), has not been well-studied. Furthermore, how beta-blockade use in a real-world situation modifies this relation between pulse rate and outcomes in HFrEF is not well-known. Hence, we identified a large, national, real-world cohort of HFrEF to examine the association of pulse rate and outcomes.

METHODS

Using Veterans Affairs (VA) national electronic health records we identified incident HFrEF cases between 2006 and 2012. We examined the associations of both baseline and serially measured pulse rates, with mortality and days hospitalized per year for heart failure and for any cause, using crude and multivariable Cox proportional hazards and Poisson or negative binomial models, respectively. The exposure was examined as continuous, dichotomous, and categorical. Post-hoc analyses addressed the interaction of pulse rate and beta-blocker target dose.

RESULTS

We identified 51,194 incident HFrEF cases (67 ± 12 years, 98% male, 77% white. A significant positive, near linear relationship was observed for both baseline and serially measured pulse rates with all-cause mortality, all-cause hospitalization and heart failure hospitalization after adjusting for covariates including beta-blocker use. Patients who had a pulse rate ≥ 70 bpm in the past 6 months had 36% (95% CI: 31-42%), 25% (95% CI: 19-32%), and 51% (95% CI: 33-72%) increased rates of mortality, all-cause hospitalization, and heart failure hospitalization, respectively, compared to patients with pulse rates < 70 bpm. A minority of subjects (15%) were treated with guideline directed beta blockade ≥50% of recommended target dose, among whom better outcomes were seen compared to those who did not achieve target dose in patients with pulse rates both above and below 70 beats per minute.

CONCLUSIONS

High pulse rate, both at the time of diagnosis and during follow-up, is strongly associated with increased risk of adverse outcomes in HFrEF patients, independent of the use of beta-blockers. In a real-world setting, the majority of HFrEF patients do not achieve target dose of beta-blockade; greater use of strategies to reduce heart rate may improve outcomes in HFrEF.

Plasma vitamin D, past chest illness, and risk of future chest illness in chronic spinal cord injury (SCI): a longitudinal observational study

STUDY DESIGN

Observational study.

OBJECTIVE

Assess associations between vitamin D levels and other risk factors on future chest illness in a chronic spinal cord injury (SCI) cohort.

SETTING

Veterans Affairs Boston and the Boston, MA community.

METHODS

Between August 2009 and August 2017, 253 participants with chronic SCI were followed over a median of 3.2 years (up to 7.4 years) with two to four visits a median of 1.7 years apart. At each visit, plasma 25-hydroxyvitamin D level was obtained, spirometry performed, and a respiratory questionnaire assessing chest illnesses since last visit was completed. Repeated measures negative binomial regression was used to assess chest illness risk longitudinally.

RESULTS

At entry, 25% had deficient vitamin D levels (<20 nanograms/milliliter (ng/ml)), 52% were insufficient (20 to <30 ng/ml), and 23% were sufficient (≥30 ng/ml). Over 545 study visits, chest illnesses (n = 106) were reported by 60 participants. In multivariable models (including previous chest illness history), deficient vitamin D levels (compared with those with sufficient levels) were associated with future chest illness though with wide confidence limits (relative risk (RR) = 1.36, 95% confidence intervals (CI) = 0.74, 2.47). The strongest association with chest illness during the follow-up period was in persons who reported pneumonia/bronchitis after injury and a chest illness in the three years before study entry (RR = 7.62; 95% CI = 3.70, 15.71).

CONCLUSION

Assessed prospectively in chronic SCI, there was a suggestive association between deficient vitamin D levels and future chest illness. Past chest illness history was also strongly associated with future chest illness.

Long-term effects of web-based pedometer-mediated intervention on COPD exacerbations

BACKGROUND

Technology-based physical activity (PA) interventions have been shown to improve daily step counts and health-related quality of life, but their effect on long-term clinical outcomes like acute exacerbations (AEs) is unknown in persons with COPD.

METHODS

U.S. Veterans with stable COPD were randomized (1:1) to either pedometer alone (control) or pedometer plus a website with feedback, goal-setting, disease education, and a community forum (intervention) for 3 months. AEs were assessed every 3 months over a follow-up period of approximately 15 months. Pedometer-assessed daily step counts, health-related quality-of-life (HRQL), and self-efficacy were assessed at baseline, end-of-intervention at 3 months, and during follow-up approximately 6 and 12 months after enrollment. Zero-inflated Poisson models assessed the effect of the intervention on risk for AEs, compared to controls. Generalized linear mixed-effects models for repeated measures examined between-group and within-group changes in daily step count, HRQL, and self-efficacy.

RESULTS

There were no significant differences in age, FEV₁% predicted, baseline daily step count, AEs the year prior to enrollment, or duration of follow-up between the intervention (n = 57) and control (n = 52) groups. The intervention group had a significantly reduced risk of AEs (rate ratio = 0.51, [95%CI 0.31-0.85]), compared to the control group. There were no significant between-group differences in change in average daily step count, HRQL, or self-efficacy at 6 and 12 months after enrollment.

CONCLUSIONS

A 3-month internet-mediated, pedometer-based PA intervention was associated with reduced risk for AEs of COPD over 12-15 months of follow-up. ClinicalTrials.gov identifier: NCT01772082.

Association Between Early Hypertension Control and Cardiovascular Disease Incidence in Veterans With Diabetes

OBJECTIVE

Guidelines for hypertension treatment in patients with diabetes diverge regarding the systolic blood pressure (SBP) threshold at which treatment should be initiated and treatment goal. We examined associations of early SBP treatment with atherosclerotic cardiovascular disease (ASCVD) events in U.S. adults with diabetes.

RESEARCH DESIGN AND METHODS

We studied 43,986 patients with diabetes who newly initiated antihypertensive therapy between 2002 and 2007. Patients were classified into categories based on SBP at treatment initiation (130-139 or ≥140 mmHg) and after 2 years of treatment (100-119, 120-129, 130-139, 140-159, and ≥160 mmHg). The primary outcome was composite ASCVD events (fatal and nonfatal myocardial infarction and stroke), estimated using inverse probability of treatment-weighted Poisson regression and multivariable Cox proportional hazards regression.

RESULTS

Relative to individuals who initiated treatment when SBP was 130-139 mmHg, those with pretreatment SBP ≥140 mmHg had higher ASCVD risk (hazard ratio 1.10 [95% CI 1.02, 1.19]). Relative to those with pretreatment SBP of 130-139 mmHg and on-treatment SBP of 120-129 mmHg (reference group), ASCVD incidence was higher in those with pretreatment SBP ≥140 mmHg and on-treatment SBP 120-129 mmHg (adjusted incidence rate difference [IRD] 1.0 [-0.2 to 2.1] events/1,000 person-years) and in those who achieved on-treatment SBP 130-139 mmHg (IRD 1.9 [0.6, 3.2] and 1.1 [0.04, 2.2] events/1,000 person-years for those with pretreatment SBP 130-139 mmHg and ≥140 mmHg, respectively).

CONCLUSIONS

In this observational study, patients with diabetes initiating antihypertensive therapy when SBP was 130-139 mmHg and those achieving on-treatment SBP <130 mmHg had better outcomes than those with higher SBP levels when initiating or after 2 years on treatment.

GENERATING SURVIVAL TIMES WITH TIME-VARYING COVARIATES USING THE LAMBERT W FUNCTION

Simulation studies provide an important statistical tool in evaluating survival methods, requiring an appropriate data-generating process to simulate data for an underlying statistical model. Many studies with time-to-event outcomes use the Cox proportional hazard model. While methods for simulating such data with time-invariant predictors have been described, methods for simulating data with time-varying covariates are sorely needed. Here, we describe an approach for generating data for the Cox proportional hazard model with time-varying covariates when event times follow an Exponential or Weibull distribution. For each distribution, we derive a closed-form expression to generate survival times and link the time-varying covariates with the hazard function. We consider a continuous time-varying covariate measured at regular intervals over time, as well as time-invariant covariates, in generating time-to-event data under a number of scenarios. Our results suggest this method can lead to simulation studies with reliable and robust estimation of the association parameter in Cox-Weibull and Cox-Exponential models.

Gender Differences in Demographic and Health Characteristics of the Million Veteran Program Cohort

BACKGROUND

The Department of Veterans Affairs Million Veteran Program (MVP) is the largest ongoing cohort program of its kind, with 654,903 enrollees as of June 2018. The objectives of this study were to examine gender differences in the MVP cohort with respect to response and enrollment rates; demographic, health, and health care characteristics; and prevalence of self-reported health conditions.

METHODS

The MVP Baseline Survey was completed by 415,694 veterans (8% women), providing self-report measures of demographic characteristics, health status, and medical history.

RESULTS

Relative to men, women demonstrated a higher positive responder rate (23.0% vs. 16.0%), slightly higher enrollment rate (13.5% vs. 12.9%), and, among enrollees, a lower survey completion rate (59.7% vs. 63.8%). Women were younger, more racially diverse, had higher educational attainment, and were less likely to be married or cohabitating with a partner than men. Women were more likely to report good to excellent health status but poorer physical fitness, and less likely to report lifetime smoking and drinking than men. Compared with men, women veterans showed an increased prevalence of musculoskeletal conditions, thyroid problems, gastrointestinal conditions, migraine headaches, and mental health disorders, as well as a decreased prevalence of gout, cardiovascular diseases, high cholesterol, diabetes, and hearing problems.

CONCLUSIONS

These results revealed some substantial gender differences in the research participation rates, demographic profile, health characteristics, and prevalence of health conditions for veterans in the MVP cohort. Findings highlight the need for tailoring recruitment efforts to ensure representation of the increasing women veteran population receiving care through the Veterans Health Administration.

Fried food consumption and risk of coronary artery disease: The Million Veteran Program

INTRODUCTION

Previous studies of the relationship between fried food consumption and coronary artery disease (CAD) have yielded conflicting results. We tested the hypothesis that frequent fried food consumption is associated with a higher risk of incident CAD events in Million Veteran Program (MVP) participants.

METHODS

Veterans Health Administration electronic health record data were linked to questionnaires completed at MVP enrollment. Self-reported fried food consumption at baseline was categorized: (<1, 1-3, 4-6 times per week or daily). The outcome of interest was non-fatal myocardial infarction or CAD events. We fitted a Cox regression model adjusting for age, sex, race, education, exercise, smoking and alcohol consumption.

RESULTS

Of 154,663 MVP enrollees with survey data, mean age was 64 years and 90% were men. During a mean follow-up of approximately 3 years, there were 6,725 CAD events. There was a positive linear relationship between frequency of fried food consumption and risk of CAD (p for trend 0.0015). Multivariable adjusted hazard ratios (95% CI) were 1.0 (ref), 1.07 (1.01-1.13), 1.08 (1.01-1.16), and 1.14 (1.03-1.27) across consecutive increasing categories of fried food intake.

CONCLUSIONS

In a large national cohort of U.S. Veterans, fried food consumption has a positive, dose-dependent association with CAD.

Validation of an Electronic Medical Record-Based Algorithm for Identifying Posttraumatic Stress Disorder in U.S. Veterans

We developed an algorithm for identifying U.S. veterans with a history of posttraumatic stress disorder (PTSD), using the Department of Veterans Affairs (VA) electronic medical record (EMR) system. This work was motivated by the need to create a valid EMR-based phenotype to identify thousands of cases and controls for a genome-wide association study of PTSD in veterans. We used manual chart review (n = 500) as the gold standard. For both the algorithm and chart review, three classifications were possible: likely PTSD, possible PTSD, and likely not PTSD. We used Lasso regression with cross-validation to select statistically significant predictors of PTSD from the EMR and then generate a predicted probability score of being a PTSD case for every participant in the study population (range: 0-1.00). Comparing the performance of our probabilistic approach (Lasso algorithm) to a rule-based approach (International Classification of Diseases [ICD] algorithm), the Lasso algorithm showed modestly higher overall percent agreement with chart review than the ICD algorithm (80% vs. 75%), higher sensitivity (0.95 vs. 0.84), and higher accuracy (AUC = 0.95 vs. 0.90). We applied a 0.7 probability cut-point to the Lasso results to determine final PTSD case-control status for the VA population. The final algorithm had a 0.99 sensitivity, 0.99 specificity, 0.95 positive predictive value, and 1.00 negative predictive value for PTSD classification (grouping possible PTSD and likely not PTSD) as determined by chart review. This algorithm may be useful for other research and quality improvement endeavors within the VA.