eTable. Main Analysis HIV and the Risk of Total HF, HFPEF, and HFREF With Added Event Data Outside VA
Customize your JAMA Network experience by selecting one or more topics from the list below.
Freiberg MS, Chang CH, Skanderson M, et al. Association Between HIV Infection and the Risk of Heart Failure With Reduced Ejection Fraction and Preserved Ejection Fraction in the Antiretroviral Therapy Era: Results From the Veterans Aging Cohort Study. JAMA Cardiol. 2017;2(5):536–546. doi:10.1001/jamacardio.2017.0264
Does HIV infection increase the risk of heart failure with reduced ejection fraction, heart failure with preserved ejection fraction, or both, and do these risks vary by age, race/ethnicity, HIV–specific biomarkers, and receipt of antiretroviral therapy?
In this cohort study of 98 015 veterans, individuals with HIV infection had a 61% increased risk of heart failure with reduced ejection fraction (ejection fraction <40%), a 21% increased risk of heart failure with preserved ejection fraction (ejection fraction ≥50%), and a 37% increased risk of borderline heart failure with preserved ejection fraction (ejection fraction 40%-49%) compared with uninfected veterans. These risks are significant, even after adjusting for possible confounders, and the association between HIV infection and types of heart failure varies by age, race/ethnicity, HIV–specific biomarkers, and receipt of antiretroviral therapy.
A strategy that encompasses HIV infection treatment, heart failure risk factor prevention and management, and the development of heart failure risk stratification tools would be beneficial for this high-risk population.
With improved survival, heart failure (HF) has become a major complication for individuals with human immunodeficiency virus (HIV) infection. It is unclear if this risk extends to different types of HF in the antiretroviral therapy (ART) era. Determining whether HIV infection is associated with HF with reduced ejection fraction (HFrEF), HF with preserved ejection fraction (HFpEF), or both is critical because HF types differ with respect to underlying mechanism, treatment, and prognosis.
To investigate whether HIV infection increases the risk of future HFrEF and HFpEF and to assess if this risk varies by sociodemographic and HIV-specific factors.
Design, Setting, and Participants
This study evaluated 98 015 participants without baseline cardiovascular disease from the Veterans Aging Cohort Study, an observational cohort of HIV-infected veterans and uninfected veterans matched by age, sex, race/ethnicity, and clinical site, enrolled on or after April 1, 2003, and followed up through September 30, 2012. The dates of the analysis were October 2015 to November 2016.
Human immunodeficiency virus infection.
Main Outcomes and Measures
Outcomes included HFpEF (EF≥50%), borderline HFpEF (EF 40%-49%), HFrEF (EF<40%), and HF of unknown type (EF missing).
Among 98 015 participants, the mean (SD) age at enrollment in the study was 48.3 (9.8) years, 97.0% were male, and 32.2% had HIV infection. During a median follow-up of 7.1 years, there were 2636 total HF events (34.6% were HFpEF, 15.5% were borderline HFpEF, 37.1% were HFrEF, and 12.8% were HF of unknown type). Compared with uninfected veterans, HIV-infected veterans had an increased risk of HFpEF (hazard ratio [HR], 1.21; 95% CI, 1.03-1.41), borderline HFpEF (HR, 1.37; 95% CI, 1.09-1.72), and HFrEF (HR, 1.61; 95% CI, 1.40-1.86). The risk of HFrEF was pronounced in veterans younger than 40 years at baseline (HR, 3.59; 95% CI, 1.95-6.58). Among HIV-infected veterans, time-updated HIV-1 RNA viral load of at least 500 copies/mL compared with less than 500 copies/mL was associated with an increased risk of HFrEF, and time-updated CD4 cell count less than 200 cells/mm3 compared with at least 500 cells/mm3 was associated with an increased risk of HFrEF and HFpEF.
Conclusions and Relevance
Individuals who are infected with HIV have an increased risk of HFpEF, borderline HFpEF, and HFrEF compared with uninfected individuals. The increased risk of HFrEF can manifest decades earlier than would be expected in a typical uninfected population. Future research should focus on prevention, risk stratification, and identification of the mechanisms for HFrEF and HFpEF in the HIV-infected population.
More than 36 million people are infected with human immunodeficiency virus (HIV) worldwide.1 Almost 17 million are receiving antiretroviral therapy (ART).1 With the improved life expectancy because of ART,2 cardiovascular disease (CVD) is now a major health complication among HIV-infected individuals.3 Acute myocardial infarction (AMI) has been studied, and the increased risk of AMI among HIV-infected individuals compared with uninfected individuals is well documented.4-7 Similarly, an excess risk of heart failure (HF) is also present for HIV-infected individuals compared with uninfected individuals; however, it is not known what types of HF are associated with this risk and whether the risk of different types of HF varies by age, race/ethnicity, HIV-specific biomarkers, and receipt of ART.8,9
Studies have shown that HIV infection increases the risk of HF independent of AMI9,10 and that the increased risk is higher among older people, individuals of black race/ethnicity, and those with obesity, hypertension, diabetes, current smoking, alcohol abuse or dependence, elevated HIV-1 RNA viral loads, or a history of AMI.8,9 The success of ART has increased life expectancy for patients with HIV, and treatment of modifiable HF risk factors, in combination with improvements in CVD care, has increased survival for patients with AMI.2,11 Consequently, many HIV-infected individuals will survive with a damaged heart, and their health care professionals will have the challenge of preventing and managing HF in this high-risk population.
To reduce the risk of HF in the HIV-infected population, there is a need to understand the epidemiological patterns surrounding HIV and the risk of HF in the ART era. Among uninfected people, differentiating between HF with reduced ejection fraction (HFrEF) and HF with preserved ejection fraction (HFpEF) is critical because these types of HF differ with respect to underlying mechanism, treatment, and prognosis.12 In the HIV-infected population, our present knowledge on HIV and type of HF in the ART era is limited to case reports, cross-sectional data, and longitudinal data linking HIV infection to echocardiographic changes consistent with HFrEF and HFpEF.13-20 To our knowledge, there are no large studies showing that HIV-infected individuals have a significantly increased risk of HFrEF and HFpEF events compared with demographically and behaviorally similar uninfected individuals in the ART era. Similarly, data describing the association between HIV infection and type of HF across age groups, by race/ethnicity, by HIV-specific biomarkers, and by receipt of ART regimens are also lacking. Yet, HIV infection is common among younger adults21 and minority populations22 and is increasingly diagnosed among older adults.21 Health care professionals do not have the information needed to advise and risk stratify HIV-infected patients who may be at risk for HF.
Therefore, we investigated whether HIV infection is associated with an increased risk of future HFrEF and HFpEF in a national cohort of HIV-infected and uninfected veterans. We evaluated whether this risk varied by age group, race/ethnicity, HIV-specific biomarkers, and receipt of ART regimens.
The Veterans Aging Cohort Study (VACS) is an observational, longitudinal cohort of HIV-infected veterans and uninfected veterans matched by age, sex, race/ethnicity, and clinical site who were enrolled in the same calendar year that has been described previously.23 Study participants are known to have been continuously enrolled each year since 1998 using a validated existing algorithm from the US Department of Veterans Affairs (VA) national electronic medical record system.23 Data for this cohort are extracted from the VA Central Data Warehouse. The Vanderbilt University and West Haven Veterans Affairs Medical Center institutional review boards approved this study. The VACS has a waiver of informed consent.
For this analysis, we included all VACS participants who were alive and enrolled in the VACS on or after April 1, 2003. The baseline date was a participant’s first clinical encounter on or after April 1, 2003. All participants were followed up from their baseline date to an HF event, death, or the last follow-up date (September 30, 2012). The dates of the analysis were October 2015 to November 2016. We excluded participants with prevalent CVD based on International Classification of Diseases, Ninth Revision (ICD-9) codes for AMI, unstable angina, other coronary heart disease, stroke or transient ischemic attack, or HF on or before their baseline date. After these exclusions (n = 35 003), our final sample included 98 015 veterans, of whom 32.2% were infected with HIV.
Using a previously validated algorithm, HIV was defined as the presence of at least 1 inpatient or at least 2 outpatient ICD-9 codes for HIV and inclusion in the VA Immunology Case Registry.23
For the dependent variables, we used the presence of at least 1 inpatient (discharge diagnosis) or at least 2 outpatient VA ICD-9 codes to identify HF events (ICD-9 codes 402.01, 402.11, 402.91, 404.01, 404.03, 404.11, 404.13, 404.91, 404.93, 428.0, 428.1, 428.20, 428.21, 428.22, 428.23, 428.30, 428.31, 428.32, 428.33, 428.40, 428.41, 428.42, 428.43, and 428.9). This definition was based on prior validation work within and outside of the VA.24 Ejection fraction (EF) measurements were used only to classify HF into HFpEF, borderline HFpEF, HFrEF, or HF of unknown type as per guidelines.25 All EF data were obtained using an automated information extraction application that was developed and validated within the VA health care system to identify, among other variables, the mention of EF in clinical notes and corresponding quantitative or qualitative values. This application was informed in part based on an earlier application that extracted EF data from the VA electronic medical record system.26 When the application used for this study was tested across multiple data sources, the application achieved, on average, positive predictive values of 0.968 to 1.000 and sensitivities of 0.801 to 0.899 for EF measurements across different document types. Using values extracted from clinical notes, we selected the EF data closest to the date on or after the incident HF event. The presence of HFpEF was defined as HF with documentation of an EF of at least 50%; when no numerical value was recorded, the left ventricular (LV) function was described as normal. Borderline HFpEF was defined as an EF between 40% and 49%. The presence of HFrEF was defined as HF with an EF less than 40%; when no numerical value was present, the LV dysfunction was described as moderate or severe. When no EF documentation was present, the HF was classified as unknown type.
We used administrative data to determine age, sex, and race/ethnicity. We assessed hypertension, diabetes, lipid levels, renal disease, body mass index (BMI), and anemia using clinical outpatient and laboratory data collected closest to the baseline date. Hydroxymethylglutaryl coenzyme A reductase inhibitor use and ART were based on pharmacy data, and smoking was measured from health factors data that are collected in a standardized form within the VA.27 Hypertension was categorized based on Joint National Committee 7 criteria.28 Our blood pressure measurement was the mean of the 3 routine outpatient clinical measurements closest to the baseline date. Diabetes was diagnosed using a validated metric that considers glucose measurements, antidiabetic agent use, and at least 1 inpatient or at least 2 outpatient ICD-9 codes for this diagnosis.29 Current hydroxymethylglutaryl coenzyme A reductase inhibitor use was defined as a prescription filled within 180 days of the baseline date. Smoking status was categorized as current, past, or never, while BMI (calculated as weight in kilograms divided by height in meters squared) was dichotomized as BMI of at least 30 or less than 30. Hepatitis C virus (HCV) infection was defined as a positive HCV antibody test result or at least 1 inpatient or at least 2 outpatient ICD-9 codes for this diagnosis.29,30 History of alcohol and cocaine abuse or dependence was defined using ICD-9 codes, as was a history of atrial fibrillation.31 We collected data (eg, on CD4 lymphocyte counts [CD4 cell counts] and HIV-1 RNA) at baseline (ie, within 180 days of our enrollment date) through September 30, 2012. Baseline ART was categorized by regimen of ART within 180 days of baseline, including a nucleoside reverse transcriptase inhibitor (NRTI) plus a protease inhibitor (PI), an NRTI plus a non-NRTI (NNRTI), other (ie, use of PI, NRTI, or NNRTI medications but not in combination as described in the other 2 categories), and no ART (reference group). All ART medications that were on VA formulary during the study period were included. Our group has previously demonstrated in a nested sample that 98% of HIV-infected veterans obtain their ART medications from the VA.23
Descriptive statistics for all variables by HIV infection status were assessed using t test or its nonparametric counterpart for continuous variables and using χ2 test or Fisher exact test for categorical variables. We calculated incident total HF, HFpEF, borderline HFpEF, HFrEF, and HF of unknown type rates per 1000 person-years and incidence rate ratios stratified by age group and HIV infection status. We constructed Cox proportional hazards regression models to estimate the hazard ratio (HR) and 95% CI for the association between HIV and the risk of each type of HF after adjusting for other covariates. We also performed sensitivity analyses that included HF events outside of the VA (ie, HF diagnosed using Medicare and VA fee-for-service HF ICD-9 codes). For these analyses, we linked those non-VA HF events to EF data within the VA after the non-VA HF event date. Proportional hazards assumption was not violated for the main predictor (HIV infection status) using the log-log survival plot.32 In secondary analyses, we adjusted our final HFrEF model for incident AMI during the follow-up period. In separate, similar analyses, we also examined the association between HIV infection status and types of HF in important subgroups (eg, those younger than 40 years). Among HIV-infected veterans, we examined the association between time-updated HIV-1 RNA, CD4 cell count, and HF type while also adjusting for potential confounders, including baseline ART. Missing covariate data were included in the analyses using multiple imputation techniques that generated 5 data sets with complete covariate values to increase the robustness of the Cox proportional hazards regression models.
In this analysis, there were 98 015 veterans (32.2% infected with HIV) who were free of baseline CVD. Their mean (SD) age at enrollment in the study was 48.3 (9.8) years, and 97.0% were male. The CVD risk factors and substance use measures varied by HIV infection status (Table 1), in part because of the large sample size. In general, uninfected veterans had a higher prevalence of traditional cardiovascular risk factors except smoking, whereas HIV-infected veterans had a higher prevalence of nontraditional CVD risk factors (eg, HCV infection). For HIV-infected veterans, the median baseline HIV-1 RNA viral load was 1357 copies/mL, the median baseline CD4 cell count was 382 cells/mm3, and 73.9% were receiving ART consisting of PIs (58.4% of those receiving ART) and NRTIs (73.6% of those receiving ART).
During a median follow-up of 7.1 years, there were 2636 total HF events. Of these events, 35.7% occurred in HIV-infected veterans (34.6% were HFpEF, 15.5% were borderline HFpEF, 37.1% were HFrEF, and 12.8% were HF of unknown type). Compared with uninfected veterans, HIV-infected veterans had higher rates of total HF and HFrEF but not HFpEF and borderline HFpEF (Table 2). Similar results were observed when rates were stratified by HIV infection status and age group categories except among those 70 years or older.
Compared with uninfected veterans, HIV-infected veterans had a significantly increased risk of total HF, HFpEF, borderline HFpEF, and HFrEF after adjusting for possible confounders (Table 3). In sensitivity analyses that included non-VA HF events and VA EF data, the association between HIV and total HF, HFpEF, borderline HFpEF, and HFrEF remained essentially unchanged (eTable in the Supplement). Similarly, the association between HIV and HF held when we restricted the sample to those without hypertension (HR, 1.32; 95% CI, 1.08-1.61), individuals without alcohol or cocaine abuse or dependence (HR, 1.43; 95% CI, 1.25-1.65), and never smokers (HR, 1.33; 95% CI, 1.05-1.70). The association between HIV infection and HFrEF persisted after further adjustment for incident AMI during the follow-up period (HR, 1.58; 95% CI, 1.37-1.82).
Among the younger veterans (<40 years at baseline) and individuals of white or black race/ethnicity, HIV infection was significantly associated with an increase in total HF and HFrEF but not HFpEF or borderline HFpEF (Table 3). When we compared uninfected veterans with HIV-infected veterans stratified by HIV-specific biomarkers, the risk of HFrEF persisted even among HIV-infected veterans with a baseline HIV-1 RNA viral load less than 500 copies/mL compared with uninfected veterans (HR, 1.41; 95% CI, 1.17-1.70) (Table 4).
When we restricted the sample to only HIV-infected veterans and adjusted for covariates, including baseline HIV-1 RNA viral load and CD4 cell count, baseline NRTI plus PI (HR, 1.80; 95% CI, 1.19-2.71), NRTI plus NNRTI (HR, 1.48; 95% CI, 1.01-2.15), and other (HR, 3.46; 95% CI, 1.79-6.72) compared with no ART were associated with an increased risk of HFpEF but not HFrEF. In time-updated analyses, CD4 cell count less than 200 cells/mm3 was associated with an increased risk of total HF, HFpEF, borderline HFpEF, and HFrEF (Table 5), whereas time-updated HIV-1 RNA viral load of at least 500 copies/mL was only associated with HFrEF.
In the VACS, HIV-infected veterans had an increased risk of HFpEF, borderline HFpEF, and HFrEF. The association between HIV and HFrEF remained significant even when the sample size was reduced for subgroup analyses that included individuals of white or black race/ethnicity and the younger veterans, as well as after adjustment for AMI in the follow-up period. Among HIV-infected veterans, time-updated HIV-1 RNA viral load of at least 500 copies/mL compared with less than 500 copies/mL was associated with an increased risk of HFrEF, whereas time-updated CD4 cell count less than 200 cells/mm3 compared with at least 500 cells/mm3 was associated with an increased risk of total HF, HFpEF, borderline HFpEF, and HFrEF.
To our knowledge, this investigation is the first large study to report that HIV-infected individuals have a significantly increased risk of HFpEF, borderline HFpEF, and HFrEF events compared with demographically and behaviorally similar uninfected individuals in the ART era. These findings are consistent with and extend earlier echocardiographic reports linking HIV infection to reduced LV systolic function and diastolic dysfunction,13-20 as well our group’s earlier work reporting an association between HIV infection and total HF.8 More specifically, we show herein that the risk of HFrEF extends beyond AMI, is present across multiple decades of age groups, and occurs among individuals of black or white race/ethnicity, those without decades-long exposure to HF risk factors, and those with high HIV-1 RNA viral load and low CD4 cell count over time. In fact, HFrEF among HIV-infected individuals in the ART era can manifest at a young age, decades earlier than might be expected among uninfected individuals.33
While the exact mechanisms underlying the association between HIV and types of HF remain unclear, the fact that time-updated low CD4 cell count was associated with HFrEF and HFpEF suggests that duration of HIV infection and, by extension, chronic inflammation, T-cell activation, and loss of adaptive immunity likely all have important roles. Individuals who are infected with HIV with low CD4 cell counts have increased levels of immune activation and inflammation,34 which are associated with an increased HF risk.35 In murine models, depletion of T-regulatory cells leads to increased myocardial fibrosis, a factor consistent with HFrEF and HFpEF phenotypes.36 Most important, our data also suggest that even HIV-infected individuals with high CD4 cell counts are likely still at risk for HF compared with uninfected individuals, in part because HIV-infected individuals with high CD4 cell counts who are rapidly diagnosed, treated, and virally suppressed do not return to their pre-HIV levels of inflammation.37 Moreover, this residual inflammation is associated with an increased risk of future non-AIDS diseases.37 In contrast, time-updated elevated HIV-1 RNA viral load was only significantly associated with HFrEF. These findings are consistent with reports before the ART era in which unsuppressed HIV viremia, perhaps through direct infection of cardiac myocytes38,39 or cardiac autoantibodies,40 results in a cardiomyopathy consistent with HFrEF.41
The role of ART in the development of HF is less clear. Cardiac mitochondrial toxic effects in the highly active ART era is well documented.42 In the present study, baseline ART use was associated with an increased HFpEF risk, whereas our time-updated data suggested that successful ART as measured by lower HIV-1 RNA viral load and higher CD4 cell count reduces the risk of HFrEF and HFpEF. As prior studies have shown, ART can simultaneously lower AMI risk through viral suppression43 and increase AMI risk likely through medication adverse effects.44 Therefore, determining if newer ART medications have a role in the development of HF should be explored because many HIV-infected individuals will be receiving ART medications for decades.
Our findings have important implications for HIV-infected individuals and their health care professionals. In the United States, 25% of all new cases of HIV are among those aged 13 to 24 years, and 25% of HIV-infected individuals are older than 55 years,21 while 44% of new HIV infections occur in persons of black race/ethnicity,22 who are at high risk for HF.45 Globally, HF is common in low-income and middle-income countries, where the burden of HIV is high and availability of ART can be limited.46 Given these facts, health care professionals should focus on guideline-recommended HIV treatment and HF risk factor prevention (including diabetes, hypertension, renal disease, smoking, alcohol abuse and dependence, and obesity), as well as screening for HIV in individuals with new-onset HF where appropriate.25 Developing tools designed to risk stratify HIV-infected individuals for HF will also be required.
Our investigation has limitations that warrant discussion. First, because HF was determined using ICD-9 codes, it is possible that some misclassification occurred (ie, some true HF events were not captured by ICD-9 codes). However, this finding would have biased our results to the null. Second, because EF data were extracted using a natural language processing application, misclassification may have occurred. However, the application was developed to capture EF data internally within the VA health care system, and its validation against manual data extraction demonstrated high accuracy (positive predictive value, 0.99-1.00). Therefore, we expect the corresponding misclassification to be small. Third, because our study population comprised mostly men, we cannot generalize our findings to women. Fourth, our ART analyses do not include ART duration, nor did we examine specific ART medications. Fifth, our analysis focused on HF events occurring in the VA because EF data outside of the VA were not available. However, when we analyzed non-VA HF event data and linked those events to EF data within the VA after the non-VA HF event date, the associations between HIV infection and types of HF remained essentially unchanged.
In summary, HIV-infected individuals have an increased risk of HFrEF, HFpEF, and borderline HFpEF. For people who are infected with HIV, CD4 cell count less than 200 cells/mm3 compared with at least 500 cells/mm3 is a risk factor for HFpEF, borderline HFpEF, and HFrEF, whereas HIV-1 RNA viral load of at least 500 copies/mL compared with less than 500 copies/mL is a risk factor for HFrEF. Most important, the risk of HFrEF in HIV-infected individuals can manifest decades earlier than would be expected among uninfected individuals. To prevent HF, a strategy focusing on guideline-recommended HIV treatment, prevention, and management of HF risk factors will be required, with screening for HIV infection where appropriate for individuals with new-onset HF, in addition to the development of HF risk stratification tools. Finally, there is a need for basic and translational science research focusing on elucidating the underlying mechanisms causing the excess risk of HFrEF and HFpEF in HIV-infected populations.
Accepted for Publication: January 25, 2017.
Corresponding Author: Matthew S. Freiberg, MD, MSc, Division of Cardiovascular Medicine, Department of Medicine, Vanderbilt University School of Medicine, 2525 W End Ave, Ste 300-A, Nashville, TN 37203 (firstname.lastname@example.org).
Published Online: April 5, 2017. doi:10.1001/jamacardio.2017.0264
Author Contributions: Dr Freiberg had full access to all of the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis.
Study concept and design: Freiberg, Vasan, Gottlieb, Leaf, Rodriguez-Barradas, Tracy, Gibert, Bedimo, Crothers, Butt.
Acquisition, analysis, or interpretation of data: Chang, Skanderson, Patterson, DuVall, Brandt, So-Armah, Oursler, Gottdiener, Gottlieb, Leaf, Rodriguez-Barradas, Rimland, Brown, Goetz, Warner, Crothers, Tindle, Alcorn, Bachmann, Justice, Butt.
Drafting of the manuscript: Freiberg, Chang, Brown, Justice, Butt.
Critical revision of the manuscript for important intellectual content: Chang, Skanderson, Patterson, DuVall, Brandt, So-Armah, Vasan, Oursler, Gottdiener, Gottlieb, Leaf, Rodriguez-Barradas, Tracy, Gibert, Rimland, Bedimo, Brown, Goetz, Warner, Crothers, Tindle, Alcorn, Bachmann, Justice, Butt.
Statistical analysis: Chang, So-Armah.
Obtained funding: Freiberg, Justice.
Administrative, technical, or material support: Skanderson, DuVall, Brandt, Leaf, Rodriguez-Barradas, Rimland, Tindle, Justice.
Study supervision: Oursler, Leaf, Justice.
Conflict of Interest Disclosures: All authors have completed and submitted the ICMJE Form for Disclosure of Potential Conflicts of Interest. Dr Freiberg reported receiving grants from the National Institutes of Health (NIH) during the conduct of the study. Dr DuVall reported receiving grants from the National Heart, Lung, and Blood Institute (NHLBI) during the conduct of the study and reported receiving grants from the following outside of the submitted work: AbbVie Inc, Amgen Inc, Anolinx LLC, Astellas Pharma Inc, AstraZeneca Pharmaceuticals LP, Boehringer Ingelheim International GmbH, F. Hoffman-La Roche Ltd, Genentech Inc, Genomic Health, Inc, Gilead Sciences Inc, GlaxoSmithKline plc, HITEKS Solutions Inc, Innocrin Pharmaceuticals Inc, Kantar Health, LexisNexis Risk Solutions, Merck & Co, Inc, Mylan Specialty LP, Myriad Genetics, Inc, Northrop Grumman Information Systems, Novartis International AG, PAREXEL International Corporation, and Shire PLC. Dr Brandt reported receiving grants from the NHLBI during the conduct of the study and reported receiving grants from the NIH and the Department of Veterans Affairs outside of the submitted work. Dr So-Armah reported receiving grants from the NIH during the conduct of the study. Dr Gottlieb reported receiving grants from Novartis International AG outside of the submitted work. Dr Tracy reported receiving grants from the NIH outside of the submitted work. Dr Bedimo reported receiving grants and other compensation from Bristol-Myers Squibb and Merck & Co, Inc, and reported receiving other compensation from Theratechnologies, Inc, and Gilead Sciences Inc, all outside of the submitted work. Dr Brown reported receiving grants from the National Institute on Alcohol Abuse and Alcoholism (NIAAA) during the conduct of the study. Dr Justice reported receiving grants from the NIAAA during the conduct of the study. Dr Butt reported receiving grants from Merck & Co, Inc, Gilead Sciences Inc, and AbbVie Inc, all outside of the submitted work. No other disclosures were reported.
Funding/Support: This work was supported by grants HL095136 (Drs Freiberg and Justice) and HL125032 (Drs Freiberg and Tracy) from the NHLBI and by grants AA020790 and AA020794 (both to Dr Justice) from the NIAAA.
Role of the Funder/Sponsor: The funding sources had no role in the design and conduct of the study; collection, management, analysis, and interpretation of the data; preparation, review, or approval of the manuscript; and decision to submit the manuscript for publication.
Disclaimer: The views expressed in this article are those of the authors and do not necessarily reflect the position or policies of the Department of Veterans Affairs.
Create a personal account or sign in to: