Safety, Efficacy, and Pharmacokinetics of Combination SARS-CoV-2 Neutralizing Monoclonal Antibodies BMS-986414 (C135-LS) and BMS-986413 (C144-LS) Administered Subcutaneously in Non-Hospitalized Persons with COVID-19 in a Phase 2 Trial.

Background: Outpatient COVID-19 monoclonal antibody (mAb) treatment via subcutaneous delivery, if effective, overcomes the logistical burdens of intravenous administration. Methods: ACTIV-2/A5401 was a randomized, masked placebo-controlled platform trial where participants with COVID-19 at low risk for progression were randomized 1:1 to subcutaneously administered BMS-986414 (C135-LS) 200 mg, plus BMS-986413 (C144-LS) 200 mg, (BMS mAbs), or placebo. Coprimary outcomes were time to symptom improvement through 28 days; nasopharyngeal SARS-CoV-2 RNA below the lower limit of quantification (LLoQ) on days 3, 7, or 14; and treatment-emergent grade 3 or higher adverse events (TEAEs) through 28 days. Results: A total of 211 participants (105 BMS mAbs and 106 placebo) initiated study product. Time to symptom improvement favored the active therapy but was not significant (median 8 vs 10 days, P=0.19). There was no significant difference in the proportion with SARS-CoV-2 RNA

Characterization of treatment resistance and viral kinetics in the setting of single- versus dual-active monoclonal antibodies against SARS-CoV-2.

BACKGROUND: Monoclonal antibodies (mAbs) represent a crucial antiviral strategy for SARS-CoV-2 infection, but it is unclear whether combination mAbs offer a benefit over single-active mAb treatment. Amubarvimab and romlusevimab significantly reduced the risk of hospitalizations or death in the ACTIV-2/A5401 trial. Certain SARS-CoV-2 variants are intrinsically resistant against romlusevimab, leading to only single-active mAb therapy with amubarvimab in these variants. We evaluated virologic outcomes in individuals treated with single- versus dual-active mAbs. METHODS: Participants were non-hospitalized adults at higher risk of clinical progression randomized to amubarvimab plus romlusevimab or placebo. Quantitative SARS-CoV-2 RNA levels and targeted S gene next-generation sequencing was performed on anterior nasal samples. We compared viral load kinetics and resistance emergence between individuals treated with effective single- versus dual-active mAbs depending on the infecting variant. RESULTS: Study participants receiving single- and dual-active mAbs had similar demographics, baseline nasal viral load, symptom score, and symptom duration. Compared to single-active mAb, treatment with dual-active mAbs led to faster viral load decline at study day 3 (p < 0.001) and day 7 (p < 0.01). Treatment-emergent resistance mutations were more likely to be detected after amubarvimab plus romlusevimab treatment than placebo (2.6% vs 0%, P < 0.001), and more frequently detected in the setting of single-active compared to dual-active mAb treatment (7.2% vs 1.1%, p < 0.01). Single-active and dual-active mAb treatment resulted in similar decrease in rates of hospitalizations or death. CONCLUSION: Compared to single-active mAb therapy, dual-active mAbs led to similar clinical outcomes, but significantly faster viral load decline and a lower risk of emergent resistance.

Systemic inflammation in pregnant women with HIV: relationship with HIV treatment regimen and preterm delivery.

OBJECTIVE: : HIV treatment regimen during pregnancy was associated with preterm delivery (PTD) in the PROMISE 1077 BF trial. Systemic inflammation among pregnant women with HIV could help explain differences in PTD by treatment regimen. We assessed associations between inflammation, treatment regimen, and PTD. DESIGN/METHODS: : A nested 1:1 case-control study (N = 362) was conducted within a multi-country randomized trial comparing three HIV regimens in pregnant women: zidovudine alone, or combination antiretroviral therapy (ART) with lopinavir/ritonavir and either zidovudine or tenofovir. Cases were women with PTD (<37 weeks of gestational age). The following inflammatory biomarkers were measured in plasma samples using immunoassays: soluble CD14 (sCD14) and sCD163, intestinal fatty acid-binding protein, interleukin (IL)-6, interferon γ, and tumor necrosis factor α. We fit regression models to assess associations between second trimester biomarkers (measured before ART initiation at 13-23 weeks of gestational age and 4 weeks later), treatment regimen, and PTD. We also assessed whether inflammation was a mediator in the relationship between ART regimen and PTD. RESULTS: : Persistently high interleukin-6 was associated with increased PTD. Compared to zidovudine alone, the difference in biomarker concentration between week 0 and week 4 was significantly higher (p < 0.05) for both PI-based regimens. However, the estimated proportion of the ART effect on increased PTD mediated by persistently high biomarker levels was ≤5% for all biomarkers. CONCLUSIONS: : Persistently high IL-6 during pregnancy was associated with PTD. While PI-based ART was associated with increases in inflammation, factors other than inflammation likely explain the increased PTD in ART-based regimens compared to zidovudine alone.

Safety and efficacy of inhaled interferon-ß1a (SNG001) in adults with mild-to-moderate COVID-19: a randomized, controlled, phase II trial.

Background: With the emergence of SARS-CoV-2 variants resistant to monoclonal antibody therapies and limited global access to therapeutics, the evaluation of novel therapeutics to prevent progression to severe COVID-19 remains a critical need. Methods: Safety, clinical and antiviral efficacy of inhaled interferon-ß1a (SNG001) were evaluated in a phase II randomized controlled trial on the ACTIV-2/A5401 platform (ClinicalTrials.govNCT04518410). Adult outpatients with confirmed SARS-CoV-2 infection within 10 days of symptom onset were randomized and initiated either orally inhaled nebulized SNG001 given once daily for 14 days (n = 110) or blinded pooled placebo (n = 110) between February 10 and August 18, 2021. Findings: The proportion of participants reporting premature treatment discontinuation was 9% among SNG001 and 13% among placebo participants. There were no differences between participants who received SNG001 or placebo in the primary outcomes of treatment emergent Grade 3 or higher adverse events (3.6% and 8.2%, respectively), time to symptom improvement (median 13 and 9 days, respectively), or proportion with unquantifiable nasopharyngeal SARS-CoV-2 RNA at days 3 (28% [26/93] vs. 39% [37/94], respectively), 7 (65% [60/93] vs. 66% [62/94]) and 14 (91% [86/95] vs. 91% [83/81]). There were fewer hospitalizations with SNG001 (n = 1; 1%) compared with placebo (n = 7; 6%), representing an 86% relative risk reduction (p = 0.07). There were no deaths in either arm. Interpretation: In this trial, SNG001 was safe and associated with a non-statistically significant decrease in hospitalization for COVID-19 pneumonia. Funding: The ACTIV-2 platform study is funded by the NIH. Research reported in this publication was supported by the National Institute of Allergy and Infectious Diseases of the National Institutes of Health under Award Number UM1 AI068634, UM1 AI068636 and UM1 AI106701. The content is solely the responsibility of the authors and does not necessarily represent the official views of the National Institutes of Health.

Association Between Anterior Nasal and Plasma SARS-CoV-2 RNA Levels and Hospitalization or Death in Nonhospitalized Adults With Mild-to-Moderate COVID-19.

BACKGROUND: There is little information regarding severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) RNA as a predictor for clinical outcomes in outpatients with mild-to-moderate coronavirus disease 2019 (COVID-19). METHODS: Anterior nasal (AN) and plasma SARS-CoV-2 RNA data from 2115 nonhospitalized adults who received monoclonal antibodies (mAbs) or placebo in the ACTIV-2/A5401 trial were analyzed for associations with hospitalization or death. RESULTS: One hundred two participants were hospitalized or died through 28 days of follow-up. Higher day 0 (pretreatment) AN RNA was associated with increasing risk of hospitalization/death (risk ratio [RR], 1.24 per log10 copies/mL [95% confidence interval {CI}, 1.04-1.49]) among placebo recipients, ranging from 3% to 16% for <2 to ≥6 log10 copies/mL. Although only 1% had quantifiable levels, there was a similar trend across day 0 plasma RNA categories. Higher day 3 AN RNA was associated with subsequent hospitalization/death among placebo recipients (RR, 1.42 per log10 copies/mL [95% CI, 1.00-2.03]), but not mAb recipients (RR, 1.02 per log10 copies/mL [95% CI, 0.68-1.56]). The proportion of treatment effect (reduction in hospitalizations/deaths after day 3 for mAb vs placebo) explained by day 3 AN RNA was 8%. CONCLUSIONS: SARS-CoV-2 RNA levels are predictive of hospitalization/death in the natural history setting, but AN RNA levels may not be a reliable surrogate marker of mAb treatment effect in COVID-19 trials. Clinical Trials Registration. NCT04518410.

Variant-Specific Viral Kinetics in Acute COVID-19.

Understanding variant-specific differences in severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) viral kinetics may explain differences in transmission efficiency and provide insights on pathogenesis and prevention. We evaluated SARS-CoV-2 kinetics from nasal swabs across multiple variants (Alpha, Delta, Epsilon, Gamma) in placebo recipients of the ACTIV-2/A5401 trial. Delta variant infection led to the highest maximum viral load and shortest time from symptom onset to viral load peak. There were no significant differences in time to viral clearance across the variants. Viral decline was biphasic with first- and second-phase decays having half-lives of 11 hours and 2.5 days, respectively, with differences among variants, especially in the second phase. These results suggest that while variant-specific differences in viral kinetics exist, post-peak viral load all variants appeared to be efficiently cleared by the host. Clinical Trials Registration. NCT04518410.

Statistical Challenges When Analyzing SARS-CoV-2 RNA Measurements Below the Assay Limit of Quantification in COVID-19 Clinical Trials.

Most clinical trials evaluating coronavirus disease 2019 (COVID-19) therapeutics include assessments of antiviral activity. In recently completed outpatient trials, changes in nasal severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) RNA levels from baseline were commonly assessed using analysis of covariance (ANCOVA) or mixed models for repeated measures (MMRM) with single imputation for results below assay lower limits of quantification (LLoQ). Analyzing changes in viral RNA levels with singly imputed values can lead to biased estimates of treatment effects. In this article, using an illustrative example from the ACTIV-2 trial, we highlight potential pitfalls of imputation when using ANCOVA or MMRM methods, and illustrate how these methods can be used when considering values

One Week of Oral Camostat Versus Placebo in Nonhospitalized Adults With Mild-to-Moderate Coronavirus Disease 2019: A Randomized Controlled Phase 2 Trial.

BACKGROUND: Camostat inhibits severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) infection in vitro. We studied the safety and efficacy of camostat in ACTIV-2/A5401, a phase 2/3 platform trial of therapeutics for COVID-19 in nonhospitalized adults. METHODS: We conducted a phase 2 study in adults with mild-to-moderate COVID-19 randomized to oral camostat for 7 days or a pooled placebo arm. Primary outcomes were time to improvement in COVID-19 symptoms through day 28, proportion of participants with SARS-CoV-2 RNA below the lower limit of quantification (LLoQ) from nasopharyngeal swabs through day 14, and grade ≥3 treatment-emergent adverse events (TEAEs) through day 28. RESULTS: Of 216 participants (109 randomized to camostat, 107 to placebo) who initiated study intervention, 45% reported ≤5 days of symptoms at study entry and 26% met the protocol definition of higher risk of progression to severe COVID-19. Median age was 37 years. Median time to symptom improvement was 9 days in both arms (P = .99). There were no significant differences in the proportion of participants with SARS-CoV-2 RNA

Safety and Efficacy of Combination SARS-CoV-2 Neutralizing Monoclonal Antibodies Amubarvimab Plus Romlusevimab in Nonhospitalized Patients With COVID-19.

BACKGROUND: Development of safe and effective SARS-CoV-2 therapeutics is a high priority. Amubarvimab and romlusevimab are noncompeting anti-SARS-CoV-2 monoclonal antibodies with an extended half-life. OBJECTIVE: To assess the safety and efficacy of amubarvimab plus romlusevimab. DESIGN: Randomized, placebo-controlled, phase 2 and 3 platform trial. (ClinicalTrials.gov: NCT04518410). SETTING: Nonhospitalized patients with COVID-19 in the United States, Brazil, South Africa, Mexico, Argentina, and the Philippines. PATIENTS: Adults within 10 days onset of symptomatic SARS-CoV-2 infection who are at high risk for clinical progression. INTERVENTION: Combination of monoclonal antibodies amubarvimab plus romlusevimab or placebo. MEASUREMENTS: Nasopharyngeal and anterior nasal swabs for SARS-CoV-2, COVID-19 symptoms, safety, and progression to hospitalization or death. RESULTS: Eight-hundred and seven participants who initiated the study intervention were included in the phase 3 analysis. Median age was 49 years (quartiles, 39 to 58); 51% were female, 18% were Black, and 50% were Hispanic or Latino. Median time from symptom onset at study entry was 6 days (quartiles, 4 to 7). Hospitalizations and/or death occurred in 9 (2.3%) participants in the amubarvimab plus romlusevimab group compared with 44 (10.7%) in the placebo group, with an estimated 79% reduction in events (P < 0.001). This reduction was similar between participants with 5 or less and more than 5 days of symptoms at study entry. Grade 3 or higher treatment-emergent adverse events through day 28 were seen less frequently among participants randomly assigned to amubarvimab plus romlusevimab (7.3%) than placebo (16.1%) (P < 0.001), with no severe infusion reactions or drug-related serious adverse events. LIMITATION: The study population was mostly unvaccinated against COVID-19 and enrolled before the spread of Omicron variants and subvariants. CONCLUSION: Amubarvimab plus romlusevimab was safe and significantly reduced the risk for hospitalization and/or death among nonhospitalized adults with mild to moderate SARS-CoV-2 infection at high risk for progression to severe disease. PRIMARY FUNDING SOURCE: National Institute of Allergy and Infectious Diseases of the National Institutes of Health.

Safety and Efficacy of Combined Tixagevimab and Cilgavimab Administered Intramuscularly or Intravenously in Nonhospitalized Patients With COVID-19: 2 Randomized Clinical Trials.

Importance: Development of effective, scalable therapeutics for SARS-CoV-2 is a priority. Objective: To test the efficacy of combined tixagevimab and cilgavimab monoclonal antibodies for early COVID-19 treatment. Design, Setting, and Participants: Two phase 2 randomized blinded placebo-controlled clinical trials within the Accelerating COVID-19 Therapeutic Interventions and Vaccines (ACTIV)-2/A5401 platform were performed at US ambulatory sites. Nonhospitalized adults 18 years or older within 10 days of positive SARS-CoV-2 test and symptom onset were eligible and were enrolled from February 1 to May 31, 2021. Interventions: Tixagevimab-cilgavimab, 300 mg (150 mg of each component) given intravenously (IV) or 600 mg (300 mg of each component) given intramuscularly (IM) in the lateral thigh, or pooled placebo. Main Outcomes and Measures: Coprimary outcomes were time to symptom improvement through 28 days; nasopharyngeal SARS-CoV-2 RNA below the lower limit of quantification (LLOQ) on days 3, 7, or 14; and treatment-emergent grade 3 or higher adverse events through 28 days. Results: A total of 229 participants were randomized for the IM study and 119 were randomized for the IV study. The primary modified intention-to-treat population included 223 participants who initiated IM tixagevimab-cilgavimab (n = 106) or placebo treatment (n = 117) (median age, 39 [IQR, 30-48] years; 113 [50.7%] were men) and 114 who initiated IV tixagevimab-cilgavimab (n = 58) or placebo treatment (n = 56) (median age, 44 [IQR, 35-54] years; 67 [58.8%] were women). Enrollment in the IV study was stopped early based on a decision to focus on IM product development. Participants were enrolled at a median of 6 (IQR, 4-7) days from COVID-19 symptom onset. Significant differences in time to symptom improvement were not observed for IM tixagevimab-cilgavimab vs placebo or IV tixagevimab-cilgavimab vs placebo. A greater proportion in the IM tixagevimab-cilgavimab arm (69 of 86 [80.2%]) than placebo (62 of 96 [64.6%]) had nasopharyngeal SARS-CoV-2 RNA below LLOQ at day 7 (adjusted risk ratio, 1.33 [95% CI, 1.12-1.57]) but not days 3 and 14; the joint test across time points favored treatment (P = .003). Differences in the proportion below LLOQ were not observed for IV tixagevimab-cilgavimab vs placebo at any of the specified time points. There were no safety signals with either administration route. Conclusions: In these 2 phase 2 randomized clinical trials, IM or IV tixagevimab-cilgavimab was safe but did not change time to symptom improvement. Antiviral activity was more evident in the larger IM trial. Trial Registration: ClinicalTrials.gov Identifier: NCT04518410.

Statistical Challenges when Analyzing SARS-CoV-2 RNA Measurements Below the Assay Limit of Quantification in COVID-19 Clinical Trials.

Most clinical trials evaluating COVID-19 therapeutics include assessments of antiviral activity. In recently completed outpatient trials, changes in nasal SARS-CoV-2 RNA levels from baseline were commonly assessed using analysis of covariance (ANCOVA) or mixed models for repeated measures (MMRM) with single-imputation for results below assay lower limits of quantification (LLoQ). Analyzing changes in viral RNA levels with singly-imputed values can lead to biased estimates of treatment effects. In this paper, using an illustrative example from the ACTIV-2 trial, we highlight potential pitfalls of imputation when using ANCOVA or MMRM methods, and illustrate how these methods can be used when considering values

Phase 2 Safety and Antiviral Activity of SAB-185, a Novel Polyclonal Antibody Therapy for Nonhospitalized Adults With COVID-19.

BACKGROUND: SAB-185, a novel fully human IgG polyclonal immunoglobulin product, underwent phase 2 evaluation for nonhospitalized adults with mild-moderate coronavirus disease 2019 (COVID-19). METHODS: Participants received intravenous SAB-185 3840 units/kg (low-dose) or placebo, or 10 240 units/kg (high-dose) or placebo. Primary outcome measures were nasopharyngeal severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) RNA < lower limit of quantification (LLOQ) at study days 3, 7, and 14, time to symptomatic improvement, and safety through day 28. RESULTS: Two-hundred thirteen participants received low-dose SAB-185/placebo (n = 107/106) and 215 high-dose SAB-185/placebo (n = 110/105). The proportions with SARS-CoV-2 RNA < LLOQ were higher for SAB-185 versus placebo at days 3 and 7 and similar at day 14, and significantly higher at day 7 for high-dose SAB-185 versus placebo only, relative risk 1.23 (95% confidence interval, 1.01-1.49). At day 3, SARS-CoV-2 RNA levels were lower with low-dose and high-dose SAB-185 versus placebo: differences in medians of -0.78 log10 copies/mL (P = .08) and -0.71 log10 copies/mL (P = .10), respectively. No difference was observed in time to symptom improvement: median 11/10 days (P = .24) for low-dose SAB-185/placebo and 8/10 days (P = .50) for high-dose SAB-185/placebo. Grade ≥3 adverse events occurred in 5%/13% of low-dose SAB-185/placebo and 9%/12% of high-dose SAB-185/placebo. CONCLUSIONS: SAB-185 was safe and generally well tolerated and demonstrated modest antiviral activity in predominantly low-risk nonhospitalized adults with COVID-19. Clinical Trials Registration. NCT04518410.

Coordination of inflammatory responses in children with perinatally acquired HIV infection.

OBJECTIVE: We investigated dynamics of inflammatory biomarkers in children with perinatally acquired HIV (PHIV) who started antiretrovirals at age less than 3 years and achieved sustained virologic control (HIV plasma RNA <400 copies/ml). DESIGN: This was a retrospective analysis of inflammatory biomarkers in children enrolled in a randomized trial of early (<3 years of age) PI-based versus NNRTI-based regimens (P1060), who achieved sustained virologic control and participated in a neurodevelopmental follow-up study (P1104s) between ages 5 and 11 years. METHODS: We measured 20 inflammatory biomarkers using ELISA or chemiluminescence at onset of sustained virologic control (Tc) and at P1104s entry (Te). RESULTS: The 213 participants had median ages of 1.2, 1.9, and 7 years at antiretroviral initiation, Tc, and Te, respectively, with 138 on protease inhibitor-based and 74 on NNRTI-based regimens at Tc. Eighteen markers decreased and two increased from Tc to Te (Te-Tc). Biomarker subsets, particularly cytokines, the chemokine IP-10, and adhesion molecules sICAM-1 and sVCAM-1, correlated at Tc, Te, and Te-Tc. At Tc, higher biomarker levels were associated with younger age, female sex, HIV plasma RNA at least 750 000 copies/ml, lower nadir CD4 + %, lower nadir weight z scores, and NNRTI-based treatment. Greater Te-Tc biomarker declines were associated with younger age, male sex, higher Tc biomarker levels, lower nadir CD4 + %, and NNRTI-based treatment. Duration of controlled viremia and nadir height z scores showed mixed associations. CONCLUSION: Biomarker expression showed substantial coordination. Most markers decreased after virologic control. Demographic and clinical variables associated with biomarker patterns were identified. Mechanistic studies of these biomarker patterns are needed to inform interventions to control inflammation.

Plasma biomarker factors associated with neurodevelopmental outcomes in children with perinatal HIV infection and controlled viremia.

OBJECTIVE: We examined relationships between plasma biomarkers and neurodevelopment in children from sub-Saharan Africa with perinatal HIV (PHIV) with controlled viremia on antiretroviral therapy (ART). DESIGN: Longitudinal retrospective cohort study of children with controlled blood HIV replication. METHODS: Children (N = 213; 57% girls) started ART at less than 3 years of age, had neurodevelopmental assessments (cognition, attention/impulsivity, motor proficiency, global executive functions) at 5-11 years, and achieved controlled viremia (HIV-1 RNA <400 copies/ml for ≥9 months before initial assessment). Twenty-three plasma biomarkers were measured at onset of controlled viremia, week 0 (first neurodevelopmental assessment), and week 48 (second neurodevelopmental assessment). Factor analysis was conducted at each time point. Multivariable linear regressions assessed associations between factors and neurodevelopmental scores. RESULTS: Median age at week 0 was 7.0 years. Eighteen biomarkers loaded on six factors: a (L-10, IFNγ, IFNα2, IL-1ß, IL-6, IP-10, TNFα); B (sCD163, sICAM-1, sVCAM-1, CRP); C (sE-selectin, sP-selectin); D [MIP-1ß, vascular endothelial growth factor (VEGF)-A]; E (sCD14, CRP); and F (CX3CL1, MCP-1). Higher factor B scores were consistently associated with worse cognition and attention/impulsivity, and higher factor D scores with better attention/impulsivity. CONCLUSION: These results suggest a detrimental effect of increased endothelial cell activation (sICAM-1, sVCAM-1) and monocyte/macrophage scavenger function (sCD163) and a beneficial effect of increased CCR5 ligand and HIV entry blocker MIP-1ß and angiogenesis stimulant-VEGF concentrations on the neurodevelopment of children with PHIV. The model that emerges is of vascular inflammation leading to neurodevelopmental deficits. The role of persistent HIV replication in the central nervous system also needs to be further explored.

ACCOUNTING FOR DEPENDENT ERRORS IN PREDICTORS AND TIME-TO-EVENT OUTCOMES USING ELECTRONIC HEALTH RECORDS, VALIDATION SAMPLES, AND MULTIPLE IMPUTATION.

Data from electronic health records (EHR) are prone to errors, which are often correlated across multiple variables. The error structure is further complicated when analysis variables are derived as functions of two or more error-prone variables. Such errors can substantially impact estimates, yet we are unaware of methods that simultaneously account for errors in covariates and time-to-event outcomes. Using EHR data from 4217 patients, the hazard ratio for an AIDS-defining event associated with a 100 cell/mm3 increase in CD4 count at ART initiation was 0.74 (95%CI: 0.68-0.80) using unvalidated data and 0.60 (95%CI: 0.53-0.68) using fully validated data. Our goal is to obtain unbiased and efficient estimates after validating a random subset of records. We propose fitting discrete failure time models to the validated subsample and then multiply imputing values for unvalidated records. We demonstrate how this approach simultaneously addresses dependent errors in predictors, time-to-event outcomes, and inclusion criteria. Using the fully validated dataset as a gold standard, we compare the mean squared error of our estimates with those from the unvalidated dataset and the corresponding subsample-only dataset for various subsample sizes. By incorporating reasonably sized validated subsamples and appropriate imputation models, our approach had improved estimation over both the naive analysis and the analysis using only the validation subsample.

Multiple-Imputation Variance Estimation in Studies With Missing or Misclassified Inclusion Criteria.

In observational studies using routinely collected data, a variable with a high level of missingness or misclassification may determine whether an observation is included in the analysis. In settings where inclusion criteria are assessed after imputation, the popular multiple-imputation variance estimator proposed by Rubin ("Rubin's rules" (RR)) is biased due to incompatibility between imputation and analysis models. While alternative approaches exist, most analysts are not familiar with them. Using partially validated data from a human immunodeficiency virus cohort, we illustrate the calculation of an imputation variance estimator proposed by Robins and Wang (RW) in a scenario where the study exclusion criteria are based on a variable that must be imputed. In this motivating example, the corresponding imputation variance estimate for the log odds was 29% smaller using the RW estimator than using the RR estimator. We further compared these 2 variance estimators with a simulation study which showed that coverage probabilities of 95% confidence intervals based on the RR estimator were too high and became worse as more observations were imputed and more subjects were excluded from the analysis. The RW imputation variance estimator performed much better and should be employed when there is incompatibility between imputation and analysis models. We provide analysis code to aid future analysts in implementing this method.

Frequency of non-communicable diseases in people 50 years of age and older receiving HIV care in Latin America.

BACKGROUND: A growing population of older adults with HIV will increase demands on HIV-related healthcare. Nearly a quarter of people receiving care for HIV in Latin America are currently 50 years or older, yet little is known about the frequency of comorbidities in this population. We estimated the prevalence and incidence of non-communicable diseases (NCDs) among people 50 years of age or older (≥50yo) receiving HIV care during 2000-2015 in six centers affiliated with the Caribbean, Central and South American network for HIV epidemiology (CCASAnet). METHODS: We estimated the annual prevalence, and overall prevalence and incidence of cardiovascular diseases, diabetes, hypertension, dyslipidemia, psychiatric disorders, chronic liver and renal diseases, and non-AIDS-defining cancers, and multimorbidity (more than one NCD) of people ≥50yo receiving care for HIV. Analyses were performed according to age at enrollment into HIV care (<50yo and ≥50yo). RESULTS: We included 3,415 patients ≥50yo, of whom 1,487(43%) were enrolled at age ≥50 years. The annual prevalence of NCDs increased from 32% to 68% and multimorbidity from 30% to 40% during 2000-2015. At the last registered visit, 53% of patients enrolled <50yo and 50% of those enrolled ≥50yo had at least one NCD. Most common NCDs at the last visit in each age-group at enrollment were dyslipidemia (36% in <50yo and 28% in ≥50yo), hypertension (17% and 18%), psychiatric disorders (15% and 10%), and diabetes (11% and 12%). CONCLUSIONS: The prevalence of NCDs and multimorbidity in people ≥50 years receiving care for HIV in CCASAnet centers in Latin America increased substantially in the last 15 years. Our results make evident the need of planning for provision of complex, primary care for aging adults living with HIV.

Self-audits as alternatives to travel-audits for improving data quality in the Caribbean, Central and South America network for HIV epidemiology.

INTRODUCTION: Audits play a critical role in maintaining the integrity of observational cohort data. While previous work has validated the audit process, sending trained auditors to sites ("travel-audits") can be costly. We investigate the efficacy of training sites to conduct "self-audits." METHODS: In 2017, eight research groups in the Caribbean, Central, and South America network for HIV Epidemiology each audited a subset of their patient records randomly selected by the data coordinating center at Vanderbilt. Designated investigators at each site compared abstracted research data to the original clinical source documents and captured audit findings electronically. Additionally, two Vanderbilt investigators performed on-site travel-audits at three randomly selected sites (one adult and two pediatric) in late summer 2017. RESULTS: Self- and travel-auditors, respectively, reported that 93% and 92% of 8919 data entries, captured across 28 unique clinical variables on 65 patients, were entered correctly. Across all entries, 8409 (94%) received the same assessment from self- and travel-auditors (7988 correct and 421 incorrect). Of 421 entries mutually assessed as "incorrect," 304 (82%) were corrected by both self- and travel-auditors and 250 of these (72%) received the same corrections. Reason for changing antiretroviral therapy (ART) regimen, ART end date, viral load value, CD4%, and HIV diagnosis date had the most mismatched corrections. CONCLUSIONS: With similar overall error rates, findings suggest that data audits conducted by trained local investigators could provide an alternative to on-site audits by external auditors to ensure continued data quality. However, discrepancies observed between corrections illustrate challenges in determining correct values even with audits.

The impact of data quality and source data verification on epidemiologic inference: a practical application using HIV observational data.

BACKGROUND: Data audits are often evaluated soon after completion, even though the identification of systematic issues may lead to additional data quality improvements in the future. In this study, we assess the impact of the entire data audit process on subsequent statistical analyses. METHODS: We conducted on-site audits of datasets from nine international HIV care sites. Error rates were quantified for key demographic and clinical variables among a subset of records randomly selected for auditing. Based on audit results, some sites were tasked with targeted validation of high-error-rate variables resulting in a post-audit dataset. We estimated the times from antiretroviral therapy initiation until death and first AIDS-defining event using the pre-audit data, the audit data, and the post-audit data. RESULTS: The overall discrepancy rate between pre-audit and audit data (n = 250) across all audited variables was 17.1%. The estimated probability of mortality and an AIDS-defining event over time was higher in the audited data relative to the pre-audit data. Among patients represented in both the post-audit and pre-audit cohorts (n = 18,999), AIDS and mortality estimates also were higher in the post-audit data. CONCLUSION: Though some changes may have occurred independently, our findings suggest that improved data quality following the audit may impact epidemiological inferences.