Confounding and regression adjustment in difference-in-differences studies.

OBJECTIVE: To define confounding bias in difference-in-difference studies and compare regression- and matching-based estimators designed to correct bias due to observed confounders. DATA SOURCES: We simulated data from linear models that incorporated different confounding relationships: time-invariant covariates with a time-varying effect on the outcome, time-varying covariates with a constant effect on the outcome, and time-varying covariates with a time-varying effect on the outcome. We considered a simple setting that is common in the applied literature: treatment is introduced at a single time point and there is no unobserved treatment effect heterogeneity. STUDY DESIGN: We compared the bias and root mean squared error of treatment effect estimates from six model specifications, including simple linear regression models and matching techniques. DATA COLLECTION: Simulation code is provided for replication. PRINCIPAL FINDINGS: Confounders in difference-in-differences are covariates that change differently over time in the treated and comparison group or have a time-varying effect on the outcome. When such a confounding variable is measured, appropriately adjusting for this confounder (ie, including the confounder in a regression model that is consistent with the causal model) can provide unbiased estimates with optimal SE. However, when a time-varying confounder is affected by treatment, recovering an unbiased causal effect using difference-in-differences is difficult. CONCLUSIONS: Confounding in difference-in-differences is more complicated than in cross-sectional settings, from which techniques and intuition to address observed confounding cannot be imported wholesale. Instead, analysts should begin by postulating a causal model that relates covariates, both time-varying and those with time-varying effects on the outcome, to treatment. This causal model will then guide the specification of an appropriate analytical model (eg, using regression or matching) that can produce unbiased treatment effect estimates. We emphasize the importance of thoughtful incorporation of covariates to address confounding bias in difference-in-difference studies.

Functional clustering methods for longitudinal data with application to electronic health records.

We develop a method to estimate subject-level trajectory functions from longitudinal data. The approach can be used for patient phenotyping, feature extraction, or, as in our motivating example, outcome identification, which refers to the process of identifying disease status through patient laboratory tests rather than through diagnosis codes or prescription information. We model the joint distribution of a continuous longitudinal outcome and baseline covariates using an enriched Dirichlet process prior. This joint model decomposes into (local) semiparametric linear mixed models for the outcome given the covariates and simple (local) marginals for the covariates. The nonparametric enriched Dirichlet process prior is placed on the regression and spline coefficients, the error variance, and the parameters governing the predictor space. This leads to clustering of patients based on their outcomes and covariates. We predict the outcome at unobserved time points for subjects with data at other time points as well as for new subjects with only baseline covariates. We find improved prediction over mixed models with Dirichlet process priors when there are a large number of covariates. Our method is demonstrated with electronic health records consisting of initiators of second-generation antipsychotic medications, which are known to increase the risk of diabetes. We use our model to predict laboratory values indicative of diabetes for each individual and assess incidence of suspected diabetes from the predicted dataset.

Streptococcus pneumoniae colonization in pneumococcal vaccine-naïve human immunodeficiency virus-exposed infected and -uninfected South African children.

Pneumococcal nasopharyngeal colonization is a pre-requisite for pneumococcal disease; the risk for pneumococcal disease is high in children born to women living with human immunodeficiency virus (HIV). We investigated pneumococcal colonization, serotype distribution and antibiotic susceptibility of Streptococcus pneumoniae isolates carried by perinatal HIV-infected and HIV-exposed-uninfected (HEU) children.Serial nasopharyngeal swabs were collected from 331 HIV-infected and 491 HEU children, at up to 6 scheduled timepoints, between median ages of 25 to 181 weeks. Pneumococcus was identified by culture; serotyping and antibiotic susceptibility testing were done by conventional methods. No pneumococcal vaccine was given.HIV-infected children were less likely to be colonized with 7-valent pneumococcal conjugate vaccine 7 serotypes than HEU at a median of 25 weeks of age (23% vs 36%; P < .001); however, no differences in colonization between the 2 groups were observed at subsequent study-visits. Over the 36-months study-period pneumococcal colonization increased in both HIV-infected (from 45% to 77%) and HEU (from 57% to 61%) children. Over the study-period, pneumococcal isolates non-susceptible to cotrimoxazole decreased from 92% to 57% and had a similar trend to penicillin (from 65% to 42%) in HIV-infected children. Similarly, pneumococcal nonsusceptible to cotrimoxazole decreased from 93% to 57% and to penicillin from 69% to 37% in HEU children.Vaccine serotype colonization was common in this population and similar rates were observed in HIV-infected and HEU children. The prevalence of pneumococcal isolates non-susceptible to cotrimoxazole and penicillin decreased with age.

A SEMIPARAMETRIC MODELING APPROACH USING BAYESIAN ADDITIVE REGRESSION TREES WITH AN APPLICATION TO EVALUATE HETEROGENEOUS TREATMENT EFFECTS.

Bayesian Additive Regression Trees (BART) is a flexible machine learning algorithm capable of capturing nonlinearities between an outcome and covariates and interactions among covariates. We extend BART to a semiparametric regression framework in which the conditional expectation of an outcome is a function of treatment, its effect modifiers, and confounders. The confounders are allowed to have unspecified functional form, while treatment and effect modifiers that are directly related to the research question are given a linear form. The result is a Bayesian semiparametric linear regression model where the posterior distribution of the parameters of the linear part can be interpreted as in parametric Bayesian regression. This is useful in situations where a subset of the variables are of substantive interest and the others are nuisance variables that we would like to control for. An example of this occurs in causal modeling with the structural mean model (SMM). Under certain causal assumptions, our method can be used as a Bayesian SMM. Our methods are demonstrated with simulation studies and an application to dataset involving adults with HIV/Hepatitis C coinfection who newly initiate antiretroviral therapy. The methods are available in an R package called semibart.

Bayesian nonparametric generative models for causal inference with missing at random covariates.

We propose a general Bayesian nonparametric (BNP) approach to causal inference in the point treatment setting. The joint distribution of the observed data (outcome, treatment, and confounders) is modeled using an enriched Dirichlet process. The combination of the observed data model and causal assumptions allows us to identify any type of causal effect-differences, ratios, or quantile effects, either marginally or for subpopulations of interest. The proposed BNP model is well-suited for causal inference problems, as it does not require parametric assumptions about the distribution of confounders and naturally leads to a computationally efficient Gibbs sampling algorithm. By flexibly modeling the joint distribution, we are also able to impute (via data augmentation) values for missing covariates within the algorithm under an assumption of ignorable missingness, obviating the need to create separate imputed data sets. This approach for imputing the missing covariates has the additional advantage of guaranteeing congeniality between the imputation model and the analysis model, and because we use a BNP approach, parametric models are avoided for imputation. The performance of the method is assessed using simulation studies. The method is applied to data from a cohort study of human immunodeficiency virus/hepatitis C virus co-infected patients.

Risk of liver decompensation with cumulative use of mitochondrial toxic nucleoside analogues in HIV/hepatitis C virus coinfection.

PURPOSE: Among patients dually infected with human immunodeficiency virus (HIV) and chronic hepatitis C virus (HCV), use of antiretroviral therapy (ART) containing mitochondrial toxic nucleoside reverse transcriptase inhibitors (mtNRTIs) might induce chronic hepatic injury, which could accelerate HCV-associated liver fibrosis and increase the risk of hepatic decompensation and death. METHODS: We conducted a cohort study among 1747 HIV/HCV patients initiating NRTI-containing ART within the Veterans Aging Cohort Study (2002-2009) to determine if cumulative mtNRTI use increased the risk of hepatic decompensation and death among HIV-/HCV-coinfected patients. Separate marginal structural models were used to estimate hazard ratios (HRs) of each outcome associated with cumulative exposure to ART regimens that contain mtNRTIs versus regimens that contain other NRTIs. RESULTS: Over 7033 person-years, we observed 97 (5.6%) decompensation events (incidence rate, 13.8/1000 person-years) and 125 (7.2%) deaths (incidence rate, 17.8 events/1000 person-years). The risk of hepatic decompensation increased with cumulative mtNRTI use (1-11 mo: HR, 1.79 [95% confidence interval (CI), 0.74-4.31]; 12-35 mo: HR, 1.39 [95% CI, 0.68-2.87]; 36-71 mo: HR, 2.27 [95% CI, 0.92-5.60]; >71 mo: HR, 4.66 [95% CI, 1.04-20.83]; P = .045) versus nonuse. Cumulative mtNRTI use also increased risk of death (1-11 mo: HR, 2.24 [95% CI, 1.04-4.81]; 12-35 mo: HR, 2.05 [95% CI, 0.68-6.20]; 36-71 mo: HR, 3.04 [95% CI, 1.12-8.26]; >71 mo: HR, 3.93 [95% CI, 0.75-20.50]; P = .030). CONCLUSIONS: These findings suggest that cumulative mtNRTI use may increase the risk of hepatic decompensation and death in HIV/HCV coinfection. These drugs should be avoided when alternatives exist for HIV/HCV patients.

Effect of Testosterone Treatment on Volumetric Bone Density and Strength in Older Men With Low Testosterone: A Controlled Clinical Trial.

Importance: As men age, they experience decreased serum testosterone concentrations, decreased bone mineral density (BMD), and increased risk of fracture. Objective: To determine whether testosterone treatment of older men with low testosterone increases volumetric BMD (vBMD) and estimated bone strength. Design, Setting, and Participants: Placebo-controlled, double-blind trial with treatment allocation by minimization at 9 US academic medical centers of men 65 years or older with 2 testosterone concentrations averaging less than 275 ng/L participating in the Testosterone Trials from December 2011 to June 2014. The analysis was a modified intent-to-treat comparison of treatment groups by multivariable linear regression adjusted for balancing factors as required by minimization. Interventions: Testosterone gel, adjusted to maintain the testosterone level within the normal range for young men, or placebo gel for 1 year. Main Outcomes and Measures: Spine and hip vBMD was determined by quantitative computed tomography at baseline and 12 months. Bone strength was estimated by finite element analysis of quantitative computed tomography data. Areal BMD was assessed by dual energy x-ray absorptiometry at baseline and 12 months. Results: There were 211 participants (mean [SD] age, 72.3 [5.9] years; 86% white; mean [SD] body mass index, 31.2 [3.4]). Testosterone treatment was associated with significantly greater increases than placebo in mean spine trabecular vBMD (7.5%; 95% CI, 4.8% to 10.3% vs 0.8%; 95% CI, -1.9% to 3.4%; treatment effect, 6.8%; 95% CI, 4.8%-8.7%; P < .001), spine peripheral vBMD, hip trabecular and peripheral vBMD, and mean estimated strength of spine trabecular bone (10.8%; 95% CI, 7.4% to 14.3% vs 2.4%; 95% CI, -1.0% to 5.7%; treatment effect, 8.5%; 95% CI, 6.0%-10.9%; P < .001), spine peripheral bone, and hip trabecular and peripheral bone. The estimated strength increases were greater in trabecular than peripheral bone and greater in the spine than hip. Testosterone treatment increased spine areal BMD but less than vBMD. Conclusions and Relevance: Testosterone treatment for 1 year of older men with low testosterone significantly increased vBMD and estimated bone strength, more in trabecular than peripheral bone and more in the spine than hip. A larger, longer trial could determine whether this treatment also reduces fracture risk. Trial Registration: clinicaltrials.gov Identifier: NCT00799617.

Association of Testosterone Levels With Anemia in Older Men: A Controlled Clinical Trial.

Importance: In one-third of older men with anemia, no recognized cause can be found. Objective: To determine if testosterone treatment of men 65 years or older with unequivocally low testosterone levels and unexplained anemia would increase their hemoglobin concentration. Design, Setting, and Participants: A double-blinded, placebo-controlled trial with treatment allocation by minimization using 788 men 65 years or older who have average testosterone levels of less than 275 ng/dL. Of 788 participants, 126 were anemic (hemoglobin ≤12.7 g/dL), 62 of whom had no known cause. The trial was conducted in 12 academic medical centers in the United States from June 2010 to June 2014. Interventions: Testosterone gel, the dose adjusted to maintain the testosterone levels normal for young men, or placebo gel for 12 months. Main Outcomes and Measures: The percent of men with unexplained anemia whose hemoglobin levels increased by 1.0 g/dL or more in response to testosterone compared with placebo. The statistical analysis was intent-to-treat by a logistic mixed effects model adjusted for balancing factors. Results: The men had a mean age of 74.8 years and body mass index (BMI) (calculated as weight in kilograms divided by height in meters squared) of 30.7; 84.9% were white. Testosterone treatment resulted in a greater percentage of men with unexplained anemia whose month 12 hemoglobin levels had increased by 1.0 g/dL or more over baseline (54%) than did placebo (15%) (adjusted OR, 31.5; 95% CI, 3.7-277.8; P = .002) and a greater percentage of men who at month 12 were no longer anemic (58.3%) compared with placebo (22.2%) (adjusted OR, 17.0; 95% CI, 2.8-104.0; P = .002). Testosterone treatment also resulted in a greater percentage of men with anemia of known cause whose month 12 hemoglobin levels had increased by 1.0 g/dL or more (52%) than did placebo (19%) (adjusted OR, 8.2; 95% CI, 2.1-31.9; P = .003). Testosterone treatment resulted in a hemoglobin concentration of more than 17.5 g/dL in 6 men who had not been anemic at baseline. Conclusions and Relevance: Among older men with low testosterone levels, testosterone treatment significantly increased the hemoglobin levels of those with unexplained anemia as well as those with anemia from known causes. These increases may be of clinical value, as suggested by the magnitude of the changes and the correction of anemia in most men, but the overall health benefits remain to be established. Measurement of testosterone levels might be considered in men 65 years or older who have unexplained anemia and symptoms of low testosterone levels. Trial Registration: clinicaltrials.gov Identifier: NCT00799617.

Testosterone Treatment and Sexual Function in Older Men With Low Testosterone Levels.

CONTEXT: The Testosterone Trials are a coordinated set of seven trials to determine the efficacy of T in symptomatic men ≥65 years old with unequivocally low T levels. Initial results of the Sexual Function Trial showed that T improved sexual activity, sexual desire, and erectile function. OBJECTIVE: To assess the responsiveness of specific sexual activities to T treatment; to relate hormone changes to changes in sexual function; and to determine predictive baseline characteristics and T threshold for sexual outcomes. DESIGN: A placebo-controlled trial. SETTING: Twelve academic medical centers in the United States. PARTICIPANTS: A total of 470 men ≥65 years of age with low libido, average T <275 ng/dL, and a partner willing to have sexual intercourse at least twice a month. METHODS: Men were assigned to take T gel or placebo for 1 year. Sexual function was assessed by three questionnaires every 3 months: the Psychosexual Daily Questionnaire, the Derogatis Interview for Sexual Function, and the International Index of Erectile Function. RESULTS: Compared with placebo, T administration significantly improved 10 of 12 measures of sexual activity. Incremental increases in total and free T and estradiol levels were associated with improvements in sexual activity and desire, but not erectile function. No threshold T level was observed for any outcome, and none of the 27 baseline characteristics predicted responsiveness to T. CONCLUSIONS: In older men with low libido and low T levels, improvements in sexual desire and activity in response to T treatment were related to the magnitude of increases in T and estradiol levels, but there was no clear evidence of a threshold effect.

Effects of Testosterone Treatment in Older Men.

BACKGROUND: Serum testosterone concentrations decrease as men age, but benefits of raising testosterone levels in older men have not been established. METHODS: We assigned 790 men 65 years of age or older with a serum testosterone concentration of less than 275 ng per deciliter and symptoms suggesting hypoandrogenism to receive either testosterone gel or placebo gel for 1 year. Each man participated in one or more of three trials--the Sexual Function Trial, the Physical Function Trial, and the Vitality Trial. The primary outcome of each of the individual trials was also evaluated in all participants. RESULTS: Testosterone treatment increased serum testosterone levels to the mid-normal range for men 19 to 40 years of age. The increase in testosterone levels was associated with significantly increased sexual activity, as assessed by the Psychosexual Daily Questionnaire (P<0.001), as well as significantly increased sexual desire and erectile function. The percentage of men who had an increase of at least 50 m in the 6-minute walking distance did not differ significantly between the two study groups in the Physical Function Trial but did differ significantly when men in all three trials were included (20.5% of men who received testosterone vs. 12.6% of men who received placebo, P=0.003). Testosterone had no significant benefit with respect to vitality, as assessed by the Functional Assessment of Chronic Illness Therapy-Fatigue scale, but men who received testosterone reported slightly better mood and lower severity of depressive symptoms than those who received placebo. The rates of adverse events were similar in the two groups. CONCLUSIONS: In symptomatic men 65 years of age or older, raising testosterone concentrations for 1 year from moderately low to the mid-normal range for men 19 to 40 years of age had a moderate benefit with respect to sexual function and some benefit with respect to mood and depressive symptoms but no benefit with respect to vitality or walking distance. The number of participants was too few to draw conclusions about the risks of testosterone treatment. (Funded by the National Institutes of Health and others; ClinicalTrials.gov number, NCT00799617.).

Low Vitamin-D Levels Combined with PKP3-SIGIRR-TMEM16J Host Variants Is Associated with Tuberculosis and Death in HIV-Infected and -Exposed Infants.

BACKGROUND: This study examined the associations of 25-hydroxyvitamin D and specific host genetic variants that affect vitamin D levels or its effects on immune function, with the risk of TB or mortality in children. METHODS: A case-cohort sample of 466 South African infants enrolled in P1041 trial (NCT00080119) underwent 25-hydroxyvitamin D testing by chemiluminescent immunoassay. Single nucleotide polymorphisms (SNPs) that alter the effect of vitamin D [e.g. vitamin D receptor (VDR)], vitamin D levels [e.g. vitamin D binding protein (VDBP)], or toll like receptor (TLR) expression (SIGIRR including adjacent genes PKP3 and TMEM16J) were identified by real-time PCR. Outcomes were time to TB, and to the composite of TB or death by 192 weeks of follow-up. Effect modification between vitamin D status and SNPs for outcomes was assessed. FINDINGS: Median age at 25-hydroxyvitamin D determination was 8 months; 11% were breastfed, 51% were HIV-infected and 26% had low 25-hydroxyvitamin D (<32ng/mL). By 192 weeks, 138 incident TB cases (43 definite/probable, and 95 possible) and 26 deaths occurred. Adjusting for HIV status and potential confounders, low 25-hydroxyvitamin D was associated with any TB (adjusted hazard ratio [aHR] 1.76, 95% CI 1.01-3.05; p = 0.046) and any TB or death (aHR 1.76, 95% CI 1.03-3.00; p = 0.038). Children with low 25-hydroxyvitamin D and TMEM 16J rs7111432-AA or PKP3 rs10902158-GG were at increased risk for probable/definite TB or death (aHR 8.12 and 4.83, p<0.05) and any TB or death (aHR 4.78 and 3.26, p<0.005) respectively; SNPs in VDBP, VDR, and vitamin D precursor or hydroxylation genes were not. There was significant interaction between low 25-hydroxyvitamin D and, TMEM 16J rs7111432-AA (p = 0.04) and PKP3 rs10902158-GG (p = 0.02) SNPs. CONCLUSIONS: Two novel SNPs, thought to be associated with innate immunity, in combination with low vitamin D levels were identified as increasing a young child's risk of developing TB disease or death. Identifying high-risk children and providing targeted interventions such as vitamin D supplementation may be beneficial. TRIAL REGISTRATION: ClinicalTrials.gov NCT00080119.

Distortion product otoacoustic emission data in perinatally HIV-infected and HIV-exposed but uninfected children and adolescents in the Pediatric HIV/AIDS Cohort Study.

The effect of perinatal HIV infection and exposure on sub-clinical auditory function can be measured with distortion product otoacoustic emissions (DPOAEs). DPOAEs were obtained at 4 frequency bins (1, 2, 3 and 4 kHz) and categorized by a signal-to-noise ratio. HIV infection was not associated with poorer DPOAEs. Among HIV-infected children, HIV viral load≥400 copies/mL had significantly lower odds of better DPOAEs.

Cardiac effects of in-utero exposure to antiretroviral therapy in HIV-uninfected children born to HIV-infected mothers.

OBJECTIVES: We evaluated the potential cardiac effects of in-utero exposures to antiretroviral drugs in HIV-exposed but uninfected (HEU) children. DESIGN AND METHODS: We compared echocardiographic parameters of left ventricular function (ejection fraction, fractional shortening, and stress-velocity index) and structure (left ventricular dimension, posterior wall/septal thickness, mass, thickness-to-dimension ratio, and wall stress) (expressed as Z-scores to account for age and body surface area) between HEU and HIV-unexposed cohorts from the Pediatric HIV/AIDS Cohort Study's Surveillance Monitoring for ART Toxicities study. Within the HEU group, we investigated the associations between the echocardiographic Z-scores and in-utero exposures to maternal antiretroviral drugs. RESULTS: There were no significant differences in echocardiographic Z-scores between 417 HEU and 98 HIV-unexposed children aged 2-7 years. Restricting the analysis to HEU children, first-trimester exposures to combination antiretroviral therapy (a regimen including at least three antiretroviral drugs) and to certain specific antiretroviral drugs were associated with significantly lower stress-velocity Z-scores (mean decreases of 0.22-0.40 SDs). Exposure to combination antiretroviral therapy was also associated with lower left ventricular dimension Z-scores (mean decrease of 0.44 SD). First-trimester exposure to combination antiretroviral therapy was associated with higher mean left ventricular posterior wall thickness and lower mean left ventricular wall stress Z-scores. CONCLUSION: There was no evidence of significant cardiac toxicity of perinatal combination antiretroviral therapy exposure in HEU children. Subclinical differences in left ventricular structure and function with specific in-utero antiretroviral exposures indicate the need for a longitudinal cardiac study in HEU children to assess long-term cardiac risk and cardiac monitoring recommendations.

Discordance of cognitive and academic achievement outcomes in youth with perinatal HIV exposure.

BACKGROUND: To evaluate achievement in youth with perinatally acquired HIV (PHIV) compared with HIV-exposed uninfected peers (HEU) and to examine differential effects of HIV on cognition-achievement concordance. METHODS: Cognition and achievement were assessed using standardized measures. Intelligence quotient-derived predicted achievement scores were subtracted from observed achievement scores to calculate discrepancy values. Linear regression models were used to compare achievement discrepancies between PHIV and HEU, adjusting for demographic covariates. PARTICIPANTS: 295 PHIV and 167 HEU youth; 71% black, 48% male, mean age 13.1 and 11.3 years, respectively. PHIV youth were relatively healthy (mean CD4%, 32%; viral load ≤400 copies/mL, 72%). PHIV and HEU youth had cognitive and achievement scores significantly below population norm means (P < 0.001), but did not differ in cognition (mean full scale IQ = 86.7 vs. 89.4, respectively). In unadjusted models, HEU outperformed PHIV youth on total achievement (mean = 89.2 vs. 86.0, P = 0.04) and numerical operations (mean = 88.8 vs. 82.9, P < 0.001); no differences remained after adjustment. Mean observed-predicted achievement discrepancies reflected "underachievement". History of encephalopathy predicted poorer achievement (P = 0.039) and greater underachievement, even after adjustment. PHIV showed greater underachievement than HEU for numerical operations (P < 0.001) and total achievement (P = 0.03), but these differences did not persist in adjusted models. CONCLUSIONS: Both PHIV and HEU youth demonstrated lower achievement than normative samples and underachieved relative to predicted achievement scores. Observed-predicted achievement discrepancies were associated with prior encephalopathy, older age and other non-HIV factors. PHIV youth with prior encephalopathy had significantly lower achievement and greater underachievement compared with PHIV without encephalopathy and HEU youth, even in adjusted models.

Biomarkers and neurodevelopment in perinatally HIV-infected or exposed youth: a structural equation model analysis.

OBJECTIVE: To examine the relationship between markers of vascular dysfunction and neurodevelopmental outcomes in perinatally HIV-infected (PHIV+) and perinatally HIV-exposed but uninfected (PHEU) youth. DESIGN: Cross-sectional design within a prospective, 15-site US-based cohort study. METHODS: Neurodevelopmental outcomes were evaluated in relation to nine selected vascular biomarkers in 342 youth (212 PHIV+, 130 PHEU). Serum levels were assessed for adiponectin, C-reactive protein (CRP), fibrinogen, interleukin-6 (IL-6), soluble vascular cell adhesion molecule-1 (sVCAM-1), E-selectin (sE-selectin), monocyte chemoattractant protein (sMCP-1), intercellular adhesion molecule-1 (sICAM-1), and P-selectin (sP-selectin). The Wechsler Intelligence Scale for Children-Fourth Edition (WISC-IV) was administered at entry, yielding a Full-Scale IQ score, and four index scores. Factor analysis was conducted to reduce the biomarkers to fewer factors with related biological roles. Structural equation models (SEMs) were used to measure associations between resulting factors and WISC-IV scores. RESULTS: Mean participant age was 11.4 years, 54% were female, 70% black. The nine biomarkers were clustered into three factor groups: F1 (fibrinogen, CRP, and IL-6); F2 (sICAM-1 and sVCAM-1); and F3 (MCP-1, sP-selectin, and sE-selectin). Adiponectin showed little correlation with any factor. SEMs revealed significant negative association of F1 with WISC-IV processing speed score in the total cohort. This effect remained significant after adjusting for HIV status and other potential confounders. A similar association was observed when restricted to PHIV+ participants in both unadjusted and adjusted SEMs. CONCLUSION: Aggregate measures of fibrinogen, CRP, and IL-6 may serve as a latent biomarker associated with relatively decreased processing speed in both PHIV+ and PHEU youth.

Evaluation of risk for late language emergence after in utero antiretroviral drug exposure in HIV-exposed uninfected infants.

BACKGROUND: Combination antiretroviral (cARV) regimens are recommended for pregnant women with HIV to prevent perinatal HIV transmission. Safety is a concern for infants who were HIV-exposed but uninfected, particularly for neurodevelopmental problems, such as language delays. METHODS: We studied late language emergence (LLE) in HIV-exposed but uninfected children enrolled in a US-based prospective cohort study. LLE was defined as a caregiver-reported score ≤10th percentile in any of 4 domains of the MacArthur-Bates Communicative Development Inventory for 1-year olds and as ≥1 standard deviation below age-specific norms for the Ages and Stages Questionnaire for 2-year olds. Logistic regression models were used to evaluate associations of in utero cARV exposure with LLE, adjusting for infant, maternal and environmental characteristics. RESULTS: 1129 language assessments were conducted among 792 1- and 2-year-old children (50% male, 62% black and 37% Hispanic). Overall, 86% had in utero exposure to cARV and 83% to protease inhibitors. LLE was identified in 26% of 1-year olds and 23% of 2-year olds, with higher rates among boys. In adjusted models, LLE was not associated with maternal cARV or ARV drug classes in either age group. Among cARV-exposed 1-year olds, increased odds of LLE was observed for those exposed to atazanavir (adjusted odds ratio = 1.83, 95% confidence interval: 1.10-3.04), particularly after the first trimester (adjusted odds ratio = 3.56, P = 0.001), compared with atazanavir-unexposed infants. No associations of individual ARV drugs with LLE were observed among 2-year olds. CONCLUSIONS: In utero cARV exposure showed little association with LLE, except for a higher risk of language delay observed in 1-year-old infants with atazanavir exposure.

Cardiac biomarkers in HIV-exposed uninfected children.

OBJECTIVES: To evaluate associations of cardiac biomarkers with in-utero antiretroviral drug exposures and cardiac function/structure measured by echocardiograms in HIV-exposed but uninfected (HEU) children. DESIGN AND METHODS: We analyzed the association of three cardiac biomarkers (cardiac troponin T, cTnT; high sensitivity C-reactive protein, hsCRP; and N-terminal pro-brain natriuretic peptide, NT-proBNP) with prenatal antiretroviral drug exposures, maternal-child characteristics, and echocardiographic parameters. RESULTS: Among 338 HEU children (mean age 4.3 years), 51% had at least one elevated cardiac biomarker. Maternal tobacco use was associated with elevated NT-proBNP [adjusted odds ratio (aOR) 2.28, P=0.02]. Maternal alcohol and abacavir use were associated with elevated cTnT levels (aOR 3.56, P=0.01 and aOR 2.33, P=0.04, respectively). Among 94 children with paired echocardiogram-biomarker measurements, cTnT measurements were correlated with increased left-ventricular thickness-to-dimension ratio (r=0.21, P=0.04); and elevated cTnT was associated with higher mean left-ventricular end-diastolic (LVED) posterior wall thickness (P=0.04). hsCRP measurements were negatively correlated with septal thickness (r=-0.22, P=0.03) and elevated hsCRP was associated with lower mean left-ventricular contractility Z-scores (P=0.04). NT-proBNP measurements were correlated with increased LVED dimension (r=0.20, P=0.05) and elevated NT-proBNP was associated with lower mean end-systolic septal thickness (P=0.03). CONCLUSION: Our findings suggest that cardiac biomarkers may help identify HEU children who require further cardiac evaluation including echocardiography. Potential cardiac effects of prenatal abacavir exposure in this population need further investigation.

Hearing loss in perinatally HIV-infected and HIV-exposed but uninfected children and adolescents.

BACKGROUND: Little is known about hearing loss in children with HIV infection (HIV+). We examined the prevalence of hearing loss in perinatally HIV+ and HIV-exposed but uninfected (HEU) children, compared these with the percentage with hearing loss in the general population and evaluated possible risk factors for hearing loss in HIV+ and HEU children. METHODS: Audiometric examinations were completed in children who met any prespecified criteria for possible hearing loss. The hearing examination consisted of a tympanogram in each ear and pure-tone air-conduction threshold testing from 500 through 4000 Hz. Hearing loss was defined as the pure-tone average over these frequencies ≥ 20 dB hearing level. The associations of demographic variables, parent/caregiver, HIV disease and HIV treatment with hearing loss were evaluated with univariate and multivariable logistic regression models. RESULTS: Hearing testing was completed in 231 children (145 HIV+ and 86 HEU). Hearing loss occurred in 20.0% of HIV+ children and 10.5% of HEU children. After adjusting for caregiver education level, HIV infection was associated with increased odds of hearing loss (adjusted odds ratio = 2.13, 95% confidence interval: 0.95-4.76, P = 0.07). Among HIV+ children, those with a Centers for Disease Control and Prevention class C diagnosis had over twice the odds of hearing loss (adjusted odds ratio = 2.47, 95% confidence interval: 1.04-5.87, P = 0.04). The prevalence of hearing loss was higher in both HIV+ and HEU children compared with National Health and Nutrition Examination Survey III children. CONCLUSIONS: Hearing loss was more common in both HIV+ and HEU children than in children from a US population sample. More advanced HIV illness increased the risk of hearing loss in HIV+ children.

Language impairment in children perinatally infected with HIV compared to children who were HIV-exposed and uninfected.

OBJECTIVE: To investigate the risk for language impairment (LI) in children perinatally infected or exposed to HIV. METHODS: We evaluated the prevalence of LI in 7- to 16-year-old children with perinatal HIV infection (HIV+) compared with HIV-exposed and uninfected children, using a comprehensive standardized language test (Clinical Evaluation of Language Functioning-Fourth Edition [CELF-4]). LI was classified as primary LI (Pri-LI) (monolingual English exposure and no cognitive or hearing impairment), concurrent LI (Con-LI) (cognitive or hearing impairment), or no LI. Associations of demographic, caregiver, HIV disease, and antiretroviral treatment factors with LI category were evaluated using univariate and multivariable logistic regression models. RESULTS: Of the 468 children with language assessments, 184 (39%) had LI. No difference was observed by HIV infection status for overall LI or for Pri-LI or Con-LI; mean (SD) CELF-4 scores were 88.5 (18.4) for HIV+ versus 87.5 (17.9) for HIV-exposed and uninfected children. After adjustment, black children had higher odds of Pri-LI versus no LI (adjusted odds ratio [aOR] = 2.43, p = .03). Children who were black, Hispanic, had a caregiver with low education or low intelligence quotient, or a nonbiological parent as caregiver had higher odds of Con-LI versus no LI. Among HIV+ children, viral load >400 copies/mL (aOR = 3.04, p < .001), Centers for Disease Control and Prevention Class C (aOR = 2.19, p = .02), and antiretroviral treatment initiation <6 months of age (aOR = 2.12, p = .02) were associated with higher odds of Con-LI versus no LI. CONCLUSIONS: Children perinatally exposed to HIV are at high risk for LI, but such risk was not increased for youth with HIV. Risk factors differed for Pri-LI and Con-LI.