Search tips
Search criteria

Results 1-25 (70)

Clipboard (0)

Select a Filter Below

more »
Year of Publication
more »
1.  Worth the Weight: Using Inverse Probability Weighted Cox Models in AIDS Research 
AIDS Research and Human Retroviruses  2014;30(12):1170-1177.
In an observational study with a time-to-event outcome, the standard analytical approach is the Cox proportional hazards regression model. As an alternative to the standard Cox model, in this article we present a method that uses inverse probability (IP) weights to estimate the effect of a baseline exposure on a time-to-event outcome. IP weighting can be used to adjust for multiple measured confounders of a baseline exposure in order to estimate marginal effects, which compare the distribution of outcomes when the entire population is exposed versus when the entire population is unexposed. For example, IP-weighted Cox models allow for estimation of the marginal hazard ratio and marginal survival curves. IP weights can also be employed to adjust for selection bias due to loss to follow-up. This approach is illustrated using an example that estimates the effect of injection drug use on time until AIDS or death among HIV-infected women.
PMCID: PMC4250953  PMID: 25183195
2.  A Flexible, Computationally Efficient Method for Fitting the Proportional Hazards Model to Interval-Censored Data 
Biometrics  2015;72(1):222-231.
The proportional hazards model (PH) is currently the most popular regression model for analyzing time-to-event data. Despite its popularity, the analysis of interval-censored data under the PH model can be challenging using many available techniques. This paper presents a new method for analyzing interval-censored data under the PH model. The proposed approach uses a monotone spline representation to approximate the unknown nondecreasing cumulative baseline hazard function. Formulating the PH model in this fashion results in a finite number of parameters to estimate while maintaining substantial modeling flexibility. A novel expectation-maximization (EM) algorithm is developed for finding the maximum likelihood estimates of the parameters. The derivation of the EM algorithm relies on a two-stage data augmentation involving latent Poisson random variables. The resulting algorithm is easy to implement, robust to initialization, enjoys quick convergence, and provides closed-form variance estimates. The performance of the proposed regression methodology is evaluated through a simulation study, and is further illustrated using data from a large population-based randomized trial designed and sponsored by the United States National Cancer Institute.
PMCID: PMC4803641  PMID: 26393917
EM algorithm; interval-censored data; latent Poisson random variables; monotone splines; proportional hazards model
3.  Reducing lost to follow-up in a large clinical trial of prevention of mother-to-child transmission of HIV: The Breastfeeding, Antiretrovirals and Nutrition (BAN) study experience 
Clinical trials (London, England)  2014;12(2):156-165.
Retaining patients in prevention of mother-to-child transmission of HIV studies can be challenging in resource limited settings, where high lost to follow-up (LTFU) rates have been reported. In this paper, we describe the effectiveness of methods used to encourage retention in the Breastfeeding, Antiretrovirals, and Nutrition (BAN) study and analyze factors associated with LTFU in the study.
The BAN clinical trial was designed to evaluate the efficacy of 3 different mother-to-child HIV transmission prevention strategies. Lower than expected participant retention prompted enhanced efforts to reduce LTFU during the conduct of the trial. Following study completion, we employed regression modeling to determine predictors of perfect attendance and variables associated with being LTFU.
During the study, intensive tracing efforts were initiated after the first 1686 mother-infant pairs had been enrolled, and 327 pairs were missing. Sixty of these pairs were located and had complete data obtained. Among the 683 participants enrolling after initiation of intensive tracing efforts, the LTFU rate was 3.4%. At study's end, 290 (12.2%) of the 2369 mother-infant pairs were LTFU. Among successfully traced missing pairs, relocation was common and three were deceased. Log-binomial regression modeling revealed higher maternal hemoglobin and older maternal age to be significant predictors of perfect attendance. These factors and the presence of food insecurity were also significantly associated with lower rates of LTFU.
In this large HIV prevention trial, intensive tracing efforts centered on reaching study participants at their homes succeeded in finding a substantial proportion of LTFU participants, and were very effective in preventing further LTFU during the remainder of the trial. The association between food insecurity and lower rates of LTFU is likely related to the study's provision of nutritional support, including a family maize supplement, which may have contributed to patient retention.
PMCID: PMC4355163  PMID: 25518956
Antiretroviral therapy; human immunodeficiency virus; lost to follow up; prevention of mother to child transmission; retention; tracing
4.  Randomization inference for treatment effects on a binary outcome 
Statistics in medicine  2014;34(6):924-935.
Two methods are developed for constructing randomization based confidence sets for the average effect of a treatment on a binary outcome. The methods are nonparametric and require no assumptions about random sampling from a larger population. Both of the resulting 1 − α confidence sets are exact in the sense that the probability of containing the true treatment effect is at least 1 − α. Both types of confidence sets are also guaranteed to have width no greater than one. In contrast, a previously proposed asymptotic confidence interval is not exact and may have width greater than one. The first approach combines Bonferroni adjusted prediction sets for the attributable effects in the treated and untreated. The second method entails inverting a permutation test. Simulations are presented comparing the two randomization based confidence sets with the asymptotic interval as well as the standard Wald confidence interval and a commonly used exact interval for the difference in binomial proportions. Results show for small to moderate sample sizes that the permutation confidence set attains the narrowest width on average among the methods that maintain nominal coverage. Extensions that allow for stratifying on categorical baseline covariates are also discussed.
PMCID: PMC4459717  PMID: 25471299
Additivity; Attributable Effects; Causal Inference; Exact Confidence Interval; Randomization Inference; Stratified Data
5.  Acquisition and Persistence of Human Papillomavirus 16 (HPV-16) and HPV-18 Among Men With High-HPV Viral Load Infections in a Circumcision Trial in Kisumu, Kenya 
The Journal of Infectious Diseases  2014;211(5):811-820.
Background. Circumcision and lower human papillomavirus (HPV) viral loads in men are possibly associated with a reduced risk of HPV transmission to women. However, the association between male circumcision and HPV viral load remains unclear.
Methods. Swab specimens from the glans and shaft of the penis were collected from men enrolled in a circumcision trial in Kisumu, Kenya. GP5+/6+ polymerase chain reaction (PCR) was used to identify HPV DNA types. HPV-16 and HPV-18 loads were measured with a LightCycler real-time PCR and classified as high (>250 copies/scrape) or low (≤250 copies/scrape).
Results. A total of 1159 men were randomly assigned to undergo immediate circumcision, and 1140 men were randomly assigned to the control arm (these individuals were asked to remain uncircumcised until the study ended). The hazard of acquisition of high-viral load infections in the glans was lower in the circumcision arm, compared with the control arm, for HPV-16 (hazard ratio [HR], 0.32 [95% confidence interval {CI}, .20–.49]) and HPV-18 (HR, 0.34 [95% CI, .21–.54]). The 6-month risk of HPV persistence among men with high-viral load infections in the glans at baseline was lower in the circumcision arm, compared with the control arm, for HPV-16 (risk ratio [RR], 0.36 [95% CI, .18–.72]) and HPV-18 (RR 0.34 [95% CI, .13–.86]). Weaker and less precise results were obtained for shaft samples.
Conclusions. Male circumcision could potentially reduce the risk of HPV transmission to women by reducing the hazard of acquisition, and the risk of persistence of high-HPV viral load infections in the glans in men.
PMCID: PMC4402376  PMID: 25261492
male circumcision; human papillomavirus (HPV); viral load; HPV-16; HPV-18; men; randomized controlled trial (RCT), incidence; persistence; Kenya
6.  Risk 
American Journal of Epidemiology  2015;181(4):246-250.
The epidemiologist primarily studies transitions between states of health and disease. The purpose of the present article is to define a foundational parameter for such studies, namely risk. We begin simply and build to the setting in which there is more than 1 event type (i.e., competing risks or competing events), as well as more than 1 treatment or exposure level of interest. In the presence of competing events, the risks are a set of counterfactual cumulative incidence functions for each treatment. These risks can be depicted visually and summarized numerically. We use an example from the study of human immunodeficiency virus to illustrate concepts.
PMCID: PMC4325680  PMID: 25660080
causal inference; cohort study; semi-Bayes method; semiparametric inference; survival analysis
7.  Effect of Directly Observed Antiretroviral Therapy Compared to Self-Administered Antiretroviral Therapy on Adherence and Virological Outcomes among HIV-Infected Prisoners: A Randomized Controlled Pilot Study 
AIDS and behavior  2015;19(1):128-136.
The effect of directly observed therapy (DOT) versus self-administered therapy (SAT) on antiretroviral (ART) adherence and virological outcomes in prison has never been assessed in a randomized, controlled trial. Prisoners were randomized to receive ART by DOT or SAT. The primary outcome was medication adherence [percent of ART doses measured by the medication event monitoring system (MEMS) and pill counts] at the end of 24 weeks. The changes in the plasma viral loads from baseline and proportion of participants virological suppressed (<400 copies/mL) at the end of 24 weeks were assessed. Sixty-six percent (90/136) of eligible prisoners declined participation. Participants in the DOT arm (n = 20) had higher viral loads than participants in the SAT (n = 23) arm (p = 0.23). Participants, with complete data at 24 weeks, were analyzed as randomized. There were no significant differences in median ART adherence between the DOT (n = 16, 99% MEMS [IQR 93.9, 100], 97.1 % pill count [IQR 95.1, 99.3]) and SAT (n = 21, 98.3 % MEMS [IQR 96.0, 100], 98.5 % pill count [95.8, 100]) arms (p = 0.82 MEMS, p = 0.40 Pill Count) at 24 weeks. Participants in the DOT arm had a greater reduction in viral load of approximately −1 log 10 copies/mL [IQR −1.75, −0.05] compared to −0.05 [IQR −0.45, 0.51] in the SAT arm (p value = 0.02) at 24 weeks. The proportion of participants achieving virological suppression in the DOT vs SAT arms was not statistically different at 24 weeks (53 % vs 32 %, p = 0.21). These findings suggest that DOT ART programs in prison settings may not offer any additional benefit on adherence than SAT programs.
PMCID: PMC4303492  PMID: 25055766
Directly observed therapy; Antiretroviral therapy; Adherence; Prisoners; HIV/AIDS; Protease inhibitors; Clinical trial
8.  Rotavirus-Specific IgG Antibodies from Mothers’ Serum may Inhibit Infant Immune Responses to the Pentavalent Rotavirus Vaccine 
PMCID: PMC4352718  PMID: 25741808
Rotavirus vaccines; Rotavirus infections; Antibodies; Mothers; Nicaragua
9.  Adherence to extended postpartum antiretrovirals is associated with decreased breastmilk HIV-1 transmission: Results of the BAN study 
AIDS (London, England)  2014;28(18):2739-2749.
Estimate association between postpartum antiretroviral adherence and breastmilk HIV-1 transmission
Prospective cohort study
Mother-infant pairs were randomized after delivery to immediately begin receiving 28 weeks of either triple maternal antiretrovirals (zidovudine, lamivudine, and either nevirapine, nelfinavir, or lopinavir-ritonavir) or daily infant nevirapine as part of the Breastfeeding, Antiretrovirals, and Nutrition study. Associations between postpartum antiretroviral adherence and rate of breastmilk HIV-1 transmission were estimated using Cox models. We measured adherence over four postpartum time intervals using pill count, suspension bottle weight, and maternal self-report. Adherence was categorized and lagged by one interval. Missing adherence measures were multiply imputed. Infant HIV-1 infection was determined by DNA PCR every 2-6 weeks. The primary endpoint was infant HIV-1 infection by 38 weeks of age among infants alive and uninfected at 5 weeks.
Analyses included 1479 mother-infant pairs and 45 transmission events. Using pill count and bottle weight information, 22-40% of mother-infant pairs at any given interval were <90% adherent. Having ≥90% adherence was associated with a 52% (95% CI 3-76%) relative reduction in the rate of breastmilk HIV-1 transmission, compared with having <90% adherence when controlling for study arm, breastfeeding status, and maternal characteristics. Complete case analysis rendered similar results (n=501; relative reduction 59%, 95% CI 6-82%).
Non-adherence to extended postpartum ART regimens in ‘real world’ settings is likely to be higher than that seen in BAN. Identifying mothers with difficulty adhering to antiretrovirals, and developing effective adherence interventions, will help maximize benefits of ARV provision throughout breastfeeding.
PMCID: PMC4397579  PMID: 25493600
adherence; antiretroviral; breastfeeding; HIV transmission; PMTCT
10.  Etiology of Childhood Diarrhea Following Rotavirus Vaccine Introduction: A Prospective, Population-Based Study in Nicaragua 
Nicaragua was the first developing nation to implement routine immunization with the pentavalent rotavirus vaccine (RV5). In this RV5-immunized population, understanding infectious etiologies of childhood diarrhea is necessary to direct diarrhea treatment and prevention efforts.
We followed a population-based sample of children less than 5 years in León, Nicaragua for diarrhea episodes through household visits. Information was obtained on RV5 history and sociodemographics. Stool samples collected during diarrhea episodes and among healthy children underwent laboratory analysis for viral, bacterial, and parasitic enteropathogens. Detection frequency and incidence of each enteropathogen was calculated.
The 826 children in the cohort experienced 677 diarrhea episodes during 607.5 child-years of exposure time (1.1 episodes per child-year). At least one enteropathogen was detected among 61.1% of the 337 diarrheal stools collected. The most common enteropathogens among diarrheal stools were: norovirus (20.4%), sapovirus (16.6%), enteropathogenic Escherichia coli (EPEC, 11.3%), Entamoeba histolytica/dispar (8.3%), Giardia lamblia (8.0%), and enterotoxigenic E.coli (ETEC, 7.7%), with rotavirus detected among 5.3% of diarrheal stools. EPEC and ETEC were frequently detected among stools from healthy children. Among children with diarrhea, norovirus was more commonly detected among younger children (< 2 years) and G. lamblia was more commonly detected among older children (2-4 years). The mean age of rotavirus detection was 34.6 months.
In this Central American community following RV5 introduction, rotavirus was not commonly detected among children with diarrhea. Prevention and appropriate management of norovirus and sapovirus should be considered to further reduce the burden of diarrheal disease.
PMCID: PMC4216626  PMID: 24879131
Childhood; Community; Diarrhea; Nicaragua; Rotavirus vaccine
11.  Nonparametric Bounds and Sensitivity Analysis of Treatment Effects 
This paper considers conducting inference about the effect of a treatment (or exposure) on an outcome of interest. In the ideal setting where treatment is assigned randomly, under certain assumptions the treatment effect is identifiable from the observable data and inference is straightforward. However, in other settings such as observational studies or randomized trials with noncompliance, the treatment effect is no longer identifiable without relying on untestable assumptions. Nonetheless, the observable data often do provide some information about the effect of treatment, that is, the parameter of interest is partially identifiable. Two approaches are often employed in this setting: (i) bounds are derived for the treatment effect under minimal assumptions, or (ii) additional untestable assumptions are invoked that render the treatment effect identifiable and then sensitivity analysis is conducted to assess how inference about the treatment effect changes as the untestable assumptions are varied. Approaches (i) and (ii) are considered in various settings, including assessing principal strata effects, direct and indirect effects and effects of time-varying exposures. Methods for drawing formal inference about partially identified parameters are also discussed.
PMCID: PMC4317325  PMID: 25663743
Causal inference; nonparametric bounds; partially identifiable models; sensitivity analysis
12.  Large sample randomization inference of causal effects in the presence of interference 
Recently, increasing attention has focused on making causal inference when interference is possible. In the presence of interference, treatment may have several types of effects. In this paper, we consider inference about such effects when the population consists of groups of individuals where interference is possible within groups but not between groups. A two stage randomization design is assumed where in the first stage groups are randomized to different treatment allocation strategies and in the second stage individuals are randomized to treatment or control conditional on the strategy assigned to their group in the first stage. For this design, the asymptotic distributions of estimators of the causal effects are derived when either the number of individuals per group or the number of groups grows large. Under certain homogeneity assumptions, the asymptotic distributions provide justification for Wald-type confidence intervals (CIs) and tests. Empirical results demonstrate the Wald CIs have good coverage in finite samples and are narrower than CIs based on either the Chebyshev or Hoeffding inequalities provided the number of groups is not too small. The methods are illustrated by two examples which consider the effects of cholera vaccination and an intervention to encourage voting.
PMCID: PMC3960089  PMID: 24659836
causal inference; confidence interval; interference; Normal mixture; randomization
13.  Assessing Effects of Cholera Vaccination in the Presence of Interference 
Biometrics  2014;70(3):731-741.
Interference occurs when the treatment of one person affects the outcome of another. For example, in infectious diseases, whether one individual is vaccinated may affect whether another individual becomes infected or develops disease. Quantifying such indirect (or spillover) effects of vaccination could have important public health or policy implications. In this paper we use recently developed inverse-probability weighted (IPW) estimators of treatment effects in the presence of interference to analyze an individually-randomized, placebo-controlled trial of cholera vaccination that targeted 121,982 individuals in Matlab, Bangladesh. Because these IPW estimators have not been employed previously, a simulation study was also conducted to assess the empirical behavior of the estimators in settings similar to the cholera vaccine trial. Simulation study results demonstrate the IPW estimators can yield unbiased estimates of the direct, indirect, total and overall effects of vaccination when there is interference provided the untestable no unmeasured confounders assumption holds and the group-level propensity score model is correctly specified. Application of the IPW estimators to the cholera vaccine trial indicates the presence of interference. For example, the IPW estimates suggest on average 5.29 fewer cases of cholera per 1000 person-years (95% confidence interval 2.61, 7.96) will occur among unvaccinated individuals within neighborhoods with 60% vaccine coverage compared to neighborhoods with 32% coverage. Our analysis also demonstrates how not accounting for interference can render misleading conclusions about the public health utility of vaccination.
PMCID: PMC4239215  PMID: 24845800
Causal inference; Interference; Inverse-probability weighted estimators; Spillover effect; Two-stage randomization; Vaccine
15.  Pooled Individual Data Analysis of 5 Randomized Trials of Infant Nevirapine Prophylaxis to Prevent Breast-Milk HIV-1 Transmission 
A pooled analysis of individual data from >5000 human immunodeficiency virus type 1 (HIV-1)–infected mothers and their infants from Africa and India who participated in 5 randomized trials shows that extended prophylaxis with nevirapine or with nevirapine and zidovudine significantly reduces postnatal HIV-1 infection.
Background. In resource-limited settings, mothers infected with human immunodeficiency virus type 1 (HIV-1) face a difficult choice: breastfeed their infants but risk transmitting HIV-1 or not breastfeed their infants and risk the infants dying of other infectious diseases or malnutrition. Recent results from observational studies and randomized clinical trials indicate daily administration of nevirapine to the infant can prevent breast-milk HIV-1 transmission.
Methods. Data from 5396 mother-infant pairs who participated in 5 randomized trials where the infant was HIV-1 negative at birth were pooled to estimate the efficacy of infant nevirapine prophylaxis to prevent breast-milk HIV-1 transmission. Four daily regimens were compared: nevirapine for 6 weeks, 14 weeks, or 28 weeks, or nevirapine plus zidovudine for 14 weeks.
Results. The estimated 28-week risk of HIV-1 transmission was 5.8% (95% confidence interval [CI], 4.3%–7.9%) for the 6-week nevirapine regimen, 3.7% (95% CI, 2.5%–5.4%) for the 14-week nevirapine regimen, 4.8% (95% CI, 3.5%–6.7%) for the 14-week nevirapine plus zidovudine regimen, and 1.8% (95% CI, 1.0%–3.1%) for the 28-week nevirapine regimen (log-rank test for trend, P < .001). Cox regression models with nevirapine as a time-varying covariate, stratified by trial site and adjusted for maternal CD4 cell count and infant birth weight, indicated that nevirapine reduces the rate of HIV-1 infection by 71% (95% CI, 58%–80%; P < .001) and reduces the rate of HIV infection or death by 58% (95% CI, 45%–69%; P < .001).
Conclusions. Extended prophylaxis with nevirapine or with nevirapine and zidovudine significantly reduces postnatal HIV-1 infection. Longer duration of prophylaxis results in a greater reduction in the risk of infection.
PMCID: PMC3518881  PMID: 22997212
breast milk; HIV; nevirapine
16.  The interplay between immune maturation, age, chronic viral infection and environment 
The worldwide increase in life expectancy has been associated with an increase in age-related morbidities. The underlying mechanisms resulting in immunosenescence are only incompletely understood. Chronic viral infections, in particular infection with human cytomegalovirus (HCMV), have been suggested as a main driver in immunosenescence. Here, we propose that rhesus macaques could serve as a relevant model to define the impact of chronic viral infections on host immunity in the aging host. We evaluated whether chronic rhesus CMV (RhCMV) infection, similar to HCMV infection in humans, would modulate normal immunological changes in the aging individual by taking advantage of the unique resource of rhesus macaques that were bred and raised to be Specific Pathogen Free (SPF-2) for distinct viruses.
Our results demonstrate that normal age-related immunological changes in frequencies, activation, maturation, and function of peripheral blood cell lymphocytes in humans occur in a similar manner over the lifespan of rhesus macaques. The comparative analysis of age-matched SPF-2 and non-SPF macaques that were housed under identical conditions revealed distinct differences in certain immune parameters suggesting that chronic pathogen exposure modulated host immune responses. All non-SPF macaques were infected with RhCMV, suggesting that chronic RhCMV infection was a major contributor to altered immune function in non-SPF macaques, although a causative relationship was not established and outside the scope of these studies. Further, we showed that immunological differences between SPF-2 and non-SPF macaques were already apparent in adolescent macaques, potentially predisposing RhCMV-infected animals to age-related pathologies.
Our data validate rhesus macaques as a relevant animal model to study how chronic viral infections modulate host immunity and impact immunosenescence. Comparative studies in SPF-2 and non-SPF macaques could identify important mechanisms associated with inflammaging and thereby lead to new therapies promoting healthy aging in humans.
Electronic supplementary material
The online version of this article (doi:10.1186/s12979-015-0030-3) contains supplementary material, which is available to authorized users.
PMCID: PMC4436863  PMID: 25991918
Aging; Immune maturation and function; Rhesus CMV infection; Rhesus macaques; Immune development and maturation; RhCMV infection; Aging; Inflammaging
18.  Hepatitis C Virus (HCV) NS3 Sequence Diversity and Antiviral Resistance-Associated Variant Frequency in HCV/HIV Coinfection 
Antimicrobial Agents and Chemotherapy  2014;58(10):6079-6092.
HIV coinfection accelerates disease progression in chronic hepatitis C and reduces sustained antiviral responses (SVR) to interferon-based therapy. New direct-acting antivirals (DAAs) promise higher SVR rates, but the selection of preexisting resistance-associated variants (RAVs) may lead to virologic breakthrough or relapse. Thus, pretreatment frequencies of RAVs are likely determinants of treatment outcome but typically are below levels at which the viral sequence can be accurately resolved. Moreover, it is not known how HIV coinfection influences RAV frequency. We adopted an accurate high-throughput sequencing strategy to compare nucleotide diversity in HCV NS3 protease-coding sequences in 20 monoinfected and 20 coinfected subjects with well-controlled HIV infection. Differences in mean pairwise nucleotide diversity (π), Tajima's D statistic, and Shannon entropy index suggested that the genetic diversity of HCV is reduced in coinfection. Among coinfected subjects, diversity correlated positively with increases in CD4+ T cells on antiretroviral therapy, suggesting T cell responses are important determinants of diversity. At a median sequencing depth of 0.084%, preexisting RAVs were readily identified. Q80K, which negatively impacts clinical responses to simeprevir, was encoded by more than 99% of viral RNAs in 17 of the 40 subjects. RAVs other than Q80K were identified in 39 of 40 subjects, mostly at frequencies near 0.1%. RAV frequency did not differ significantly between monoinfected and coinfected subjects. We conclude that HCV genetic diversity is reduced in patients with well-controlled HIV infection, likely reflecting impaired T cell immunity. However, RAV frequency is not increased and should not adversely influence the outcome of DAA therapy.
PMCID: PMC4187920  PMID: 25092699
19.  Risk of HIV acquisition among circumcised and uncircumcised young men with penile human papillomavirus infection 
AIDS (London, England)  2014;28(5):745-752.
There are very few data from men on the risk of HIV acquisition associated with penile human papillomavirus (HPV) infection and no data on the potential modifying effect of male circumcision. Therefore, this study evaluated whether HPV is independently associated with risk of HIV.
A cohort study of HPV natural history nested within a randomized control trial of male circumcision to reduce HIV incidence in Kisumu, Kenya.
Prospective data from 2519 men were analyzed using 6-month discrete-time Cox models to determine if HIV acquisition was higher among circumcised or uncircumcised men with HPV compared to HPV-uninfected men.
Risk of HIV acquisition was nonsignificantly increased among men with any HPV [adjusted hazard ratio (aHR) 1.72; 95% confidence interval (CI) 0.94–3.15] and high-risk HPV (aHR 1.92; 95% CI 0.96–3.87) compared to HPV-uninfected men, and estimates did not differ by circumcision status. Risk of HIV increased 27% with each additional HPV genotype infection (aHR 1.27; 95% CI 1.09–1.48). Men with persistent (aHR 3.27; 95% CI 1.59–6.72) or recently cleared (aHR 3.05; 95% CI 1.34–6.97) HPV had a higher risk of HIV acquisition than HPV-uninfected men.
Consistent with the findings in women, HPV infection, clearance, and persistence were associated with an increased risk of HIV acquisition in men. Given the high prevalence of HPV in populations at risk of HIV, consideration of HPV in future HIV-prevention studies and investigation into mechanisms through which HPV might facilitate HIV acquisition are needed.
PMCID: PMC4074250  PMID: 24149088
cervical cancer; co-infection; HIV; human papillomavirus; Kenya; male circumcision; vaccination
20.  Parametric likelihood inference for interval censored competing risks data 
Biometrics  2014;70(1):1-9.
Parametric estimation of the cumulative incidence function (CIF) is considered for competing risks data subject to interval censoring. Existing parametric models of the CIF for right censored competing risks data are adapted to the general case of interval censoring. Maximum likelihood estimators for the CIF are considered under the assumed models, extending earlier work on nonparametric estimation. A simple naive likelihood estimator is also considered that utilizes only part of the observed data. The naive estimator enables separate estimation of models for each cause, unlike full maximum likelihood in which all models are fit simultaneously. The naive likelihood is shown to be valid under mixed case interval censoring, but not under an independent inspection process model, in contrast with full maximum likelihood which is valid under both interval censoring models. In simulations, the naive estimator is shown to perform well and yield comparable efficiency to the full likelihood estimator in some settings. The methods are applied to data from a large, recent randomized clinical trial for the prevention of mother-to-child transmission of HIV.
PMCID: PMC4004384  PMID: 24400873
Competing risks; Cumulative incidence function; Gompertz; HIV/AIDS; Maximum likelihood
21.  Antiretroviral pharmacokinetics in mothers and breastfeeding infants from 6 to 24 weeks post partum: results of the BAN Study 
Antiviral therapy  2014;19(6):587-595.
An intensive, prospective, open-label pharmacokinetic (PK) study in a subset of HIV-infected mothers and their uninfected infants enrolled in the Breastfeeding, Antiretroviral, and Nutrition study was performed to describe drug exposure and antiviral response.
Women using Combivir®[zidovudine (ZDV)+ lamivudine (3TC)]+Aluvia®[lopinavir/ritonavir(LPV/RTV)] were enrolled. Breast milk (BM) and mother and infant plasma (MP, IP) samples were obtained over 6hrs after observed dosing at 6, 12, or 24wks post-partum for drug concentrations and HIV RNA.
30 mother/infant pairs (10 each at 6, 12,and 24wks post-partum) were enrolled. Relative to MP, BM concentrations of ZDV and 3TC were 35% and 21% higher, while LPV and RTV were 80% lower. Only 3TC was detected in IP with concentrations 96% and 98% lower than MP and BM, respectively. Concentrations in all matrices were similar at 6-24wks. The majority (98.3%) of BM concentrations were >HIVwt IC50, with one having detectable virus. There was no association between PK parameters and MP or BM HIV RNA.
ZDV and 3TC concentrated in BM while LPV and RTV did not, possibly due to protein binding and drug transporter affinity. Undetectable to low ARV concentrations in IP suggests prevention of transmission while breast feeding may be due to ARV effects on systemic or BM HIV RNA in the mother. Low IP 3TC exposure may predispose an infected infant to HIV resistance, necessitating testing and treating infants early.
PMCID: PMC4110187  PMID: 24464632
22.  Causal Inference for Vaccine Effects on Infectiousness 
The International Journal of Biostatistics  2012;8(2):10.2202/1557-4679.1354 /j/ijb.2012.8.issue-2/1557-4679.1354/1557-4679.1354.xml.
If a vaccine does not protect individuals completely against infection, it could still reduce infectiousness of infected vaccinated individuals to others. Typically, vaccine efficacy for infectiousness is estimated based on contrasts between the transmission risk to susceptible individuals from infected vaccinated individuals compared with that from infected unvaccinated individuals. Such estimates are problematic, however, because they are subject to selection bias and do not have a causal interpretation. Here, we develop causal estimands for vaccine efficacy for infectiousness for four different scenarios of populations of transmission units of size two. These causal estimands incorporate both principal stratification, based on the joint potential infection outcomes under vaccine and control, and interference between individuals within transmission units. In the most general scenario, both individuals can be exposed to infection outside the transmission unit and both can be assigned either vaccine or control. The three other scenarios are special cases of the general scenario where only one individual is exposed outside the transmission unit or can be assigned vaccine. The causal estimands for vaccine efficacy for infectiousness are well defined only within certain principal strata and, in general, are identifiable only with strong unverifiable assumptions. Nonetheless, the observed data do provide some information, and we derive large sample bounds on the causal vaccine efficacy for infectiousness estimands. An example of the type of data observed in a study to estimate vaccine efficacy for infectiousness is analyzed in the causal inference framework we developed.
PMCID: PMC3348179  PMID: 22499732
causal inference; principal stratification; interference; infectious disease; vaccine
23.  Change-Point Models to Estimate the Limit of Detection 
Statistics in medicine  2013;32(28):4995-5007.
In many biological and environmental studies, measured data is subject to a limit of detection. The limit of detection is generally defined as the lowest concentration of analyte that can be differentiated from a blank sample with some certainty. Data falling below the limit of detection is left-censored, falling below a level that is easily quantified by a measuring device. A great deal of interest lies in estimating the limit of detection for a particular measurement device. In this paper we propose a change-point model to estimate the limit of detection using data from an experiment with known analyte concentrations. Estimation of the limit of detection proceeds by a two-stage maximum likelihood method. Extensions are considered that allow for censored measurements and data from multiple experiments. A simulation study is conducted demonstrating that in some settings the change-point model provides less biased estimates of the limit of detection than conventional methods. The proposed method is then applied to data from an HIV pilot study.
PMCID: PMC3858526  PMID: 23784922
change point; limit of detection; linear calibration curve; two-stage maximum likelihood
24.  A Peer-Educator Network HIV Prevention Intervention Among Injection Drug Users: Results of a Randomized Controlled Trial in St. Petersburg, Russia 
AIDS and behavior  2013;17(7):2510-2520.
We evaluated the efficacy of a peer-educator network intervention as a strategy to reduce HIV acquisition among injection drug users (IDUs) and their drug and/or sexual networks. A randomized controlled trial was conducted in St. Petersburg, Russia among IDU index participants and their risk network participants. Network units were randomized to the control or experimental intervention. Only the experimental index participants received training sessions to communicate risk reduction techniques to their network members. Analysis includes 76 index and 84 network participants who were HIV uninfected. The main outcome measure was HIV sero-conversion. The incidence rates in the control and experimental groups were 19.57 (95 % CI 10.74–35.65) and 7.76 (95 % CI 3.51–17.19) cases per 100 p/y, respectively. The IRR was 0.41 (95 % CI 0.15–1.08) without a statistically significant difference between the two groups (log rank test statistic X2 = 2.73, permutation p value = 0.16). Retention rate was 67 % with a third of the loss due to incarceration or death. The results show a promising trend that this strategy would be successful in reducing the acquisition of HIV among IDUs.
PMCID: PMC3950300  PMID: 23881187
Injection drug users; Russia; HIV prevention; Network
25.  Community Diarrhea Incidence Before and After Rotavirus Vaccine Introduction in Nicaragua 
We estimated the incidence of watery diarrhea in the community before and after introduction of the pentavalent rotavirus vaccine in León, Nicaragua. A random sample of households was selected before and after rotavirus vaccine introduction. All children < 5 years of age in selected households were eligible for inclusion. Children were followed every 2 weeks for watery diarrhea episodes. The incidence rate was estimated as numbers of episodes per 100 child-years of exposure time. A mixed effects Poisson regression model was fit to compare incidence rates in the pre-vaccine and vaccine periods. The pre-vaccine cohort (N = 726) experienced 36 episodes per 100 child-years, and the vaccine cohort (N = 826) experienced 25 episodes per 100 child-years. The adjusted incidence rate ratio was 0.60 (95% confidence interval [CI] 0.40, 0.91) during the vaccine period versus the pre-vaccine period, indicating a lower incidence of watery diarrhea in the community during the vaccine period.
PMCID: PMC3741244  PMID: 23817336

Results 1-25 (70)