Reduced Bayesian Hierarchical Models: Estimating Health Effects of Simultaneous Exposure to Multiple Pollutants

Quantifying the health effects associated with simultaneous exposure to many air pollutants is now a research priority of the US EPA. Bayesian hierarchical models (BHM) have been extensively used in multisite time series studies of air pollution and health to estimate health effects of a single pollutant adjusted for potential confounding of other pollutants and other time-varying factors. However, when the scientific goal is to estimate the impacts of many pollutants jointly, a straightforward application of BHM is challenged by the need to specify a random-effect distribution on a high-dimensional vector of nuisance parameters, which often do not have an easy interpretation. In this paper we introduce a new BHM formulation, which we call "reduced BHM", aimed at analyzing clustered data sets in the presence of a large number of random effects that are not of primary scientific interest. At the first stage of the reduced BHM, we calculate the integrated likelihood of the parameter of interest (e.g. excess number of deaths attributed to simultaneous exposure to high levels of many pollutants). At the second stage, we specify a flexible random-effect distribution directly on the parameter of interest. The reduced BHM overcomes many of the challenges in the specification and implementation of full BHM in the context of a large number of nuisance parameters. In simulation studies we show that the reduced BHM performs comparably to the full BHM in many scenarios, and even performs better in some cases. Methods are applied to estimate location-specific and overall relative risks of cardiovascular hospital admissions associated with simultaneous exposure to elevated levels of particulate matter and ozone in 51 US counties during the period 1999-2005.

Seasonal Analyses of Air Pollution and Mortality in 100 U.S. Cities

Time series models relating short-term changes in air pollution levels to daily mortality counts typically assume that the effects of air pollution on the log relative rate of mortality do not vary with time. However, these short-term effects might plausibly vary by season. Changes in the sources of air pollution and meteorology can result in changes in characteristics of the air pollution mixture across seasons. The authors develop Bayesian semi-parametric hierarchical models for estimating time-varying effects of pollution on mortality in multi-site time series studies. The methods are applied to the updated National Morbidity and Mortality Air Pollution Study database for the period 1987--2000, which includes data for 100 U.S. cities. At the national level, a 10 micro-gram/m3 increase in PM(10) at lag 1 is associated with a 0.15 (95% posterior interval: -0.08, 0.39),0.14 (-0.14, 0.42), 0.36 (0.11, 0.61), and 0.14 (-0.06, 0.34) percent increase in mortality for winter, spring, summer, and fall, respectively. An analysis by geographical regions finds a strong seasonal pattern in the northeast (with a peak in summer) and little seasonal variation in the southern regions of the country. These results provide useful information for understanding particle toxicity and guiding future analyses of particle constituent data.

On Time Series Analysis of Public Health and Biomedical Data

A time series is a sequence of observations made over time. Examples in public health include daily ozone concentrations, weekly admissions to an emergency department or annual expenditures on health care in the United States. Time series models are used to describe the dependence of the response at each time on predictor variables including covariates and possibly previous values in the series. Time series methods are necessary to account for the correlation among repeated responses over time. This paper gives an overview of time series ideas and methods used in public health research.

Treatment-naive individuals are the major source of transmitted HIV-1 drug resistance in men who have sex with men in the Swiss HIV Cohort Study

BACKGROUND Human immunodeficiency virus type 1 (HIV-1) transmitted drug resistance (TDR) can compromise antiretroviral therapy (ART) and thus represents an important public health concern. Typically, sources of TDR remain unknown, but they can be characterized with molecular epidemiologic approaches. We used the highly representative Swiss HIV Cohort Study (SHCS) and linked drug resistance database (SHCS-DRDB) to analyze sources of TDR. METHODS ART-naive men who have sex with men with infection date estimates between 1996 and 2009 were chosen for surveillance of TDR in HIV-1 subtype B (N = 1674), as the SHCS-DRDB contains pre-ART genotypic resistance tests for >69% of this surveillance population. A phylogeny was inferred using pol sequences from surveillance patients and all subtype B sequences from the SHCS-DRDB (6934 additional patients). Potential sources of TDR were identified based on phylogenetic clustering, shared resistance mutations, genetic distance, and estimated infection dates. RESULTS One hundred forty of 1674 (8.4%) surveillance patients carried virus with TDR; 86 of 140 (61.4%) were assigned to clusters. Potential sources of TDR were found for 50 of 86 (58.1%) of these patients. ART-naive patients constitute 56 of 66 (84.8%) potential sources and were significantly overrepresented among sources (odds ratio, 6.43 [95% confidence interval, 3.22-12.82]; P < .001). Particularly large transmission clusters were observed for the L90M mutation, and the spread of L90M continued even after the near cessation of antiretroviral use selecting for that mutation. Three clusters showed evidence of reversion of K103N or T215Y/F. CONCLUSIONS Many individuals harboring viral TDR belonged to transmission clusters with other Swiss patients, indicating substantial domestic transmission of TDR in Switzerland. Most TDR in clusters could be linked to sources, indicating good surveillance of TDR in the SHCS-DRDB. Most TDR sources were ART naive. This, and the presence of long TDR transmission chains, suggests that resistance mutations are frequently transmitted among untreated individuals, highlighting the importance of early diagnosis and treatment.

Clustering of HCV coinfections on HIV phylogeny indicates domestic and sexual transmission of HCV

BACKGROUND: HCV coinfection remains a major cause of morbidity and mortality among HIV-infected individuals and its incidence has increased dramatically in HIV-infected men who have sex with men(MSM). METHODS: Hepatitis C virus (HCV) coinfection in the Swiss HIV Cohort Study(SHCS) was studied by combining clinical data with HIV-1 pol-sequences from the SHCS Drug Resistance Database(DRDB). We inferred maximum-likelihood phylogenetic trees, determined Swiss HIV-transmission pairs as monophyletic patient pairs, and then considered the distribution of HCV on those pairs. RESULTS: Among the 9748 patients in the SHCS-DRDB with known HCV status, 2768(28%) were HCV-positive. Focusing on subtype B(7644 patients), we identified 1555 potential HIV-1 transmission pairs. There, we found that, even after controlling for transmission group, calendar year, age and sex, the odds for an HCV coinfection were increased by an odds ratio (OR) of 3.2 [95% confidence interval (CI) 2.2, 4.7) if a patient clustered with another HCV-positive case. This strong association persisted if transmission groups of intravenous drug users (IDUs), MSMs and heterosexuals (HETs) were considered separately(in all cases OR >2). Finally we found that HCV incidence was increased by a hazard ratio of 2.1 (1.1, 3.8) for individuals paired with an HCV-positive partner. CONCLUSIONS: Patients whose HIV virus is closely related to the HIV virus of HIV/HCV-coinfected patients have a higher risk for carrying or acquiring HCV themselves. This indicates the occurrence of domestic and sexual HCV transmission and allows the identification of patients with a high HCV-infection risk.

Higher Risk of Incident HCV-Coinfections in MSM with a Wide HIV Transmission Bottleneck

BACKGROUND  High-risk sexual behaviors have been suggested as drivers of the recent dramatic increase of sexually-transmitted HCV among HIV-infected men who have sex with men(MSM). METHODS  We assessed the association between the HIV-transmission-bottleneck and the prevalence and incidence of HCV-coinfections in HIV-infected MSM from the Swiss-HIV-Cohort-Study(SHCS). As a proxy for the width of the transmission bottleneck we used the fraction of ambiguous nucleotides in Genotypic-Resistance-Tests(GRTs) from recent HIV-infections. We defined a broad bottleneck as a fraction of ambiguous nucleotides exceeding a previously-established threshold(0.5%). RESULTS  From the SHCS, we identified 671 MSMs with available HCV-serologies and with a HIV-GRT sampled during recent infection. Of those, 161(24.0%) exhibited a broad HIV-transmission-bottleneck, 38(5.7%) had at least one positive HCV test, and 26(3.9%) had an incident HCV infection. Individuals with broad HIV-transmission bottlenecks exhibited a twofold-higher odds of having ever experienced an HCV coinfection(OR[95%CI]=2.2[1.1, 4.3]) and a threefold-higher hazard of an incident HCV infection(HR[95%CI]= 3.0[1.4, 6.6]) than individuals with narrow HIV-transmission-bottlenecks. CONCLUSIONS  Our results indicate that the currently occurring sexual spread of HCV is focused on those MSMs that are prone to exhibit broad HIV-transmission-bottlenecks. This is consistent with an important role of high-risk behavior and mucosal-barrier-impairment in the transmission of HCV among MSM.

Low-frequency drug-resistant HIV-1 and risk of virological failure to first-line NNRTI-based ART: a multicohort European case-control study using centralized ultrasensitive 454 pyrosequencing.

OBJECTIVES It is still debated if pre-existing minority drug-resistant HIV-1 variants (MVs) affect the virological outcomes of first-line NNRTI-containing ART. METHODS This Europe-wide case-control study included ART-naive subjects infected with drug-susceptible HIV-1 as revealed by population sequencing, who achieved virological suppression on first-line ART including one NNRTI. Cases experienced virological failure and controls were subjects from the same cohort whose viraemia remained suppressed at a matched time since initiation of ART. Blinded, centralized 454 pyrosequencing with parallel bioinformatic analysis in two laboratories was used to identify MVs in the 1%-25% frequency range. ORs of virological failure according to MV detection were estimated by logistic regression. RESULTS Two hundred and sixty samples (76 cases and 184 controls), mostly subtype B (73.5%), were used for the analysis. Identical MVs were detected in the two laboratories. 31.6% of cases and 16.8% of controls harboured pre-existing MVs. Detection of at least one MV versus no MVs was associated with an increased risk of virological failure (OR = 2.75, 95% CI = 1.35-5.60, P = 0.005); similar associations were observed for at least one MV versus no NRTI MVs (OR = 2.27, 95% CI = 0.76-6.77, P = 0.140) and at least one MV versus no NNRTI MVs (OR = 2.41, 95% CI = 1.12-5.18, P = 0.024). A dose-effect relationship between virological failure and mutational load was found. CONCLUSIONS Pre-existing MVs more than double the risk of virological failure to first-line NNRTI-based ART.

Prognostic significance of multidetector CT in normotensive patients with pulmonary embolism: results of the protect study.

BACKGROUND In patients with acute pulmonary embolism (PE), rapid and accurate risk assessment is paramount in selecting the appropriate treatment strategy. The prognostic value of right ventricular dysfunction (RVD) assessed by multidetector CT (MDCT) in normotensive patients with PE has lacked adequate validation. METHODS The study defined MDCT-assessed RVD as a ratio of the RV to the left ventricle short axis diameter greater than 0.9. Outcomes assessed through 30 days after the diagnosis of PE included all-cause mortality and 'complicated course', which consisted of death from any cause, haemodynamic collapse or recurrent PE. RESULTS MDCT detected RVD in 533 (63%) of the 848 enrolled patients. Those with RVD on MDCT more frequently had echocardiographic RVD (31%) than those without RVD on MDCT (9.2%) (p<0.001). Patients with RVD on MDCT had significantly higher brain natriuretic peptide (269±447 vs 180±457 pg/ml, p<0.001) and troponin (0.10±0.43 vs 0.03±0.24 ng/ml, p=0.001) levels in comparison with those without RVD on MDCT. During follow-up, death occurred in 25 patients with and in 13 patients without RVD on MDCT (4.7% vs 4.3%; p=0.93). Those with and those without RVD on MDCT had a similar frequency of complicated course (3.9% vs 2.3%; p=0.30). CONCLUSIONS The PROgnosTic valuE of CT study showed a relationship between RVD assessed by MDCT and other markers of cardiac dysfunction around the time of PE diagnosis, but did not demonstrate an association between MDCT-RVD and prognosis.

Derivation and validation of multimarker prognostication for normotensive patients with acute symptomatic pulmonary embolism.

RATIONALE Not all patients with acute pulmonary embolism (PE) have a high risk of an adverse short-term outcome. OBJECTIVES This prospective cohort study aimed to develop a multimarker prognostic model that accurately classifies normotensive patients with PE into low and high categories of risk of adverse medical outcomes. METHODS The study enrolled 848 outpatients from the PROTECT (PROgnosTic valuE of Computed Tomography) study (derivation cohort) and 529 patients from the Prognostic Factors for Pulmonary Embolism (PREP) study (validation cohort). Investigators assessed study participants for a 30-day complicated course, defined as death from any cause, hemodynamic collapse, and/or adjudicated recurrent PE. MEASUREMENTS AND MAIN RESULTS A complicated course occurred in 63 (7.4%) of the 848 normotensive patients with acute symptomatic PE in the derivation cohort and in 24 patients (4.5%) in the validation cohort. The final model included the simplified Pulmonary Embolism Severity Index, cardiac troponin I, brain natriuretic peptide, and lower limb ultrasound testing. The model performed similarly in the derivation (c-index of 0.75) and validation (c-index of 0.85) cohorts. The combination of the simplified Pulmonary Embolism Severity Index and brain natriuretic peptide testing showed a negative predictive value for a complicated course of 99.1 and 100% in the derivation and validation cohorts, respectively. The combination of all modalities had a positive predictive value for the prediction of a complicated course of 25.8% in the derivation cohort and 21.2% in the validation cohort. CONCLUSIONS For normotensive patients who have acute PE, we derived and validated a multimarker model that predicts all-cause mortality, hemodynamic collapse, and/or recurrent PE within the following 30 days.

Increases in Condomless Sex in the Swiss HIV Cohort Study

Condomless sex is a key driver of sexually transmitted diseases. In this study, we assess the long-term changes (2000-2013) of the occurrence of condomless sex among human immunodeficiency virus (HIV)-infected individuals enrolled in the Swiss HIV Cohort study. The frequencies with which HIV-infected individuals reported condomless sex were either stable or only weakly increasing for 2000-2008. For 2008-2013, these rates increased significantly for stable relationships among heterosexuals and men who have sex with men (MSM) and for occasional relationships among MSM. Our results highlight the increasing public health challenge posed by condomless sex and show that condomless sex has been increasing even in the most recent years.

HIV-1 Transmission During Recent Infection and During Treatment Interruptions as Major Drivers of New Infections in the Swiss HIV Cohort Study.

BACKGROUND  Reducing the fraction of transmissions during recent human immunodeficiency virus (HIV) infection is essential for the population-level success of "treatment as prevention". METHODS  A phylogenetic tree was constructed with 19 604 Swiss sequences and 90 994 non-Swiss background sequences. Swiss transmission pairs were identified using 104 combinations of genetic distance (1%-2.5%) and bootstrap (50%-100%) thresholds, to examine the effect of those criteria. Monophyletic pairs were classified as recent or chronic transmission based on the time interval between estimated seroconversion dates. Logistic regression with adjustment for clinical and demographic characteristics was used to identify risk factors associated with transmission during recent or chronic infection. FINDINGS  Seroconversion dates were estimated for 4079 patients on the phylogeny, and comprised between 71 (distance, 1%; bootstrap, 100%) to 378 transmission pairs (distance, 2.5%; bootstrap, 50%). We found that 43.7% (range, 41%-56%) of the transmissions occurred during the first year of infection. Stricter phylogenetic definition of transmission pairs was associated with higher recent-phase transmission fraction. Chronic-phase viral load area under the curve (adjusted odds ratio, 3; 95% confidence interval, 1.64-5.48) and time to antiretroviral therapy (ART) start (adjusted odds ratio 1.4/y; 1.11-1.77) were associated with chronic-phase transmission as opposed to recent transmission. Importantly, at least 14% of the chronic-phase transmission events occurred after the transmitter had interrupted ART. CONCLUSIONS  We demonstrate a high fraction of transmission during recent HIV infection but also chronic transmissions after interruption of ART in Switzerland. Both represent key issues for treatment as prevention and underline the importance of early diagnosis and of early and continuous treatment.

Contribution of APOBEC3G/F Activity to the Development of Low-Abundance Drug-Resistant HIV-1 variants.

Plasma drug-resistant minority HIV-1 variants (DRMV) increase the risk of virological failure to first-line NNRTI antiretroviral therapy (ART). The origin of DRMVs in ART-naive patients, however, remains unclear. In a large pan-European case-control study investigating the clinical relevance of pre-existing DRMVs using 454 pyrosequencing, the six most prevalent plasma DRMVs detected corresponded to G-to-A nucleotide mutations (V90I, V106I, V108I, E138K, M184I and M230I). Here, we evaluated if such DRMVs could have emerged from APOBEC3G/F activity. Out of 236 ART-naïve evaluated subjects, APOBEC3G/F hypermutation signatures were detected in plasma viruses of 14 (5.9%) individuals. Samples with minority E138K, M184I, and M230I mutations, but not those with V90I, V106I, or V108I were significantly associated with APOBEC3G/F activity (Fisher's p<0.005), defined as presence of >0.5% of sample sequences with an APOBEC3G/F signature. Mutations E138K, M184I and M230I co-occurred in the same sequence as APOBEC3G/F signatures in 3/9 (33%), 5/11 (45%) and 4/8 (50%) of samples, respectively; such linkage was not found for V90I, V106I or V108I. In-frame STOP codons were observed in 1.5% of all clonal sequences; 14.8% of them co-occurred with APOBEC3G/F signatures. APOBEC3G/F-associated E138K, M184I and M230I appeared within clonal sequences containing in-frame STOP codons in 2/3 (66%), 5/5 (100%) and 4/4 (100%) of the samples. In a reanalysis of the parent case-control study, presence of APOBEC3G/F signatures was not associated with virological failure. In conclusion, the contribution of APOBEC3G/F editing to the development of DRMVs is very limited and does not affect the efficacy of NNRTI ART.

Assessing efficacy of different nucleos(t)ide backbones in NNRTI-containing regimens in the Swiss HIV Cohort Study.

BACKGROUND The most recommended NRTI combinations as first-line antiretroviral treatment for HIV-1 infection in resource-rich settings are tenofovir/emtricitabine, abacavir/lamivudine, tenofovir/lamivudine and zidovudine/lamivudine. Efficacy studies of these combinations also considering pill numbers, dosing frequencies and ethnicities are rare. METHODS We included patients starting first-line combination ART (cART) with or switching from first-line cART without treatment failure to tenofovir/emtricitabine, abacavir/lamivudine, tenofovir/lamivudine and zidovudine/lamivudine plus efavirenz or nevirapine. Cox proportional hazards regression was used to investigate the effect of the different NRTI combinations on two primary outcomes: virological failure (VF) and emergence of NRTI resistance. Additionally, we performed a pill burden analysis and adjusted the model for pill number and dosing frequency. RESULTS Failure events per treated patient for the four NRTI combinations were as follows: 19/1858 (tenofovir/emtricitabine), 9/387 (abacavir/lamivudine), 11/344 (tenofovir/lamivudine) and 45/1244 (zidovudine/lamivudine). Compared with tenofovir/emtricitabine, abacavir/lamivudine had an adjusted HR for having VF of 2.01 (95% CI 0.86-4.55), tenofovir/lamivudine 2.89 (1.22-6.88) and zidovudine/lamivudine 2.28 (1.01-5.14), whereas for the emergence of NRTI resistance abacavir/lamivudine had an HR of 1.17 (0.11-12.2), tenofovir/lamivudine 11.3 (2.34-55.3) and zidovudine/lamivudine 4.02 (0.78-20.7). Differences among regimens disappeared when models were additionally adjusted for pill burden. However, non-white patients compared with white patients and higher pill number per day were associated with increased risks of VF and emergence of NRTI resistance: HR of non-white ethnicity for VF was 2.85 (1.64-4.96) and for NRTI resistance 3.54 (1.20-10.4); HR of pill burden for VF was 1.41 (1.01-1.96) and for NRTI resistance 1.72 (0.97-3.02). CONCLUSIONS Although VF and emergence of resistance was very low in the population studied, tenofovir/emtricitabine appears to be superior to abacavir/lamivudine, tenofovir/lamivudine and zidovudine/lamivudine. However, it is unclear whether these differences are due to the substances as such or to an association of tenofovir/emtricitabine regimens with lower pill burden.