Search

Background. Recent updates of human immunodeficiency virus (HIV) treatment guidelines have raised the CD4+ cell count thresholds for antiretroviral therapy initiation from 350 to 500 cells/mm(3) in the United States and from 200 to 350 cells/mm(3) in mid- and low-income countries. Robust data of time from HIV seroconversion to CD4+ cell counts of 200, 350, and 500 cells/mm(3) are lacking but are needed to inform health care planners of the likely impact and cost effectiveness of these and possible future changes in CD4+ cell count initiation threshold. Methods. Using Concerted Action on Seroconversion to AIDS and Death in Europe data from individuals with well-estimated dates of HIV seroconversion, we fitted mixed models on the square root of CD4+ cell counts measured before combined antiretroviral therapy (cART) initiation. Restricting analyses to adults (age >16 years), we predicted time between seroconversion and CD4+ cell count

Competing risks data arise naturally in medical research, when subjects under study are at risk of more than one mutually exclusive event such as death from different causes. The competing risks framework also includes settings where different possible events are not mutually exclusive but the interest lies on the first occurring event. For example, in HIV studies where seropositive subjects are receiving highly active antiretroviral therapy (HAART), treatment interruption and switching to a new HAART regimen act as competing risks for the first major change in HAART. This article introduces competing risks data and critically reviews the widely used statistical methods for estimation and modelling of the basic (estimable) quantities of interest. We discuss the increasingly popular Fine and Gray model for subdistribution hazard of interest, which can be readily fitted using standard software under the assumption of administrative censoring. We present a simulation study, which explores the robustness of inference for the subdistribution hazard to the assumption of administrative censoring. This shows a range of scenarios within which the strictly incorrect assumption of administrative censoring has a relatively small effect on parameter estimates and confidence interval coverage. The methods are illustrated using data from HIV-1 seropositive patients from the collaborative multicentre study CASCADE (Concerted Action on SeroConversion to AIDS and Death in Europe).

When competing risks data arise, information on the actual cause of failure for some subjects might be missing. Therefore, a cause-specific proportional hazards model together with multiple imputation (MI) methods have been used to analyze such data. Modelling the cumulative incidence function is also of interest, and thus we investigate the proportional subdistribution hazards model (Fine and Gray model) together with MI methods as a modelling approach for competing risks data with missing cause of failure. Possible strategies for analyzing such data include the complete case analysis as well as an analysis where the missing causes are classified as an additional failure type. These approaches, however, may produce misleading results in clinical settings. In the present work we investigate the bias of the parameter estimates when fitting the Fine and Gray model in the above modelling approaches. We also apply the MI method and evaluate its comparative performance under various missing data scenarios. Results from simulation experiments showed that there is substantial bias in the estimates when fitting the Fine and Gray model with naive techniques for missing data, under missing at random cause of failure. Compared to those techniques the MI-based method gave estimates with much smaller biases and coverage probabilities of 95 per cent confidence intervals closer to the nominal level. All three methods were also applied on real data modelling time to AIDS or non-AIDS cause of death in HIV-1 infected individuals.

BACKGROUND Studies based on seroconverters have increased our understanding of HIV disease. It is not clear, however, whether their disease progression differs from that of the general HIV population, given their reasons for presenting for testing. METHODS Using linear mixed models we compared CD4 decline rates for a seroconverter (CASCADE) and seroprevalent group (Concorde trial) with time origin being dates of seroconversion and randomization, respectively. Follow-up was censored at the earlier of last alive date and 1 January 1996. Analyses were adjusted for risk group, age and sex. To explore the role of symptomatic seroconversion we further categorized seroconverters into two groups: with and without an HIV test interval below 30 days as proxy. RESULTS The 7226 seroconverter and 1746 seroprevalent eligible individuals were mainly men (78 and 85%, respectively) infected through sex between men (52 and 63%) with mean [95% confidence interval (CI)] baseline CD4 cell count of 610 (602, 619) and 492 (479, 505) cells/μl, respectively. There was no evidence that rate of CD4 decline differs between the two groups even after adjusting for potential confounders (P = 0.67). Estimated loss in the year after reaching an arbitrary threshold of 400 cells/μl was 67 (95% CI 65, 69) and 67 (64, 69) cells/μl in the seroconverter and seroprevalent group, respectively. Whereas seroconverters with test interval below 30 days (n = 310) experienced faster decline, there was no difference in rates between other seroconverters and seroprevalent individuals (P = 0.87). CONCLUSIONS These data suggest that estimates of HIV progression derived from seroconverters are likely to hold more generally for the HIV-infected population.

INTRODUCTION: Migrants are overrepresented in the European HIV epidemic. We aimed to understand the barriers and facilitators to HIV testing and current treatment and healthcare needs of migrants living with HIV in Europe. METHODS: A cross-sectional study was conducted in 57 HIV clinics in nine countries (Belgium, Germany, Greece, Italy, The Netherlands, Portugal, Spain, Switzerland and United Kingdom), July 2013 to July 2015. HIV-positive patients were eligible for inclusion if they were as follows: 18 years or older; foreign-born residents and diagnosed within five years of recruitment. Questionnaires were completed electronically in one of 15 languages and linked to clinical records. Primary outcomes were access to primary care and previous negative HIV test. Data were analysed using random effects logistic regression. Outcomes of interest are presented for women, heterosexual men and gay/bisexual men. RESULTS: A total of 2093 respondents (658 women, 446 heterosexual men and 989 gay/bisexual men) were included. The prevalence of a previous negative HIV test was 46.7%, 43.4% and 82.0% for women, heterosexual and gay/bisexual men respectively. In multivariable analysis previous testing was positively associated with: receipt of post-migration antenatal care among women, permanent residency among heterosexual men and identifying as gay rather than bisexual among gay/bisexual men. Access to primary care was found to be high (>83%) in all groups and was strongly associated with country of residence. Late diagnosis was common for women and heterosexual men (60.8% and 67.1%, respectively) despite utilization of health services prior to diagnosis. Across all groups almost three-quarters of people on antiretrovirals had an HIV viral load

Many studies in various research areas have designs that involve repeated measurements over time of a continuous variable across a group of subjects. A frequent and serious problem in such studies is the occurrence of missing data. In many cases, missing data are caused by an event that leads to a premature termination of the series of repeated measurements on some subjects. When the probability of the occurrence of this event is related to the subject-specific underlying trend of the variable of interest, this missingness process is called informative censoring or informative dropout. Standard likelihood-based methods (for example, linear mixed models) fail to give consistent estimates. In such cases, one needs to apply methods that simultaneously model the observed data and the missingness process. In this article, we review a method proposed by Touloumi et al. (1999, Statistics in Medicine 18: 1215–1233) to adjust for informative dropout in longitudinal data analysis. We also present the jmre1 command, which can be used to fit the proposed model. The estimation method combines the restricted iterative generalized least-squares method with a nested expectation-maximization algorithm. The method is implemented mainly using Stata's matrix programming language, Mata. Our example is derived from the epidemiology of the HIV infection.

Repeated measurements of surrogate markers are frequently used to track disease progression, but these series are often prematurely terminated due to disease progression or death. Analysing such data through standard likelihood-based approaches can yield severely biased estimates if the censoring mechanism is non-ignorable. Motivated by this problem, we have proposed the bivariate joint multivariate random effects (JMRE) model, which has shown that when correctly specified it performs well in terms of bias reduction and precision. The bivariate JMRE model is fully parametric and belongs to the class of shared parameters joint models where a survival model for the dropouts and a mixed model for the markers' evolution are linked through a multivariate normal distribution of random effects. As in every parametric model, robustness under violations of its distributional assumptions is of great importance. In this study we generated 500 simulated data sets assuming that random effects jointly follow a heavy-tailed distribution, two skewed distributions or a mixture of two normal distributions. Moreover, we generated data where level-1 errors or residuals in the survival part of the model follow a skewed distribution. Further sensitivity analysis on the effects of reduced sample size, increased level-1 variances and altered fixed effects values was also performed. We found that fixed effects estimates are almost unaffected, but their standard errors (SEs) may be underestimated especially under heavily skewed distributions. The proposed model seems robust enough, but its performance on smaller data sets or under more extreme departures of its assumptions needs further investigation.

OBJECTIVES: We investigated the short-term effects of carbon monoxide on total and cardiovascular mortality in 19 European cities participating in the APHEA-2 (Air Pollution and Health: A European Approach) project.METHODS: We examined the association using hierarchical models implemented in two stages. In the first stage, data from each city were analyzed separately, whereas in the second stage the city-specific air pollution estimates were regressed on city-specific covariates to obtain overall estimates and to explore sources of possible heterogeneity. We evaluated the sensitivity of our results by applying different degrees of smoothing for seasonality control in the city-specific analysis.RESULTS: We found significant associations of CO with total and cardiovascular mortality. A 1-mg/m(3) increase in the 2-day mean of CO, levels was associated with a 1.20% [95% confidence interval (CI), 0.63-1.77%] increase in total deaths and a 1.25% (95% CI, 0.30-2.21%) increase in cardiovascular deaths. There was indication of confounding with black smoke and nitrogen dioxide, but the pollutant-adjusted effect of CO on mortality remained at least marginally statistically significant. The effect of CO on total and cardiovascular mortality was observed mainly in western and southern European cities and was larger when the standardized mortality rate was lower.CONCLUSIONS: The results of this large study are consistent with an independent effect of CO on mortality. The heterogeneity found in the effect estimates among cities may be explained partly by specific city characteristics.

Some individuals remain AIDS-free with a high and stable CD4 cell count without antiretroviral therapy (ART) for many years. We estimated long-term progression-free survival after HIV seroconversion and aimed to identify factors associated with loss of long-term non-progression (LTNP) status.For this cohort study, we used data for individuals with well-estimated dates of HIV-1 seroconversion from the CASCADE Collaboration a network of 28 HIV seroconverter cohort studies in Europe, Australia, Canada, and sub-Saharan Africa. The first cohort began enrolling patients in 1979, and for this analysis we used data pooled in May 1, 2011. We defined non-progression as being HIV-positive without AIDS, ART-naive, and with CD4 counts of 500 cells per μL or higher. We defined LTNP as non-progression during the first 10 years after seroconversion. We used longitudinal methods to characterise LTNP.Of the 4979 HIV seroconverters in our dataset, 3708 (75%) were men. Median time to progression was 2·07 years (95% CI 1·96-2·17), giving estimated progression-free survivals of 18·4% (17·2-19·6) 5 years, 4·0% (3·6-4·5) 10 years, and 1·4% (0·9-1·5) 15 years after seroconversion. The rate of progression did not change beyond 10 years after seroconversion (0·28 [95%CI 0·26-0·31] per person-year at 10 years after seroconversion, 0·24 [0·19-0·29] per person-year at 15 years, and 0·18 [0·10-0·33] per person-year at 20 years). At 10 years since HIV seroconversion, 283 individuals had LTNP, of whom 202 subsequently lost this status (median time to loss of status 2·49 years [2·05-2·92]). In univariable analyses, loss of LTNP status was associated with CD4 cell count at 10 years after seroconversion (p0·0001) and HIV RNA load at 10 years after seroconversion (p = 0·005), but not age (p = 0·544), mode of infection (p = 0·621), sex (p = 0·676), or calendar year of seroconversion (p = 0·397). In the multivariable analyses, loss of LTNP status was associated with lower CD4 counts at 10 years after seroconversion (p0·0001). After exclusion of CD4 cell counts from the model, higher HIV RNA load at 10 years after seroconversion was independently associated with loss of LTNP status (p = 0·009).Progression-free survival is rare. Most individuals with LTNP eventually lose immunological and clinical control of HIV infection eventually.European Union Seventh Framework Programme.

Combined antiretroviral treatment (cART) modifications are often required due to treatment failure or side effects. We investigate cART regimens' durability, frequency of treatment-limiting adverse events, and potential risk factors and temporal trends. Data were derived from the Athens Multicenter AIDS Cohort Study (AMACS). Statistical analyses were based on survival techniques, allowing for multiple contributions per individual. Overall, 2,756 individuals, aged15 years, initiated cART. cART regimens were grouped by their initiation date into four calendar periods (1995-1998, 1999-2002, 2003-2006, and 2007+). Median [95% confidence interval (CI)] time to first treatment modification was 2.11 (1.95-2.33) years; cumulative probabilities at 1 year were 31.6%, 29.0%, 33.1%, and 29.6% for the four periods, respectively. cART modifications were less frequent in more recent years (adjusted HR=0.96 per year; p0.001). Longer treatment duration was associated with lower HIV-RNA, higher CD4 counts, and being previously ART naive. cART modifications due to treatment failure became less frequent in recent years (adjusted HR=0.91 per year; p0.001). Estimated (95% CI) 1 year cumulative probabilities of treatment-limiting side effects were 16.4% (12.0-21.3%), 19.3% (15.6-23.3%), 24.9% (20.3-29.7%), and 21.1% (13.4-29.9%) for the four periods, respectively, with no significant temporal trends. Risk of side effects was lower in nonnucleoside reverse transcriptase inhibitor (NNRTI)-based regimens or triple nucleoside reverse transcriptase inhibitor (NRTI)-based cART regimens. Treatment modifications have become less frequent in more recent years. This could be partly attributed to the lower risk for side effects of NNRTI-based cART regimens and mainly to the improved efficacy of newer drugs. However, the rate of drugs substitutions due to adverse events remains substantially high.

The possibility of safe discontinuation of therapy with nucleos(t)ide analogues (NAs) remains one of the most controversial topics in the management of chronic hepatitis B. Therefore, we systematically reviewed the existing data on NA discontinuation in this setting and tried to identify factors affecting the probability of posttherapy remission. A literature search was performed in order to identify all published studies including patients who discontinued NAs in virological remission (VR) and were followed for ≥12 months thereafter. Twenty-five studies with 1716 patients were included. The pooled rates of durable VR remission were 51.4%, 39.3%, and 38.2% at 12, 24, and 36 months, respectively, after NA discontinuation, being relatively higher in initially hepatitis B e antigen (HBeAg)-positive patients (62.5%, 53.4%, 51.5%) than HBeAg-negative patients (43.7%, 31.3%, 30.1%) (P = 0.064). The weighted probability of durable biochemical remission was 65.4%, being numerically higher in HBeAg-positive than HBeAg-negative patients (76.2% versus 56.7%, P = 0.130). The weighted probability of hepatitis B surface antigen loss was 2.0%. The rates of durable VR did not significantly differ according to the VR definition (hepatitis B virus DNA200, 2000, 20,000 IU/mL) or duration of on-therapy VR in HBeAg-positive patients, but they were significantly higher in studies with HBeAg-negative patients and on-therapy VR 24 than ≤ 24 months (VR at 12 months off-NAs: 75.0% versus 35.6%, P = 0.005). The weighted probability of durable HBeAg seroconversion was 91.9% and 88.0% at 12 and 24 months, respectively, after NA discontinuation without being affected by the duration of on-therapy VR or consolidation therapy (6 months in all studies).Durable VR seems to be feasible in a substantial proportion of patients who discontinue long-term NA therapy; on-therapy VR 24 months offers higher chances of off-NA VR in patients with HBeAg-negative chronic hepatitis B.

Background: Particulate air pollution is associated with increased mortality. There is a need for European results from multicountry databases concerning cause-specific mortality to obtain more accurate effect estimates.Methods: We report the estimated effects of ambient particle concentrations (black smoke and particulate matter less than 10 Am [PM10]) on cardiovascular and respiratory mortality, from 29 European cities, within the Air Pollution and Health: a European Approach (APHEA2) project. We applied a 2-stage hierarchical modeling approach assessing city-specific effects first and then overall effects. City characteristics were considered as potential effect modifiers.Results: An increase in PM10 by 10 mu g/m(3) (lag 0 + 1) was associated with increases of 0.76% (95% confidence interval = 0.47 to 1.05%) in cardiovascular deaths and 0.58% (0.21 to 0.95%) in respiratory deaths. The same increase in black smoke was associated with increases of 0.62% (0.35 to 0.90%) and 0.84% (0.11 to 1.57%), respectively.Conclusions: These effect estimates are appropriate for health impact assessment and standard-setting procedures.

Summary Background Factors that predict response and breakthrough phenomenon to lamivudine monotherapy in patients with HBeAg-negative chronic hepatitis B have not been well defined. Aim To determine pre-treatment and on treatment variables that predict initial response and breakthrough in patients with HBeAg-negative chronic hepatitis B receiving long-term lamivudine. Methods Seventy-nine patients, with chronic HBeAg-negative hepatitis B, who received lamivudine for a median of 31 months were included in the study. Results Initial virologic and biochemical response was observed in 73 (92%) and 70 (89%) patients, respectively, while 34 (47%) and 15 (21%) patients developed virological and biochemical breakthrough, respectively. High levels of necroinflammation in liver biopsy were associated with a higher probability of initial virological and biochemical response. Patients with pre-treatment serum hepatitis B virus DNA concentrations of more than 106 copies/mL were three times more likely to develop virologic breakthrough. Two patients died, one with baseline cirrhosis because of liver failure during biochemical breakthrough while the second death was liver and treatment unrelated. Conclusions In HBeAg-negative chronic hepatitis B, initial response to lamivudine therapy is associated with necroinflammation, while baseline serum hepatitis B virus DNA exceeding 106 copies/mL is a strong predictor for breakthrough because of drug-resistant mutations. Severe complications are uncommon and are associated with biochemical breakthrough and pre-existing cirrhosis.

Summary. The epidemic of hepatitis C virus (HCV) infection is a major public health issue. We conducted a comprehensive analysis to estimate future HCV-related morbidity and mortality, using a model which is the first to take into account currently available treatments. We reconstructed the incident infections per year in the past that progressed to chronic hepatitis C (CHC) in Greece. Then, the natural history of the disease was simulated in subcohorts of newly infected subjects in the presence or absence of treatment using yearly estimates of the number of treated patients obtained from national databases. Annual estimates of the incidence and prevalence of CHC by fibrosis stage, hepatocellular carcinoma (HCC) and mortality were obtained up to 2030. The current proportion of naive CHC patients receiving treatment in Greece is 1.2% per year. Treatment of 1.2–10% of naive CHC patients per year would reduce the cumulative number of incident cirrhosis and HCC cases from 2002 to 2030 by 10.8–39.4% and 12.8–39.8%, respectively and decrease the number of prevalent cirrhosis and HCC cases in 2030 by approximately 17–48% compared with the number estimated under the assumption of no treatment. Approximately 17 cirrhosis cases or six HCC cases or 10 premature deaths would be prevented for every 100 treated patients. However, the prevalent cirrhotic/HCC cases because of HCV and HCV-related deaths would not plateau until 2030. Despite the introduction of effective treatment, HCV-related morbidity and mortality will likely increase during the next 20–30 years in Greece. Intensive primary prevention efforts coupled with increased access to the currently available treatments are necessary to control the chronic consequences of HCV epidemic.

Summary. Aim: The study aims to describe the course of HIV-1 infection in the pre- and post-HAART period in a cohort of HIV+ haemophilia patients followed up for up to 21 years. Methods: The cohort includes 158 haemophilic men with known seroconversion dates followed up prospectively for a median time of 12 and 5.7 years in the pre- (1980–96) and post-HAART period (1997–2003), respectively. Results: The risk of developing AIDS was lowered by 56% in the post- as compared to the pre-HAART period. Of the 158 patients 69 developed AIDS in the pre-HAART period while of the 59 subjects still alive and AIDS free on 1/1/1997 six developed AIDS. The rate of PCP (12.0 cases per 1000 person-years) and NHL (5.4 cases per 1000 person-years), the most common causes of AIDS diagnosis in the pre-HAART era, were remarkably reduced in the post-HAART era (both rates: 2.8 cases per 1000 person-years). On the contrary, the corresponding risk for non-AIDS deaths was fourfold increased in the post-HAART period. Of the 38 non-AIDS related deaths in both periods, 13 occurred post-HAART. The predominant cause of non-AIDS mortality in both periods was end-stage liver disease (ESLD) (7 pre- and 4 post-HAART). The rate of non-AIDS related cancers was also increased during the post-HAART period. Conclusion: In this haemophilia cohort the risk of AIDS has substantially reduced in the post-HAART period, but the rate of non-AIDS mortality tended to increase. Among haemophilia subjects, due to the high rates of HCV/HIV coinfection, ESLD, the predominant cause of non-AIDS mortality, will become an increasingly important clinical problem.

Summary The main statistical problem in many epidemiological studies which involve repeated measurements of surrogate markers is the frequent occurrence of missing data. Standard likelihood-based approaches like the linear random-effects model fail to give unbiased estimates when data are non-ignorably missing. In human immunodeficiency virus (HIV) type 1 infection, two markers which have been widely used to track progression of the disease are CD4 cell counts and HIV–ribonucleic acid (RNA) viral load levels. Repeated measurements of these markers tend to be informatively censored, which is a special case of non-ignorable missingness. In such cases, we need to apply methods that jointly model the observed data and the missingness process. Despite their high correlation, longitudinal data of these markers have been analysed independently by using mainly random-effects models. Touloumi and co-workers have proposed a model termed the joint multivariate random-effects model which combines a linear random-effects model for the underlying pattern of the marker with a log-normal survival model for the drop-out process. We extend the joint multivariate random-effects model to model simultaneously the CD4 cell and viral load data while adjusting for informative drop-outs due to disease progression or death. Estimates of all the model’s parameters are obtained by using the restricted iterative generalized least squares method or a modified version of it using the EM algorithm as a nested algorithm in the case of censored survival data taking also into account non-linearity in the HIV–RNA trend. The method proposed is evaluated and compared with simpler approaches in a simulation study. Finally the method is applied to a subset of the data from the ‘Concerted action on seroconversion to AIDS and death in Europe’ study.

Air pollution is associated with total mortality. This association may be confounded by uncontrolled time-varying risk factors such as influenza epidemics.We analyzed independent data on influenza epidemics from 7 European cities that also had data on mortality associated with particulates (PM10). We used 10 methods to control for epidemics (5 derived from influenza data and 5 from respiratory mortality series) and compared those results with analyses that did not control for these epidemics.Adjustment for influenza epidemics increased the PM10 effect estimate in most cases (% change in the pooled regression coefficient: -1.9 to 38.9 for total mortality and 1.3 to 25.5 for cardiovascular mortality). A 10-microg/m increase in PM10 concentrations (lag 0-1) was associated with a 0.48% (95% confidence interval=0.27-0.70%) increase in daily mortality unadjusted for influenza epidemics, whereas under the various methods to control for epidemics the increase ranged from 0.45% (0.26-0.69%) to 0.67% (0.46-0.89%). The corresponding figures for cardiovascular mortality were 0.85% (0.53-1.18%) with no adjustment and from 0.86% (0.53-1.19%) to 1.06% (0.74-1.39%) with the methods of control.The association between air pollution and mortality is not weakened by control for influenza epidemic irrespective of the method used. To adjust for influenza epidemics, one can use methods based on respiratory mortality counts instead of counts of influenza cases if the latter are not available. However, adjustment for influenza by any method tested did not markedly alter the air pollution effect estimate.

In the Air Pollution and Health: A European Approach (APHEA2) project, the effects of ambient ozone concentrations on mortality were investigated. Data were collected on daily ozone concentrations, the daily number of deaths, confounders, and potential effect modifiers from 23 cities/areas for at least 3 years since 1990. Effect estimates were obtained for each city with city-specific models and were combined using second-stage regression models. No significant effects were observed during the cold half of the year. For the warm season, an increase in the 1-hour ozone concentration by 10 mug/ml was associated with a 0.33% (95% confidence interval [CI], 0.17-0.52) increase in the total daily number of deaths, 0.45% (95% CI, 0.22-0.69) in the number of cardiovascular deaths, and 1.13% (95% CI, 0.62-1.48) in the number of respiratory deaths. The corresponding figures for the 8-hour ozone were similar. The associations with total mortality were independent of SO2 and particulate matter with aerodynamic diameter less than 10 mum (PM10) but were somewhat confounded by NO2 and CO. Individual city estimates were heterogeneous for total (a higher standardized mortality rate was associated with larger effects) and cardiovascular mortality (larger effects were observed in southern cities). The dose-response curve of ozone effects on total mortality during the summer did not deviate significantly from linearity.

Summary. In this study, a comprehensive methodology for modelling the hepatitis C virus (HCV) epidemic is proposed to predict the future disease burden and assess whether the recent decline in the incidence of HCV may affect the future occurrence of cirrhosis and hepatocellular carcinoma (HCC) cases. Using the prevalence of HCV, the distribution of chronic hepatitis C (CHC) patients within the various transmission groups and their infection-onset times, it was possible to reconstruct the incident infections per year in the past that progressed to CHC in Greece. The natural history of the disease was simulated in subcohorts of newly infected subjects using transition probabilities derived either empirically between fibrosis stages 0–4 or from literature review. Annual estimates of the incidence and prevalence of CHC by fibrosis stage, HCC and mortality in Greece were obtained up to 2030. HCV incidence peaked in the late 1980s at five new infections/10 000 person-years. Under the assumption of 20–100% decline in HCV incidence after 1990, the cumulative number of incident cirrhosis and HCC cases from 2002–2030 was projected to be lower by 9.6–48.2% and 5.9–29.5%, respectively, than that estimated under the assumption of no decline. However, the prevalent cirrhotic/HCC cases and HCV-related deaths are predicted to decline in the next 30 years only under the assumption of complete elimination of new HCV infections after 1990. Despite the progress in the reduction of HCV transmission, primary prevention does not seem adequate to reverse the rise in the incidence of cirrhosis and HCC.

Several recent studies have reported significant health effects of air pollution even at low levels of air pollutants. These studies have been criticized for the statistical methods and for inconsistency in results between cities. An important development in air pollution epidemiology has come from multicenter studies. Within the APHEA-2 project we have developed a statistical methodology to evaluate short-term health effects of air pollution using data from 30 cities across Europe. For the analysis, a hierarchical modelling approach was adopted and implemented in two stages: (a) data from each city were analyzed separately to allow for local differences, using generalized additive Poisson regression models; (b) city-specific effects estimates were regressed on city-specific covariates to obtain an overall estimate and to explore heterogeneity across cities. In order to illustrate our methodology we present results for PM10 effects. It was found that a 10 mug/m(3) increase in PM10 or NO2 concentrations is associated with a 0.67% (95% Cl: 0.50 to 0.90) and 0.33% (0.20 to 0.40) increase in total mortality, respectively. After mutual adjustment, the PM10 effect was reduced by 40% and that of NO2 by 20%, but both pooled estimates remained significant. Long-term mean NO2 concentrations act as an effect modifier for PM10 effects, even after adjustment for NO2 confounding effects. In the second stage we explored two different models for combining the adjusted for NO2, PM10 effects across cities: bivariate, which accounts for within-city correlation of PM10 and NO2; and univariate, which ignores this correlation. Both models gave broadly the same results. Copyright (C) 2004 John Wiley Sons, Ltd.

Background: Several recent studies have reported significant health effects of air pollution even at low levels of air pollutants, but in most of these studies linear non-threshold relations were assumed. Aims: To investigate the NO2 mortality dose-response association in nine cities participating in the APHEA-2 project using two different methods: the meta-smooth and the cubic spline method. Methods: The meta-smooth method developed by Schwartz and Zanobetti is based on combining individual city non-parametric smooth curves; the cubic spline method developed within the APHEA-2 project combines individual city estimates of cubic spline shaped dose-response relations. The meta- smooth method is easier and faster to implement, but the cubic spline method is more flexible for further investigation of possible heterogeneity in the dose-response curves among cities. Results: In the range of the pollutant common to all cities the two methods gave similar and comparable curves. Using the cubic spline method it was found that smoking prevalence acts as an effect modifier with larger NO2 effects on mortality at lower smoking prevalence. Conclusions: The NO2-mortality association in the cities included in the present analysis, could be adequately estimated using the linear model. However, investigation of the city specific dose-response curves should precede the application of linear models.

Several methods for the estimation and comparison of rates of change in longitudinal Studies with staggered entry and informative drop-outs have been recently proposed. For multivariate normal linear models, REML estimation is used. There are various approaches to maximizing the corresponding log-likelihood; in this paper we use a restricted iterative generalized least squares method (RIGLS) combined with a nested EM algorithm. An important statistical problem in such approaches is the estimation of the standard errors adjusted for the missing data (observed data information matrix). Louis has provided a general technique for computing the observed data information in terms of completed data quantities within the EM framework. The multiple imputation (MI) method for obtaining variances can be regarded as an alternative to this. The aim of this paper is to develop, apply and compare the Louis and a modified MI method in the setting of longitudinal studies where the source of missing data is either death or disease progression (informative) or end of the study (assumed non-informative). Longitudinal data are simultaneously modelled with the missingness process. The methods are illustrated by modelling CD4 count data from an HIV-1 clinical trial and evaluated through simulation studies. Both methods, Louis and MI, are used with Monte Carlo simulations of the missing data using the appropriate conditional distributions, the former with 100 simulations, the latter with 5 and 10. It is seen that naive SEs based on the completed data likelihood can be seriously biased. This bias was largely corrected by Louis and modified MI methods, which gave broadly similar estimates. Given the relative simplicity of the modified MI method, it may be preferable. Copyright (C) 2003 John Wiley Sons, Ltd.

BACKGROUND: Although studies have reported large reductions in the risks of AIDS and death since the introduction of potent anti-retroviral therapies, few have evaluated whether this has been similar for all AIDS-defining diseases. We wished to evaluate changes over time in the risk of specific AIDS-defining diseases, as first events, using data from individuals with known dates of HIV seroconversion. METHODS: Using a competing risks proportional hazards model on pooled data from 20 cohorts (CASCADE), we evaluated time from HIV seroconversion to each first AIDS-defining disease (16 groups) and to death without AIDS for four calendar periods, adjusting for exposure category, age, sex, acute infection, and stratifying by cohort. We compared results to those obtained from a cause-specific hazards model. RESULTS: Of 6,941, 2,021 (29%) developed AIDS and 437 (6%) died without AIDS. The risk of AIDS or death remained constant to 1996 then reduced; relative hazard = 0.89 (95% CI: 0.77-1.03); 0.90 (95% CI: 0.81-1.01); and 0.32 (95% CI: 0.28-0.37) for 1979-1990, 1991-1993, and 1997-2001, respectively, compared to 1994-1996. Significant risk reductions in 1997-2001 were observed in all but two AIDS-defining groups and death without AIDS in a competing risks model (with similar results from a cause-specific model). There was significant heterogeneity in the risk reduction across events; from 96% for cryptosporidiosis, to 17% for death without AIDS (P < 0.0001). CONCLUSION: These findings suggest that studies reporting a stable trend for particular AIDS diseases over the period 1979-2001 may not have accounted for the competing risks among other events or lack the power to detect smaller trends.

Chronic hepatitis C is associated with more severe liver disease in patients coinfected with HIV, but the pathogenic mechanism of this more aggressive course is still unclear. The aim of this study was to assess the relationship of HCV genotype, viral load and epidemiological factors with the histological severity of chronic hepatitis in haemophilia patients with HCV/HIV coinfection, taking into consideration the immune status of the patients. Twenty-one HIV/HCV coinfected haemophilia patients, with mean age +/- SD 35.7 +/- 8.7 years, underwent transcutaneous liver biopsy 6-15 years (median 12 years) after HIV and 6-32 (median 21.5 years) years after HCV infection. Twelve patients were stage A (CDC), six stage B and three stage C. CD4 cells were 200 in 7(33.3%). Mean +/- SD log(10) HCV-RNA was 6.87 +/- 0.7 copies mL(-1) (range 5.4-7.9), and mean +/- SD log(10) HIV-RNA was 3.75 +/- 0.98 copies mL(-1) (range 2.7-6), at the time of liver biopsy. Minimal hepatitis was diagnosed in five patients (24%), mild in 10 (48%) and moderate in six (28%). Hepatitis stage 0-2 was found in seven cases (33%) and cirrhosis in six (29%). Statistical analysis showed a significant association of CD4 count 200 with mild and moderate hepatitis (P = 0.033). In addition, minimal hepatitis was found only in patients with stage C, while the majority of subjects with HIV stage A showed mild and moderate hepatitis (P = 0.003). Moreover genotype 1 was independently associated with advanced hepatitis stage (P = 0.04). No relationship was found between hepatitis severity, HIV or HCV RNA levels, patient's age and duration of HIV or HCV infection. Our results suggest that HCV/HIV coinfection may aggravate the course of hepatitis in the phase of immunocompetence, most probably through an immune mediated process. Genotype 1 seems to be associated with advanced liver disease.

Background. Many cohort studies and clinical trials use repeated measurements of laboratory markers to track disease progression and to evaluate new therapies. A major problem in the analysis of such studies is that marker data are censored in some patients owing to withdrawal, loss to follow-up, or death. The objective of this paper is to evaluate the impact of selective dropouts attributable to death or disease progression on the estimates of marker change among different groups. Methods. Data on CD4 cell count in human immunodeficiency virus 1-infected individuals from a clinical trial and a cohort study are used to illustrate this problem and a possible solution. Simulation studies are also presented. Results. When the rate of dropout is greater in subjects whose marker status is declining rapidly, commonly used methods, like random effects models, that ignore informative dropouts lead to overoptimistic statements about the marker trends in all compared groups, because subjects with steeper marker drops tend to have shorter follow-up times and hence are weighted less in the estimation of the group rate of the average marker decline. Conclusions. The potential biases attributable to incomplete data require greater recognition in longitudinal studies. Sensitivity analyses to assess the effect of dropouts are important.

The APHEA 2 project investigated short-term health effects of particles in eight European cities. In each city associations between particles with an aerodynamic diameter of less than 10 microm (PM(10)) and black smoke and daily counts of emergency hospital admissions for asthma (0-14 and 15-64 yr), chronic obstructive pulmonary disease (COPD), and all-respiratory disease (65+ yr) controlling for environmental factors and temporal patterns were investigated. Summary PM(10) effect estimates (percentage change in mean number of daily admissions per 10 microg/m(3) increase) were asthma (0-14 yr) 1.2% (95% CI: 0.2, 2.3), asthma (15-64 yr) 1.1% (0.3, 1.8), and COPD plus asthma and all-respiratory (65+ yr) 1.0% (0.4, 1.5) and 0.9% (0.6, 1.3). The combined estimates for Black Smoke tended to be smaller and less precisely estimated than for PM(10). Variability in the sizes of the PM(10) effect estimates between cities was also investigated. In the 65+ groups PM(10) estimates were positively associated with annual mean concentrations of ozone in the cities. For asthma admissions (0-14 yr) a number of city-specific factors, including smoking prevalence, explained some of their variability. This study confirms that particle concentrations in European cities are positively associated with increased numbers of admissions for respiratory diseases and that some of the variation in PM(10) effect estimates between cities can be explained by city characteristics.

To determine virological and immunological response to highly active antiretroviral therapy (HAART) and to investigate factors influencing response in a community-based setting.Plasma HIV RNA levels and CD4 cell counts were studied in 168 unselected individuals starting HAART including indinavir or ritonavir or hard-gel saquinavir-containing regimens.Overall, 60% of the patients reduced their HIV RNA to below 500 Eq/mL, but half of them experienced a subsequent virologic rebound. Patients with higher baseline HIV RNA, higher baseline CD4 cell count, and simultaneous initiation of combination therapy and patients on indinavir or ritonavir regimen were more likely to have virologic response within 6 months since HAART initiation. Patients with lower baseline CD4 cell count and with lower rates of viral clearance had a higher probability of a subsequent virologic rebound. Forty percent of the patients had increased their CD4 cell counts by more than 100 cells/microL (immunologic response). The probability of immunologic response was independent of baseline HIV RNA levels and CD4 cell count; however, the more complete the virologic suppression, the higher the probability of immunologic response. Thirty percent of the patients had discordance between virologic and immunologic responses.The rate of virologic failure in this unselected group of patients was higher than that observed in randomized clinical trials, but only a minority (11%) of the patients were treatment naïve. Starting combination therapy simultaneously and initiating antiretroviral therapy before advanced HIV disease has developed predict virologic response, whereas the magnitude of viral suppression predicts mid to long immunological response.

Short-term effects of air pollution on daily mortality in eight western and five central-eastern European countries have been reported previously, as part of the APHEA project. One intriguing finding was that the effects were lower in central-eastern European cities. The analysis used sinusoidal terms for seasonal control and polynomial terms for meteorologic variables, but this is a more rigid approach than the currently accepted method, which uses generalized additive models (GAM). We therefore reanalyzed the original data to examine the sensitivity of the results to the statistical model. The data were identical to those used in the earlier analyses. The outcome was the daily total number of deaths, and the pollutants analyzed were black smoke (BS) and sulfur dioxide (SO(2)). The analyses were restricted to days with pollutant concentration < 200 microg/m(3) and < 150 microg/m(3) alternately. We used Poisson regression in a GAM model, and combined individual city regression coefficients using fixed and random-effect models. An increase in BS by 50 microg/m(3) was associated with a 2.2% and 3.1% increase in mortality when analysis was restricted to days < 200 microg/m(3) and < 150 microg/m(3), respectively. The corresponding figures were 5.0% and 5.6% for a similar increase in SO(2). These estimates are larger than the ones published previously: by 69% for BS and 55% for SO(2). The increase occurred only in central-eastern European cities. The ratio of western to central-eastern cities for estimates was reduced to 1.3 for BS (previously 4.8) and 2.6 for SO(2) (previously 4.4). We conclude that part of the heterogeneity in the estimates of air pollution effects between western and central-eastern cities reported in previous publications was caused by the statistical approach used and the inclusion of days with pollutant levels above 150 microg/m(3). However, these results must be investigated further.

PURPOSE: We sought to identify the predictors of clinical outcome and of the evolution of cerebral abnormalities in patients with neuropsychiatric systemic lupus erythematosus (SLE). SUBJECTS AND METHODS: Thirty-two patients with SLE (including 14 with the antiphospholipid syndrome) who had been hospitalized with primary neuropsychiatric disease were observed prospectively for at least 2 years. Laboratory and clinical characteristics and data from magnetic resonance imaging (MRI) studies obtained during the hospitalization and 2 years later were evaluated. We ascertained nonreversible or new MRI changes and clinical outcomes, including neuropsychiatric events, during follow-up. RESULTS: Cranial MRI scans on admission were abnormal in 26 (81%) of the 32 patients. Patients with the antiphospholipid syndrome were more likely to have focal cerebral white matter lesions (odds ratio [OR] = 12, 95% confidence interval [CI]: 2.0 to 72). After 2 years, neuropsychiatric deficits substantially improved in 22 (69%) of the patients, stabilized in 6 (19%), and deteriorated in 4 (12%). The number of prior neuropsychiatric events was associated with persistent MRI lesions (OR = 4.8 per each event, 95% CI: 1.1 to 21) and unfavorable clinical outcome (OR = 4.3 per each event, 95% CI: 1.4 to 13) at 2 years. The antiphospholipid syndrome also predicted an unfavorable clinical outcome at 2 years (OR = 11, 95% CI: 1.7 to 65). CONCLUSIONS: Among patients with SLE who have neuropsychiatric disease, prior neuropsychiatric events and the antiphospholipid syndrome increase the risk of adverse outcomes.

The human immunodeficiency virus type 1 (HIV-1) infection is characterized by a long and often prolonged asymptomatic period after initial infection. A number of early epidemiological studies, and particularly prospective incident cohort studies (ie, cohort studies with known seroconversion dates), provided estimates of the AIDS incubation period in different population groups and investigated several host and virus factors for their potential prognostic value. The development of new techniques for sensitive and accurate quantification of HIV-RNA in plasma and/or serum has revolutionized studies of the natural history of the HIV-1 infection. It is now realized that the viral pathogenesis is a highly dynamic process reflecting the various properties of HIV-1 and the host’s immune response to the virus. The increased clinical experience and the introduction of primary prophylactic therapy have altered the spectrum of AIDS-related conditions and the natural history of the infection. Antiretroviral treatment with reverse transcriptase inhibitors (RTI) has shown to have a sizable but transient beneficial effect. The impressive recent advances in treatment with the availability of protease inhibitors (PI) may radically change the natural history of HIV-1 infection. Initial results from clinical trials suggest that highly active antiretroviral therapy (HAART), which combines RTI and PI, results in viral-load reduction and possibly long-lasting immune recovery. For adults in developed countries, in the absence of antiviral therapy, the median time from initial infection to the development of AIDS is about 10 to 11 years; however, there are individuals who progress to AIDS within 5 years of infection and others who remain AIDS-free for more than 15 years. Factors influencing the HIV-1 disease progression have received important consideration. In this review, the natural history of the HIV-1 infection in different populations and the effect of several host and/or virus factors on it will be discussed. The Course of HIV-1 Infection

Summary Background We used data from Europe, North America, and Australia to assess the effect of exposure category on the AIDS incubation period and HIV-1 survival and whether the effect of age at seroconversion varies with exposure category and with time since seroconversion. Methods 38 studies of HIV-1-infected individuals whose dates of seroconversion could be reliably estimated were included in the analysis. Individual data on 13 030 HIV-1-infected individuals from 15 countries were collated, checked, and analysed centrally. We calculated estimates of mortality and AIDS incidence rates and estimated the proportions of individuals surviving and developing AIDS at each year after seroconversion from the numbers of observed deaths or cases of AIDS and the corresponding person-years at risk. Analyses were adjusted for age at seroconversion, time since seroconversion, and other factors as appropriate. Findings Mortality and AIDS incidence increased strongly with time since seroconversion and age at seroconversion. Median survival varied from 12·5 years (95% Cl 12·1–12·9) for those aged 15–24 years at seroconversion to 7·9 years (7·4–8·5) for those aged 45–54 years at seroconversion, whereas for development of AIDS the corresponding values were 11·0 years (10·7–11·7) and 7·7 years (7·1–8·6). There was no appreciable effect of exposure category on survival. For AIDS incidence, the exposure category effect that we noted was explained by the high incidence of Kaposi's sarcoma in those infected through sex between men. We estimated that among people aged 25–29 years at seroconversion 90% (89–91) and 60% (57–62) survived to 5 years and 10 years, respectively, after seroconversion, whereas 13% (12–15) and 46% (44–49), respectively, developed AIDS (excluding Kaposi's sarcoma). Interpretation Before widespread use of highly-active antiretroviral therapy (before 1996), time since seroconversion and age at seroconversion were the major determinants of survival and development of AIDS in Europe, North America, and Australia.

The concentration of T-cell receptor-rearrangement excision DNA circles (TREC) in peripheral-blood T cells is a marker of recent thymic emigrant alphabeta T cells. We studied the predictive ability of measurements of TREC for clinical outcome in HIV-1-infected individuals.We measured TREC in peripheral-blood mononuclear cells with a real-time PCR assay. We studied 131 Greek participants in the Multicenter Hemophilia Cohort Study who had known HIV-1 seroconversion dates. The prognostic value of baseline TREC, CD4 T-cell count, and HIV-1 RNA concentration was assessed by Kaplan-Meier and Cox's regression analysis.Four participants had progressed to AIDS by first blood sampling. Among the remaining 127 individuals, the median value of TREC per 10(6) cells was 6900 (IQR 2370-15604). Baseline TREC values were lower in the 53 who progressed to AIDS than in those who did not (geometric mean 2843 [95% CI 1468-5504] vs 6560 [4723-9113] per 10(6) cells; p=0.017). The relative hazard of AIDS, adjusted for plasma viral load, CD4 T-cell count, and age at seroconversion was 1.44 (95% CI 1.04-2.01; p=0.031) per ten-fold increase in TREC; that for death was 1.52 (1.12-2.06; p=0.007). The adjusted relative hazards of death were 2.91 (1.91-4.44; p0.001) per ten-fold increase in plasma HIV-1 RNA load and 1.20 (1.04-1.38; p=0.014) per 100-cell decrease in CD4 T-cell count.The concentration of TREC in the peripheral T-cell pool complements HIV-1 RNA load and CD4 T-cell count in predicting the rate of HIV-1 disease progression. Recent thymic emigrants have a role in the pathogenesis of HIV-1 disease.

Ten large European cities provided data on daily air pollution as well as mortality from respiratory and cardiovascular mortality. We used Poisson autoregressive models that controlled for trend, season, influenza epidemics, and meteorologic influences to assess the short-term effects of air pollution at each city. We then compared and pooled the city-specific results in a meta-analysis. The pooled relative risks of daily deaths from cardiovascular conditions were 1.02 [95% confidence interval (CI) = 1.01-1.04] for a 50 microg/m3 increment in the concentration of black smoke and 1.04 (95% CI = 1.01-1.06) for an increase in sulfur dioxide levels in western European cities. For respiratory diseases, these figures were 1.04 (95% CI = 1.02-1.07) and 1.05 (95% CI = 1.03-1.07), respectively. These associations were not found in the five central European cities. Eight-hour averages of ozone were also moderately associated with daily mortality in western European cities (relative risk = 1.02; 95% CI = 1.00-1.03 for cardiovascular conditions and relative risk = 1.06; 95% CI = 1.02-1.10 for respiratory conditions). Nitrogen dioxide did not show consistent relations with daily mortality. These results are similar to previously published data and add credence to the causal interpretation of these associations at levels of air pollution close to or lower than current European standards.

Our objectives are to describe the progression of HIV disease and to assess the influence of hemophilia-related variables, age at infection, and antibodies to cytomegalovirus infection (anti-CMV) in a Greek cohort of 158 HIV-1-positive hemophilic men, who received prospective follow-up for up to 16 years after infection. A total of 79 patients had died, representing a cumulative progression rate of 72.4% (95% confidence interval [CI], 56.6-83.3). A significant proportion of the mortality (30%) resulted from conditions not formally related to AIDS, with liver failure and cerebral hemorrhage predominant. At 16 years after seroconversion, 66 patients had developed clinical AIDS, a cumulative progression rate of 58.2% (95% CI, 47.1%-86.3%) whereas 15 years after infection 81.5% (95% CI, 74.2%-87.9%) of the patients had AIDS or a CD4 cell count200 cells/mm3. Hemophilia-related variables or presence of anti-CMV were not significantly associated with disease progression. Age at infection was a strong prognostic factor for all three endpoints. Appropriate modeling showed a nonlinear age effect, with a steeper increase of relative hazard for patients40 years of age at seroconversion. The age effect remained significant even after controlling for current CD4 cell count. Further investigation is required to elucidate the mechanisms of the age effect and the contribution of HCV coinfection on the disease progression.The Greek Hemophilia Cohort Study encompasses 158 HIV-positive men with documented seroconversion dates. The present study estimated the rate of progression to all-cause mortality, clinical AIDS, or advanced immunodeficiency 16 years after seroconversion and evaluated the independent effects of hemophilia-related factors, age at seroconversion, and cytomegalovirus status early in the course of infection. Seroconversion dates extended from September 1980 to December 1985. By 1996, after a median follow-up of 11.6 years, 117 of the 158 men had developed AIDS or had a CD4 cell count of 20 cells/cu. mm, and 79 had died. The estimated cumulative incidence rates of clinical AIDS and death over 16 years since infection were 58.1% and 72%, respectively. 30% of the mortality was due to diseases not formally associated with AIDS (e.g., liver failure and cerebral hemorrhage). A significant association existed between older age at seroconversion and more rapid progression to both AIDS and death, with a particularly steep gradient for patients over 40 years old at seroconversion. The age effect remained significant even after controlling for current CD4 cell count. Increases of 1.32-fold and 1.33-fold in the risks of dying and developing AIDS, respectively, were obtained with every one unit decrease in CD4 cell count on the square root scale. Severity of hemophilia, dosage of clotting factor concentrates, and antibodies to cytomegalovirus were not associated with either AIDS risk or mortality. Further investigations are urged to clarify the mechanisms underlying the age effect observed in this study.

Objective: To assess the impact of age at seroconversion and HIV RNA level in serum during early chronic infection on the initial values and subsequent trends (slopes) of CD4+ lymphocyte counts. Design and methods: In a cohort of 137 HIV-1-positive hemophiliacs with well-estimated dates of seroconversion, baseline HIV RNA level was measured by reverse transcription PCR in serum specimens collected 12-36 months after the estimated date of seroconversion. Baseline values, 24 months after seroconversion, and slopes of CD4+ lymphocyte counts by age and HIV RNA quartile were examined by fitting random effects models that allowed for intrasubject variability. Results: Both age at seroconversion and HIV RNA level were associated with the CD4+ lymphocyte count at baseline and its subsequent slope. The baseline median CD4+ lymphocyte count was 620 x 10 6 /l. Within each HIV RNA quartile, the median CD4+ cell count of the oldest subjects (age 30-58 years) was about 200 x 10 6 /l lower and at least 350 x 10 6 /l lower than the median counts of the younger (age 11-29 years) and youngest (age 2-10 years) subjects, respectively. Within each age-group, the median CD4+ cell count differed by about 200 x 10 6 /l between subjects in the lowest compared with the highest HIV RNA quartiles. The mean slope of the CD4+ lymphocyte count after month 24 was linear on the square-root scale, steeper in children, and did not vary significantly by baseline HIV RNA quartile. There was large variation between subjects that was unexplained by differences in age and HIV RNA level. Conclusions: By 24 months after HIV seroconversion, the oldest subjects and those with the highest HIV RNA levels during early chronic infection had experienced the most severe depletion of CD4+ cells. Subsequent declines in CD4+ cells varied little by early chronic HIV RNA level or age.

The Air Pollution and Health: a European Approach (APHEA) project is a coordinated study of the short-term effects of air pollution on mortality and hospital admissions. Five West European cities (i.e., London, Amsterdam, Rotterdam, Paris, Milano) contributed several years of hospital admissions data for all respiratory causes. In the current study, the authors describe the results obtained from the quantitative pooling (meta-analysis) of local analyses. The diagnostic group was defined by ICD 460-519. The age groups studied were 15-64 y (i.e., adults) and 65 + y (elderly). The air pollutants studied were sulfur dioxide; particles (i.e., Black Smoke or total suspended particles); ozone; and nitrogen dioxide. The pollutants were obtained from existing fixed-site monitors in a standardized manner. We used Poisson models and standardized confounder models to examine the associations between daily hospital admissions and air pollution. We conducted quantitative pooling by calculating the weighted means of local regression coefficients. We used a fixed-effects model when no heterogeneity could be detected; otherwise, we used a random-effects model. When possible, the authors investigated the factors correlated with heterogeneity. The most consistent and strong finding was a significant increase of daily admissions for respiratory diseases (adults and elderly) with elevated levels of ozone. This finding was stronger in the elderly, had a rather immediate effect (same or next day), and was homogeneous over cities. The elderly were affected more during the warm season. The Sulfur dioxide daily mean was available in all cities, and it was not associated consistently with an adverse effect. Effects were present in areas in which more than one station was used in the assessment of daily exposure. Some significant associations were observed, although no conclusion that related to an overall particle effect could be drawn. The effect of Black Smoke was significantly stronger with high nitrogen dioxide levels on the same day, but nitrogen dioxide itself was not associated with admissions. The ozone results were in good agreement with the results of similar U.S. studies. The coherence of the results of this study and other results gained under different conditions strengthens the argument for causality.

To determine the incidence rates and to describe the epidemiological patterns of non-AIDS Kaposi's sarcoma in the central southern area of Greece during the period 1974–1989, all 473 incidence cases reported to Pathology Departments were studied. The mean age (SD) was 67.6 (12.9) years among 297 males and 66.1 (15.9) years among 176 females. The mean age-standardized (Greek population 1981) incidence rate was 0.47 cases per 100,000 total population per year (males 0.62, females 0.32). The standardized incidence rates increased over time for males, with the incidence-rate ratios relative to the earliest period, 1974–1978, being 1.44 (95% CI, 1.02–2.04) for the 1979–1983 interval and 2.12 (95% CI, 1.55–2.90) for the 1984–1989 interval. However, the rates for females did not show a similar pattern. The age-adjusted male:female ratio was 1.6 in 1974–1983 and 2.6 in 1984–1989. Poisson-regression modelling suggested a shift in the age-specific incidence rate in men, towards younger ages during the last period, 1984–1989.Int. J. Cancer 70:538–541. © 1997 Wiley-Liss Inc.

Marginal structural models (MSMs) have been proposed for estimating a treatment's effect, in the presence of time-dependent confounding. We aimed to evaluate the performance of the Cox MSM in the presence of missing data and to explore methods to adjust for missingness. We simulated data with a continuous time-dependent confounder and a binary treatment. We explored two classes of missing data: (i) missed visits, which resemble clinical cohort studies; (ii) missing confounder's values, which correspond to interval cohort studies. Missing data were generated under various mechanisms. In the first class, the source of the bias was the extreme treatment weights. Truncation or normalization improved estimation. Therefore, particular attention must be paid to the distribution of weights, and truncation or normalization should be applied if extreme weights are noticed. In the second case, bias was due to the misspecification of the treatment model. Last observation carried forward (LOCF), multiple imputation (MI), and inverse probability of missingness weighting (IPMW) were used to correct for the missingness. We found that alternatives, especially the IPMW method, perform better than the classic LOCF method. Nevertheless, in situations with high marker's variance and rarely recorded measurements none of the examined method adequately corrected the bias.

Study objective - To review the issues and methodologies in epidemiologic time series studies of daily counts of mortality and hospital admissions and illustrate some of the methodologies.Design - This is a review paper with an example drawn from hospital admissions of the elderly in Cleveland, Ohio, USA.Main results - The central issue is control for seasonality. Both over and under control are possible, and the use of diagnostics, including plots, is necessary. Weather dependence is probably non-linear, and adequate methods are necessary to adjust for this. In Cleveland, the use of categorical variables for weather and sinusoidal terms for filtering season are illustrated. After control for season, weather, and day of the week effects, hospital admission of persons aged 65 and older in Cleveland for respiratory illness was associated with ozone (RR = 1.09, 95% CI 1.02, 1.16) and particulates (PM(10) (RR = 1.12, 95% CI 1.01, 1.24), and marginally associated with sulphur dioxide (SO2) (RR = 1.03, 95% CI = 0.99, 1.06). All of the relative risks are for a 100 mu g/m(3) increase in the pollutant.Conclusions - Several adequate methods exist to control for weather and seasonality while examining the associations between air pollution and daily counts of mortality and morbidity. In each case, care and judgement are required.

Mortality and morbidity indices are known to depend on changes in meteorological conditions. In Athens, severe adverse health effects following extreme heat conditions have been reported. The usefulness has been investigated of the complex thermohygrometric index (THI), a simple index based on maximum daily temperature and relative humidity, in predicting the health effects of specific meteorological conditions. The values of THI were found to correlate well with more complex bioclimatic indices; the THI could successfully replace temperature and humidity in predicting the daily number of deaths through multiple linear regression modelling. Thus the introduction of THI levels more than 28.5 degrees C and between 26.5 and 28.5 degrees C, through dummy variables, in a regression model explained 40% of the variability in the number of deaths during the months of July and August. During days with THI values less than 26.5 degrees C the mean number of deaths was 33.5, compared to 41.8 when THI was between 26.5 and 28.5 degrees C. The daily number of deaths increased to 108.2 when THI exceeded 28.5 degrees C. From this study, the exact level of THI at which public health measures must be taken was not clear and more work is needed to identify it. However, given its simplicity, the use of THI for predicting meteorological conditions which are adverse to health would appear to be promising in preventive medicine and in health services planning.

Background: Chronic kidney disease (CKD) in HIV-infected patients is associated with both HIV and non-HIV-related factors. Initial renal dysfunction is silent and detectable only by laboratory tests such as the glomerular filtration rate (GFR) estimated by the Cockcroft-Gault equation. Our objective was to assess possible risk factors for CKD in a cohort of Greek HIV-1-infected adults. Methods: Patients in the AMACS (Athens Multicenter AIDS Cohort Study) cohort with at least two available creatinine values were enrolled in the study. Renal dysfunction was defined as eGFR below 90 mL/min/1.73 m 2 . The Kaplan-Meier estimator and the Cox proportional hazards model were used to analyze the occurrence and predictors of renal dysfunction. Results: A total of 1073 patients were enrolled in the study; 255 (23.76%) had baseline eGFR below 90 mL/min/1.73 m 2 and were excluded. Characteristics of the study population: men 88.4%, MSM 62.6%, median baseline age, CD4+ count and viral load were 32.6 years, 413 cells/μL and 3.77 log 10 copies/mL, respectively. 240 (29.3%) patients experienced an eGFR decrease below 90 mL/min/1.73 m 2 during follow-up period. Older age, female gender, heterosexual mode of transmission, lower baseline eGFR (all p

Objective: To evaluate the long-term effects of different boosted protease inhibitors (bPIs) or non-nucleoside reverse transcriptase inhibitors (NNRTIs)-based antiretroviral regimens on lipid levels in HIV seropositive individuals who have not received lipid-lowering agents. Methods: Data consisted of 595 patients participating in the population-based Athens Multicenter Cohort Study who were consistently followed up during 1996–2008. Results: In naive patients, lipid parameters increased sharply during the first 3 months of antiretroviral therapy and reached a plateau level approximately 6–9 months after therapy initiation. The plateau levels remained almost stable for up to 3·5 years. In general, bPIs exerted a more pronounced effect compared to NNRTIs. Conclusions: The administration of PI- or NNRTI-based regimens especially in naive but also in unboosted PI experienced patients provoked a sharp increase in lipid levels that remained stable in higher levels for more than 3 years.

The seasonal variation of neonatal and infant deaths in Greece was analyzed for nine consecutive years (1979-1987) by cause of death, age of death and urbanization of permanent residence. Data were supplied by the National Statistical Service of Greece. Statistical analysis was done using the Edward's method. The seasonal patterns of the number of deaths and death rates were similar. Neonatal deaths in total did not show significant seasonality but postneonatal deaths showed seasonal variation with a peak in the winter, more evident in rural areas. Neonatal deaths from prematurity showed statistically significant seasonal variation with a peak in May. Postneonatal deaths from infections and mainly those from pneumonia showed very significant seasonal variation with a peak in February that was more prominent in rural areas. Seasonal pattern with peak in late winter was also found for postneonatal deaths from injuries. The seasonal patterns for neonatal and postneonatal deaths from sudden infant death syndrome were suggestive of an increased occurrence during the winter months mainly in urban areas.The data were provided by a search in the National Statistical Service of Greece (NSSG), and covered the period from January 1, 1979 through December 31, 1987, for all infant deaths in Greece. Data on live births were taken from the annual statistical reports of NSSG. Statistical analysis was done by means of the Edward's method. The seasonal patterns of the number of deaths and death rates (per 1000 live-born) were almost identical for the 2 parts of the period studied, for the years 1979 and 1987. 1979-83 and 1984-87 were treated separately for the neonatal period and for the postneonatal period. The number of neonatal and early neonatal deaths did not show significant seasonality in the total period, in either the urban or the rural areas, although the peaks for early neonatal deaths in 8 out of 9 studied years were in the spring and summer. The maximum number of postneonatal deaths was observed during January-February, representing a 60-90% increase compared to the minimum number of deaths, and the difference was more evident in the rural areas of residence in 1979-83. Neonatal deaths from prematurity showed statistically significant seasonal variation with a peak in May, more prominent in urban areas. Postneonatal deaths from infections showed statistically significant seasonal variation with a peak in February more prominent in rural areas and in the 1979-83 period. Postneonatal deaths from pneumonia showed very significant seasonal variation, with a peak in February more prominent in rural areas and in the 1979-83 period. Neonatal deaths from sepsis showed increased occurrence in May, whereas postneonatal deaths from sepsis and from enteric infections did not show significant seasonality. Deaths from injuries showed a statistically significant peak during January-February, in both urban and rural areas, in the postneonatal period. Neonatal and postneonatal deaths from sudden infant death syndrome were more common during the winter (December-January-February) in urban areas.

Athens has a serious air pollution problem which became evident in the early 1970s. Studies for the years 1975-1982 have indicated a positive association of sulphur dioxide (SO2) with total daily mortality. Since 1983 the pollution profile in Athens has gradually changed but the levels of smoke, SO2 and carbon monoxide (CO) remain relatively high.The association of air pollution with daily all-cause mortality in Athens for the years 1984-1988 was investigated using daily values of SO2, smoke and CO. Autoregressive models with log-transformed daily mortality as the dependent variable, were used to adjust for temperature and relative humidity (both lagged by 1 day), year, season and day of week, as well as for serial correlations in mortality.Graphic analysis revealed non-linear monotonically increasing relationships between total mortality and SO2, smoke and CO, with steeper exposure-response slopes at lower air pollution levels. Air pollution data lagged by 1 day had the strongest association with daily mortality. In three separate autoregression models for log(SO2), log(smoke) and log(CO) the regression coefficients for each were highly statistically significant (P0.001). Further multiple regression modelling showed that SO2 and smoke are both independent predictors of daily mortality, though to a lesser extent than temperature and relative humidity. The inclusion of CO in the model did not further improve the prediction of daily mortality. The magnitude of association is small, for instance, a 10% reduction in smoke is estimated to decrease daily mortality by 0.75% (95% confidence interval [CI]: 0.51-0.99). However, it cannot be accounted for by climatic and seasonal effects, so that a causal influence of air pollution on daily mortality seems plausible.These findings suggest that current air pollution levels in Athens (and many other industrialized cities) may be responsible for substantial numbers of premature deaths, and hence remain an important public health issue.

Recent updates of human immunodeficiency virus (HIV) treatment guidelines have raised the CD4+ cell count thresholds for antiretroviral therapy initiation from 350 to 500 cells/mm(3) in the United States and from 200 to 350 cells/mm³ in mid- and low-income countries. Robust data of time from HIV seroconversion to CD4+ cell counts of 200, 350, and 500 cells/mm³ are lacking but are needed to inform health care planners of the likely impact and cost effectiveness of these and possible future changes in CD4+ cell count initiation threshold.Using Concerted Action on Seroconversion to AIDS and Death in Europe data from individuals with well-estimated dates of HIV seroconversion, we fitted mixed models on the square root of CD4+ cell counts measured before combined antiretroviral therapy (cART) initiation. Restricting analyses to adults (age16 years), we predicted time between seroconversion and CD4+ cell count200,350, and500 cells/mm³ as well as CD4+ cell count distribution and proportions reaching these thresholds at 1, 2, and 5 years after seroconversion.Median (interquartile range [IQR]) follow-up for the 18495 eligible individuals from seroconversion while cART-free was 3.7 years (1.5, 7). Most of the subjects were male (78%), had a median age at seroconversion of 30 years (IQR, 25-37 years), and were infected through sex between men (55%). Estimated median times (95% confidence interval [CI]) from seroconversion to CD4+ cell count500,350, and200 cells/mm(3) were 1.19 (95% CI, 1.12-1.26), 4.19 (95% CI, 4.09-4.28), and 7.93 (95% CI, 7.76-8.09) years, respectively. Almost half of infected individuals would require treatment within 1 year of seroconversion for guidelines recommending its initiation at 500 cells/mm³, compared with 26% and 9% for guidelines recommending initiation at 350 and 200 cells/mm³, respectively.These data suggest substantial increases in the number of individuals who require treatment and call for early HIV testing.

Background Data from randomized trials on the development of anaemia after interruption of therapy are not well-described. Methods A total of 2,248 patients from the SMART study were included. We used Cox proportional hazards models to investigate development of new (≤12 mg/dl for females and ≤14 mg/dl for males) or worsening (≤8 mg/dl if anaemic at randomization) anaemia and Poisson regression analyses to explore the relationship between anaemia and the development of AIDS, death or non-AIDS events. Results Overall, 759 patients developed new or worsening anaemia: 420/1,106 (38.0%) in the drug conservation (DC) arm and 339/1127 (30.1%) in the viral suppression (VS) arm ( PConclusions Patients who interrupted combination anti-retroviral therapy had a higher risk of new or worsening anaemia. Anaemic patients had a higher incidence of AIDS, non-AIDS defining events or deaths, possibly due to deteriorating health and subclinical disease.