Inmaculada Ferreros

Changes in the incidence of tuberculosis in a cohort of HIV-seroconverters before and after the introduction of HAART.

Objective:To analyse incidence and determinants of tuberculosis in HIV-seroconverters before and after the introduction of HAART. Methods:Data from a multicenter cohort study of 2238 HIV-seroconverters between the 1980s and 2004 were analysed and censored by December 2004. Calendar year at risk intervals were pre-1992, 1992–1996 and 1997–2004. Incident tuberculosis was calculated as cases per 1000 person-years (p-y). Survival analyses using Kaplan–Meier and multivariate Cox regression allowing for late-entry were used. Proportional hazards assumptions were checked with tests based on Schoenfeld residuals. Results:Overall, 173 (7.7%) patients developed tuberculosis over 23 698 p-y at a rate of 7.3 cases per 1000 p-y [95% confidence interval (CI), 6.3–8.5]. Incident tuberculosis was higher in intravenous drug-users (IDUs), 12.3 per 1000 p-y compared with persons infected sexually, 3.8 per 1000 p-y (P < 0.001), and persons with clotting disorders (PCD), 2.7 per 1000 p-y (P < 0.001). A decreasing tuberculosis incidence trend was observed from 1995 in all categories. Highest tuberculosis rates, 44 per 1000 p-y, were observed prior to 1997 in IDUs infected with HIV for 11 years. In multivariable analyses women were less likely to develop tuberculosis [relative hazard (RH), 0.62; 95% CI, 0.41–0.96; P < 0.05) and IDUs were more likely to develop tuberculosis (RH, 3.0; 95% CI, 1.72–5.26, P < 0.001). In the HAART era, the hazard of developing tuberculosis was 70% lower (RH, 0.31; 95% CI, 0.17–0.54; P < 0.001). Before 1997, the risk of tuberculosis increased with time since HIV seroconversion, whereas it remained nearly constant in the HAART era. Conclusions:Since the mid-1990s important decreases in tuberculosis have been observed in HIV-seroconverters that probably reflect the impact of both HAART and tuberculosis control programmes.

Impact of hepatitis C infection on long-term mortality of injecting drug users from 1990 to 2002: differences before and after HAART.

Objective:To assess the impact of HIV and hepatitis C virus (HCV) infection on long-term mortality in injecting drug users (IDU). Design:Community-based prospective cohort study. Methods:Mortality data from follow-up in clinical sites and the Mortality Registry by December 2002 were collected for 3247 IDU who attended three centres for voluntary counselling and testing for HIV/AIDS, HCV and hepatitis B virus (HBV) in 1990–1996. Mortality rates by Poisson regression were adjusting for age, sex, duration of drug use, education, HBV and calendar period (1990–1997 and 1998–2002). Results:Overall, 11.2% were HIV/HCV negative, 43.7% positive only for HCV and 45.1% positive for both. During 26 772 person-years of follow-up, 585 deaths were detected (2.19/100 person-years). Before 1997, HIV/HCV-positive subjects had a five-fold increase in risk of death [relative risk (RR), 5.4; 95% confidence interval (CI), 2.5–11.4] compared with those negative for both; after 1997, a three-fold increase was observed (RR, 2.7; 95% CI, 1.7–4.2). Being HCV positive/HIV negative was not associated with an increase in the risk of death either before (RR, 1.3; 95% CI, 0.6–2.9) or after (RR, 1.2; 95% CI, 0.8–1.9) 1997 compared with HCV/HIV negative. While increases in mortality were seen in those HCV/HIV negative (RR, 1.6; 95% CI, 0.7–3.7) and those only positive for HCV (RR, 1.5; 95% CI, 1.0–2.1), a 20% reduction among coinfected IDUs was observed after 1997 (interaction P = 0.033). Conclusions:HCV/HIV coinfection has had a large impact on mortality in IDU. After 1997, mortality increased in HIV negative/HCV positive subjects and decreased in HIV positive/HCV positive.

Gender differences in progression to AIDS and death from HIV seroconversion in a cohort of injecting dug users from 1986 to 2001

Background: Although the consensus is that gender does not influence HIV progression, its relevance may depend on the setting. Aim: To study gender differences in HIV progression to AIDS and death from 1986 to 2001 in a cohort of injecting drug user (IDU) seroconverters in Spain. Methods: Risk of AIDS and death in persons infected for the same length of time were compared through Kaplan-Meier, allowing for late entry, and Cox regression adjusting for gender, age, and calendar period (before 1992, 1992–1995, 1996–1998, 1999–2001) fitted as time dependent covariates. Results: Of 929 IDU, 24.7% were women. Median seroconversion year was 1993.3 for men and women. 44% of women and 34% of men received antiretroviral therapy. Risk of AIDS was lower in women in univariate (hazard ratio (HR) 0.72; 95%CI:0.51 to 1.01) and multivariate analyses (HR 0.73 95%CI:0.52 to 1.03). A 46% reduction in risk of AIDS for period 1999–2001 compared with 1992–1995 was seen in both men and women (HR: 0.56 (95%CI:0.36 to 0.87). As for mortality, women’s risk of death was lower univariately (HR 0.67 95%CI:0.45 to 0.99) although compared with 1992–95, men experienced a 34% reduction in mortality during 1999–2001 (HR 0.66 95%CI:0.40 to 1.01), which was not statistically significant in women. Conclusions: HIV progression was lower in female IDU before and after 1997 and their uptake of antiretroviral therapy was higher than male IDU. The inability to detect a reduction in mortality for women during 1999–2001 is probably attributable to lack of power. Differences in severity of addiction, drug using patterns, and competing causes of death may explain these findings.

Trends in HIV testing, serial HIV prevalence and HIV incidence among people attending a Center for AIDS Prevention from 1988 to 2003

Aim: To analyse trends in HIV testing, serial HIV prevalence and HIV incidence among people who underwent voluntary testing in a Center for AIDS Prevention in Valencia, Spain. Methods: Open cohort study including all subjects who went to the Center for AIDS Prevention from 1988 to 2003. Information on sociodemographic variables and HIV test results was collected. Serial prevalence and incidence rates were calculated, and joinpoint regression was used to identify changes in trends over time. Results: 21 241 subjects were analysed; 67% men, 27% injecting drug users (IDUs), 43% heterosexuals and 13% men who have sex with men (MSM). From 1988 to 1990, IDUs accounted for 57% of clinic attenders, decreasing to 14% by 1997–2003, accompanied by an increase in heterosexuals. Overall, HIV prevalence for the whole period was 15%, dropping from 35% to <10% after 1999 and to 3% by 2003, when HIV prevalence was 26% in IDUs, 6% in MSM and 2% in heterosexuals. Total HIV incidence was 2.5%. From 1988 to 1990, HIV incidence ranged from 6% to 8%, and a gradual and progressive decline observed from 1990 onwards. From 1995 onwards, HIV incidence was <2%. The highest incidence rate is seen in IDUs, 7–12% in the first period and 4–5% at the end. Among MSM, a change in the decreasing trend is seen by 1998, and increases in incidence are detected by 2002–3. Conclusions: Serial HIV prevalence has markedly decreased from 1988 in all transmission categories, although it is still high. With regard to HIV incidence, the drop has been marked too, although a worrying increase, that requires further follow-up, has been detected in MSM in the past 2 years.

The shifting pattern of cause-specific mortality in a cohort of human immunodeficiency virus-infected and non-infected injecting drug users

AIMS To monitor changes in cause-specific mortality before and after 1997 according to human immunodeficiency virus (HIV) serological status in a cohort of injecting drug users (IDUs) observed for a 17-year period (1987--2004). DESIGN Community-based prospective cohort study of IDUs recruited in three acquired immunodeficiency virus (AIDS) prevention centres (1987--96) and followed-up until to 2004. METHODS We obtained annual overall mortality rates and mortality rates by specific causes according to HIV status. Poisson regression models were adjusted to compare mortality rates between calendar periods. Significant changes in slope trends were evaluated by join-point regression. Disease-specific mortality rates were estimated using competing risk models. FINDINGS From 7186 IDUs recruited (80677.218 person-years), 1589 deaths were observed with an overall mortality rate of 19.7 per 1000 person-years (95% CI, 18.8-20.7). This rate decreased from 22.9 per 1000 (95% CI, 21.4-24.7) before 1997 to 17.4 per 1000 (95% CI, 16.3-18.6) after 1997 [relative risk (RR) 0.83; 95% confidence interval (CI), 0.75-0.92]. Risk of death for HIV-positive was four times higher than for HIV-negative (RR 4.08; 95% CI, 3.63-4.58). Among HIV-positive individuals a significantly decreased change point in trend was found in 1997 for both total and AIDS mortality. HIV-negative individuals showed a similar pattern for drug overdose, suicide and accident mortality. Both groups showed an increase in proportional mortality by liver-related causes, cardiovascular diseases and cancer. Furthermore, a progressively increasing trend was observed for the three causes. However, there were no significant differences according to serological groups. CONCLUSIONS Cardiovascular and cancer mortality are increasing among IDUs, but the increases are not related to HIV infection. We have not found a link between highly active antiretroviral therapy (HAART) introduction and increases in mortality for specific causes.

Progression of HIV infection and mortality by hepatitis C infection in patients with haemophilia over 20 years

Summary.  Hepatitis C virus (HCV) infection is an important cause of mortality in human immune deficiency virus (HIV)‐positive haemophiliacs. This study describes progression to AIDS, death from HCV end‐stage liver disease (ESLD) and all‐cause mortality over 20 years. All HIV‐positive haemophiliacs in La Paz University Hospital were included in this cohort. HIV seroconversion was estimated using mathematical techniques for interval‐censored data from 1979 to 1985. Poisson regression was used to estimate rates of AIDS, death from ESLD and all causes in different periods: before 1988, 1988–89, 1990–91, 1992–93, 1994–95, 1996–97 and 1998–2001 using competing risk models. Among 383 cohort members, global AIDS incidence was 9.7 per 100 person‐years, peaking in 1992–93 and dropping by 87% in 1998–2001 compared with before 1988 [incidence rate ratio (IRR) 0.13; 95% CI: 0.03–0.53]. Overall mortality was 7.5 per 100 person‐years, was highest from 1992 to 1997, and fell by 66% in 1998–2001 compared with before 1988 (IRR 0.34; 95% CI: 0.14–0.81). Eighteen (5%) persons died of ESLD which represented 19% of deaths before 1988, 4% during 1988–89, 1990–91 and 1992–93, 2% in 1994–95, 10% in 1996–97 and 33% in 1998–2001. Overall death rate from ESLD was 0.5 cases per 100 person‐years with no statistically significant trend observed over time. Important reductions in HIV disease progression to AIDS and death have been observed from 1998 to 2001, and can be attributed to highly active antiretroviral therapy. Although no increase in the rate of HCV‐related deaths can be demonstrated, HCV accounts for an increasing proportion of deaths in the recent years.

Educational level and HIV disease progression before and after the introduction of HAART: a cohort study in 989 HIV seroconverters in Spain

Objectives To analyse the effect of educational level on the progression from HIV seroconversion to highly active antiretroviral therapy (HAART) requirement, HAART initiation, AIDS and death from any cause at different periods of the HIV epidemic in Spain. Methods Open, prospective, multicentre cohort of HIV seroconverters set up in 1983. The risk of progression was calculated by the multiple decrements method. Effect of educational level was estimated by Fine and Gray model, adjusting for sex, HIV transmission category, age and method to estimate seroconversion. Calendar period was introduced as a variable that could change over time (<1997; 1997–2003; >2003). Results Up to 2009, 989 HIV seroconverters with information on educational level were identified. Some 52% and 48% had a low and a high educational level respectively. Persons with higher education had 32% lower risk of death (HR: 0.68; 95% CI 0.45 to 1.03). Regarding progression to AIDS, educational level had no effect in the pre-HAART era (HR: 1.47; 95% CI 0.91 to 2.38), but did show an effect in the period 1997–2003 (HR: 0.58; 95% CI 0.34 to 0.99), which was accentuated after 2004 (HR: 0.26; 95% CI 0.10 to 0.68). No difference was found in time to HAART requirement or initiation. Conclusions Results suggest that, despite similar access to HAART, persons with low educational level are at increased risk of HIV disease progression, highlighting the impact of social inequities on health. The availability of more effective treatments over time will strengthen the protective effect of higher education on the development of AIDS.

Imputación del instante de seroconversión al VIH en cohortes de hemofílicos

Objectives: To describe the methods used to impute HIV seroconversion date in the haemophiliac cohorts from GEMES project and to validate its use. Method: 632 haemophiliacs coming from three hemophilia units identified as HIV+ and 1.092 individuals coming from 5 project GEMES cohorts with a seroconversion window (time among test HIV‐ and HIV+) less than 3 years where mid point (PM) was assumed as seroconversion date. For both groups, seroconversion date was imputed after estimating the probability distribution of seroconversion by means of the EM algorithm. Two imputation methods are used: one obtained from the expected value and the other from the geometric mean of 5 random samples. from the estimated distribution. Imputations have been validated in the non haemophiliacs cohorts comparing with the PM seroconversion date. Also AIDS free time and survival from the different seroconversion imputed dates were compared. Results: Median seroconversion date is located in May of 1993 for the non haemophiliacs and in 1982 for the haemophiliacs. Not big differences are observed among the imputed seroconversion dates and the mid-point seroconversion date in the non-haemophiliac cohorts. Similar results are found for the haemophiliac cohorts. Also no differences are observed in the estimated AIDS-free time for both groups of cohorts. Conclusions: Geometric mean imputation from several random samples provides a good estimate of the HIV seroconversion date that can be used to estimate AIDS-free time and survival in haemophiliac cohorts where seroconversion date is ignored.

Marginal structural models application to estimate the effects of antiretroviral therapy in 5 cohorts of HIV seroconverters

OBJECTIVES Standard methods to evaluate population effectiveness of treatments in observational studies have important limitations to appropriately adjust for time-dependent confounders. In this paper, we describe a recently developed methodological approach, marginal structural models (MSM), and use it to estimate the effectiveness of highly active antiretroviral therapy (HAART) on AIDS or death incidence. SUBJECTS AND METHODS We analyzed all subjects followed after 1997 as part of the GEMES project (comprised by several cohorts of HIV seroconverters in Spain) and who had not used HAART before the start of follow-up. To estimate the effect of HAART on AIDS or death incidence, we estimated the parameters of a marginal structural Cox model by fitting an inverse probability weighted logistic regression model. The estimation of the weights was based on CD4 count, time since seroconversion, sex, age, transmission category and previous treatment. RESULTS 917 eligible subjects were followed for an average of 3.4 years and we observed 139 events. 42.1% of the participants received HAART during the study. The estimated rate ratio was 1.01 (95% confidence interval [CI], 0.68-1.49) using a Cox model without covariates and 0.90 (95% CI, 0.61-1.32) using a Cox model with time-dependent covariates. The causal rate ratio estimated for MSM was 0.74, (95% CI, 0.49-1.12). CONCLUSIONS The beneficial effect of HAART estimated by the MSM, but largely missed by conventional methods, is consistent with the findings of previous randomized studies. The MSM appropriately adjusted for the time-dependent covariate CD4 count, which is both a time-varying confounder and is affected by prior treatment.

Aplicación de modelos estructurales marginales para estimar los efectos de la terapia antirretroviral en 5 cohortes de seroconvertores al virus de la inmunodeficiencia humana

Los metodos convencionales tienen limitaciones para ajustar por factores de confusion dependientes del tiempo para evaluar la efectividad poblacional de tratamientos en estudios observacionales. En este trabajo se muestra un nuevo tipo de metodologia, los modelos estructurales marginales (MEM), y se estima la efectividad de la terapia antirretroviral de gran actividad (TARGA) sobre la incidencia de sida o muerte. Sujetos y metodos: Se identificaron los sujetos sin TARGA seguidos a partir de 1997 en las cohortes de seroconvertores al virus de la inmunodeficiencia humana (VIH) del proyecto GEMES (Grupo de Estudio Multicentrico Espanol de Seroconvertores). Para estimar el efecto sobre la incidencia de sida o muerte, se obtuvieron los parametros de un MEM mediante una regresion logistica ponderada por probabilidad inversa. La estimacion de los pesos se baso en el recuento de CD4, el tiempo desde la seroconversion, el sexo, la edad, la categoria de trasmision y el tratamiento previo. Resultados: Los 917 sujetos elegibles se siguieron durante una media de 3,4 anos, durante los cuales se observaron 139 desenlaces de interes. El 42,1% de los participantes recibio TARGA durante el estudio. La tasa relativa fue de 1,01 (intervalo de confianza &(IC&) del 95%, 0,68-1,49) mediante un modelo de Cox convencional sin covariables, y de 0,90 (IC del 95%, 0,61-1,32) mediante un modelo de Cox convencional con covariables cambiantes en el tiempo. La tasa relativa causal estimada por un MEM fue de 0,74 (IC del 95%, 0,49-1,12). Conclusiones: El efecto beneficioso del TARGA encontrado por los MEM esta bien establecido, pero los modelos convencionales no pudieron detectarlo. El uso de un MEM permitio ajustar apropiadamente por la variable CD4, que es a la vez una variable de confusion dependiente del tiempo y esta afectada por el uso previo de tratamiento.