Does antiretroviral treatment increase the infectiousness of smear-positive pulmonary tuberculosis?

BACKGROUND: Understanding of the effects of human immunodeficiency virus (HIV) infection and antiretroviral treatment (ART) on Mycobacterium tuberculosis transmission dynamics remains limited. We undertook a cross-sectional study among household contacts of smear-positive pulmonary tuberculosis (TB) cases to assess the effect of established ART on the infectiousness of TB. METHOD: Prevalence of tuberculin skin test (TST) positivity was compared between contacts of index cases aged 2-10 years who were HIV-negative, HIV-positive but not on ART, on ART for <1 year and on ART for 1 year. Random-effects logistic regression was used to take into account clustering within households. RESULTS: Prevalence of M. tuberculosis infection in contacts of HIV-negative patients, HIV-positive patients on ART 1 year and HIV-positive patients not on ART/on ART <1 year index cases was respectively 44%, 21% and 22%. Compared to contacts of HIV-positive index cases not on ART or recently started on ART, the odds of TST positivity was similar in contacts of HIV-positive index cases on ART 1 year (adjusted OR [aOR] 1.0, 95%CI 0.3-3.7). The odds were 2.9 times higher in child contacts of HIV-negative index cases (aOR 2.9, 95%CI 1.0-8.2). CONCLUSIONS: We found no evidence that established ART increased the infectiousness of smear-positive, HIV-positive index cases.

Risk factors for Mycobacterium tuberculosis infection in 2-4 year olds in a rural HIV-prevalent setting.

BACKGROUND: Mycobacterium tuberculosis infection in children acts as a sentinel for infectious tuberculosis. OBJECTIVE: To assess risk factors associated with tuberculous infection in pre-school children. METHOD: We conducted a population-wide tuberculin skin test (TST) survey from January to December 2012 in Malawi. All children aged 2-4 years residing in a demographic surveillance area were eligible. Detailed demographic data, including adult human immunodeficiency virus (HIV) status, and clinical and sociodemographic data on all diagnosed tuberculosis (TB) patients were available. RESULTS: The prevalence of M. tuberculosis infection was 1.1% using a TST induration cut-off of 15 mm (estimated annual risk of infection of 0.3%). The main identifiable risk factors were maternal HIV infection at birth (adjusted OR [aOR] 3.6, 95%CI 1.1-12.2), having three or more adult members in the household over a lifetime (aOR 2.4, 95%CI 1.2-4.8) and living in close proximity to a known case of infectious TB (aOR 1.6, 95%CI 1.1-2.4), modelled as a linear variable across categories (>200 m, 100-200 m, <100 m, within household). Less than 20% of the infected children lived within 200 m of a known diagnosed case. CONCLUSION: Household and community risk factors identified do not explain the majority of M. tuberculosis infections in children in our setting.

Large-scale whole genome sequencing of M. tuberculosis provides insights into transmission in a high prevalence area.

To improve understanding of the factors influencing tuberculosis transmission and the role of pathogen variation, we sequenced all available specimens from patients diagnosed over 15 years in a whole district in Malawi. Mycobacterium tuberculosis lineages were assigned and transmission networks constructed, allowing ≤10 single nucleotide polymorphisms (SNPs) difference. We defined disease as due to recent infection if the network-determined source was within 5 years, and assessed transmissibility from forward transmissions resulting in disease. High-quality sequences were available for 1687 disease episodes (72% of all culture-positive episodes): 66% of patients linked to at least one other patient. The between-patient mutation rate was 0.26 SNPs/year (95% CI 0.21-0.31). We showed striking differences by lineage in the proportion of disease due to recent transmission and in transmissibility (highest for lineage-2 and lowest for lineage-1) that were not confounded by immigration, HIV status or drug resistance. Transmissions resulting in disease decreased markedly over time.

Impact of changing diagnostic criteria for smear-positive tuberculosis: a cohort study in Malawi.

We assessed the impact on measured burden and outcomes of the revised World Health Organization and Malawi guidelines reclassifying people with single (including 'scanty') positive smears as smear-positive pulmonary tuberculosis cases. In a retrospective cohort in rural Malawi, 567 (34%) of 1670 smear-positive episodes were based on single positive smears (including 176 with scanty smears). Mortality rates and the proportion starting treatment were similar in those with two positive smears or single, non-scanty smears. Those with single scanty smears had higher mortality and a lower proportion starting treatment. The reclassification will increase the reported burden substantially, but should improve treatment access.

The benefits to communities and individuals of screening for active tuberculosis disease: a systematic review.

BACKGROUND: Screening for tuberculosis (TB) disease aims to improve early TB case detection. The ultimate goal is to improve outcomes for people with TB and to reduce Mycobacterium tuberculosis transmission in the community through improved case detection, reduction in diagnostic delays and early treatment. Before screening programmes are recommended, evidence is needed of individual and/or community-level benefits. METHODS: We conducted a systematic review of the literature to assess the evidence that screening for TB disease 1) initially increases the number of TB cases initiated on anti-tuberculosis treatment, 2) identifies cases earlier in the course of disease, 3) reduces mortality and morbidity, and 4) impacts on TB epidemiology. RESULTS: A total of 28 798 publications were identified by the search strategy: 27 087 were excluded on initial screening and 1749 on full text review, leaving 62 publications that addressed at least one of the study questions. Screening increases the number of cases found in the short term. In many settings, more than half of the prevalent TB cases in the community remain undiagnosed. Screening tends to find cases earlier and with less severe disease, but this may be attributed to case-finding studies using more sensitive diagnostic methods than routine programmes. Treatment outcomes among people identified through screening are similar to outcomes among those identified through passive case finding. Current studies provide insufficient evidence to show that active screening for TB disease impacts on TB epidemiology. CONCLUSION: Individual and community-level benefits from active screening for TB disease remain uncertain. So far, the benefits of earlier diagnosis on patient outcomes and transmission have not been established.

Reproductive preferences and contraceptive use: a comparison of monogamous and polygamous couples in northern Malawi.

There is now widespread agreement on the importance of men's role in reproductive decision-making. Several studies have argued that fertility preferences and their translation into behaviour differ between polygamous and monogamous unions. Studies investigating the dominance of men's preferences over women's preferences, in cases of couple disagreement, found mixed evidence of the effect of polygamy. However, an often cited limitation of these studies has been the inability to link husband's intention with each of his wives in a polygamous union. By adding fertility-intention questions to an on-going Demographic Surveillance Site in Karonga District in northern Malawi the fertility preferences and contraceptive use of husbands and wives were investigated. An analysis of the relationship between the level of agreement and disagreement between husbands' and wives' fertility preferences was then performed to gain insight into the reproductive decision-making process of polygamous couples.

Quantifying errors in the estimation of tuberculosis mortality in a population of South African miners.

BACKGROUND: All-cause mortality, based on national tuberculosis programme (NTP) register deaths, may under- or overestimate tuberculosis (TB) specific mortality in the population. OBJECTIVE: To assess the factors influencing this measurement in a single large population with high TB prevalence and mortality. METHODS: Routinely collected data on TB cases and treatment outcomes were linked to population data from a cohort of South African miners from 1995 to 2008. Vital status and cause of death were determined from multiple sources, including the TB programme, death register and autopsy. RESULTS: The TB mortality rate, based on 430 deaths on the TB register, was 192/100,000 person-years (py). Many of these deaths (57%) were not caused by TB, and 483 TB deaths were identified outside the programme. Overall, there were 674 TB-specific deaths; the TB-specific mortality rate was 302/100,000 py. These deaths included 191 (28%) on the TB register, 23 (3%) among defaulters/transfers, 153 (23%) after anti-tuberculosis treatment and 307 (46%) in men who had never been on the programme. CONCLUSIONS: This study highlights methodological issues in estimating TB mortality. In this population, a method using the product of TB incidence and case fatality consistently underestimated TB mortality. Accurate estimates of TB-specific mortality are crucial for the proper evaluation of TB control programmes.

Interferon release does not add discriminatory value to smear-negative HIV-tuberculosis algorithms.

Clinical algorithms for evaluating HIV-infected individuals for tuberculosis (TB) prior to isoniazid preventive therapy (IPT) perform poorly, and interferon-γ release assays (IGRAs) have moderate accuracy for active TB. It is unclear whether, when used as adjunct tests, IGRAs add any clinical discriminatory value for active TB diagnosis in the pre-IPT assessment. 779 sputum smear-negative HIV-infected persons, established on or about to commence combined antiretroviral therapy (ART), were screened for TB prior to IPT. Stepwise multivariable logistic regression was used to develop clinical prediction models. The discriminatory ability was assessed by receiver operator characteristic area under the curve (AUC). QuantiFERON-TB Gold in-tube (QFT-GIT) was evaluated. The prevalence of smear-negative TB by culture was 6.4% (95% CI 4.9-8.4%). Used alone, QFT-GIT and the tuberculin skin test (TST) had comparable performance; the post-test probability of disease based on single negative tests was 3-4%. In a multivariable model, the QFT-GIT test did not improve the ability of a clinical algorithm, which included not taking ART, weight <60 kg, no prior history of TB, any one positive TB symptom/sign (cough ≥ 2 weeks) and CD4+ count <250 cells per mm(3), to discriminate smear-negative culture-positive and -negative TB (72% to 74%; AUC comparison p=0.33). The TST marginally improved the discriminatory ability of the clinical model (to 77%, AUC comparison p=0.04). QFT-GIT does not improve the discriminatory ability of current TB screening clinical algorithms used to evaluate HIV-infected individuals for TB ahead of preventive therapy. Evaluation of new TB diagnostics for clinical relevance should follow a multivariable process that goes beyond test accuracy.

Married to M. tuberculosis: risk of infection and disease in spouses of smear-positive tuberculosis patients.

OBJECTIVES: To quantify the risk of infection and disease in spouses of tuberculosis patients and the extent to which intervention could reduce the risk in this highly exposed group. METHODS: We compared HIV prevalence, TB prevalence and incidence and tuberculin skin test (TST) results in spouses of TB patients and community controls. HIV-positive spouses were offered isoniazid preventive therapy (IPT), and TST was repeated at 6, 12 and 24 months. RESULTS: We recruited 148 spouses of smear-positive patients ascertained prospectively and 3% had active TB. We identified 203 spouses of previously diagnosed smear-positive patients, 11 had already had TB, and the rate of TB was 2.4 per 100 person years(py) over 2 years (95% CI 1.15-5.09). 116 were found alive and recruited. HIV prevalence was 37% and 39% in the prospective and retrospective spouse groups and 17% in controls. TST was ≥10 mm in 80% of HIV negative and in 57% of HIV-positive spouses ascertained retrospectively; 74% HIV negative and 62% HIV-positive spouses ascertained prospectively, and 48% HIV negative and 26% HIV-positive community controls. Of 54 HIV-positive spouses, 18 completed 6-month IPT. At 2 year follow-up, 87% of surviving spouses had TST ≥10 mm and the rate of TB was 1.1 per 100 py (95% CI 0.34-3.29). CONCLUSIONS: Spouses are a high-risk group who should be screened for HIV and active TB. TST prevalence was already high by the time the spouses were approached but further infections were seen to occur. Uptake and adherence to IPT was disappointing, lessening the impact of short-duration therapy.

Human immunodeficiency virus associated tuberculosis more often due to recent infection than reactivation of latent infection.

BACKGROUND: It is unclear whether human immunodeficiency virus (HIV) increases the risk of tuberculosis (TB) mainly through reactivation or following recent Mycobacterium tuberculosis (re)infection. Within a DNA fingerprint-defined cluster of TB cases, reactivation cases are assumed to be the source of infection for subsequent secondary cases. As HIV-positive TB cases are less likely to be source cases, equal or higher clustering in HIV-positives would suggest that HIV mainly increases the risk of TB following recent infection. METHODS: A systematic review was conducted to identify all studies on TB clustering and HIV infection in HIV-endemic populations. Available individual patient data from eligible studies were pooled to analyse the association between clustering and HIV. RESULTS: Of seven eligible studies, six contributed individual patient data on 2116 patients. Clustering was as, or more, likely in the HIV-positive population, both overall (summary OR 1.26, 95%CI 1.0-1.5), and within age groups (OR 1.50, 95%CI 0.9-2.3; OR 1.00, 95%CI 0.8-1.3 and OR 2.57, 95%CI 1.4-5.7) for ages 15-25, 26-50 and >50 years, respectively. CONCLUSIONS: Our results suggest that HIV infection mainly increases the risk of TB following recent M. tuberculosis transmission, and that TB control measures in HIV-endemic settings should therefore focus on controlling M. tuberculosis transmission rather than treating individuals with latent M. tuberculosis infection.

Human immunodeficiency virus increases the risk of tuberculosis due to recent re-infection in individuals with latent infection.

BACKGROUND: Human immunodeficiency virus associated tuberculosis (TB) disease can follow reactivation of latent Mycobacterium tuberculosis infection or recent (re-)infection with M. tuberculosis. If contemporary TB cases share identical M. tuberculosis strains (i.e., are 'clustered'), the episode is likely to have followed recent (re-)infection, irrespective of evidence of previous latent infection. METHODS: Individuals experiencing a first TB episode between 1996 and 2008 in Karonga District, Northern Malawi, were included if information on M. tuberculosis infection status (from tuberculin tests) before 1990 and a DNA fingerprint from the TB episode were available. We explored differences in proportion clustered by prior M. tuberculosis infection status and HIV status, adjusting for age, sex, bacille Calmette-Guérin scar status and time since tuberculin testing. RESULTS: Of 79 HIV-negative TB cases, those with previous M. tuberculosis infection were much less likely to be clustered than cases without prior infection (29% vs. 77%, adjusted OR = 0.15, 95%CI 0.04-0.59). Among 119 HIV-positive TB cases, clustering was similar in both groups (88% vs. 84%, adjusted OR = 1.85, 95%CI 0.41-8.29). DISCUSSION: HIV infection appears to increase the risk of TB following recent re-infection in patients with latent M. tuberculosis infection. Our results add to the mounting evidence that HIV-associated TB mainly follows recent M. tuberculosis infection.

Mortality and health among internally displaced persons in western Kenya following post-election Violence; 2008: novel use of demographic surveillance

Objective To evaluate mortality and morbidity among internally displaced persons (IDPs) who relocated in a demographic surveillance system (DSS) area in western Kenya following post-election violence. Methods In 2007; 204 000 individuals lived in the DSS area; where field workers visit households every 4 months to record migrations; births and deaths. We collected data on admissions among children 5 years of age in the district hospital and developed special questionnaires to record information on IDPs. Mortality; migration and hospitalization rates among IDPs and regular DSS residents were compared; and verbal autopsies were performed for deaths. Findings Between December 2007 and May 2008; 16 428 IDPs migrated into the DSS; and over half of them stayed 6 months or longer. In 2008; IDPs aged 15.49 years died at higher rates than regular residents of the DSS (relative risk; RR: 1.34; 95confidence interval; CI: 1.004.1.80). A greater percentage of deaths from human immunodeficiency virus (HIV) infection occurred among IDPs aged . 5 years (53) than among regular DSS residents (25.29) (P 0.001). Internally displaced children 5 years of age did not die at higher rates than resident children but were hospitalized at higher rates (RR: 2.95; 95CI: 2.44.3.58). Conclusion HIV-infected internally displaced adults in conflict-ridden parts of Africa are at increased risk of HIV-related death. Relief efforts should extend to IDPs who have relocated outside IDP camps; particularly if afflicted with HIV infection or other chronic conditions

What has Karonga taught us? Tuberculosis studied over three decades.

This paper summarises tuberculosis (TB) research over almost 30 years in Karonga District, northern Malawi, an area typical of much of rural Africa. The dominant factor has been the human immunodeficiency virus (HIV), which arrived in the district about 1980, leading to an increase in TB incidence to a peak of approximately 65 smear-positive pulmonary cases per 100000 population in 2000. Tuberculin surveys indicate annual risks of Mycobacterium tuberculosis infection of approximately 1%; thus, most of the population is uninfected and at risk of primary infection and disease. Molecular epidemiological studies demonstrate that about two thirds of TB arises from recent infection, but recognisable recent contact is responsible for only about 10% of disease. By 2001, 57% of TB was directly attributable to HIV, implying that it would have declined were it not for HIV. HIV infection increases the risk of TB most among young adults, and greatly increases the risk of recurrence from new infection after treatment. Mortality rates in the HIV-infected are high, but there is no association of HIV with drug resistance. Other risk factors with relatively smaller effects include age and sex, contact, several genetic polymorphisms and area. Neither one nor two doses of the bacille Calmette-Guérin (BCG) vaccine provides protection against adult pulmonary TB, despite protecting against leprosy. Skin test surveys, cohort studies and comparative immunological studies with the UK suggest that exposure to environmental mycobacteria provides some protection against TB and that BCG's failure is attributable partly to this widespread heterologous exposure masking effects of the vaccine. Drug resistance has remained constant (<10%) over more than 20 years. Immunotherapy with M. vaccae provided no benefits, but treatment of HIV-positive patients with cotrimoxazole reduced mortality. The Karonga programme illustrates the value of long-term population-based studies to investigate the natural history of TB and to influence TB control policy. Current studies focus on immunological markers of infection, disease and protection, and on elucidating the impact of antiretroviral treatment on TB incidence at population level.

Molecular epidemiology of tuberculosis in England, 1998.

SETTING: England. OBJECTIVE: To investigate the proportion of tuberculosis (TB) cases attributable to recent transmission and factors associated with clustering. DESIGN: Demographic, clinical and microbiological surveillance data were collated from all new culture-confirmed cases in 1998. Using insertion sequence (IS) 6110 restriction fragment length polymorphism (RFLP) typing, strains were classified as clustered (identical patterns) or unique and risk factors were determined using multivariable logistic regression. RESULTS: RFLP patterns were available for 2265 of 3713 (61%) cases: 1808 had >or=5 IS6110 copies, while 372 cases were in 152 clusters, giving an estimated proportion due to recent transmission of 12.2%.Pulmonary disease (aOR 1.6; 95%CI 1.1-2.2), previous treatment (aOR 3.7; 2.2-6.5) and homelessness (aOR 5.5; 1.2-24.1) were independent risk factors for clustering. Fourteen per cent of patients of Indian subcontinent origin were clustered compared with 27% of white patients. Many clusters spanned ethnic groups (45%) and geographical regions (47%). CONCLUSION: The calculated proportion of TB cases due to recent transmission is low.Adjusting for missed cases and study duration, it increases to 27.6%. Many cases may arise from reactivation or acquisition outside England. Transmission within England accounted for approximately one in four cases and occurred over wide geographic areas, between ethnic groups and among the homeless. Molecular epidemiology can inform local and national public health action.

Population-level effect of HSV-2 therapy on the incidence of HIV in sub-Saharan Africa.

BACKGROUND: Herpes simplex virus type 2 (HSV-2) infection increases acquisition and transmission of HIV, but the results of trials measuring the impact of HSV-2 therapy on HIV genital shedding and HIV acquisition are mixed, and the potential impact of HSV-2 therapy on the incidence of HIV at the population level is unknown. METHODS: The effects of episodic and suppressive HSV-2 therapy were simulated using the individual-level model STDSIM fitted to data from Cotonou, Benin (relatively low HIV prevalence) and Kisumu, Kenya (high HIV prevalence). Clinician- and patient-initiated episodic therapy, started when symptomatic, were assumed to reduce ulcer duration. Suppressive therapy, given regardless of symptoms, was also assumed to reduce ulcer frequency and HSV-2 infectiousness. RESULTS: Clinician-initiated episodic therapy in the general population had almost no effect on the incidence of HIV. The impact of patient-initiated therapy was higher because of earlier treatment initiation, but still low (<5%) unless symptom recognition and treatment-seeking behaviour were very high. Suppressive therapy given to female sex workers (FSW) in Kisumu had little effect on population HIV incidence. In Cotonou, suppressive therapy in FSW with high coverage and long duration reduced population HIV incidence by >20% in the long term. Impact was increased in both cities by also treating a proportion of their clients. Long-term suppressive therapy with high coverage in the general population could reduce HIV incidence by more than 30%. CONCLUSIONS: These results show that HSV-2 therapy could potentially have a population-level impact on the incidence of HIV, especially in more concentrated epidemics. However, a substantial impact requires high coverage and long duration therapy, or very high symptom recognition and treatment-seeking behaviour.

Herpes simplex virus type 2 trends in relation to the HIV epidemic in northern Malawi.

OBJECTIVES: It is unclear whether the high prevalence of herpes simplex virus type 2 (HSV-2) found in much of Africa predates the HIV epidemic or is, to some extent, a consequence of it. HSV-2 prevalence trends in a rural African community were assessed over a period in which HIV prevalence rose sharply, and antenatal clinic (ANC) surveillance was explored as a method of estimating community HSV-2 prevalence. METHODS: HSV-2 seroprevalence was determined among community controls seen for case-control studies of mycobacterial disease in Karonga district, Malawi, in 1988-90, 1998-2001 and 2002-5, and in women attending ANC as part of surveillance for HIV in 1999-2000. Over this period HIV prevalence rose from 4% to 12%. RESULTS: HSV-2 prevalence in all periods increased sharply with age and was higher in women than in men. After excluding migrants, there was no evidence of change in HSV-2 prevalence in the different periods. Women in the ANC group had lower HSV-2 prevalence than those in the community, but the ANC prevalence was a good approximation to the combined male and female prevalence for the same age group. CONCLUSIONS: This study suggests that HSV-2 was already widespread before the HIV epidemic and has not been greatly influenced by it. It also demonstrates that ANC surveillance may be useful for estimating community HSV-2 prevalence.

Assessment and evaluation of contact as a risk factor for tuberculosis in rural Africa.

SETTING: A rural district in Malawi. OBJECTIVE: To determine the effect of inaccurate recall on estimates of the proportion of tuberculosis (TB) cases attributable to contact with identifiable prior cases. DESIGN: Case-control study of laboratory-confirmed TB cases and community controls, comparing family, household and area contacts identified from a database of TB cases with those named at interview. Estimation of prior contact as a risk factor for TB and identified factors associated with being a named contact. RESULTS: Ninety-five per cent of named contacts were known TB cases. The proportion of total identified contacts who were named at interview was 75%, and was similar for cases and controls. Cases were twice as likely as controls to identify prior contacts. Adding database information did not affect odds ratios, but increased the proportion of TB cases attributable to prior contact. Smear-positive, male and human immunodeficiency virus (HIV) negative TB patients were more likely to be named by subsequent cases. Identifiable recent contact with known smear-positive cases accounted for 12.5% of the TB burden. CONCLUSIONS: Reporting of putative source contacts showed little evidence of recall bias and gave estimates of the relative risk of TB associated with identifiable contact. The lower likelihood of HIV-positive cases being named as contacts may reflect reduced infectiousness.

The association between HIV and antituberculosis drug resistance.

In the UK, HIV is considered to be a risk factor for antituberculosis drug resistance. Evidence of the association is, however, inconclusive and there are few population-level data. The present study investigated the association in England and Wales during the period 1999-2005. National tuberculosis surveillance data for adults were matched to HIV/AIDS reports. Unmatched cases were assumed to be HIV-negative. Separate analyses were conducted on new tuberculosis cases and those with a previous diagnosis. Logistic regression was used for univariable and multivariable analyses. There were 1,657 previously diagnosed cases (80 HIV-positive) and 18,130 new cases (1,156 HIV-positive). Isoniazid resistance was found in 8.1% of previously diagnosed cases and 6.6% of new cases, and multidrug resistance in 2.8% and 0.7%, respectively. There was no evidence of an association between HIV and antituberculosis drug resistance among previously diagnosed cases. Among new cases, there was no overall association between HIV and isoniazid or multidrug resistance after adjusting for confounding factors. White HIV-positive patients were more likely to have multidrug resistance, but numbers were small. In contrast to some previous studies, this large, up-to-date study provides little evidence that HIV co-infected tuberculosis patients in England and Wales are at increased risk of firstline antituberculosis drug resistance.

Population differences in death rates in HIV-positive patients with tuberculosis.

SETTING: Randomised controlled clinical trial of Mycobacterium vaccae vaccination as an adjunct to anti-tuberculosis treatment in human immunodeficiency virus (HIV) positive patients with smear-positive tuberculosis (TB) in Lusaka, Zambia, and Karonga, Malawi. OBJECTIVE: To explain the difference in mortality between the two trial sites and to identify risk factors for death among HIV-positive patients with TB. DESIGN: Information on demographic, clinical, laboratory and radiographic characteristics was collected. Patients in Lusaka (667) and in Karonga (84) were followed up for an average of 1.56 years. Cox proportional hazard analyses were used to assess differences in survival between the two sites and to determine risk factors associated with mortality during and after anti-tuberculosis treatment. RESULTS: The case fatality rate was 14.7% in Lusaka and 21.4% in Karonga. The hazard ratio for death comparing Karonga to Lusaka was 1.47 (95% confidence interval [CI] 0.9-2.4) during treatment and 1.76 (95%CI 1.0-3.0) after treatment. This difference could be almost entirely explained by age and more advanced HIV disease among patients in Karonga. CONCLUSION: It is important to understand the reasons for population differences in mortality among patients with TB and HIV and to maximise efforts to reduce mortality.