A cautionary note about estimating effects of secondary exposures in cohort studies.

Cohort studies are often enriched for a primary exposure of interest to improve cost-effectiveness, which presents analytical challenges not commonly discussed in epidemiology. In this paper, we use causal diagrams to represent exposure-enriched cohort studies, illustrate a scenario wherein the risk ratio for the effect of a secondary exposure on an outcome is biased, and propose an analytical method for correcting for such bias. In our motivating example, maternal smoking (Z) is a cause of fetal growth restriction (X), which subsequently affects preterm birth (Y) (i.e., Z → X → Y); strong positive associations exist between both Z, X and X, Y; and enrichment for X increases its prevalence from 10% to 50%. In the X-enriched cohort, unadjusted and X-adjusted analyses lead to bias in the risk ratio for the total effect of Z on Y. After application of inverse probability weights, the bias is corrected, with a small loss of efficiency in comparison with a same-sized study without X-enrichment. With increasing interest in conducting secondary analyses to reduce research costs, caution should be employed when analyzing studies that have already been enriched, intentionally or unintentionally, for a primary exposure of interest. Causal diagrams can help identify scenarios in which secondary analyses may be biased. Inverse probability weights can be used to remove the bias.

Impact of three empirical anti-tuberculosis treatment strategies for people initiating antiretroviral therapy.

BACKGROUND: Early mortality in people initiating antiretroviral treatment (ART) remains high. Empirical anti-tuberculosis treatment strategies aim to reduce early mortality by initiating anti-tuberculosis treatment in individuals at high risk of death from undiagnosed TB. METHODS: Using data from 16 913 individuals starting ART under program conditions, we simulated the impact of three empirical treatment strategies (two clinical trials and a pragmatic approach), assuming that 50% of early deaths and 100% of incident TB are averted in those eligible. RESULTS: Compared to starting anti-tuberculosis treatment on clinical or mycobacteriological grounds, 4.4-31.4% more individuals were eligible for anti-tuberculosis treatment, 5.5-25.4% of deaths were averted and 10.9-57.3% of incident TB cases were prevented under empirical anti-tuberculosis treatment strategies. The proportion receiving any anti-tuberculosis treatment during the first 6 months of ART increased from the observed 24.0% to an estimated 27.5%, 40.4% and 51.3%, under the PrOMPT, REMEMBER and pragmatic approach, respectively. CONCLUSION: The impact of empirical anti-tuberculosis treatment strategies depends greatly on the eligibility criteria chosen. The additional strain placed on anti-tuberculosis treatment facilities and the relatively limited impact of some empirical TB strategies raise the question as to whether the benefits will outweigh the risks at population level.

Antiretroviral therapy initiation during tuberculosis treatment and HIV-RNA and CD4 T-lymphocyte responses.

SETTING: A large human immunodeficiency virus (HIV) clinic in South Africa. OBJECTIVE: To examine the effect of initiating antiretroviral therapy (ART) on CD4 and viral response at different time periods during anti-tuberculosis treatment (<14 days, 15-60 days, or ≥60 days) using prospectively collected clinical data. METHODS: Cohort data analysis for 1499 patients with tuberculosis (TB) and HIV co-infection classified according to timing of ART after the initiation of anti-tuberculosis treatment. RESULTS: In adjusted modified Poisson regression models, CD4 and viral responses showed no significant differences according to timing of ART initiation (failure to increase CD4 by 6 months, <14 days vs. >60 days: RR 1.02, 95%CI 0.85-1.22; 15-60 days vs. >60 days: RR 1.00, 95%CI 0.86-1.15; failure to suppress virus by 6 months, <14 days vs. >60 days: RR 0.98, 95%CI 0.59-1.63; 15-60 days vs. >60 days: RR 0.96, 95%CI 0.66-1.41 and viral rebound at 12 months, 14 days vs. >60 days: RR 1.43, 95%CI 0.50-4.12; 15-60 days vs. >60 days: RR 1.14, 95%CI 0.39-3.34). Similar estimates were found in analysis restricted to patients with severe immunosuppression. CONCLUSION: Concerns over the overlapping impact of anti-tuberculosis treatment with ART on ART response should not be a reason for delaying ART in patients with HIV-associated TB.

Herpes simplex virus type 2 antibody detection performance in Kisumu, Kenya, using the Herpeselect ELISA, Kalon ELISA, Western blot and inhibition testing.

BACKGROUND: In certain parts of Africa, type-specific herpes simplex virus type 2 (HSV-2) ELISAs may have limited specificity. To date, no study has been conducted to validate HerpeSelect and Kalon type-specific HSV-2 ELISAs using both the Western blot and recombinant gG ELISA inhibition testing as reference standards. METHODS: A total of 120 men who were HIV seronegative (aged 18-24 years) provided blood samples. HSV-2 IgG serum antibodies were detected using four different methods: HerpeSelect HSV-2 ELISA (n = 120), Kalon HSV-2 ELISA (n = 120), University of Washington Western blot (n = 101) and a recombinant inhibition test (n = 93). RESULTS: HSV-2 seroprevalence differed significantly by HSV-2 detection method, ranging from 24.8% with the Western blot to 69.8% with the HerpeSelect ELISA. Using the Western blot as the reference standard, the HerpesSelect had the highest sensitivity for HSV-2 antibody detection (100%) yet lowest specificity (40%). Similar results were obtained using the inhibition test as the reference standard. The sensitivity and specificity of the Kalon test versus the Western blot were 92% and 79%, respectively, and 80% and 82% versus the inhibition test. Using the inhibition test as the reference standard, the sensitivity of the Western blot appeared low (49%). CONCLUSIONS: In men in western Kenya who were HIV seronegative, the HerpeSelect and Kalon type-specific ELISAs had high sensitivities yet limited specificities using the Western blot as reference standard. Overall, the Kalon ELISA performed better than the HerpeSelect ELISA in these young men from Kisumu. Further understanding is needed for the interpretation of HSV-2 inhibition or ELISA test positive/ Western blot seronegative results. Before HSV-2 seropositivity may be reliably reported in selected areas of Africa, performance studies of HSV-2 serological assays in individual geographical areas are recommended.

Counseling and testing TB patients for HIV: evaluation of three implementation models in Kinshasa, Congo.

SETTING: Kinshasa, Democratic Republic of Congo. OBJECTIVES: To evaluate the implementation of three models of provider-initiated HIV counseling and testing (CT) for tuberculosis (TB) patients. METHODS: HIV CT was offered to all TB patients aged > or =18 months registered for treatment at three project clinics between August 2004 and June 2005. HIV CT was performed at the TB clinic, the health center or the freestanding voluntary counseling and testing (VCT) center. HIV-infected patients received cotrimoxazole prophylaxis. RESULTS: Uptake of HIV CT was high (95-98%) when performed at the TB clinic or primary health care center, but significantly lower (68.5%) among patients referred to a free-standing VCT center. The overall HIV prevalence among the 1088 patients tested for HIV was 18.8%. HIV was associated with female sex (aOR 1.91), recurrent TB (aOR 2.74), extra-pulmonary TB (aOR 1.97) and age. CONCLUSIONS: Implementation of provider-initiated routine HIV CT by the TB nurse or health care worker at the primary health care center results in a higher uptake compared to referral of patients with TB to freestanding VCT clinics. Provider-initiated HIV CT is only a first step and needs to be linked to access to HIV care, support and treatment.