Relative ORs Research Articles

Evaluating bias with loss to follow-up in a community-based cohort: empirical investigation from the CARRS Study

BackgroundRetention of participants is a challenge in community-based longitudinal cohort studies. We aim to evaluate the factors associated with loss to follow-up and estimate attrition bias.MethodsData are from an ongoing...

Initial non-participation and loss to follow-up in a Danish youth cohort: implications for relative risk estimates

BackgroundInitial non-participation and loss to follow-up in the Danish youth cohort Vestliv could introduce selection bias of the measured risk estimates.ObjectiveTo investigate the impact of initial non-participation and loss to...

Tests for Latent Tuberculosis in People With ESRD: A Systematic Review

The relative diagnostic accuracy of interferon γ release assays (IGRAs; based on ELISA [enzyme-linked immunosorbent assay] or ELISPOT [enzyme-linked immunosorbent spot], ie, the QuantiFERON and T-SPOT.TB tests, respectively) and the tuberculin skin test (TST) for latent tuberculosis (TB) infection in people with end-stage kidney disease is uncertain and national guidelines for their use are inconsistent. Systematic review. Evaluated performance of tests for latent TB with clinical risk-factor assessment. People with end-stage kidney disease (chronic kidney disease stage 5 [eGFR <15] or kidney transplant recipients). No limits on setting. ELISA- or ELISPOT-based IGRAs, TST, assays to detect antimycobacterial antibodies, and flow cytometry-based tests. Odds of test positivity with clinical risk factor for latent TB, expressed as ORs and relative ORs (RORs). 47 studies (6,828 participants) were included, but only 30 studies (4,546 participants) contained sufficient data to contribute to meta-analysis. Studies were predominately in the dialysis population (23/30; 3,700 participants) in countries with low to moderate TB prevalence (0.0-50.0 cases/10(5) persons). BCG vaccination rate was variable (2.7%-100.0%). 9 studies compared IGRAs with the TST directly, 17 studies evaluated the TST only, and the other 4 studies evaluated other tests. Compared to a positive TST result, a positive ELISA-based IGRA result was associated more strongly with radiologic evidence of past TB (ROR, 4.29; 95% CI, 1.83-10.3; P = 0.001) and contact with active TB (ROR, 3.36; 95% CI, 1.61-7.01; P = 0.001). Compared to a negative TST result, a negative ELISA-based IGRA result was associated more strongly with BCG vaccination (ROR, 0.30; 95% CI, 0.14-0.63; P = 0.002). There were insufficient data to compare performance of the ELISPOT-based IGRA with the TST or ELISA-based IGRA. 17 of 47 included studies (36.2%) did not contain sufficient data to contribute to meta-analysis. Compared to the TST, the ELISA-based IGRA was associated more strongly with risk factors for latent TB in end-stage kidney disease.