Health care index score and risk of death following tuberculosis diagnosis in HIV-positive patients

Abstract

To assess health care utilisation for patients co-infected with TB and HIV (TB-HIV), and to develop a weighted health care index (HCI) score based on commonly used interventions and compare it with patient outcome. A total of 1061 HIV patients diagnosed with TB in four regions, Central/Northern, Southern and Eastern Europe and Argentina, between January 2004 and December 2006 were enrolled in the TB-HIV study. A weighted HCI score (range 0-5), based on independent prognostic factors identified in multivariable Cox models and the final score, included performance of TB drug susceptibility testing (DST), an initial TB regimen containing a rifamycin, isoniazid and pyrazinamide, and start of combination antiretroviral treatment (cART). The mean HCI score was highest in Central/Northern Europe (3.2, 95%CI 3.1-3.3) and lowest in Eastern Europe (1.6, 95%CI 1.5-1.7). The cumulative probability of death 1 year after TB diagnosis decreased from 39% (95%CI 31-48) among patients with an HCI score of 0, to 9% (95%CI 6-13) among those with a score of ≥4. In an adjusted Cox model, a 1-unit increase in the HCI score was associated with 27% reduced mortality (relative hazard 0.73, 95%CI 0.64-0.84). Our results suggest that DST, standard anti-tuberculosis treatment and early cART may improve outcome for TB-HIV patients. The proposed HCI score provides a tool for future research and monitoring of the management of TB-HIV patients. The highest HCI score may serve as a benchmark to assess TB-HIV management, encouraging continuous health care improvement.