Novel health system strategies for tuberculin skin test-guided Isoniazid preventive therapy

• Published in: European journal of public health 2020-09, Vol.30 (Supplement_5)
• Main Authors: Van Ginderdeuren, E, Bassett, J, Hanrahan, C F, Mutunga, L, Van Rie, A
• Format: Article
• Language: English
• Subjects: Adults; Antiretroviral agents; Antiretroviral therapy; Cost analysis; Costs; Economic impact; Effectiveness; Health care; Health services; HIV; Human immunodeficiency virus; Human performance; Human resources; Impact analysis; Isoniazid; Medical personnel; Medical treatment; Patients; Public health; Reading; Skin tests; Tracking control; Tuberculin; Workers
• ISSN: 1101-1262; 1464-360X
• DOI: 10.1093/eurpub/ckaa166.557

Abstract Background Global TB elimination demands a scale-up of Isoniazid preventive therapy (IPT) but tuberculin skin test (TST)-guided IPT poses great logistical and human resource challenges. Objectives Performance of TST self-reading by patients and fast-track TST reading by trained low cadre health care workers (task-shifting) was compared to formal TST reading by high cadre staff in a cohort of 278 South African adults living with HIV. Health economic impact of these novel strategies was assessed from a provider and societal perspective and simulations were performed for 5 other countries (USA, Germany, Brazil, India, Russia) to evaluate generalizability. In addition, accuracy of TST at antiretroviral treatment (ART) initiation was assessed by a repeat TST 6 and 12 month later. Results TST self-reading was highly accurate, with 89% sensitivity (95% CI 80, 95) and 100% specificity (95% CI 97,100) for detecting presence/absence of any induration. Agreement in TST reading between low and high cadre health care workers was very high (kappa 0.97). Compared to standard of care, a combined fast-track, task-shifting and self-reading strategy reduced TST reading costs in South Africa from a patient perspective by 81% and from a provider perspective by 92%. In all 5 countries simulated, TST reading cost was reduced by ≥ 78 % from a provider perspective. Repeat testing at 6 and 12 months showed high (31%, 95% CI 23, 40) TST conversion during the first 12 months of ART. Conclusions Empiric IPT for all people living with HIV followed by TST assessment after 6 or 12 months to identify those in need for lifelong IPT could increase the effectiveness of IPT programs. TST self-reading to reduce the number of patients that need to return for TST reading (to only those patients with self-determined presence of any induration) together with fast-tracking and task-shifting of TST reading could increase cost-effectiveness and reduce patient costs associated with IPT programs. Key messages Novel strategies are essential to control TB. Task-shifting, fast-tracking and patient TST self-reading empower patients and HCWs; optimal TST timing can increase cost-effectiveness of IPT programs.