Since its introduction as a first-line tuberculosis (TB) diagnostic test in South Africa in 2011, Xpert was expected to accelerate treatment for multidrug-resistant TB (MDR-TB). However, studies showed that its implementation did not significantly impact TB-related morbidity, mortality, or time to treatment for drug-sensitive TB (DS-TB). While Xpert reduced the time to appropriate treatment for MDR-TB, the expected same-day or same-week treatment initiation was not achieved. The XTEND economic evaluation concluded that Xpert implementation was cost- and effect-neutral, with key challenges in health system integration and patient linkage to treatment. See also: Lin TB Lab
Cost analysis found that provider costs per symptomatic individual tested were $89.66, while societal costs reached $169.94. Reducing initial loss to follow-up (iLTFU) slightly increased treatment costs but had minimal impact on health outcomes. Immediate treatment initiation after diagnosis provided minor mortality benefits, especially for HIV-positive patients. Supporting same-day clinical diagnosis following a negative test increased costs by $21.12 per symptomatic individual, while additional diagnostic testing raised costs by $35 per patient due to extra visits and treatment delays. See also: Australian Scholarships
The most cost-effective approach depended on the cost-effectiveness threshold. At lower thresholds, reducing iLTFU was preferred, while the negative pathway was more effective at higher thresholds. However, as per-transaction costs increased, empirical treatment became the preferred option due to fewer healthcare visits. These findings suggest that in high-TB-prevalence settings with well-developed laboratory infrastructure, the introduction of new TB diagnostics should be accompanied by additional investments in the health system. Current international policy focuses on expanding TB detection, but without support for decision-making after a negative test result, these efforts alone are unlikely to significantly impact the TB epidemic.
Diagnostic strategies vary in effectiveness depending on HIV prevalence, drug-resistant TB levels, and healthcare infrastructure. Tests that minimize patient visits can reduce costs and follow-up losses, while early TB detection improves treatment outcomes. Although new diagnostic tools may reduce lab delays, they can create bottlenecks elsewhere in the healthcare system. Accurate diagnostics alone are insufficient for TB control—their true impact depends on whether they expedite effective treatment. Evaluating their epidemiological effects is challenging due to TB’s slow progression, but operational and dynamic models can help assess their overall impact.
A study in Antanimora prison in Madagascar found a high TB prevalence among detainees, with confirmed active TB cases at 0.5% (4/748) and probable cases at 1.3% (10/748), resulting in a total active TB prevalence of 1.9%. Latent TB was significantly higher at 69.6% (517/743; 95% CI: 66.27–72.89). HIV prevalence was low at 0.4% (3/745), and no TB/HIV co-infection was detected. Key risk factors identified in univariable analysis included age ≥40 years (OR = 5.6), previous incarceration (OR = 7.1), prior TB history (OR = 8.4), and TB treatment history (OR = 9.7). Multivariable regression confirmed that older detainees were 4.4 times more likely to have active TB, while those with prior TB treatment had a 6.3-fold increased risk. Although confidence intervals were wide, the associations remained significant.
These findings highlight the urgent need for targeted TB screening and prevention strategies in prison settings, particularly for older detainees and those with prior TB treatment. The study successfully addressed TB and HIV prevalence and identified key risk factors, aligning with its research objectives. With a high latent TB burden and notable risk concentration among older detainees, the results underscore the importance of enhanced TB surveillance and intervention efforts.
Social network analysis (SNA), enriched by ethnographic data on human interactions, can enhance the realism of compartmental models by capturing the impact of social structures on disease transmission. Another study in Madagascar found that despite 15 years of intervention, latent TB infection prevalence showed only a slight decline, highlighting the persistence of TB reservoirs even after systematic treatment of active cases. The intensity of social contacts plays a crucial role in TB exposure, yet conventional transmission models often overlook these inter-community differences, underscoring the need for more nuanced approaches to understanding and controlling TB spread.
A study in Tanzania included a large number of participants, mostly adults aged 25–49, with a high proportion being male. The coastal and lake regions had the most participants. A significant portion was HIV-positive, and the majority had pulmonary TB. Most patients were self-referred and managed at hospitals, with nearly all treated using community-based DOT and first-line TB treatment. Bacteriological diagnosis was more common. Newly diagnosed TB patients were the vast majority, while recurrent TB cases were rare. Key risk factors for TB recurrence included older age, male sex, HIV positivity, referral from CTC, bacteriological diagnosis, and facility-based DOT. Patients in Zanzibar had a notably higher recurrence risk. Among recurrent TB cases, some experienced poor treatment outcomes, with death being the most common. Risk factors for poor outcomes included HIV positivity, treatment in certain regions (central, coastal, Zanzibar), bacteriological diagnosis, and facility-based DOT.
Expanding new diagnostic methods and algorithms could enhance TB detection while reducing delays in treatment initiation. Among available options, the full rollout of Xpert (B1) offers the most significant patient benefits. It decreases the number of visits required for diagnosis, shortens the time to treatment by nearly a week, and reduces diagnostic loss to follow-up, ultimately increasing successful treatment completion. At the health-system level, scaling up Xpert significantly lowers the need for sputum samples and laboratory staff time, easing resource burdens. Additionally, its implementation is expected to have the greatest impact on reducing TB prevalence, mortality, and incidence. Over a decade, Xpert could prevent tens of thousands of TB cases and related deaths, particularly improving survival rates for TB/HIV co-infected patients by expanding access to antiretroviral therapy.
Despite its advantages, Xpert's implementation requires substantial financial investment. However, it remains one of the three most cost-effective diagnostic strategies in Tanzania. Full Xpert rollout is estimated to cost $169 per DALY averted, making it a viable option despite higher initial resource demands. Alternative strategies, such as same-day LED fluorescence microscopy (A3) and standard LED fluorescence microscopy (A2), offer lower-cost solutions at $45 and $29 per DALY averted, respectively. While these approaches may be more affordable, they do not match Xpert's comprehensive benefits in improving patient outcomes and reducing TB burden. Balancing cost-effectiveness with epidemiological impact will be crucial in determining the optimal diagnostic strategy for widespread implementation.
A study aimed to assess the extent of pre-treatment loss to follow-up (PTLFU) among adults with pulmonary TB (PTB) in western Kenya and to identify associated patient factors. The research utilized a retrospective record review from January 2018 to December 2021, examining laboratory and treatment registers at Jaramogi Oginga Odinga Teaching and Referral Hospital (JOOTRH) in Kisumu. The study population comprised adults (≥18 years) with bacteriologically confirmed PTB. This method proved suitable for determining PTLFU rates and associated factors, though it depended on the accuracy of recorded data and did not account for patient behaviors or external systemic influences.
The study reviewed independent variables including demographics, contact information, residence, HIV status, TB history, diagnosis methods, and linkage to treatment. The primary dependent variable was the time from diagnosis to treatment initiation. The analysis found a PTLFU rate of 42.4% among the 476 participants studied. Significant risk factors included limited contact details, with those having only a physical address or a telephone number facing markedly higher odds of PTLFU compared to those with both types of contact information. Additionally, older adults (≥55 years) were more likely to experience PTLFU. Factors such as sex, HIV status, place of residence, and prior TB treatment did not significantly impact PTLFU after adjusting for confounders. The study concluded that a significant proportion of adults with PTB in western Kenya are lost to follow-up before treatment, with restricted contact details and older age being key risk factors.
Reading notes by Yoseph Leonardo Samodra.
References:
- Foster, N., Cunnama, L., McCarthy, K., Ramma, L., Siapka, M., Sinanovic, E., Churchyard, G., Fielding, K., Grant, A.D. and Cleary, S., 2021. Strengthening health systems to improve the value of tuberculosis diagnostics in South Africa: A cost and cost-effectiveness analysis. Plos one, 16(5), p.e0251547.
- Lin, H.H., Langley, I., Mwenda, R., Doulla, B., Egwaga, S., Millington, K.A., Mann, G.H., Murray, M., Squire, S.B. and Cohen, T., 2011. A modelling framework to support the selection and implementation of new tuberculosis diagnostic tools. The International journal of tuberculosis and lung disease, 15(8), pp.996-1004.
- Rakotomanana, F., Dreyfus, A., Randrianarisoa, M.M., Raberahona, M., Chevallier, E., Andriamasy, H.E., Bernardson, B.A., Ranaivomanana, P., Ralaitsilanihasy, F., Rasoamaharo, M. and Randrianirisoa, S.A., 2024. Prevalence of pulmonary tuberculosis and HIV infections and risk factors associated to tuberculosis in detained persons in Antananarivo, Madagascar. Scientific Reports, 14(1), p.8640.
- Pando, C., Hazel, A., Tsang, L.Y., Razafindrina, K., Andriamiadanarivo, A., Rabetombosoa, R.M., Ambinintsoa, I., Sadananda, G., Small, P.M., Knoblauch, A.M. and Rakotosamimanana, N., 2023. A social network analysis model approach to understand tuberculosis transmission in remote rural Madagascar. BMC Public Health, 23(1), p.1511.
- Njiro, B.J., Kisonga, R., Joachim, C., Sililo, G.A., Nkiligi, E., Ibisomi, L., Chirwa, T. and Francis, J.M., 2024. Epidemiology and treatment outcomes of recurrent tuberculosis in Tanzania from 2018 to 2021 using the National TB dataset. PLOS Neglected Tropical Diseases, 18(2), p.e0011968.
- Langley, I., Lin, H.H., Egwaga, S., Doulla, B., Ku, C.C., Murray, M., Cohen, T. and Squire, S.B., 2014. Assessment of the patient, health system, and population effects of Xpert MTB/RIF and alternative diagnostics for tuberculosis in Tanzania: an integrated modelling approach. The Lancet Global Health, 2(10), pp.e581-e591.
- Mulaku, M.N., Ochodo, E., Young, T. and Steingart, K.R., 2024. Pre-treatment loss to follow-up in adults with pulmonary TB in Kenya. Public Health Action, 14(1), pp.34-39.
- Kirimi, E.M., Muthuri, G.G., Ngari, C.G. and Karanja, S., 2024. A Model for the Propagation and Control of Pulmonary Tuberculosis Disease in Kenya. Discrete Dynamics in Nature and Society, 2024(1), p.5883142.
- Abdullahi, L.H., Oketch, S., Komen, H., Mbithi, I., Millington, K., Mulupi, S., Chakaya, J. and Zulu, E.M., 2024. Gendered gaps to tuberculosis prevention and care in Kenya: a political economy analysis study. BMJ open, 14(4), p.e077989.
TBC 046
