Designing an optimized diagnostic network to improve access to TB diagnosis and treatment in Lesotho

A tuberculosis elimination-focused geospatial approach to optimising access to diagnostic GeneXpert machines in Fiji

OBJECTIVES

Fiji could be the first country to eliminate tuberculosis. To inform this strategy, we aimed to identify how many GeneXpert® machines are required to enable over 90% of Fijians to be within one-hour easy access.

METHODS

We used Geographic Information System (Quantum GIS; QGIS), OpenStreetMap and population data (Kontur) to map possible facilities in relation to QGIS generated 60-min drive-time isochrones, with correction for missing road data. For outer islands, we calculated a distance to nearest hub operation.

RESULTS

The solution comprised 24 GeneXpert® machines, allocating 7 GeneXpert® to Viti Levu, 6 GeneXpert® to Vanua Levu and 11 to other islands. This resulted in 827,810 people, 93.6% of Fiji's population, being within 1 h of a machine. Twenty-one thousand four hundred seventy-nine people on outer islands were an average of 43 km by water from the nearest facility.

CONCLUSIONS

We conclude that over 90% of Fijians could be within an hour of a GeneXpert® machine with placement of 24 machines.

Application of diagnostic network optimization in Kenya and Nepal to design integrated, sustainable and efficient bacteriology and antimicrobial resistance surveillance networks

Antimicrobial resistance (AMR) is a major global public health concern, particularly in low- and middle-income countries, which experience the highest burden of AMR. Critical to combatting AMR is ensuring there are effective, accessible diagnostic networks in place to diagnose, monitor and prevent AMR, but many low- and middle-income countries lack such networks. Consequently, there is substantial need for approaches that can inform the design of efficient AMR laboratory networks and sample referral systems in lower-resource countries. Diagnostic network optimization (DNO) is a geospatial network analytics approach to plan diagnostic networks and ensure greatest access to and coverage of services, while maximizing the overall efficiency of the system. In this intervention, DNO was applied to strengthen bacteriology and AMR surveillance network design in Kenya and Nepal for human and animal health, by informing linkages between health facilities and bacteriology testing services and sample referral routes between farms, health facilities and laboratories. Data collected from the target settings in each country were entered into the open-access DNO tool OptiDx, to generate baseline scenarios, which depicted the current state of AMR laboratory networks and sample referral systems in the countries. Subsequently, baselines were adjusted to evaluate changing factors such as samples flows, transport frequency, transport costs, and service distances. Country stakeholders then compared resulting future scenarios to identify the most feasible solution for their context. The DNO analyses enabled a wealth of insights that will facilitate strengthening of AMR laboratory and surveillance networks in both countries. Overall, the project highlights the benefits of using a data-driven approach for designing efficient diagnostic networks, to ensure better health resource allocation while maximizing the impact and equity of health interventions. Given the critical need to strengthen AMR laboratory and surveillance capacity, DNO should be considered an integral part of diagnostic strategic planning in the future.

Missed opportunities for integrated testing of HIV and tuberculosis on the GeneXpert platform in Lesotho

Background

Integrated testing, treatment and care are key strategies for addressing the dual burdens of tuberculosis and HIV. The GeneXpert instrument allows simultaneous HIV and tuberculosis testing, but its utilisation for integrated testing remains suboptimal.

Objective

The study determined the extent to which tuberculosis testing and HIV early infant detection (EID) were integrated on the GeneXpert platform, or the potential for integration at selected health facilities.

Methods

A mixed methods evaluation was conducted using retrospective secondary data analysis of laboratory records from 2017 to 2019, and semi-structured interviews. Data were collected between January 2020 and March 2020 in Lesotho.

Results

Forty-four health staff were interviewed across 13 health facilities: one regional, nine district, and three clinic level. Six were government facilities, six were mission hospitals, and one was a non-profit clinic. All facilities selected had at least one GeneXpert instrument used for tuberculosis or HIV testing; none included simultaneous testing for tuberculosis and HIV. In 2017, the average utilisation rate for the GeneXpert instrument for tuberculosis and EID testing was 63% and 24%, while in 2019, the average utilisation rate was 61% for tuberculosis testing and 27% for EID.

Conclusion

Except for three sites where the testing rates were high, utilisation rates were sufficiently low that all the HIV EID and tuberculosis tests undertaken in 2017 and 2019 could have been performed using only the instruments currently dedicated to tuberculosis testing. There is a missed opportunity for the integration of testing for tuberculosis and HIV on the GeneXpert instrument.

What this study adds

This study adds to the body of evidence on the need for integration of testing and highlights some practical and technical considerations for successful implementation of integrated tuberculosis and HIV testing.

Understanding barriers to tuberculosis diagnosis and treatment completion in a low-resource setting: A mixed-methods study in the Kingdom of Lesotho

BACKGROUND

Lesotho is one of the 30 countries with the highest tuberculosis incidence rates in the world, estimated at 650 per 100,000 population. Tuberculosis case detection is extremely low, particularly with the rapid spread of COVID-19, dropping from an estimated 51% in 2020 to 33% in 2021. The aim of this study is to understand the barriers to tuberculosis diagnosis and treatment completion.

METHODS

We used a convergent mixed methods study design. We collected data on the number of clients reporting symptoms upon tuberculosis screening, their sputum test results, the number of clients diagnosed, and the number of clients who started treatment from one district hospital and one health center in Berea district, Lesotho. We conducted in-depth interviews and focus group discussions with 53 health workers and patients. We used a content analysis approach to analyze qualitative data and integrated quantitative and qualitative findings in a joint display.

FINDINGS

During March-August, 2019, 218 clients at the hospital and 292 clients at the health center reported tuberculosis symptoms. The full diagnostic testing process was completed for 66% of clients at the hospital and 68% at the health center. Among clients who initiated tuberculosis treatment, 68% (61/90) at the hospital and 74% (32/43) at the health center completed treatment. The main barriers to testing and treatment completion were challenges at sample collection, lack of decentralized diagnostic services, and socioeconomic factors such as food insecurity and high patient movement to search for jobs.

CONCLUSIONS

Tuberculosis diagnosis could be improved through the effective decentralization of laboratory services at the health facility level, and treatment completion could be improved by providing food and other forms of social support to patients.

Linkage to care and treatment outcomes for patients diagnosed with drug-susceptible tuberculosis using Xpert MTB/RIF assay in Thaba-Tseka district in Lesotho

•

Patients with tuberculosis (TB) with evidence of linkage to care is not optimal in Thaba-Tseka district.

•

Same-day TB treatment initiation is feasible for facilities with onsite GeneXpert services.

•

The TB death rate in Thaba-Tseka district does not align with the World Health Organization's End TB strategy target.

Objectives

To evaluate linkage to care and treatment outcomes of patients with tuberculosis (TB) confirmed by Xpert MTB/RIF assay in Thaba-Tseka district, Lesotho.

Design

This was a retrospective cohort study of adult patients diagnosed with drug-susceptible TB using the Xpert MTB/RIF assay at two laboratories in Thaba-Tseka district from January 2016 to December 2020.

Results

Six hundred and fifty-five eligible participants were identified for inclusion in this study. Their median age was 40 [interquartile range (IQR) 32–54] years, and 468 (71.45%) were male. Evidence of linkage to care was found for 459 (70.08%) participants, but there was no documentation on treatment initiation for 196 (29.92%) participants. The median time to treatment initiation was 0 days (same-day initiation) (IQR 0–4) and the treatment success rate was 86%. Treatment success was associated with negative sputum smear results after 2, 5 and 6 months (χ², P<0.001). The overall mortality rate was 10%, with no trend of mortality reduction.

Conclusion

There is a need to address the issue of linkage to care of patients diagnosed with TB in Thaba-Tseka district. Efforts should be made to reduce TB mortality in line with the World Health Organization's ‘End TB strategy’ target.

Gaps and Interventions across the Diagnostic Care Cascade of TB Patients at the Level of Patient, Community and Health System: A Qualitative Review of the Literature

Tuberculosis (TB) continues to be one of the important public health concerns globally, and India is among the seven countries with the largest burden of TB. There has been a consistent increase in the notifications of TB cases across the globe. However, the 2018 estimates envisage a gap of about 30% between the incident and notified cases of TB, indicating a significant number of patients who remain undiagnosed or ‘missed’. It is important to understand who is ‘missed’, find this population, and provide quality care. Given these complexities, we reviewed the diagnostic gaps in the care cascade for TB. We searched Medline via PubMed and CENTRAL databases via the Cochrane Library. The search strategy for PubMed was tailored to individual databases and was as: ((((((tuberculosis[Title/Abstract]) OR (TB[Title/Abstract])) OR (koch *[Title/Abstract])) OR (“tuberculosis”[MeSH Terms]))) AND (((diagnos *) AND (“diagnosis”[MeSH Terms])))). Furthermore, we screened the references list of the potentially relevant studies to seek additional studies. Studies retrieved from these electronic searches and relevant references included in the bibliography of those studies were reviewed. Original studies in English that assessed the causes of diagnostic gaps and interventions used to address them were included. Delays in diagnosis were found to be attributable to both the individuals’ and the health system’s capacity to diagnose and promptly commence treatment. This review provides insights into the diagnostic gaps in a cascade of care for TB and different interventions adopted in studies to close this gap. The major diagnostic gaps identified in this review are as follows: people may not have access to TB diagnostic tests, individuals are at a higher risk of missed diagnosis, services are available but people may not seek care with a diagnostic facility, and patients are not diagnosed despite reaching health facilities. Therefore, reaching the goal to End TB requires putting in place models and methods to provide prompt and quality assured diagnosis to populations at par.

Use of Physical Accessibility Modelling in Diagnostic Network Optimization: A Review

Diagnostic networks are complex systems that include both laboratory-tested and community-based diagnostics, as well as a specimen referral system that links health tiers. Since diagnostics are the first step before accessing appropriate care, diagnostic network optimization (DNO) is crucial to improving the overall healthcare system. The aim of our review was to understand whether the field of DNO, and especially route optimization, has benefited from the recent advances in geospatial modeling, and notably physical accessibility modeling, that have been used in numerous health systems assessment and strengthening studies. All publications published in English between the journal’s inception and 12 August 2021 that dealt with DNO, geographical accessibility and optimization, were systematically searched for in Web of Science and PubMed, this search was complemented by a snowball search. Studies from any country were considered. Seven relevant publications were selected and charted, with a variety of geospatial approaches used for optimization. This paucity of publications calls for exploring the linkage of DNO procedures with realistic accessibility modeling framework. The potential benefits could be notably better-informed travel times of either the specimens or population, better estimates of the demand for diagnostics through realistic population catchments, and innovative ways of considering disease epidemiology to inform DNO.

Understanding reasons for suboptimal tuberculosis screening in a low-resource setting: A mixed-methods study in the Kingdom of Lesotho

Lesotho has one of the highest tuberculosis (TB) incidence rates in the world, estimated at 654/100,000 population. However, TB detection remains low, with only 51% of people with TB being diagnosed and treated. The aim of this study was to evaluate implementation of TB screening and identify drivers of suboptimal TB screening in Lesotho. We used a convergent mixed methods study design. We collected data on the number of health facility visits and the number of clients screened for TB during March-August, 2019 from one district hospital and one health center. We conducted interviews and focus group discussions with patients and health workers to elucidate the mechanisms associated with suboptimal screening. Out of an estimated 70,393 visitors to the two health facilities, only 22% of hospital visitors and 48% of health center visitors were asked about TB symptoms. Only 2% of those screened at each facility said that they had TB symptoms, comprising a total of 510 people. Lack of training on tuberculosis screening, overall staff shortages, barriers faced by patients in accessing care, and health care worker mistrust of tuberculosis screening procedures were identified as drivers of suboptimal TB screening. TB screening could be improved by ensuring the availability of well-trained, incentivized, and dedicated screeners at health facilities, and by providing TB screening services in community settings.

Developing Strategies for Onchocerciasis Elimination Mapping and Surveillance Through The Diagnostic Network Optimization Approach

Background

Onchocerciasis (river blindness) is a filarial disease targeted for elimination of transmission. However, challenges exist to the implementation of effective diagnostic and surveillance strategies at various stages of elimination programs. To address these challenges, we used a network data analytics approach to identify optimal diagnostic scenarios for onchocerciasis elimination mapping (OEM).

Methods

The diagnostic network optimization (DNO) method was used to model the implementation of the old Ov16 rapid diagnostic test (RDT) and of new RDTs in development for OEM under different testing strategy scenarios with varying testing locations, test performance and disease prevalence. Environmental suitability scores (ESS) based on machine learning algorithms were developed to identify areas at risk of transmission and used to select sites for OEM in Bandundu region in the Democratic Republic of Congo (DRC) and Uige province in Angola. Test sensitivity and specificity ranges were obtained from the literature for the existing RDT, and from characteristics defined in the target product profile for the new RDTs. Sourcing and transportation policies were defined, and costing information was obtained from onchocerciasis programs. Various scenarios were created to test various state configurations. The actual demand scenarios represented the disease prevalence at IUs according to the ESS, while the counterfactual scenarios (conducted only in the DRC) are based on adapted prevalence estimates to generate prevalence close to the statistical decision thresholds (5% and 2%), to account for variability in field observations. The number of correctly classified implementation units (IUs) per scenario were estimated and key cost drivers were identified.

Results

In both Bandundu and Uige, the sites selected based on ESS had high predicted onchocerciasis prevalence &gt;10%. Thus, in the actual demand scenarios in both Bandundu and Uige, the old Ov16 RDT correctly classified all 13 and 11 IUs, respectively, as requiring CDTi. In the counterfactual scenarios in Bandundu, the new RDTs with higher specificity correctly classified IUs more cost effectively. The new RDT with highest specificity (99.8%) correctly classified all 13 IUs. However, very high specificity (e.g., 99.8%) when coupled with imperfect sensitivity, can result in many false negative results (missing decisions to start MDA) at the 5% statistical decision threshold (the decision rule to start MDA). This effect can be negated by reducing the statistical decision threshold to 2%. Across all scenarios, the need for second stage sampling significantly drove program costs upwards. The best performing testing strategies with new RDTs were more expensive than testing with existing tests due to need for second stage sampling, but this was offset by the cost of incorrect classification of IUs.

Conclusion

The new RDTs modelled added most value in areas with variable disease prevalence, with most benefit in IUs that are near the statistical decision thresholds. Based on the evaluations in this study, DNO could be used to guide the development of new RDTs based on defined sensitivities and specificities. While test sensitivity is a minor driver of whether an IU is identified as positive, higher specificities are essential. Further, these models could be used to explore the development and optimization of new tools for other neglected tropical diseases.

A Framework for Assessing Import Costs of Medical Supplies and Results for a Tuberculosis Program in Karakalpakstan, Uzbekistan

Background. Import of medical supplies is common, but limited knowledge about import costs and their structure introduces uncertainty to budget planning, cost management, and cost-effectiveness analysis of health programs. We aimed to estimate the import costs of a tuberculosis (TB) program in Uzbekistan, including the import costs of specific imported items.Methods. We developed a framework that applies costing and cost accounting to import costs. First, transport costs, customs-related costs, cargo weight, unit weights, and quantities ordered were gathered for a major shipment of medical supplies from the Médecins Sans Frontières (MSF) Procurement Unit in Amsterdam, the Netherlands, to a TB program in Karakalpakstan, Uzbekistan, in 2016. Second, air freight, land freight, and customs clearance cost totals were estimated. Third, total import costs were allocated to different cargos (standard, cool, and frozen), items (e.g., TB drugs), and units (e.g., one tablet) based on imported weight and quantity. Data sources were order invoices, waybills, the local MSF logistics department, and an MSF standard product list.Results. The shipment contained 1.8 million units of 85 medical items of standard, cool, and frozen cargo. The average import cost for the TB program was 9.0% of the shipment value. Import cost varied substantially between cargos (8.9-28% of the cargo value) and items (interquartile range 4.5-35% of the item value). The largest portion of the total import cost was caused by transport (82-99% of the cargo import cost) and allocated based on imported weight. Ten (14%) of the 69 items imported as standard cargo were associated with 85% of the standard cargo import cost. Standard cargo items could be grouped based on contributing to import costs predominantly through unit weight (e.g., fluids), imported quantity (e.g., tablets), or the combination of unit weight and imported quantity (e.g., items in powder form).Conclusion. The cost of importing medical supplies to a TB program in Karakalpakstan, Uzbekistan, was sizable, variable, and driven by a subset of imported items. The framework used to measure and account import costs can be adapted to other health programs.

Bringing Data Analytics to the Design of Optimized Diagnostic Networks in Low- and Middle-Income Countries: Process, Terms and Definitions

Diagnostics services are an essential component of healthcare systems, advancing universal health coverage and ensuring global health security, but are often unavailable or under-resourced in low- and middle-income (LMIC) countries. Typically, diagnostics are delivered at various tiers of the laboratory network based on population needs, and resource and infrastructure constraints. A diagnostic network additionally incorporates screening and includes point-of-care testing that may occur outside of a laboratory in the community and clinic settings; it also emphasizes the importance of supportive network elements, including specimen referral systems, as being critical for the functioning of the diagnostic network. To date, design and planning of diagnostic networks in LMICs has largely been driven by infectious diseases such as TB and HIV, relying on manual methods and expert consensus, with a limited application of data analytics. Recently, there have been efforts to improve diagnostic network planning, including diagnostic network optimization (DNO). The DNO process involves the collection, mapping, and spatial analysis of baseline data; selection and development of scenarios to model and optimize; and lastly, implementing changes and measuring impact. This review outlines the goals of DNO and steps in the process, and provides clarity on commonly used terms.

Islands of Tuberculosis Elimination: An Evaluation of Community-Based Active Case Finding in North Sumatra, Indonesia

Community-based active case finding (ACF) is needed to reach key/vulnerable populations with limited access to tuberculosis (TB) care. Published reports of ACF interventions in Indonesia are scarce. We conducted an evaluation of a multicomponent community-based ACF intervention as it scaled from one district to nine in Nias and mainland North Sumatra. Community and health system support measures including laboratory strengthening, political advocacy, sputum transport, and community awareness were instituted. ACF was conducted in three phases: pilot (18 months, 1 district), intervention (12 months, 4 districts) and scale-up (9 months, 9 districts). The pilot phase identified 215 individuals with bacteriologically positive (B+) TB, representing 42% of B+ TB notifications. The intervention phase yielded 509, representing 54% of B+ notifications and the scale-up phase identified 1345 individuals with B+ TB (56% of notifications). We observed large increases in B+ notifications on Nias, but no overall change on the mainland despite district variation. Overall, community health workers screened 377,304 individuals of whom 1547 tested positive, and 95% were initiated on treatment. Our evaluation shows that multicomponent community-based ACF can reduce the number of people missed by TB programs. Community-based organizations are best placed for accessing and engaging hard to reach populations and providing integrated support which can have a large positive effect on TB notifications.