Independent evaluation of 12 artificial intelligence solutions for the detection of tuberculosis.

There have been few independent evaluations of computer-aided detection (CAD) software for tuberculosis (TB) screening, despite the rapidly expanding array of available CAD solutions. We developed a test library of chest X-ray (CXR) images which was blindly re-read by two TB clinicians with different levels of experience and then processed by 12 CAD software solutions. Using Xpert MTB/RIF results as the reference standard, we compared the performance characteristics of each CAD software against both an Expert and Intermediate Reader, using cut-off thresholds which were selected to match the sensitivity of each human reader. Six CAD systems performed on par with the Expert Reader (Qure.ai, DeepTek, Delft Imaging, JF Healthcare, OXIPIT, and Lunit) and one additional software (Infervision) performed on par with the Intermediate Reader only. Qure.ai, Delft Imaging and Lunit were the only software to perform significantly better than the Intermediate Reader. The majority of these CAD software showed significantly lower performance among participants with a past history of TB. The radiography equipment used to capture the CXR image was also shown to affect performance for some CAD software. TB program implementers now have a wide selection of quality CAD software solutions to utilize in their CXR screening initiatives.

The health policy response to COVID-19 in Malawi.

Malawi declared a state of national disaster due to the COVID-19 pandemic on 20th March 2020 and registered its first confirmed coronavirus case on the 2 April 2020. The aim of this paper was to document policy decisions made in response to the COVID-19 pandemic from January to August 2020. We reviewed policy documents from the Public Health Institute of Malawi, the Malawi Gazette, the Malawi Ministry of Health and Population and the University of Oxford Coronavirus Government Response Tracker. We found that the Malawi response to the COVID-19 pandemic was multisectoral and implemented through 15 focused working groups termed clusters. Each cluster was charged with providing policy direction in their own area of focus. All clusters then fed into one central committee for major decisions and reporting to head of state. Key policies identified during the review include international travel ban, school closures at all levels, cancellation of public events, decongesting workplaces and public transport, and mandatory face coverings and a testing policy covering symptomatic people. Supportive interventions included risk communication and community engagement in multiple languages and over a variety of mediums, efforts to improve access to water, sanitation, nutrition and unconditional social-cash transfers for poor urban and rural households.

How to reduce household costs for people with tuberculosis: a longitudinal costing survey in Nepal.

The aim of this study was to compare costs and socio-economic impact of tuberculosis (TB) for patients diagnosed through active (ACF) and passive case finding (PCF) in Nepal. A longitudinal costing survey was conducted in four districts of Nepal from April 2018 to October 2019. Costs were collected using the WHO TB Patient Costs Survey at three time points: intensive phase of treatment, continuation phase of treatment and at treatment completion. Direct and indirect costs and socio-economic impact (poverty headcount, employment status and coping strategies) were evaluated throughout the treatment. Prevalence of catastrophic costs was estimated using the WHO threshold. Logistic regression and generalized estimating equation were used to evaluate risk of incurring high costs, catastrophic costs and socio-economic impact of TB over time. A total of 111 ACF and 110 PCF patients were included. ACF patients were more likely to have no education (75% vs 57%, P = 0.006) and informal employment (42% vs 24%, P = 0.005) Compared with the PCF group, ACF patients incurred lower costs during the pretreatment period (mean total cost: US$55 vs US$87, P < 0.001) and during the pretreatment plus treatment periods (mean total direct costs: US$72 vs US$101, P < 0.001). Socio-economic impact was severe for both groups throughout the whole treatment, with 32% of households incurring catastrophic costs. Catastrophic costs were associated with 'no education' status [odds ratio = 2.53(95% confidence interval = 1.16-5.50)]. There is a severe and sustained socio-economic impact of TB on affected households in Nepal. The community-based ACF approach mitigated costs and reached the most vulnerable patients. Alongside ACF, social protection policies must be extended to achieve the zero catastrophic costs milestone of the End TB strategy.

Study protocol: analysis of regional lung health policies and stakeholders in Africa.

BACKGROUND: Lung health is a critical area for research in sub-Saharan Africa. The International Multidisciplinary Programme to Address Lung Health and TB in Africa (IMPALA) is a collaborative programme that seeks to fill evidence gaps to address high-burden lung health issues in Africa. In order to generate demand for and facilitate use of IMPALA research by policy-makers and other decision-makers at the regional level, an analysis of regional lung health policies and stakeholders will be undertaken to inform a programmatic strategy for policy engagement. METHODS AND ANALYSIS: This analysis will be conducted in three phases. The first phase will be a rapid desk review of regional lung health policies and stakeholders that seeks to understand the regional lung health policy landscape, which issues are prioritised in existing regional policy, key regional actors, and opportunities for engagement with key stakeholders. The second phase will be a rapid desk review of the scientific literature, expanding on the work in the first phase by looking at the external factors that influence regional lung health policy, the ways in which regional bodies influence policy at the national level, investments in lung health, structures for discussion and advocacy, and the role of evidence at the regional level. The third phase will involve a survey of IMPALA partners and researchers as well as interviews with key regional stakeholders to further shed light on regional policies, including policy priorities and gaps, policy implementation status and challenges, stakeholders, and platforms for engagement and promoting uptake of evidence. DISCUSSION: Health policy analysis provides insights into power dynamics and the political nature of the prioritisation of health issues, which are often overlooked. In order to ensure the uptake of new knowledge and evidence generated by IMPALA, it is important to consider these complex factors.

Structural coercion in the context of community engagement in global health research conducted in a low resource setting in Africa.

BACKGROUND: While community engagement is increasingly promoted in global health research to improve ethical research practice, it can sometimes coerce participation and thereby compromise ethical research. This paper seeks to discuss some of the ethical issues arising from community engagement in a low resource setting. METHODS: A qualitative study design focusing on the engagement activities of three biomedical research projects as ethnographic case studies was used to gain in-depth understanding of community engagement as experienced by multiple stakeholders in Malawi. Data was collected through participant observation, 43 In-depth interviews and 17 focus group discussions with community leaders, research staff, community members and research participants. Thematic analysis was used to analyse and interpret the findings. RESULTS: The results showed that structural coercion arose due to an interplay of factors pertaining to social-economic context, study design and power relations among research stakeholders. The involvement of community leaders, government stakeholders, and power inequalities among research stakeholders affected some participants' ability to make autonomous decisions about research participation. These results have been presented under the themes of perception of research as development, research participants' motivation to access individual benefits, the power of vernacular translations to influence research participation, and coercive power of leaders. CONCLUSION: The study identified ethical issues in community engagement practices pertaining to structural coercion. We conclude that community engagement alone did not address underlying structural inequalities to ensure adequate protection of communities. These results raise important questions on how to balance between engaging communities to improve research participation and ensure that informed consent is voluntarily given.

A comparative impact evaluation of two human resource models for community-based active tuberculosis case finding in Ho Chi Minh City, Viet Nam.

BACKGROUND: To achieve the WHO End TB Strategy targets, it is necessary to detect and treat more people with active TB early. Scale-up of active case finding (ACF) may be one strategy to achieve that goal. Given human resource constraints in the health systems of most high TB burden countries, volunteer community health workers (CHW) have been widely used to economically scale up TB ACF. However, more evidence is needed on the most cost-effective compensation models for these CHWs and their potential impact on case finding to inform optimal scale-up policies. METHODS: We conducted a two-year, controlled intervention study in 12 districts of Ho Chi Minh City, Viet Nam. We engaged CHWs as salaried employees (3 districts) or incentivized volunteers (3 districts) to conduct ACF among contacts of people with TB and urban priority groups. Eligible persons were asked to attend health services for radiographic screening and rapid molecular diagnosis or smear microscopy. Individuals diagnosed with TB were linked to appropriate care. Six districts providing routine NTP care served as control area. We evaluated additional cases notified and conducted comparative interrupted time series (ITS) analyses to assess the impact of ACF by human resource model on TB case notifications. RESULTS: We verbally screened 321,020 persons in the community, of whom 70,439 were eligible for testing and 1138 of them started TB treatment. ACF activities resulted in a + 15.9% [95% CI: + 15.0%, + 16.7%] rise in All Forms TB notifications in the intervention areas compared to control areas. The ITS analyses detected significant positive post-intervention trend differences in All Forms TB notification rates between the intervention and control areas (p = 0.001), as well as between the employee and volunteer human resource models (p = 0.021). CONCLUSIONS: Both salaried and volunteer CHW human resource models demonstrated additionality in case notifications compared to routine case finding by the government TB program. The salaried employee CHW model achieved a greater impact on notifications and should be prioritized for scale-up, given sufficient resources.

The role of active case finding in reducing patient incurred catastrophic costs for tuberculosis in Nepal.

BACKGROUND: The World Health Organization (WHO) End TB Strategy has established a milestone to reduce the number of tuberculosis (TB)- affected households facing catastrophic costs to zero by 2020. The role of active case finding (ACF) in reducing patient costs has not been determined globally. This study therefore aimed to compare costs incurred by TB patients diagnosed through ACF and passive case finding (PCF), and to determine the prevalence and intensity of patient-incurred catastrophic costs in Nepal. METHODS: The study was conducted in two districts of Nepal: Bardiya and Pyuthan (Province No. 5) between June and August 2018. One hundred patients were included in this study in a 1:1 ratio (PCF: ACF, 25 consecutive ACF and 25 consecutive PCF patients in each district). The WHO TB patient costing tool was applied to collect information from patients or a member of their family regarding indirect and direct medical and non-medical costs. Catastrophic costs were calculated based on the proportion of patients with total costs exceeding 20% of their annual household income. The intensity of catastrophic costs was calculated using the positive overshoot method. The chi-square and Wilcoxon-Mann-Whitney tests were used to compare proportions and costs. Meanwhile, the Mantel Haenszel test was performed to assess the association between catastrophic costs and type of diagnosis. RESULTS: Ninety-nine patients were interviewed (50 ACF and 49 PCF). Patients diagnosed through ACF incurred lower costs during the pre-treatment period (direct medical: USD 14 vs USD 32, P = 0.001; direct non-medical: USD 3 vs USD 10, P = 0.004; indirect, time loss: USD 4 vs USD 13, P <  0.001). The cost of the pre-treatment and intensive phases combined was also lower for direct medical (USD 15 vs USD 34, P = 0.002) and non-medical (USD 30 vs USD 54, P = 0.022) costs among ACF patients. The prevalence of catastrophic direct costs was lower for ACF patients for all thresholds. A lower intensity of catastrophic costs was also documented for ACF patients, although the difference was not statistically significant. CONCLUSIONS: ACF can reduce patient-incurred costs substantially, contributing to the End TB Strategy target. Other synergistic policies, such as social protection, will also need to be implemented to reduce catastrophic costs to zero among TB-affected households.

Use of GeneXpert and the role of an expert panel in improving clinical diagnosis of smear-negative tuberculosis cases.

SETTING: A high proportion of notified tuberculosis cases in the Philippines are clinically diagnosed (63%) as opposed to bacteriologically confirmed. Better understanding of this phenomenon is required to improve tuberculosis control. OBJECTIVES: To determine the percentage of smear negative presumptive tuberculosis patients that would be diagnosed by GeneXpert; compare clinical characteristics of patients diagnosed as tuberculosis cases; and review the impact that the current single government physician and a reconstituted Tuberculosis Diagnostic committee (expert panel) may have on tuberculosis over-diagnosis. DESIGN: This a cross-sectional study of 152 patients 15-85 years old with two negative Direct Sputum Smear Microscopy results, with abnormal chest X-ray who underwent GeneXpert testing and review by an expert panel. RESULTS: Thirty-two percent (48/152) of the sample were Xpert positive and 93% (97/104) of GeneXpert negatives were clinically diagnosed by a single physician. Typical symptoms and X-ray findings were higher in bacteriologically confirmed tuberculosis. When compared to the GeneXpert results the Expert panel's sensitivity for active tuberculosis was high (97.5%, 39/40), specificity was low (40.2%, 35/87). CONCLUSION: Using the GeneXpert would increase the level of bacteriologically confirmed tuberculosis substantially among presumptive tuberculosis. An expert panel will greatly reduce over-diagnosis usually seen when a decision is made by a single physician.

Design and protocol for a pragmatic randomised study to optimise screening, prevention and care for tuberculosis and HIV in Malawi (PROSPECT Study).

Background: Adults seeking diagnosis and treatment for tuberculosis (TB) and HIV in low-resource settings face considerable barriers and have high pre-treatment mortality. Efforts to improve access to prompt TB treatment have been hampered by limitations in TB diagnostics, with considerable uncertainty about how available and new tests can best be implemented. Design and methods: The PROSPECT Study is an open, three-arm pragmatic randomised study that will investigate the effectiveness and cost-effectiveness of optimised HIV and TB diagnosis and linkage to care interventions in reducing time to TB diagnosis and prevalence of undiagnosed TB and HIV in primary care in Blantyre, Malawi. Participants (≥ 18 years) attending a primary care clinic with TB symptoms (cough of any duration) will be randomly allocated to one of three groups: (i) standard of care; (ii) optimised HIV diagnosis and linkage; or (iii) optimised HIV and TB diagnosis and linkage. We will test two hypotheses: firstly, whether prompt linkage to HIV care should be prioritised for adults with TB symptoms; and secondly, whether an optimised TB triage testing algorithm comprised of digital chest x-ray evaluated by computer-aided diagnosis software and sputum GeneXpert MTB/Rif can outperform clinician-directed TB screening. The primary trial outcome will be time to TB treatment initiation by day 56, and secondary outcomes will include prevalence of undiagnosed TB and HIV, mortality, quality of life, and cost-effectiveness. Conclusions: The PROSPECT Study will provide urgently-needed evidence under "real-life" conditions to inform clinicians and policy makers on how best to improve TB/HIV diagnosis and treatment in Africa. Clinical trial registration: NCT03519425 (08/05/2018).

Health seeking for chronic lung disease in central Malawi: Adapting existing models using insights from a qualitative study.

BACKGROUND: Chronic lung diseases contribute to the growing non-communicable disease (NCD) burden and are increasing, particularly in many low and middle-income countries (LMIC) in sub-Saharan African. Early engagement with health systems in chronic lung disease management is critical to maintain quality of life and prevent further damage. Our study sought to understand health seeking behaviour in relation to chronic lung disease and TB in a rural district in Malawi. METHODS: Qualitative data was collected between March-May 2015, exploring patterns of health seeking for lung disease amongst residents of two districts in rural Malawi. Participants included those with and without lung disease, health workers and village leaders. Participants with a history of TB were included in the sample due to similarities in clinical presentation and in view of potential to cause long-term damage to lung tissue. RESULTS: Our findings are ordered around a specific model of health seeking devised by adapting previous models. The model and findings span three broad areas that were found to influence health seeking: understandings of health and disease which shaped whether, when and where to seek care; the care seeking decision which was influenced by social and structural factors; and the care seeking experience which impacted future care decisions creating 'feedback loops'. DISCUSSION: Efforts to improve effective and accessible healthcare provision for chronic lung disease need to address all the determinants of health seeking behaviour identified. This may include: enhancing the structural and financial accessibility of health services, through the strengthening of community linkages; improving communication between formal health providers, patients and communities around symptoms, diagnosis and management of chronic lung diseases; and improving the quality of diagnostic and management services through the strengthening of health systems 'hardware' (equipment availability) and 'software' (development of trusting and respectful relationships between providers and patients).

'We are the eyes and ears of researchers and community': Understanding the role of community advisory groups in representing researchers and communities in Malawi.

Community engagement to protect and empower participating individuals and communities is an ethical requirement in research. There is however limited evidence on effectiveness or relevance of some of the approaches used to improve ethical practice. We conducted a study to understand the rationale, relevance and benefits of community engagement in health research. This paper draws from this wider study and focuses on factors that shaped Community Advisory Group (CAG) members' selection processes and functions in Malawi. A qualitative research design was used; two participatory workshops were conducted with CAG members to understand their roles in research. Workshop findings were triangulated with insights from ethnographic field notes, key informant interviews with stakeholders, focus group discussions with community members and document reviews. Data were coded manually and thematic content analysis was used to identify main issues. Results have shown that democratic selection of CAG members presented challenges in both urban and rural settings. We also noted that CAG members perceived their role as a form of employment which potentially led to ineffective representation of community interests. We conclude that democratic voting is not enough to ensure effective representation of community's interests of ethical relevance. CAG members' abilities to understand research ethics, identify potential harms to community and communicate feedback to researchers is critical to optimise engagement of lay community and avoid tokenistic engagement.

Economic evaluation of the practical approach to lung health and informal provider interventions for improving the detection of tuberculosis and chronic airways disease at primary care level in Malawi: study protocol for cost-effectiveness analysis.

BACKGROUND: Chronic airway diseases pose a big challenge to health systems in most developing countries, particularly in Sub-Saharan Africa. A diagnosis for people with chronic or persistent cough is usually delayed because of individual and health system barriers. However, delayed diagnosis and treatment facilitates further transmission, severity of disease with complications and mortality. The objective of this study is to assess the cost-effectiveness of the practical approach to lung health strategy, a patient-centred approach for diagnosis and treatment of common respiratory illnesses in primary healthcare settings, as a means of strengthening health systems to improve the quality of management of respiratory diseases. METHODS/DESIGN: Economic evaluation nested in a cluster randomised controlled trial with three arms will be performed. Measures of effectiveness and costs for all arms of the study will be obtained from the cluster randomised controlled clinical trial. The main outcome measures are a combined rate of major respiratory diseases milestones and process indicators extracted from the practical approach to lung health strategy. For analysis, descriptive as well as regression techniques will be used. A cost-effectiveness analysis will be performed according to intention-to-treat principle and from a societal perspective. Cost-effectiveness ratios will be calculated using bootstrapping techniques. DISCUSSION: We hope to demonstrate the cost-effectiveness of the practical approach to lung health and informal healthcare providers, see an improvement in patients' quality of life, achieve a reduction in the duration and occurrence of episodes and the chronicity of respiratory diseases, and are able to report a decrease in the social cost. If the practical approach to lung health and informal healthcare provider's interventions are cost-effective, they could be scaled up to all primary healthcare centres. TRIAL REGISTRATION: PACTR: PACTR201411000910192.

Improving access to effective care for people with chronic respiratory symptoms in low and middle income countries.

Chronic respiratory symptoms are amongst the most common complaints among low and middle-income country (LMICs) populations and they are expected to remain common over the 10 to 20 year horizon. The underlying diseases (predominantly chronic obstructive pulmonary disease, asthma and tuberculosis) cause, and threaten to increasingly cause, substantial morbidity and mortality. Effective treatment is available for these conditions but LMICs health systems are not well set up to provide accessible clinical diagnostic pathways that lead to sustainable and affordable management plans especially for the chronic non communicable respiratory diseases. There is a need for clinical and academic capacity building together with well-conducted health systems research to underpin health service strengthening, policy and decision-making. There is an opportunity to integrate solutions for improving access to effective care for people with chronic respiratory symptoms with approaches to tackle other major population health issues that depend on well-functioning health services such as chronic communicable (e.g. HIV) and non-communicable (e.g. cardiovascular and metabolic) diseases.

Infectious disease and health systems modelling for local decision making to control neglected tropical diseases.

Most neglected tropical diseases (NTDs) have complex life cycles and are challenging to control. The "2020 goals" of control and elimination as a public health programme for a number of NTDs are the subject of significant international efforts and investments. Beyond 2020 there will be a drive to maintain these gains and to push for true local elimination of transmission. However, these diseases are affected by variations in vectors, human demography, access to water and sanitation, access to interventions and local health systems. We therefore argue that there will be a need to develop local quantitative expertise to support elimination efforts. If available now, quantitative analyses would provide updated estimates of the burden of disease, assist in the design of locally appropriate control programmes, estimate the effectiveness of current interventions and support 'real-time' updates to local operations. Such quantitative tools are increasingly available at an international scale for NTDs, but are rarely tailored to local scenarios. Localised expertise not only provides an opportunity for more relevant analyses, but also has a greater chance of developing positive feedback between data collection and analysis by demonstrating the value of data. This is essential as rational program design relies on good quality data collection. It is also likely that if such infrastructure is provided for NTDs there will be an additional impact on the health system more broadly. Locally tailored quantitative analyses can help achieve sustainable and effective control of NTDs, but also underpin the development of local health care systems.

Modelling the impacts of new diagnostic tools for tuberculosis in developing countries to enhance policy decisions.

The introduction and scale-up of new tools for the diagnosis of Tuberculosis (TB) in developing countries has the potential to make a huge difference to the lives of millions of people living in poverty. To achieve this, policy makers need the information to make the right decisions about which new tools to implement and where in the diagnostic algorithm to apply them most effectively. These decisions are difficult as the new tools are often expensive to implement and use, and the health system and patient impacts uncertain, particularly in developing countries where there is a high burden of TB. The authors demonstrate that a discrete event simulation model could play a significant part in improving and informing these decisions. The feasibility of linking the discrete event simulation to a dynamic epidemiology model is also explored in order to take account of longer term impacts on the incidence of TB. Results from two diagnostic districts in Tanzania are used to illustrate how the approach could be used to improve decisions.