Recommendations

Key Principles and Approaches

Align recommendations with country's vision and challenges

One of the other core principles followed is to align the recommendations with the country's existing vision and priorities in order to address the key challenges faced by the country. These challenges can differ significantly from country to country based on different factors discussed below.

❖ Stage and maturity of TB surveillance system

The recommendations vary from country to country based on the stage and maturity of the TB surveillance system. For example, for countries that have only an aggregated TB notification system, the typical recommendation is to first create a case-based TB surveillance system for the entire continuum of care, whereas for countries that already have a case-based TB surveillance system, the recommendation is to explore other advanced features/modules such as contact investigation, GxAlert integration, etc.

❖ Alignment with roadmap and vision

The approach is also to keep the recommendations aligned with the country's existing vision and roadmap, shared as part of the assessment exercise.

❖ Prioritization based on challenges

The recommendations also reflect the challenges highlighted by the countries as part of the assessment. While many of the basic challenges are similar across the countries, the countries also vary significantly in terms of how and when they can resolve the challenges.

Non-prescriptive

Another core principle of the recommendations is to keep them non-prescriptive. The idea is to share the best practices based on our combined understanding from multiple countries and stakeholders. However, the countries should evaluate these recommendations based on their country context, budget availability, priorities and mandates.

Key Generic Recommendations

While the recommendations vary from country to country based on the principles indicated in the previous section, some of the common recommendations provided to achieve the ideal functional and technical architecture for a case-based TB surveillance system are as follows:

Interoperability and system integration

One of the biggest opportunities and enabling environments is the digital ecosystem that already exists within the national programme – whether fully owned and managed by the NTP or managed by an external private or public entity, including other programmes in the Ministry of Health, and across Ministries such as Information or Telecommunications.

These existing data ecosystems form a strong foundation for promoting interoperability, which would enable countries to integrate and facilitate seamless data exchange between different data systems such as medicine inventory, laboratory information management systems, GeneXpert (GxAlert) system, CLM and more. The idea is to create an integrated and harmonized data exchange system in which each part is complementary and duplication of effort is avoided. This ensures that existing efforts are better optimized and utilized, instead of introducing a new system.

As part of the assessment, we recommend that countries implement the following broad measures to ensure interoperability:

❖ Leverage interoperability standards like FHIR (Fast Healthcare Interoperability Resources). FHIR is a standard describing data formats and elements and an application programming interface for exchanging electronic health records. This would enable NTPs to exchange data with other health records or data systems such as diagnostic devices, logistics management systems and other data systems. It would also ensure high-level data security and privacy during the data exchange processes.

❖ Creation of centralized data warehouse: A strong data warehouse, data governance and analytics framework are core to ensuring interoperability. These aspects will not only enable the sharing and reporting of TB data with other national health information systems, but also facilitate the consumption and triangulation of data from other data sources, including open data repositories.

❖ Recommended exchange / ETL tools: Tools like Talend or Informatica that include these features make data management and data governance tasks much easier.

❖ Develop the necessary application program interface (API) and single sign-on: TB surveillance systems that have already been developed or are in the process of development should ensure development of the necessary APIs so that they can integrate with other systems.

Enhancing hardware infrastructure

All countries without exception expressed the need for and highlighted the importance of devices to ensure real-time TB surveillance and data use at the lowest levels. Providing mobile devices and mobile Internet could be a crucial and cost-effective way to ensure successful implementation and adoption of these tools, and investing in hardware as the backbone of the case-based TB surveillance system and other ancillary data systems. The investment made in devices can be optimized by maximizing their use in other health programmes, as most health functionaries at the lowest level units are often tasked with responsibilities from multiple programmes.

Hardware includes mobile devices/tablets/laptops for outreach/facility staff for real-time data entry and access to analytical dashboards/job aids, etc.

Although servers are also an important component of the hardware infrastructure, in most of the countries assessed, servers and deployment infrastructure are already available either directly within the TB programme or within the broader Ministry of Health or national government. Cloud-based hosting is generally recommended, as it offers appropriate security and ease of operation.

Case based monitoring across the continuum of care

While digital case-based TB notification is the most basic and important data variable and early indicator for the development of a digital TB surveillance system that all countries should strive to achieve, one of the key processes and functional recommendations common to all countries is to achieve a comprehensive workflow-based system to monitor individual cases throughout the entire continuum of care – from client enrolment and risk assessment through to screening, referral, testing, treatment, treatment adherence, treatment outcome, comorbidity management and post-treatment follow-up. This should also ideally include the complete screening pathway for TB disease and infection for vulnerable groups such as HIV(Human Immunodeficiency Virus) patients and contacts of people with TB in order to identify and refer clients for TPT.

Advanced analytics for better data use

One of the most crucial aspects of the assessment was to review the status of data use in the countries. Most country programme leaderships unequivocally consider this to be one of the most crucial areas. Although most countries have some form of data and data use tool, very few countries have a real-time dashboard with data at the lowest levels, not only for M&E purposes but also for better patient management and care. There are various innovations that can result in the most effective use of data for programmatic improvement. Some of the recommendations include:

• Enabling real-time monitoring of the cascade of care pathway;
• Big data analytics and AI-based predictive modelling, mapping, risk profiling and prioritization for actions of people with TB using appropriate machine learning and AI systems;
• Geospatial analytics;
• Integration of CLM indicators with programme indicators;
• Automated AI-based job aids for programme staff, etc.

Given that there are different data systems working in conjunction, it is also recommended to explore creation of a centralized data repository/data warehouse system, bringing data from multiple sources into one place.

Existing country tools like DHIS2 have strong data analytics and dashboard functionality with robust indicator configuration. It is also recommended to explore best-of-breed tools such as Tableau and Power BI, which offer these features. APIs can be generated and connected with these applications, and these can be used as an extended analytical component of the data analysis framework.

Community Led Monitoring (CLM)

Community-based monitoring (CBM)/CLM is one way of generating granular data to provide feedback to service providers and decision-makers in order to collaboratively solve barriers and bottlenecks to services and improve the quality of services. Examples of CBM models include community treatment observatories, human rights complaints mechanisms, and scorecards. Users and communities gather, analyse and use information to improve access to services, better target resources, and address human rights and gender barriers. This is crucial for the overall success of the programme. One of the key recommendations is to fully integrate CLM into the programmatic framework of the country. This can be achieved in different ways, as outlined below:

• Include CLM indicators among the core indicators of the NTP – mainly in terms of how different barriers to services reported by the community are impacting programmatic targets and outcomes.
• Promote CLM data collection by CSOs and other human rights organizations, or include key CLM data variables as part of the M&E of the national programme.

e - Training

Providing training to health staff at the lowest level units has always been a challenge. The COVID-19 crisis has only broadened the training gaps, resulting in huge data quality issues related to the completion of data, timeliness of data, and authenticity of data in the M&E process across the globe.

Developing comprehensive e-Training/e-Learning platforms to facilitate synchronous and asynchronous video and audio training sessions, dynamic training content management systems and the use of gamification has the potential to improve data collection and data use processes and contribute to CME and Manuals of Procedures for TB programmes.