← Back to in the News
November 1, 2019

Q&A with Neelima Ramaraju: Solving the TB Crisis Using Tech

LLamasoft, a supply chain design software designer, and a FIND non-profit healthcare organization, have recently partnered to use AI technology to find millions of undiagnosed patients impacted by tuberculosis

Tuberculosis is one of the top 10 killers in the world affecting 10 million people annually. Of those 10 million, 3.6 million were “missing cases,” - undiagnosed or undetected and not reported due to a lack of access to diagnostics. As part of the new partnership, by using LLamasoft technology, the Foundation for Innovative New Diagnostics (FIND) was able to identify the distribution gaps in existing services to help effectively reach the missing millions, and model a set of new network paradigms to advise governments how to reach these patients.

To discover more about this project, Digital Journal spoke with Neelima Ramaraju, Senior Director Global Impact, LLamasoft about how this program is working in terms of finding missing TB cases in the major regions where there is a high burden for TB.

Digital Journal: What are the current rates for tuberculosis in the world?

Neelima Ramaraju: Tuberculosis (TB) is one of the top 10 causes of death worldwide and in 2018 alone 10 million people globally fell ill with TB. It’s a communicable disease, meaning it can be spread from person to person. If the disease is diagnosed in time and treated, most patients can be cured, and transmission rates can be controlled, however, nearly a quarter of the world’s populations are exposed to the disease putting them at risk of developing it.

DJ: Are cases increasing and what are the areas most impacted? If so, what are the triggers?

Ramaraju: While TB is found worldwide, a majority of the burden occurs in low and middle-income countries in South-East Asia (44%), Africa (24%) and Western Pacific (18%). Although the infection rates for TB have been somewhat steady over the years, drug-resistant TB poses a large threat to global public health. Accurate diagnosis for TB is a complex, multi-step process. These testing options are rarely available in a majority of health facilities, especially in rural areas.

Additionally, treatment for TB is very long, spanning 6 months or more, and expensive. If treatment is not accessible (either because it’s not physically there or is too expensive), the risk of drug-resistance is higher. If the disease goes undetected for long, the chance of it spreading to others is increased.

DJ: How many cases go undiagnosed and what are the reasons for this?

Ramaraju: Of the reported TB cases, many of them go “missing” from the system and are not treated – in previous years up to 40% of the cases if infections have been classified as missing. The most obvious reason for this is the lack of access to quality healthcare is resource constraints (HR, equipment, supplies, etc.), political or geographical constraints or health system limitations.

DJ: What are the biggest challenges in optimizing a supply chain geared toward tackling a public health crisis?

Ramaraju: The health systems, including the supply chain and diagnostics networks, play a crucial role in tackling complex public health challenges. In the area of designing effective diagnostic networks, the primary challenges include the lack of holistic data, large stakeholder community, geographical barriers and the changing landscape of the technologies introduced for testing solutions. Frequently the testing centers are too far from the population or do not have the proper resources to be functional. This means that many people who need to be tested and want the services do not have access to them.

The key elements of data required include not just the historical testing rates rather the number and location of the missing cases. Additionally, information about the existing footprint of the network, where the key resources are located and which organizations are responsible for them, is needed. When new and improved testing technology is introduced, it’s not always clear where to place them and how to integrate them into the existing network leading to duplication of efforts and cost. With dozens of stakeholders (donors, Ministries of Health, implementing partners, NGOs), coordination amongst them all is another key challenge.

DJ: How does the LLamasoft technology help to design these networks?

Ramaraju: Several elements of the LLamasoft technology suite are used to design diagnostic networks that are effective, reliable and sustainable in the long term. The first step is typically to get a holistic picture of the existing network. This is done through LLamasoft’s Data Guru ™ tool that is used to compile the data from various sources, identify gaps, and work to create a complete and clean data set. Elements of Demand Forecasting are also used to help generate the future-state demand profiles.

LLamasoft’s Network Optimization tools are used to design the optimal systems with the right resources in the right place at the lowest cost.

The detailed scenario modeling or “what-if analysis,” hones in on the right solution for the given context. This ensures that the customer, a country team, can serve their population most effectively and any budget that is saved can go back to the other critical programs. Another element of the design is to reduce the overall turn-around-time. The faster the case can be diagnosed, and the patient brought in to get treated, the better the chances of curtailing the disease.

The detailed maps of the current and proposed networks and visual dashboards to identify the pain points in the system are novel to this space and enable the diverse stakeholder groups to make a data-driven decision together.

DJ: How was the technology developed?

Ramaraju: LLamasoft has worked on global health problems for over a decade. The early projects were mostly focused on the more traditional supply chain challenges of getting medicines from the point of origin to the final customer at the service delivery point. In doing that work, the value of a holistic network analysis for the diagnostics system was discovered.

Through several of our partners, we engaged in projects to re-designed diagnostic systems in a few countries and went on to develop a rapid analysis tool, easier to use than our traditional Supply Chain Guru™ tool, to make the analysis more accessible to those that need it most. More recently, our work with FIND (The Foundation for Innovative New Diagnostics), a global non-profit focused on affordable diagnostic tests for poverty-related diseases, has led to more nuanced and detailed analyses.

DJ: What success has FIND had with this technology to date?

Partners such as FIND bring a deep expertise in the global diagnostics programs. Through our collaboration, we have been able to work on diagnostic network analyses in countries such as Kenya, Lesotho, and the Philippines, to optimize for resource and budget allocations and sample referral patterns. These activities resulted in reduction in procurement and operational costs, and an overall increase in the number of people diagnosed and treated.

Another realization from these engagements was that country stakeholders were engaged in the network design process and eager for more analytical tools to continue to assess and improve their diagnostics networks.

DJ: Are there any future plans to expand the program with new initiatives that can increase the number of people helped?

Ramaraju: The primary goal for the Global Impact Team at LLamasoft is to positively impact lives across the globe through the work that we do. LLamasoft’s Digital Design and Decision Center allows for a platform of continuous improvement and makes the powerful algorithms and analytics accessible to all users. We are very excited about some new partnerships to develop custom applications for targeted use-cases on LLamasoft’s platform. In this instance, a targeted solution that will allow stakeholders in various countries across the globe login, upload their data, and start to visualize and model scenarios on their own.
Original Source