1. Abstract Building an Open Access Software Tool to allow countries to design patient-centered and cost-effective Diagnostic Networks
Ryan Purcell, LLamasoft
Heidi Albert, FIND
November 21st, 2019

2. Gaps in access and efficiency of diagnostic services
24% patients with TB initiated care seeking in a health facility with diagnostic capacity1
66% patients sought care in private sector where diagnostic capacity is even lower1
Where testing capacity is available, it is often under- utilized and of variable quality.
Significant investments made, but largely focused on device procurement.
Patient pathway data from Ethiopia, Indonesia, Kenya, Pakistan and Philippines
How do we increase patient access to quality diagnosis while improving service delivery efficiency?
Hanson, J Infect Dis

3. Diagnostic network planning today
Diagnostics are essential to guide patient care and treatment, but diagnostic services are often lacking where patients first seek care
Network design that is better aligned with patient health seeking behavior is expected to reduce barriers to care and improve efficiency of investments and service delivery
Improved access and efficiency are both key to achieving national and global disease control targets.
Currently, decisions are often based on MS Excel work with current testing and case-detection data (usually for individual disease programs), possibly with simple visualization tools like Google Maps

4. Diagnostics network design and optimization
Objectives
Map TB burden and current demand for TB diagnostic services
Map current TB diagnostic network structure
Identify the extent and distribution of gaps in services
Develop a set of diagnostic network designs defined by NTP and partners that better and more efficiently reach “missing” TB cases (unmet demand) using existing infrastructure
Model a set of new network paradigms, including new product and/or services investment, to inform government strategic planning and budgeting processes
Example questions:
What proportion of presumptive TB cases are reached by the existing TB diagnostic network? What is the capacity for expansion?
Where is the trade-off between increasing point of care and establishing a better sample referral mechanism? What would be the most efficient design for sample referral to minimize turnaround times and cost?
When rolling out new instruments, which sites will show greatest impact in terms of case detection, and thus should be prioritized for early implementation?

5. Increase network efficiency Greater visibility of network
Expected outputs
Improve access
Reduce loss to follow up and diagnostic delay
More people diagnosed and treated
Increase network efficiency
Reduce procurement and operating costs
Easier-to-manage device footprint
Greater visibility of network
More efficient funding allocation
Country empowerment & better decision-making

6. Vision for diagnostic network optimization tool
A standardized methodology that is accepted as a key component of a data-driven approach to national strategic planning (across diseases), as well as the design and optimization of diagnostic networks using existing data sources.
To reduce barriers to uptake and facilitate sustainable use, it is imperative that a user friendly tool to facilitate this analysis is openly accessible, freely available to national programs and partners, and whose use is endorsed and supported by major donors and global organizations.

7. What’s next for laboratory network optimization?
Purpose-built point solution app for Diagnostic Networks

8. Supply Chain Guru® (SCG)
Open Access tool: bringing together network optimization capabilities with access and usability in LMICs
Proprietary software requiring license; desktop and web versions
Leading supply chain management software in corporate sector and significant experience in global health applications
Multi-disease
Mapping and visualization of current and future demand and network structure, incl. multi-echelon
Flexible input tables
Robust performance for large data sets
Customizable outputs/ data dashboards
Ability to link to external data systems
Supply Chain Guru® (SCG)
LabEQIP
Open Access tool
Open access, web-based tool
Multi-disease (TB/HIV and other diseases)
Designed for national programs
Includes key SCG functionalities – flexible inputs, customizable, robust with large datasets
Set up to be able to link to external data systems, e.g. ForLab (not part of initial build)
Open source, desktop software
Mapping and visualization of current and future demand and network structure (no multi-echelon)
Network optimization and EQA functionality
Simple to use
Primarily HIV
Limited demand inputs
No costing functionality
Side-to-side comparison of multiple scenario outputs not available
Performance not guaranteed beyond 3000 sites
Costing of different network options (including site costs and sample transport costs)
Multi-echelon network optimization to incorporate referrals and sample transportation (e.g. hubs)
Data-size limitations –performance is not guaranteed past sites
More detailed inputs required for demand calculations
Side-by-side comparison of multiple scenario outputs for network optimization

9. Open Access Tool Development Overview
Strategic partnership established between FIND, PSM/Chemonics and LLamasoft for open access tool development and roll out, including:
Joint product requirements for open access tool including TB and HIV
Web-based tool, available free of cost to end users in LMICs
Tool development funded by a Bill & Melinda Gates Foundation grant to FIND
Hosting/maintenance supported by PSM/Chemonics with USAID/PEPFAR funding
Steering committee mechanism established to guide tool development and roll out with provision of ad hoc members for advice from various stakeholders
Tool build started Sept 2019 and expected to be completed by mid 2020
Country pilots of open access tool planned for late 2020
Who is on the steering committee? Gates, USAID (TB + HIV), CDC, FIND, PSM. Monthly/bimonthly meetings
Technical Sub-Committee: LL, FIND, PSM. Every two weeks
Country Pilot options: Kenya, PHL, India

10. Key functions
Demand: How many tests do we expect at each health facility in the future? What about as we change testing algorithm and demand assumptions?
Options for simplified and detailed demand estimation
 Network Optimization: how can a full diagnostic network be organized to most efficiently offer services?
Example questions include:
How many machines should we have? Where should they be located?
How should samples be referred from health facilities to testing locations? Flows could be multi-echelon (e.g. Health Facility to Hub to Lab to Higher Level Lab)
 Vehicle Route Optimization (VRO) (basic costing estimate for multi-stop routes)
If the samples are transported using multi-stop routes, what will be the expected cost?
Three options for demand data:
Upload demand data directly to the template – Simple Demand Calculation
Scale up/down current demand using a scaling factor (only available where current demand data has already been provided for baseline model) – Simple Demand Calculation
Generate demand from parameters – Detailed Demand Calculation (based on developed algorithmic approach for TB)

11. Illustrative example of workflow and screens: Inputs
Access via username/login based on user profile
Customized dashboard for different users
Ability to share results, link to external data sources
Models can be created and analyzed separately for each country
Repeatable workflow guides the users through data inputs and model build

12. Illustrative example of workflow and screens: Outputs
Multiple data visualization formats
Customizable dashboard
Customized view for different users

13. Thank you! Questions? Sidharth Rupani, Neelima Ramaraju, LLamasoft
Zachary Katz, Kekeletso Kao, FIND
Lillian Gu, Clement Ndongmo, Matthew C Wattleworth, Andrew Inglis, Shadrack Were, PSM
Christy Hanson, BMGF
NTLP & partners, Kenya
NTP & partners, Philippines
Financial support:
The Bill & Melinda Gates Foundation
USAID/PEPFAR
USAID GLOBAL HEALTH SUPPLY CHAIN PROGRAM
Procurement and Supply Management

14. Thanks to our generous sponsors
2019 Global Health Supply Chain Summit, Johannesburg, S. Africa