OpenIDEO is an open innovation platform. Join our global community to solve big challenges for social good. Sign Up / Login or Learn more

Modeling EPI program for the sake of explaining performances through a causal model

Identification of the underlying causes through innovative statistical approaches, namely data mining and modeling based on Bayesian network

Photo of Raphael Girod
0 0

Written by

Describe what you intend to do and how you'll do it in one to two sentences (required 350 Characters)

The modeling project is aiming at mapping point densities, as one could use statistics to test whether EPI related performances were significantly associated with other aspects related to human, supply chain, logistics, etc... Under specific assumptions, such statistical associations based upon “Bayesian network” can even determine causation.

Explain the innovation (2,500 characters)

As showed in the other MAHA projects, GIS can be used in a variety of ways to visually display health service utilization as well as take into account the many factors related to locational constraints, which may limit an individual from acquiring proper immunization. The envisioned location-allocation model is going to be used to identify and more importantly, explain why and how gaps in health services among specific communities can be reduced. The bayesian network framework helps in providing justification on how inputs and outputs and intermediary results linked to health outcomes, can contribute to improving specific health results related to EPI program, and, how they can have a broader health impact on coverage and equity rates. In addition to this, it is possible to formally explain how the health system-wide interventions can be quantitatively assessed, thanks to the theory of change model laid down based upon the framework of the result chain for GAVI HSS grant. Bayesian networks make it possible to identify relevant drivers for Health System Support in health database and establish priorities. With this approach, public health evaluator can quickly and transparently, generate clear recommendations for decision makers. The proposed mathematical model or statistical tool, thanks to Bayesian network approach, is aiming at setting up that causal chain of results or “theory of change” model. Bayesian networks make it possible to identify relevant drivers for Health System Support in health database and establish priorities. With this approach, public health evaluator can quickly and transparently, generate clear recommendations for decision makers. The proposed mathematical model or statistical tool, thanks to Bayesian network approach, is aiming at setting up that causal chain of results or “theory of change” model. The theory of change or causal model to be built up will duly take into consideration the result chain for GAVI HSS. In the present case, it is possible to start setting up a model using the template provided in GAVI HSS ( See attached file “Results chain for GAVI HSS grant”)

Which part(s) of the world does this innovation target?

  • North Africa
  • Eastern Africa
  • Middle Africa
  • Southern Africa
  • Western Africa
  • Caribbean
  • Central Asia
  • Eastern Asia
  • South-eastern Asia
  • Southern Asia
  • Western Asia
  • Eastern Europe
  • Southern Europe
  • Oceania

Geographic Focus

Urban areas and slums

Stage of Innovation

  • Prototype

Who will work alongside your organization in the project idea? (1,000 characters)

It is worth defining the stakeholders who are impacted, or can impact that innovative project, so be sure they if will be consulted / informed / engaged: Ministry of Health and In-country statistical Health Service; United Nations Organizations (UNICEF, WHO). - Partners: strong partnership engagement with DHIS2 team & Country Statistical Service through their respective focal points, will be required for business, technical/ technological, legal aspects. Data governance, management and integration represent an important managerial and technical challenges, because of the nature of the data to be processed. The project abide by the data management principles (including interoperability with DHIS2 data warehouses and Country statistical service master data management initiatives) aiming at applying a scalable integration process of dataset and benefiting from manageable data flows and security procedures.

How is your idea unique? (750 characters)

In regards to the setting up of a realistic model, the software utilized has key functionalities so to apply a “Probabilistic Structural Equations” methodology (See One pager technical brief Modeling), as a consequence the following steps will be respected: 1- Unsupervised learning to discover all the direct probabilistic relations 2- Supervised learning to characterize the targeted variables 3- Variable clustering to induce “factors” made of highly connected variables The model will mirror the GAVI HSS diagram and corresponding indicators so to demonstrate how activities contribute to achieving intermediate results and how intermediate results contribute to achieving immunization outcomes or impact indicators.

What is the name of your organization

MAHA - Mapping & Analytics for Health Activities

Explain your organization (250 characters)

MAHA is an organization tasked with supplying its in-country partners with IT services including mapping and analytical services. The shared service approach aims at increasing the quality and professionalism of GIS services for the EPI program.

Type of Submitter

  • We are a For-Profit Startup or Startup Social Enterprise

Organizational Characteristics

  • MAHA is composed of GIS, IT infrastructure and developer experts located mainly in France and Malawi

Gender and Diversity (500 characters)

MAHA is developing Geographical Information System (GIS) based applications and services. The gender is always taken into consideration in the indicators mapped and analyzed. Every time database are disaggregated by gender, base maps will take benefit of disaggregated information (to the lowest possible granular/administrative unit) in order to better understand specificities in terms of gender, and if possible by urban poor individuals, households and communities.

Organization Location (less than 250 Characters)

Raphael Girod Founder MAHA Mapping & Analytics for Health Activities 59, vieille route des Pensières 74290 Veyrier du lac - France mob. (33) 6 33 58 65 53

Size of organization (number of employees):

  • Less than 5 people

Website URL

https://mahacenter.com

Scale of organizational work

  • Global (within 2 or more global regions)

Tell us more about you

Mapping and Analytics for Health Activities (MAHA), is a new French start up linking public health and information technology since March 2018. MAHA was contracted to provide innovative GIS Services to Ghana Health Service in 2018 with funding from GAVI. MAHA is providing GIS and mapping services in the framework of a “Technical Assistance for development of National health care delivery system optimization plan for the Kyrgyz Republic (Master Plan)” led by a German consulting firm, namely Management for Health (M4H) funded by World Bank.

Applying to Gavi INFUSE

  • Referred by a friend / colleague

0 comments

Join the conversation:

Comment