Towards a Regional Framework for Geostatistical Modelling in Schistosomiasis Programmes

19 March 2026  |  Virtual |  Hosted by WHO/AFRO and ESPEN, Brazzaville, Republic of Congo

Brazzaville (Congo) - On 19 March 2026, ESPEN convened a regional webinar on Geostatistical Modelling for Schistosomiasis Mapping and Impact Assessments, bringing together national programmes, technical partners, research institutions, and WHO teams to explore how model-based approaches can strengthen data-driven decision-making for schistosomiasis control and elimination, and how analytical innovation can be aligned with country-led data systems and the WHO strategic framework for monitoring and evaluation.

Persistent data gaps despite expanded coverage

Opening the session, Dr Jorge Cano (ESPEN) reported that the Schistosomiasis Community Data Workbook (SCH CDW) now covers 58,840 sub-implementation units across 38 countries. Only 42.7% of sub-IUs are informed by survey data and 31.1% by recent impact assessments, while 68.9% remain anchored to historical baseline mapping. More than half of sub-IUs across the Region therefore lack contemporary empirical evidence, underscoring the need for more efficient, targeted approaches to data generation and analysis.

From method to application

Dr Claudio Fronterre (University of Birmingham) and Prof Rachel L. Pullan (LSHTM) introduced model-based geostatistics as a statistically principled method for synthesising existing survey observations with spatial and environmental information, generating fine-resolution prevalence surfaces with quantified uncertainty and informing adaptive, more efficient survey designs. Dr Stella Kepha (KEMRI) illustrated the approach in practice: combining systematic empirical mapping across 32 Kenyan counties (2020–2025) with geostatistical prediction on a 5×5 km grid enabled ward-level classification against the 2% and 10% prevalence thresholds, distinguishing transmission hotspots from low-risk areas and flagging regions of higher uncertainty for targeted follow-up.

Lessons from malaria and the use of outputs

Reflecting on the malaria experience, Dr Victor Alegana (WHO/AFRO–DPC/TVD) noted that successful uptake of model-based methods has depended on standardised methodologies, transparent validation, sustained investment in data systems, and structured collaboration between programmes and partners — institutional pre-conditions that schistosomiasis programmes will need to develop in parallel. Dr Fiona Flemming (Unlimit Health) emphasised that effective use of geostatistical outputs depends on clear communication of both predicted prevalence and associated uncertainty, and on integration into existing tools (such as the SCH CDW) rather than parallel systems. Areas of higher uncertainty are best understood as priorities for additional empirical data collection, not as direct triggers for programmatic action.

A programmatic perspective

Complementary remarks by Dr Pauline Mwinzi (ESPEN) framed the appropriate scope of geostatistical modelling at different phases of programme evolution — control, elimination as a public health problem (EPHP), and EPHP surveillance. As programmes transition towards EPHP and surveillance, decisions must increasingly rely on measured and verifiable data at sub-IU and community levels. Investment in diagnostics (RDTs, Kato–Katz), routine surveillance, and decentralised analytical capacity should be prioritised to safeguard programme ownership and sustainability.

Key takeaways

• Geostatistical analysis is most usefully framed as a method for extracting maximum information from imperfect empirical data, supporting more efficient targeting of further survey work.
• Fine-scale prediction can reveal focal transmission patterns masked by IU-level aggregation, while adaptive sampling can reduce sample sizes without compromising sub-district reliability.
• Outputs should be embedded into existing tools and planning cycles, with national programmes remaining central to interpretation and decision-making.

Way forward

Participants converged on the need for a regional framework for the appropriate use of geostatistical methods in schistosomiasis programmes, balancing analytical innovation with strengthened country-level data systems and capacity. This work will be advanced through a dedicated in-person technical meeting. As emphasised throughout the session, the central question is not whether to use geostatistical methods, but how to use them appropriately — in support of programme-led empirical data collection.