Health financing and governance
Facility decision space, reimbursement delays, and how devolved systems translate policy into actual care.
About Me
From the field
From the field
to systems that
programmes can trust.
My name is Nichodemus Amollo. I grew up near Lake Victoria in Homa Bay, Western Kenya, where I learned to pay attention — to the weather, to the lake, to what people say and what they don’t. That instinct has served me well in data work.
I started at KEMRI managing oncology data. It taught me something that no methods course quite captures: data quality is a moral act. Every row is a person waiting for a diagnosis. That sense of responsibility has traveled with me into every job since.
Over eight years I have worked across the full research lifecycle — designing digital survey tools for communities without internet, building the databases that receive the data, running high-frequency quality systems, leading field teams across three countries, and delivering reports and apps that programme teams can actually use. I currently work with Georgetown University gui2de on health financial diaries (300 households over one year: HFCs, automation, live field apps) and on the Re:BUiLD cash-plus targeting Shiny application for IRC partners. I am also a part-time lecturer in Biostatistics at JOOUST and a candidate for an MSc in Biostatistics and Epidemiology.
What makes me useful is that I can operate at both ends. I can sit with a community health worker and explain why data completeness matters. Then walk into a boardroom — or ship a caseworker tool — so findings and models become decisions. That translation — between the field and the institution — is where I work best.
Outside of data: I keep goats and tend a kitchen garden in Homa Bay. There is something honest about agriculture — it keeps development work grounded in reality. I also love football, running, and long conversations over tea that go further than they were supposed to.
Currently
Lead Research Data Manager
Georgetown University gui2de · Remote
MSc Candidate — Biostatistics & Epidemiology
JOOUST · Final thesis stage · Expected graduation Dec 2026
Part-Time Lecturer — Biostatistics
JOOUST
At a Glance
📍 Nairobi, Kenya. Open to remote and on-site.
🗣️ English, Kiswahili, Dholuo
🏛️ NGO · Corporate · Research · Government
🎓 MSc Thesis: NCD financing, Kisumu County
📝 First-author paper forthcoming in BMC Public Health
📚 Co-authored publications in oncology and epidemiology
Outside Work
🐐 Goat farming — Homa Bay County
⚽ Football — watching and playing
🏃 Running & fitness
📚 Agricultural economics & development
🌍 Open governance & civic data
🤖 Staying current with AI & systems thinking
What drives the work
I believe data has a side — it either serves the people it was collected from, or it serves the institutions that collected it. My goal is always the former.
That shows up in how I approach M&E: not as a compliance checklist, but as a learning system. It shows up in how I train enumerators: I want them to understand why the question is structured the way it is, not just how to tap the screen. And it shows up in how I write reports: I want to change a decision, not demonstrate that I can run a regression.
I bring the same philosophy to AI systems: the systems we build encode decisions. They should be interpretable, fair, and accountable to the people they affect.
Research interests
NCDs in primary care
Hypertension and diabetes management, medicine continuity, and service delivery quality in rural settings.
Mixed-methods research
Using records, interviews, and operational data together to explain not just what failed, but why.
Implementation and community pathways
Designing practical interventions that fit community realities rather than importing solutions unchanged.
AI fairness and systems design
How to build responsible AI systems that account for real-world constraints, equity, and interpretability—especially in low-resource health settings.
Core Competencies
Research data architecture & engineering
PostgreSQLSQLSchema & instrument designETL designREDCap architectureReproducible pipelinesGit workflows
Data integrity & quality governance
High-frequency check frameworksAutomated validationMulti-language QCDuplicate & ID auditsAudit trails
Statistical, causal & machine learning methods
RStataPythonSPSSSurvival analysisForecastingDifference-in-differencesCausal forest / CATE
Health surveillance & field data infrastructure
ODKKoboToolboxSurveyCTOREDCapCommCareXLSFormDHIS2 / KHISOffline-first CAPI
Interactive decision support & evidence translation
R ShinyStreamlitPower BITableauQuartoR MarkdownExecutive and policy briefing
A few highlights
- Architected the end-to-end data system for a 300-household, twelve-month health financial diaries panel at gui2de, pairing automated high-frequency checks with an RA performance leaderboard and a live issue-aggregation app so supervisors resolved field problems mid-collection rather than at cleaning.
- Delivered the production application layer for Re:BUiLD cash-plus targeting, turning gui2de causal forest models over 6,260 participants into a caseworker-facing recommender with cost-aware rationale for every recommendation.
- Designed the real-time high-frequency check systems in R and Stata that multi-country research operations run on, catching data defects before they propagate into analysis.
- Led distributed field teams across Kenya, Uganda, and Tanzania, holding quality standards through remote coordination during the COVID-19 pandemic.
- Owned the KEMRI oncology data systems — REDCap registry architecture, quality control, and survival analysis — behind the Kisumu cancer surveillance dashboard launched with county health leadership.
- First-author paper forthcoming in BMC Public Health on financing constraints and chronic NCD care in rural Kisumu.
How I can join a team
I am strongest owning a programme’s data architecture end to end — as an embedded systems lead or a senior collaborator accountable for the whole chain, not as a dashboard vendor.
| Engagement | What I take responsibility for |
|---|---|
| Research data architecture | Instrument logic, study database design, REDCap/ODK/Kobo backends, pipelines, and the live field applications on top |
| Data integrity governance | Quality architecture, automated high-frequency checks, validation rules, and audit trails that hold under funder scrutiny |
| Study & evaluation leadership | Design input, analysis plans, causal and biostatistical method, and reproducible reporting partners can re-run |
| Decision products | Shiny, Streamlit, and Power BI tools that carry models and indicators into operational hands with their rationale attached |
| Team capability | R, Stata, survey design, and reproducible workflow standards that leave the team able to run the system without me |
Get in touch about a role or collaboration →
Open to
Research data manager and biostatistics roles in health and development programmes; partnerships on health financing and service delivery evidence; and capacity building that leaves teams with systems they can run without me.