A research initiative to quantify the systemic health impacts of periodontal disease through health economic modeling.
This repository contains two major health economic modeling projects examining the relationship between periodontal disease and systemic health conditions:
- Periodontal Disease and Alzheimer's Disease/Dementia (AD) — Individual-based microsimulation
- Periodontal Disease and Cardiovascular Disease (CVD) — Markov cohort model
Model Type: Markov cohort model Time Horizon: 10 years Population: 65-year-old adults with severe periodontal disease
Key Features:
- Eight health states including post-stroke, post-MI, and combinations
- Intervention: Non-surgical periodontal therapy
- Cost-effectiveness analysis against NICE thresholds (£20,000–£30,000/QALY)
- Tunnel states for acute vs. chronic event phases
- Probabilistic sensitivity analysis (10,000 Monte Carlo simulations)
Status: ✅ Submitted for peer review
Location: CVD_Study/
| File | Description |
|---|---|
Main_Text_CVD_Paper_finalised.tex |
Finalised manuscript |
Supplementary_Material_CVD.tex |
Technical appendix |
PD_CVD_markov - PSA On.xlsm |
Excel-based Markov model |
generate_cvd_figures.py |
Figure generation script |
images_CVD/ |
Publication-ready figures |
EXCEL_MODEL_EXECUTION_GUIDE.md |
Model execution instructions |
CONTRIBUTING.md |
Contributor guidelines |
Model Type: Individual-level microsimulation Time Horizon: 2023–2040 Population: England
Key Features:
- Four dementia stages: mild → moderate → severe → death
- Baseline periodontal prevalence scenarios: 25%, 50%, 75%
- Co-morbidity modelling: smoking, diabetes, cerebrovascular disease, cardiovascular disease
- Dual cost perspective: NHS and informal caregivers
- Outcomes: QALYs, costs, dementia cases prevented
- External validation against ONS and NHS prevalence data
v2 — AD_Model_v2/
| File | Description |
|---|---|
IBM_PD_AD_v2.py |
Core microsimulation model |
AD_FullText_v2.tex / AD_FullText_v2_65.md |
Manuscript drafts |
AD_Microsimulation_results/ |
PSA results (25%, 50%, 75% scenarios) |
run_psa_direct_v2.py |
PSA runner |
v3 — AD_Model_v3/ (current)
| File | Description |
|---|---|
IBM_PD_AD_v3.py |
Updated microsimulation with PSA and sensitivity analysis |
AD_FullText_v3.tex / AD_FullText_v3.md |
Current manuscript |
Supplementary_Material_AD_65V3.docx |
Technical appendix |
external_validation.py |
External validation against published data |
pd_sensitivity_analysis.py |
One-way sensitivity analysis |
run_pd_tornado.py |
Tornado plot generation |
run_psa_direct_v3.py |
PSA runner |
psa_results_25_v3/ psa_results_50_v3/ psa_results_75_v3/ |
PSA outputs |
plots/ |
Generated figures |
periodontal/
├── README.md
├── LICENSE
├── pytest.ini
├── requirements.txt
├── requirements-dev.txt
├── AD_Model_v2/ # AD microsimulation v2
├── AD_Model_v3/ # AD microsimulation v3 (current)
├── CVD_Study/ # CVD Markov model and manuscript
├── tests/ # Unit test suite
│ ├── test_ibm_pd_ad.py # AD microsimulation tests (97 tests)
│ ├── test_cvd_figures.py # CVD figure generation tests (26 tests)
│ └── test_psa_workflow.py # PSA workflow tests (14 tests)
├── plots/ # Root-level generated figures
├── psa_results_25_v3/ # PSA outputs (root-level copies)
├── psa_results_50_v3/
└── psa_results_75_v3/
# Install development dependencies
pip install -r requirements-dev.txt
# Run all tests
pytest
# Run with coverage
pytest --cov
# Run a specific suite
pytest tests/test_ibm_pd_ad.pyTest suite: 137 tests — 132 passing, 5 skipped (pending PSA module)
- Design: Hazard-based microsimulation
- Cycle: Annual time steps
- Discount Rate: 3.5% (costs and QALYs)
- Data Sources: NHS England Primary Care Dementia Data, ONS population projections
- Key Parameters:
- PD hazard ratio for dementia onset: 1.47 (95% CI: 1.32–1.65)
- PD hazard ratio for mortality: 1.36 (95% CI: 1.10–1.69)
- Design: State-transition Markov model
- Cycle: 1 year
- Discount Rate: 3.5% (costs and QALYs)
- Data Sources: UK Biobank, NHS cost collections, ONS life tables
- Treatment Effects:
- Stroke hazard reduction: 0.40–0.78
- MI hazard reduction: 0.54–0.90
- Economic Evaluation: Aligned with NICE guidelines for health technology assessment
- Reporting: CHEERS 2022
- Cost Year: 2024 GBP (HM Treasury GDP deflators)
- Utility Values: EQ-5D derived from UK population norms
- NHS England Primary Care Dementia Data
- UK Biobank
- Office for National Statistics (ONS) population projections and life tables
- NHS National Cost Collection
- Adult Oral Health Survey 2021
- English Longitudinal Study of Ageing (ELSA)
[Citation to be added upon publication]
MIT License — see LICENSE for details.
Copyright (c) 2025 Kwaai, Personal AI Lab
[Contact information to be added]
Project Status: Active Development Last Updated: February 2026