Predictive Models in Health and Morbidity Trends Related to Insurance

Heriot-Watt University, 2019-2022

Research project funded by the Society of Actuaries (SOA) through the Centers of Actuarial Excellence (CAE) programme

Project description

Over recent decades, medical interventions and advances have become important drivers of health and morbidity trends.

In this project we will develop, evaluate and assess statistical predictive models for morbidity risk and underlying trends,

relating to particular major conditions such as heart disease, diabetes and cancer, using data from major USA- and

UK-based sources. The principal aim is to address the timely need to develop robust predictive models for rapidly

changing morbidity risks and relevant impact on health-related insurance. A Bayesian approach will be employed, to

allow for uncertainty quantification, under which model diagnostics, assessment and selection will be considered for a

range of models including machine learning-based approaches. The application of such methodology in the context of

this project is novel and at the forefront of current practice.

Research Team

· Prof. George Streftaris, Principal Investigator

Heriot-Watt U, School of MACS

· Prof. Angus Macdonald, CoI, HWU

· Dr. Torsten Kleinow, CoI, HWU

· Prof Ian Duncan, CoI, U California Santa Barbara

· Dr. Erengul Dodd, CoI, U Southampton

· Alex Jose (PhD, HWU)

Publications

· Jose, A., Macdonald, A.S., Tzougas, G and Streftaris, G. (2022) A combined neural network approach for the prediction

of admission rates related to respiratory diseases. Risks, accepted for publication.

· Arik, A., Dodd, E., Cairns, A., Shao, A., Streftaris, G. (2022) Uneven outcomes: findings on cancer mortality.

The Actuary, https://www.theactuary.com/features/2022/05/30/uneven-outcomes-findings-cancer-mortality

· Arık, A., Cairns, A., Dodd, E., Macdonald, A.S. and Streftaris, G. (2022) Estimating the impact of the COVID-19

pandemic on breast cancer deaths among older women, Living to 100 Research Symposium.

· Yiu, A.M.T.L., Kleinow, T. and Streftaris, G. (2022) Cause-of-death contributions to declining mortality

improvements and life expectancies using cause-specific scenarios. arXiv:2210.12442 (and under review).

· Kwok, W. M., Dass, S. C., and Streftaris, G. (2022). Deep Learning Aided Laplace

Based Bayesian Inference for Epidemiological Systems. arXiv:2210.08865 (and under review).

· Arik, A., Dodd, E., Cairns, A., STREFTARIS, G. (2021) Socioeconomic disparities in cancer incidence and mortality in

England and the impact of age-at-diagnosis on cancer mortality.

PLoS ONE 16, 7. https://doi.org/10.1371/journal.pone.0253854

· Arık A., Dodd E., Streftaris G. (2020) Cancer morbidity trends and regional differences in England – A Bayesian analysis.

PLoS ONE 15(5): e0232844. https://doi.org/10.1371/journal.pone.0232844

· Dimakakou, E., Johnston, H.J., Streftaris, G., Cherrie, J.W. (2020) Is environmental and occupational particulate air

pollution exposure related to diabetes and dementia? A cross-sectional analysis in UK Biobank.

International Journal of Environmental Research and Public Health. 17(24), 9581. https://doi.org/10.3390/ijerph17249581

Related publications

· Oytun Haçarız, Torsten Kleinow & Angus S. Macdonald (2021) Genetics, insurance and hypertrophic cardiomyopathy,

Scandinavian Actuarial Journal, 2021:1, 54-81, DOI: 10.1080/03461238.2020.1795714

· Ungolo, F., Kleinow, T. & Macdonald, A.S. (2021). Parametric bootstrap estimation of standard errors in survival models

when covariates are missing. In Mathematical and Statistical Methods for Actuarial Sciences and Finance,

eds. Corazza, M., Gilli, M., Perna, C., Pizzi, C. & Sibillo, M. Springer.

· Schnürch S, Kleinow T, Korn R. (2021) Clustering-Based Extensions of the Common Age Effect Multi-Population

Mortality Model. Risks. 2021; 9(3):45. https://doi.org/10.3390/risks9030045

· Wen, J., Cairns, A., & Kleinow, T. (2021). Fitting multi-population mortality models to socio-economic groups.

Annals of Actuarial Science, 15(1), 144-172. doi:10.1017/S1748499520000184

· Haçarız, O., Kleinow, T., Macdonald, A.S. (2021) An actuarial model of arrhythmogenic right ventricular cardiomyopathy

and life insurance, Scandinavian Actuarial Journal. https://doi.org/10.1080/03461238.2021.1930136

· Haçarız, O., Kleinow, T., Macdonald, A.S., Tapadar, P., Thomas, R.G. (2020) Will genetic test results be monetized in life

insurance? Risk Management and Insurance Review. 23:379–399. https://doi.org/10.1111/rmir.12159

Presentations

· Insurance Data Science Conference, Milan, June 2022.

Predictive modelling for admission rates related to respiratory diseases in the US.

· University of Piraeus colloquia, Greece, April 2022.

Bayesian modelling for cancer risk – regional and socioeconomic disparities.

· Actuarial Research Conference 2022, Chicago, August 2022.

Estimating the Impact of COVID- 19 Health Disruptions on Breast Cancer Risk.

· Actuarial Research Centre webinar series. September 2021.

Modelling cancer risk: regional and socioeconomic disparities. Invited seminar.

· Sixteenth International Longevity Risk (Longevity 16) Conference. August 2021 (Copenhagen, Demark).

Socioeconomic Disparities in Cancer Risk. Contributed seminar.

· 24th International Congress on Insurance: Mathematics and Economics. July 2021.

Modelling cancer risk: regional and socioeconomic disparities. Contributed seminar.

· HWU Inaugural lecture. February 2021.

Cancer trends by region and deprivation – how big is the gap? Invited seminar.

· Protection, Health and Care Conference, IFoA, Webinar, August 2020.

Cancer morbidity risk modelling – regional variation over time.

Working papers

· Arık, A., Cairns, A., Dodd, E., Macdonald, A.S., and Streftaris, G. (2022)

A semi-Markov approach for investigating the impact of COVID-19 health disruptions on breast cancer.

In preparation, to be submitted December 2022.

· Streftaris, G., Xie, C., Dodd, E. (2021, in preparation) Bayesian modelling of critical illness insurance claim rates.

Back to My Home Page

---