Use this URL to cite or link to this record in EThOS: http://ethos.bl.uk/OrderDetails.do?uin=uk.bl.ethos.626199
Title: Individualized modelling of mortality by cause based upon risk factors
Author: Martin, C. J.
Awarding Body: University College London (University of London)
Current Institution: University College London (University of London)
Date of Award: 2013
Availability of Full Text:
Access through EThOS:
Full text unavailable from EThOS. Please try the link below.
Access through Institution:
Abstract:
This thesis describes the development and evaluation of a stochastic mortality model that is intended to support shared decision-making between health care professionals and patients; health care planning; and actuarial analysis for life insurance, health insurance and pensions. The model uses risk factors and cause of death to calculate all-cause mortality in a modularised Markov Chain Monte Carlo approach. This modularisation allows the substitution of different cause of death models and hazard ratios, enabling customisation, calibration to different populations, and further research. The cause-specific sub-models used in this evaluation incorporate a method of auto-calibration to different populations using baseline population mortality rates, risk factor distributions, and individual risk factor values. The model can investigate the impact of changing trends in risk factors and population mortality rates. Trends in lifestyle factors such as smoking cessation, medical interventions that adjust risk factors such as blood pressure or serum cholesterol reduction, or interventions that reduce mortality from specific causes, can be examined. Non-specific, observed trends like the select transition effect seen in life insurance portfolios, or cohort effects relating to year of birth, can also be applied. A thorough internal and external evaluation of the model is presented, including its performance in simulating three randomised controlled trials (two of the treatment of hypertension, and one of cholesterol reduction in subjects experiencing a cardiovascular event), as well as three simulations of historical prospective cohort data projected over at least 16 years. A demonstration of the application of the model in the assessment of the impact of changing trends in obesity rates in the UK over the next 50 years suggests that rising obesity makes a modest negative contribution to mortality improvement, but not enough to reverse current improvement rates.
Supervisor: Not available Sponsor: Not available
Qualification Name: Thesis (Ph.D.) Qualification Level: Doctoral
EThOS ID: uk.bl.ethos.626199  DOI: Not available
Share: