Updating the ‘Vitality Age’ Calculator

Lazy man on the couch watching TV

Fotolia

The Vitality Age Calculator gives users a ‘snapshot’ of their overall health based on lifestyle choices and clinical factors. RAND Europe helped to update the calculator's methodology and use of available data.

The calculator is now more applicable in different geographical areas, takes into account six important pre-existing conditions, and also incorporates the concept of Health Adjusted Life Expectancy.

Background

The Vitality Age calculator gives individuals a snapshot of their overall health. Calculations are based on lifestyle choices (diet, alcohol consumption, exercise etc.) and clinical factors, such as blood pressure or body-mass index.

Goals

RAND Europe was commissioned by Discovery group of companies, part of Discovery Limited, to update the existing algorithm and underlying data of the calculator in order to improve its accuracy. In particular, the study looked at:

  • the accuracy of existing mortality and disease risk calculators
  • the risk factors that should be included in the updated Vitality Age calculator and how to quantify them, and
  • the best approaches for updating the calculator.

Methodology of the model

The model collects data from the Global Burden of Disease Study (GBD), a large database quantifying modifiable risks of death, disease and disability through systematic reviews of the existing literature. It uses this methodology to estimate the joint risk factor burden for combinations of risks using a set of identified mediation factors.

Updates to the model

VA.3

The first update to the model was the Vitality Age (VA.3) methodology, which is more detailed and applicable to varying geographical contexts.

In this version, the underlying concept of the model remained unchanged – the model still covers behavioural and metabolic risk factors and reflects an individual’s overall health status through a single number. The new model does however build on this by expanding the number of input factors used in the analysis, and builds on a more comprehensive and consistent data source.

The new model incorporates an age-related, cause-specific mortality rate rather than the purely cause-specific mortality rates used in the previous version. It also considers the impacts of individual health and lifestyle on each identified disease separately, providing more transparency and information on the estimated effects.

Vitality Age with pre-existing conditions (VAPC)

The Vitality Age with pre-existing conditions (VAPC) is an extension to VA.3, allowing explicit differentiation between individuals with one or more selected pre-existing conditions that negatively affect their health and life expectancy.

The VAPC extension allows for taking into account six pre-existing conditions, namely breast cancer, colon and rectum oesophageal and prostate cancers, as well as ischaemic heart diseases and stroke.

Data from the Surveillance, Epidemiology, and End Results Program of the National Cancer Institute in the United States were used in the computation for the four types of cancers, differentiating by stage of cancer and years since diagnosis.

Steps must be taken when feeding back the resulting VAPC estimate to individuals with pre-existing conditions to ensure that they are aware of the limitations and to inform them that, given their condition(s), a more accurate estimation of their prognosis can be provided by a specialist.

Healthy Vitality Age (HVA)

While VA.3 compares the individual life expectancy (remaining years of life) to the average population life expectancy, this extension of the calculator, called Healthy Vitality Age (HVA), uses the concept of Health Adjusted Life Expectancy (HALE). HALE takes into account both mortality and morbidity by adjusting the life expectancy by the amount of time lived in less than perfect health. A Healthy Vitality Age higher than an actual age signals a lower-than-average number of years of healthy life remaining, with the difference between Healthy Vitality Age and actual age being equal to the estimated change in health adjusted life expectancy.

To supplement the data used by VA.3, data from the Global Burden of Diseases on years lived with disabilities, prevalence rates and disability weights are used in HVA. More research would be needed on the variance in disability weights for the same outcome across individuals, however.


Read the full study