Development of new predictive energy equations in adults aged 65-79 years and ≥80 years
Principal Investigator
Name
Judi Porter
Degrees
Ph.D, M.H.Sc., Grad.Dip.Nut.Diet.,B.App.Sc.
Institution
Deakin University
Position Title
Professor in Dietetics
Email
About this CDAS Project
Study
IDATA
(Learn more about this study)
Project ID
IDATA-42
Initial CDAS Request Approval
Mar 23, 2021
Title
Development of new predictive energy equations in adults aged 65-79 years and ≥80 years
Summary
As greater proportions of populations not only live longer, but stay healthy and active for longer, consideration needs to be given to the differing energy requirements across the ageing continuum (Porter, 2018). Most equations and national-level recommendations estimating energy intake do not account for subgroups within the 65-100+ year age category (Porter, 2019).
Our recent analysis conducted utilizing RMR data of 988 participants and TEE data of 1488 participants identified that the Ikeda, Livingston and Ikeda equations most closely agreed with measured RMR and TEE in all adults aged 65 years and over, and in subgroup analyses undertaken in 65-79 and ≥80 year age groups. Although all three equations showed proportional bias, Mifflin had the lowest mean difference across the ≥65 y age cohort, whilst the Livingston equation showed the least bias in the 65–79 year age group (Porter, 2019).
This study aims to develop predictive equations for older adults for age brackets within 65-79years and ≥80 years continuum.
Methods:
We will derive this analysis from the dataset compiled from our recently published research (Porter et al, 2018; Porter et al, 2019). This includes our pre-existing database of studies comprising 988 participant level RMR data measured using indirect calorimetry and 1488 participant level TEE data measured using DLW in adults aged 65-79 years and ≥80 years. Twenty (n=20) measured RMR and TEE data points from participants 80+ years will be added to this data set from a recently conducted study by our team. Additional data points from any recently published hand searched articles will also be included in our analyses where obtainable (currently n=157).
With the database established, defined methods for energy equation development will be followed. These include exploring relationships between variables that may influence energy expenditure (e.g. gender, age, weight, height and FFM) using correlation analysis; conducting multiple linear regression analysis to evaluate the contribution of these variables to the prediction of RMR, and develop best fit equation/s to estimate RMR from contributing variables identified using multiple linear regression analysis. Equations will be validated using either a split-group, jack-knife or leave one out approach. Using one of these approaches, the following steps will be undertaken. Bland Altman plots will be developed to compare the newly created equations with those currently in use.
Steps will be repeated with dataset stratified by the subgroups 65-79 years old and 80+ years old.
References
Porter J et al. Nutrition Journal 2018, 17:40.
Porter J et al. American Journal of Clinical Nutrition 2019; 110(6):1353-1361.
Our recent analysis conducted utilizing RMR data of 988 participants and TEE data of 1488 participants identified that the Ikeda, Livingston and Ikeda equations most closely agreed with measured RMR and TEE in all adults aged 65 years and over, and in subgroup analyses undertaken in 65-79 and ≥80 year age groups. Although all three equations showed proportional bias, Mifflin had the lowest mean difference across the ≥65 y age cohort, whilst the Livingston equation showed the least bias in the 65–79 year age group (Porter, 2019).
This study aims to develop predictive equations for older adults for age brackets within 65-79years and ≥80 years continuum.
Methods:
We will derive this analysis from the dataset compiled from our recently published research (Porter et al, 2018; Porter et al, 2019). This includes our pre-existing database of studies comprising 988 participant level RMR data measured using indirect calorimetry and 1488 participant level TEE data measured using DLW in adults aged 65-79 years and ≥80 years. Twenty (n=20) measured RMR and TEE data points from participants 80+ years will be added to this data set from a recently conducted study by our team. Additional data points from any recently published hand searched articles will also be included in our analyses where obtainable (currently n=157).
With the database established, defined methods for energy equation development will be followed. These include exploring relationships between variables that may influence energy expenditure (e.g. gender, age, weight, height and FFM) using correlation analysis; conducting multiple linear regression analysis to evaluate the contribution of these variables to the prediction of RMR, and develop best fit equation/s to estimate RMR from contributing variables identified using multiple linear regression analysis. Equations will be validated using either a split-group, jack-knife or leave one out approach. Using one of these approaches, the following steps will be undertaken. Bland Altman plots will be developed to compare the newly created equations with those currently in use.
Steps will be repeated with dataset stratified by the subgroups 65-79 years old and 80+ years old.
References
Porter J et al. Nutrition Journal 2018, 17:40.
Porter J et al. American Journal of Clinical Nutrition 2019; 110(6):1353-1361.
Aims
Primary aim:
To develop predictive equations to estimate resting metabolic rate (RMR) in adults 65 + years and in the subgroup 65-79 years and ≥80 years.
Secondary aim:
To determine if the inclusion of the body composition variable fat free mass (FFM) improves the accuracy of these equations to predict RMR.
Hypothesis:
Total energy expenditure (TEE) estimated from our equations will have a better agreement with measured TEE (from doubly labelled water) than TEE estimated from the Mifflin, Livingston and Ikeda equations.
Collaborators
Dr Kay Nguo, Monash University
Dr Simone Gibson, Monash University
Dr Zoe Davidson, Monash University
Professor Helen Truby, University of Queensland
Professor Leigh Ward, University of Queensland
Professor Ross Prentice, Women's Health Initiative