'Characterising the Healthy Ageing Phenotype'

Assessing nutritional status in older adults: Developing
tools to measure healthy ageing
Dr Jose Lara
Institute of Cellular Medicine
Newcastle University
Workshop: “Examining diet and physical
activity behaviours amongst older women
through the lens of superdiversity”
University of Birmingham
September 18, 2014
Remit of LIVEWELL Programme
The process of ageing is complex
Langie, Lara, Mathers. Best Pract Res Clin Endocrinol Metab 2012
Prevalence of healthy ageing
•
•
•
USA Health and Retirement Study (1998-2004) of adults ≥65 yrs of age
Survey of Health, Ageing, and Retirement in Europe (SHARE) of adults ≥65 yrs of age
Criteria used:
–
–
–
–
–
No major disease (DM, Cancer, Lung disease, Heart disease or stroke)
no activity of daily living disability
no more than one difficulty with 7 measures of physical functioning
≥ median score cognition
being actively engaged
USA Health and Retirement study
Survey of Health, Ageing, and Retirement in
Europe (SHARE)
Overall
11.9%
Overall
8.5%
Men
12.2%
Denmark
21.1%
Women
11.0%
Sweden
17.4%
Non-Hispanic white
12.2%
Netherlands
17.0%
Non-Hispanic Black
6.0%
Switzerland
16.1%
Hispanic
5.3%
Ireland
15.7%
Poland
1.6%
McLaughlin et al. J Gerontol B Psychol Sci Soc Sci. 2010
Hank K. J Gerontol B Psychol Sci Soc Sci. 2011
Measuring change in intervention studies to promote
healthy ageing
Baseline
measures of
healthy
ageing
phenotype
Intervention
Behaviour change
Functional benefit
Outcome
measures of
healthy
ageing
phenotype
Tools to measure healthy ageing
phenotype
↑ pulmonary function predicts ↑
longevity
Low
High
906 people (mean age 81 y)
without dementia
Buchman AS et al. (2008) Mech. Ageing Dev. 129, 625-631
Physical capability predicts mortality
Grip strength
predicts
mortality even
in younger
(<60) people
Cooper R et al. (2010) BMJ 341,
C4467
Tools to measure healthy ageing
phenotype
Characteristics:
 Robust
 Validated
 Comprehensive
 Appropriate for lifestage
 Capable of measuring difference in response to an
intervention
 Cost effective
 Suitable for use in community settings
Healthy Ageing Phenotype: How did we proceed?
Aim
• Identify Systematic Reviews and/or Meta-analysis of cohort studies
• Identify Clinical guidelines (e.g. NICE- National Institute for Health and
Clinical Excellence)
Consideration for Selection criteria
• Frequently used in longitudinal studies
• Expected to change with age
• Evidence for strong association/prediction of ageing-related phenotypes
such as morbidity, mortality, lifespan
• Focus on studies measuring these outcomes in healthy populations
rather than on individuals selected on the basis of a particular disease
• Biomarkers that respond to intervention
Some markers of Physiological and metabolic function
(Bio)marker
Syst-Review/
meta-analysis
Change
with age
Predicts ageing
phenotypes
Modified by
intervention
Blood Pressure


+++++
+++++
BP arm difference


+++++
+++++
Total Cholesterol


++++
+++++
HDL cholesterol


++++
+++++
LDL cholesterol


++++
+++++
Cholesterol ratios


++++
+++++
Triglycerides


++
+++++
Fasting glucose


+++++
+++++
HbA1C


+++++
+++++
1-2 hr OGTT


+++++
+++++
Metabolic Syndrome


++++
++++
Coagulation
Fibrinogen


+++
+++
Inflammation
CRP


+++
+++
Lung function
FEV1


++++++
++++++
Body composition
BMI


++
++++++
WC


+++
++++++
WHR


+++
+++++
CV/Metabolic
Workshop on The Healthy Ageing Phenotype Newcastle, UK held in October 2012
(i)
(ii)
Identify the most important features of the HAP
Identify or developing tools for measurement of those features specifically in the context of
community-based intervention studies.
Healthy Ageing Phenotype (HAP)
Lara et al., Maturitas 2013
Informative biomarkers of ageing in the Newcastle 85+ study
Martin-Ruiz, et al., Mech Ageing Dev. 2011
Ageing studies with Biomarkers
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
China Health and Retirement Longitudinal Study (CHARLS)
Chinese Longitudinal Healthy Longevity Survey (CLHLS)
Costa Rican Longevity and Healthy Aging Study (CRELES)
English Longitudinal Study of Ageing (ELSA)
Health and Retirement Study (HRS)
Indonesia Family Life Survey (IFLS)
INDEPTH Network
Los Angeles Family and Neighborhood Survey (LA FANS)
Longitudinal Aging Study in India (LASI)
Mid-Life in the U.S. Study (MIDUS)
National Health and Nutrition Examination Survey (NHANES)
National Health and Aging Trends Study (NHATS)
National Longitudinal Study of Adolescent Health (Add Health)
National Longitudinal Survey 1979 Cohort (NLSY79)
National Social Life, Health and Aging Project (NSHAP)
•
•
•
•
•
Newcastle 85+ 74 BIOMARKERS
Social Environment and Biomarkers of Aging Study (SEBAS)
Survey of Health, Ageing and Retirement in Europe (SHARE)
Tsimane Health and Life History Project
Wisconsin Longitudinal Study (WLS)
Conclusions
• We propose five domains to characterise healthy ageing and
tools to measure specific subdomains
• Pilot study testing interventions and outcome measures to
be completed December 2014
• Novel biomarkers validated in longitudinal studies of ageing
could be incorporated as available
LiveWell: ‘Where are we now’?
• Pilot RTC to be completed December 2014
• Panel of outcome measures developed for each
domain
Acknowledgements 1
Investigators:
John Mathers
Thomas Meyer
Paula Moynihan
Lynn Rochester
Falko Sniehotta
Martin White
Research team:
Elizabeth Evans
Alan Godfrey
Ben Heaven
Nicki Hobbs
Alexandra Munro
Shakir Chowdhury
Caroline Wiuff
Jose Lara
PhD Students:
Lynn Barron
Caroline Shaw
Suzanne McDonald
Sanchia Coatsworth
Collaborators:
Ashley Adamson
Vera Araujo-Soares
Mike Catt
Lynne Corner
John Matthews
Patrick Olivier
Mike Trenell
Thomas von Zglinicki
Suzanne Moffatt
Acknowledgements 2
Lara et al., BMC Med 2014
F&V intakes in 22 RCTs by the presence or absence of
the “barrier identification/problem solving” BCT
Lara et al., BMC Med 2014 (In Press)
F&V intakes in 22 RCTs by the presence or absence of
the “plan for social support/social change” BCT
Lara et al., BMC Med 2014 (In Press)
Table 2. Association of behaviour change techniques (BCTs) with Fruit and Vegetable intakes in 22 intervention studies
BCT
No. of
studies or
subgroups
BCT present
Yes (No)
Sample size
BCT present
Yes (No)
BCT present
Mean difference
(95% CI)
I2
(95% CI)
BCT absent
Mean difference
(95% CI)
I2 (95% CI)
Motivational interviewing
7 (18)
10183 (49503)
52.8 (33.1 to 72.8)
81 (61 to 90)
105.1 (71.4 to 138.7)
97 (95 to 97)
Provide information on
consequences of behaviour
to the individual
Prompt review of
behavioural goals
Goal setting (behaviour)
11 (14)
12488 (9394)
60.1 (45.0 to 75.3)
62 (27 to 80)
109.6 (64.6 to 154.9)
96 (95 to 97)
5 (20)
7163 (52562)
69.1 (55.8 to 82.3)
53 (0 to 83)
92.9 (59.7 to 126.1)
97 (96 to 98)
23 (2)
59323 (402)
89.0 (63.6 to 114.5)
97 (96 to 98)
95.7 (-63.1 to 254.4)
88 (N/A)
Model/Demonstrate the
behaviour
Provide instruction on how
to perform the behaviour
Provide information on
consequences of behaviour
in general
Provide feedback on
performance
Goal setting (outcome)
6 (19)
3258 (56467)
91.3 (56.9 to 125.7)
71 (32 to 87)
85.0 (55.9 to 114.2)
95 (97 to 98)
15 (10)
53717 (6008)
99.5 (69.6 to 129.5)
97 (96 to 98)
67.8 (34.5 to 101.2)
78 (59 to 88)
18 (7)
53298 (6427)
97.8 (67.6 to 127.9)
97 (96 to 98)
66.7 (32.3 to 101.1)
86 (74 to 93)
14 (11)
54563 (5162)
102.5 (68.9 to 136.0)
98 (97 to 98)
63 (37.9 to 88.0)
78 (60 to 87)
11 (14)
43702 (16023)
118.1 (74.3 to 161.9)
97 (96 to 98)
62.8 (43.5 to 82.2)
83 (73 to 90)
9 (16)
41627 (18098)
127.4 (74.1 to 180..7)
95 (92 to 97)
60.9 (43.4 to 78.4)
86 (78 to 91)
Use of follow-up prompts
Lara et al., BMC Med 2014 (In Press)
(Mean difference
of Mean
differences
(95%CI))
P
(-52 (-91 to -13))
0.009
(-49 (-97 to -2))
0.040
(-24 (-60 to 12))
0.190
(-7 (-166 to 153))
0.940
(6 (-39 to 51))
0.790
(32 (-13 to 77))
0.170
(31 (-15 to 77))
0.180
(39 (-2 to 81))
0.060
(55 (7 to 103))
0.020
(66 (10 to 123))
0.020
Meta-regression of number of BCTs on overall F&V
intake
Number of BCTs. Slope = 8.28, Q = 3.84, d.f. = 1, P = 0.049. The circle size reflects the weight
that a study obtained in the meta-regression.
Lara et al., BMC Med 2014 (In Press)
Table 2. Behaviour change techniques (BCTs) associated with intervention effectiveness
Reference
Outcome
Lara et al 201438
Increasing fruit
and
vegetables
among older
adults
Olander et al
2013 51
French et al
2014 52
BCTs associated with greater
effectiveness of interventions
“Barrier identification/problem solving”
BCTs associated with lower
effectiveness of interventions
“Plan for social support/social change”
“Goal setting (outcome)”
“Use of follow-up prompts”
Increasing
physical
activity in
obese
individuals
“Teach to use prompts/cues”
Increasing
physical
activity in
obese
individuals
None
“Prompt practice”
“Prompt rewards contingent on effort or
progress towards behaviour”
“Setting behavioural goals”
“Prompting self-monitoring of
behaviour”
“Planning for relapses”
“Providing normative information”
“Providing feedback on performance”
Lara et al., Proc Nutr Soc
2014 (In Press)
Gardner et al
2011
53
Hill et al
2013
54
Reduce
gestational
weight gain
No obvious differences in the behaviour
change techniques employed between
effective and ineffective interventions
Reduce
gestational
weight gain
“Provision of information on the
consequences of behaviour to the
individual”
“Motivational interviewing”
“Behavioural self-monitoring”
“Providing rewards contingent on successful
behaviour”
Hatmann-Boyce
et al 2014 55
Promoting
weight loss in
adults
“Provide information about others’ approval”
“Provide normative information about others
behaviour”
“Model/demonstrate the behaviour”
“Facilitate social comparison”
Martin et al
2013 56
Preventing
and managing
childhood
obesity
“Provide information on the consequences of
behaviour to the individual”
“Environmental restructuring”
“Prompt practice”
“Prompt identification as role model/position
advocate”
“Stress management/emotional control
training”
“General communication skills training”
“Prompting focus on past success”
“Prompt self-talk”
F&V intakes according to the presence or absence of
the behaviour theories
Lara et al., BMC Med 2014 (In Press)
F&V intakes in 22 RCTs according to the presence of
single vs multiple behaviour theories
Lara et al., BMC Med 2014 (In Press)