Measuring Quality of Life

Tuesday 10th May 2011

www.ilcuk.org.uk

Baroness Sally Greengross

House of Lords

Measuring Quality of Life

Professor Ann Bowling

St. George’s, University of London and Kingston

UniversityMeasuring Quality of Life

Quality of Life in older age

A. Bowling, D. Banister, P. Stenner, H. Titheridge, K. Sproston, T. McFarquhar

To measure QoL in people 65+ in BritainTo measure QoL in people 65+ in Britain

To develop & test new ‘bottom up’ To develop & test new ‘bottom up’ measure of QoL (OPQOL)measure of QoL (OPQOL)

Aims of StudiesAims of Studies

Why another measure of QOL?Why another measure of QOL?QoL:QoL:

Complex: objective & subjective conceptComplex: objective & subjective concept

Existing measures developed ‘top down’ - are we Existing measures developed ‘top down’ - are we measuring the right things?measuring the right things?

MultidimensionalMultidimensional measure with social relevance measure with social relevance needed for use in needed for use in multimulti-sector policy evaluation-sector policy evaluation

Gap in market: global pop. ageing & increasing Gap in market: global pop. ageing & increasing longevity longevity

12.4

16.0

16.0

16.4

16.4

16.5

16.7

16.9

17.0

17.1

17.2

17.3

17.8

18.6

19.5

19.5

United States

Finland

United Kingdom

France

Croatia

Latvia

Estonia

Spain

Portugal

Belgium

Bulgaria

Sweden

Greece

Germany

Italy

Japan

Sources: Carl Haub, 2006 World Population Data Sheet.

15 ‘Oldest’ Countries % age 65+

Existing QoL measures (top-down):

Life satisfaction & well-being (single QoL domain)

Physical & mental functioning (2 health-QoL domains)

Broader health status (SF-36) (health-QoL domains)

Economic utility measures (EQ-5D) (narrow health)

CASP-19 (single theory based: needs satisfaction & self-actualisation)

WHOQOL/WHOQOL-OLD(-24) multi-dimensional WHO perspective; tested across countries – convenience

samples; ‘OLD’ ‘functioning added on’ by focus groups

ONS Omnibus 1999-2000: interviewed 999 people: aged 65+ (77% r/r). ‘Bottom-up’ Qs on QoL:

‘Thinking about your life as a whole, what is it that makes your life good - that is, the things that give your life quality? You may mention as many things as you like.’

‘What is it that makes your life bad - that is the things that reduce the quality in your life? You may mention as many things as you like.’ ‘Thinking about all these good and bad things you have just mentioned which one is the most important to you?’

7-point QoL self-rating scale: ‘QoL so good, could not be better’ – ‘QoL so bad, could not be worse’

National random sample, private PC files:77% r/r; 80 re-interviewed in depth

Measures also included:

Psychological: self-efficacy (mastery and control over life); social comparisons, expectations; optimism-

pessimism

Health & functioning: Townsend ADL; Health status; health perceptions (SF-36), diagnosed conditions,

longstanding illness

Psychological morbidity: General Health Questionnaire-12

Social network: contacts & support: family/friends/neighbours, social participation;

perceived neighbourhood social capital

ONS: socio-demographic & socio-economic Qs.

Main QoL themes mentioned & used to develop Main QoL themes mentioned & used to develop OPQOL:OPQOL:

* Social & family relationships* Social roles & activities

* Health & functional ability* Home & neighbourhood (perceived social capital)

* Psychological well-being & outlook (life satisfaction; contentment; optimism; social

comparisons) Income

Independence & being in control over one’s life

Plus: religion, culture, children prioritised by 4 Plus: religion, culture, children prioritised by 4 ethnically diverse focus groupsethnically diverse focus groups

* Independently predicted global self-assessed QoL * Independently predicted global self-assessed QoL

Example: 81% said social relationships gave quality to life:

• ‘for companionship’• ‘to do things with’ • ‘to take me out’ • ‘to make life bearable’ • ‘to know there is someone there willing to help me’ • ‘to look after me’ • ‘for ‘confidence’.

• ….Oh, and my little cat. I talk to her a lot, she’s just like a little child. She doesn’t like being left alone, I love her to bits. Now and again I give her a little kiss.’

Social relationships - neighbours & family:

“Four doors down the man called me to give me broad beans. When I did not put

my washing line up he came round to see if there was any problem. The lady

two doors down does my eye drops three times a week. They are all very

good.”

“The quality of my life now is my family - my children and grandchildren. My life surrounds them. I go at weekends, they visit every week. Sometimes I have the younger grandchild staying overnight…

I’m there if they need me.”

For 12% poor social relationships took quality away from life – e.g. difficulties maintaining contacts/relationships, due to:

•geographical distance

•families ‘too busy’ to visit

•family feuds (‘If only we could be friends with our children.’)

•Ill health/difficulties getting out

200+ lay items reduced to 50 & pre-tested with 100 survey volunteers, & re-reduced: OPQOL-32 & -35:

Life overall (4) Health (4)

Social relationships & participation (8)Independence, control over life, freedom (5)

Area: home & neighbourhood (4)Psychological & emotional well-being (4)

Financial circumstances (4) Religion & culture (2)

5-point Strongly agree to Strongly disagree response scales; reverse coding of positive responses & summed: higher scores = higher QoL

Scale ranges: 35 (QoL so bad could not be worse) - 175 (QoL so good could not be better) PLUS IMPORTANCE RATINGS

Three QoL surveys 2007-8 to test Older People’s

QoL questionnaire (OPQOL)

National ONS Omnibus Survey (sifted 65+: 589/61% r/r) [94% white British; 45% aged 75+)]

National Ethnibus Survey (sifted 65+:400/70% r/r) [Indian (38%), Pakistani (29%), Black Caribbean (22%),

Chinese (11%) people; 9% aged 75+ ]

Postal follow-up of 1999-2000 ONS Omnibus QoL survey respondents (74+:287/58% r/r)

[100% white British; 83% 75+ at follow-up]

Analyses controlled for age, sex, SES

OPQOL item: 4. Life gets me down

Ethnibus ONS QoL follow-up

% % %

Strongly agree 14 1 2

Agree 47 12 7

Neither agree

nor disagree 23 17 25

Disagree 12 48 44

Strongly disagree 4 22 22

OPQOL item: 11. I have someone who gives me love and affection

Ethnibus ONS QoL follow-up % % %

Strongly agree 10 50 45

Agree 45 38 35 Neither agree

nor disagree --- 5 13 Disagree 43 5 6

Strongly disagree 2 2 1

OPQOL item: 25. I have enough money to pay for household bills

Ethnibus ONS QoL f/up

% % %

Strongly agree 17 25 29 Agree 17 66 59 Neither agree nor disagree 37 5 9 Disagree 25 3 3 Strongly disagree 4 1 ---

OPQOL Total ScoreOPQOL Total Score

Ethnibus ONS QoL f/upEthnibus ONS QoL f/up

% % %% % %QoL bad as can be ≤99 6 1 7 QoL bad as can be ≤99 6 1 7

100-119 100-119 67 67 11 11 3838120-139 25 52 43120-139 25 52 43

140-159 2 32 12 140-159 2 32 12 QoL good as can be 160-175 --- 4 --- QoL good as can be 160-175 --- 4 ---

Cronbach’s alpha ofinternal consistency 0.75 0.88 0.90

[Cronbach’s alpha threshold for consistency 0.70<0.90][Cronbach’s alpha threshold for consistency 0.70<0.90]Current Gerontology and Geriatrics Research. Open access ‘Volume 2009 (2009). Current Gerontology and Geriatrics Research. Open access ‘Volume 2009 (2009).

CASP-19 Total Score (2 new samples only)CASP-19 Total Score (2 new samples only)

EthnibusEthnibus ONS ONS % % % % ≤ ≤ 19 ‘Absence of QoL’ ---- 119 ‘Absence of QoL’ ---- 120-29 20-29 23 23 7 730-39 68 2730-39 68 2740-49 8 4640-49 8 4650-57 ‘Satisfaction in all domains’ 1 1950-57 ‘Satisfaction in all domains’ 1 19

Cronbach’s alphaCronbach’s alpha 0.55 0.55 0.870.87

Scale range 0-57 (response scales 0-3; - reversed so positive=better & summed)Scale range 0-57 (response scales 0-3; - reversed so positive=better & summed)

WHOQOL-OLD Total Score (2 new samples only)(2 new samples only)

Ethnibus ONS % % ≤ 69 Lowest possible QoL 2 470-79 23 1180-89 58 2490-99 15 40100-120 Highest possible QoL 2 27

Cronbach’s alpha 0.42 0.85

Scale range 24-120 (24 x 5-point response scales 1-5; - reversed so positive=better & summed)

WHOQOL QoL scale module: death & dying:

Ethnibus ONS % % 7. Extreme fear of not being able to control death 43 17

9. Extreme fear of pain before death 52 34

Extreme fears on any 1 of 4 DD items 77 41

& items: 8. Fear of dying; 6. Fear of way in which will die; Postgraduate Medical Journal, 2010: 86: 197-202.

Adjusted odds of OPQOL score being good (1 referent) vs. not good (0) (all p<0.001) ONS sample (65+) QoL follow-up (74+) O.R. (95% CI) O.R. (95% CI) Unable to walk 400 yards without help or at all vs. rest 0.128 (0.070 – 0.236) 0.443 (0.312-0.631)

Actual number of supporters who would help in a personal crisis 1.159 (1.062 – 1.265) 1.183 (1.070 – 1.308)

Self-efficacy+ High vs. rest N/A 3.449 (1.681 – 7.078)

(+our belief in our ability to succeed)

Validity: OPQOL items (n. 273)

8. ‘I am healthy enough to get out & about’

Strongly disagree/ Neither Agree/ Disagree Strongly agree

% % %10. ‘I would like more companionship/contact with other people’

Strongly agree/Agree 48 21 19***Neither 32 57 43 Strongly disagree/Disagree 20 22 39

Essential requirement for coping with challenges of older age:

Build up reserves of social support & psychological resources (self-efficacy) to optimise:

Skills Opportunities

Abilities

To help compensate when unable to do things

SOC: 84 year old widower (carpenter):SOC: 84 year old widower (carpenter): After his wife died he had no-one to help him put on his socks, so he made a ‘sock horn’; ... to dry between his unreachable toes he made a V-

shaped wooden implement.

“…I can’t reach. I made this up…this frame… stretches out, tightens up … I have given some

of these to other people to help them….a couple of people at the Church ... this woman ..keeps

making me cakes (laughs).”

SOC: A keen bowler: “The beauty of the bowling, of course, is the

fact is that if a partner dies they’ve got somewhere to go.

I mean they literally … play at our bowling club till 95 and even some of them have got new knees … some of them can hardly see, they have

binoculars to see where the jack is, but there’s that companionship, somewhere to go...”

Cont.

“We’ve got a section for blind bowlers ... Amazing what they can do. We put a string down the centre ... so they can feel initially

where they’ve got to go…

We’ve actually got somebody that could beat most of the club members …. he can’t see the jack, so we put the jack up for him. He then

bowls against the string.

Then there’s another one who’s got tunnel vision, he uses binoculars. And he will put on

these binoculars- there’s two of them - they will see where the jack is, and bowl….”

Cont.

“We’ve had so many people having new knees …

The Australians have come out with an assistance for knees, bad arthritic knees, and it’s an extension of your arm… …

mechanical, it fits the bowl at the end and you swing it. You don’t have to bend and

you just release it. And you’d be surprised how proficient they are."

Summary

OPQOL: good reliability & validity in British pop. & ethnically diverse samples

There were marked differences between British pop. & ethnically diverse samples’ OPQOL items & score

(& controlling x age)

Independent predictors of good Qol (OPQOL) in each sample:

Good health/functioning – esp. mobility, More supporters,

Perceived self-efficacy (belief in our ability to succeed)

Mr Paul Allin

Office of National Statistics

Measuring Quality of Life

Delivering wider measures of national well-being and quality of life – Paul Allin, Programme Director

Overview

• What is national well-being (and how different from quality of life)?

• Why measure national well-being?

• How to measure it?

• What matters to people?

• What next?

What is national well-being?

• Sum of individual well-being• How’s the UK doing? (more to life than GDP)• Measuring progress, true wealth and the well-

being of the UK• Overall societal development and progress• The state of the nation• The ‘health’ of the nation• Economic performance and social progress

(not forgetting sustainability and pressures on the environment)

Why measure it?

• Better decision making (government, markets, public)

• “not focussing on the right set of statistical indicators”

• “to steer our economies better through and out of crises”

• “facing a looming environmental crisis”

• What does this mean in practice?

How to measure national well-being?Stiglitz measurement framework:

Economic Measures, e.g.Net national income per

householdDistribution of wealth, income,

disposable income Effects of tax and benefits on

distributionHuman capital and growth rate

Environmental Measures, e.g.

Stocks of natural resources and depletion rates

Indicators of climate change

Quality of Life e.g.Health: life expectancy, expected disability free life years,

Indicators of family life e.g. Single parent householdsProblem indicators e.g. Crime rate, Children in care, drugs,

imprisonment rateSubjective wellbeing

Well-being

New Economy

Resources

Environment

Well-being

Green Growth

Quality Growth

How to measure national well-being?

• Aggregate subjective well-being (only)“Happiness is the new GDP” (discuss!)

• Wider measuresMany separate numbers – indicator sets

Few key numbers – based on consensus of what is important – dashboard

Combined index, using fixed weights?

Help-yourself index? - starting point: UK personal inflation calculator

Adjusted national accounts measures? - eg index of sustainable economic welfare

Measuring subjective well-being

• Four overall monitoring questions from April 2011 in IHS (with socio-demographics)

• Sample of 200,000 directly questioned adults each year across UK

• Detailed questions – sample of 1,000 adults per month

• Questions drawn from research and tested for use in IHS

• Annual data available from July 2012• Experimental

Measuring subjective well-being

• Overall, how satisfied are you with your life nowadays?

• Overall, how happy did you feel yesterday?• Overall, how anxious did you feel yesterday?• Overall, to what extent do you feel the things

you do in your life are worthwhile?

(all on scale 0 – 10)

What matters to people?

• Wide range• Most frequently voted for:

• Health (83% of survey monkey responses)

• Friends & relatives (82%)

• Job satisfaction (77%)

• Partner/spouse relationship (71%)

• Economic security (68%)

• Environment now and future (67%)

• *

• *

• Income and wealth (48%)

• *

• *

What next?

ONS work programme• Testing out what we’ve learned from the

national debate and how to take forward• What do the subjective well-being data tell

us?

Issues• Engagement• Meeting differing requirements• Well-being measurement and policy

Professor Emily Grundy

London School of Hygiene and Tropical Medicine

Measuring Quality of Life

Professor Emily Grundy

(collaborators George Ploubidis and Harriet Young)

London School of Hygiene and Tropical Medicine

Older Europeans’ happiness and well-being

Measuring Quality of Life

Country level differences in well-being

Why of interest?

• Large differences between European countries in, for example, older people’s incomes; living arrangements and extent of family social activity – do these lead to differences in indicators of subjective well-being?

• Might reveal policy relevant factors that potentially could be modified

• Might increase our understanding of ageing and well-being processes

Problems and limitations:

• Even if you ask the same questions, there may be differences between and within countries in how people answer them (c.f. Ann Bowling’s ‘bottom up’ approach).

• Methods which aim to produce measures less subject to measurement bias are quite complex

Living arrangements of Europeans aged 60+ and 80+ by region.

Source: Analysis of ESS 2002/4. North: DK, Fin, Norw, Swe; West: Aust, Belg, Ger, Neths, UK;East: CzR, Est, Hung, Pol, SlovK, Sloven, Ukr; South: Gre, Port, Esp.

Dataset 1: European Social Survey

• Two cross-sectional rounds of data – 2002 and 2004

• Used data from 19 countries• Sample size 18,131 people aged 60+• Scale: ‘Taking all things together, how happy

would you say you are?’ – respondents rated their answer on a scale of 0 (extremely unhappy) to 10 (extremely happy).

Country groupingsNorthNorth West West SouthSouth EastEast

SwedenSweden

NorwayNorway

FinlandFinland

DenmarkDenmark

GermanyGermany

BelgiumBelgium

UKUK

AustriaAustria

NetherlandsNetherlands

PortugalPortugal

GreeceGreece

SpainSpain

PolandPoland

SloveniaSlovenia

SlovakiaSlovakia

HungaryHungary

UkraineUkraine

EstoniaEstonia

Czech Czech RepublicRepublic

N=3621N=3621 N=5867N=5867 N=3857N=3857 N=4786N=4786

Happiness among unmarried older women by European region: results from ordinal logistic models (Higher =happier), 2002-4

North West South East

Low education

1.27 (1.17) 1.19 (1.15) 0.76 (0.73) 1.10 (1.02)

Low income 0.75 (0.74) 0.86 (0.87) 0.94 (0.81) 0.53** (0.45**)

Moderate social ties

0.76 0.69** 0.62 0.92

Low social ties

0.41** 0.56** 0.34** 0.48**

Least social ties

0.08** 0.19** 0.45** 0.24**

N 840 1664 1055 1507

Analysis of European Social Survey; models control for age, widowhood indicator, long term illness & Whether living alone or with others. ( ) results from models not including social ties.

Associations between living arrangements and happiness (higher=better) by region; ref. group=living alone: results from analysis of ESS 2002-4.

***P<.001, *P<0.05

Men Women

Spouse only

Spouse+others

Others only

Spouse only

Spouse+others

Others only

North 2.59*** 3.36*** 0.73 1.94*** 1.86* 1.09

West 2.11*** 1.82* 1.34 2.11*** 2.69*** 2.06***

East 1.04 1.09 0.60 1.81* 1.50 1.36*

South 1.61 1.47 1.10 1.77* 1.82* 2.08*

Controlling for age, education, income, social meetings, social activities, long term illness, availability of confidante, widowhood.

What is associated with happiness in the older population of Europe? (from ESS)

• Region: N(best)W S E (worst)• Comfortable income• Frequent social meetings and social activities• Availability of confidante• No long-term illness• Older age• Lower education• Living with spouse; • For unmarried women those living with others

(children) in S, E & W Europe happier than those living alone; not the case in Northern Europe.

Analysis of ESS, combined sample 18 countries fully adjusted ordinal logistic regression.

Dataset 2

• We used data from 14 European countries included in the second wave of the Survey of Health and Retirement in Europe SHARE (n = 33,528)

• The countries included are drawn from Northern (Denmark and Sweden), Western (Austria, France, Ireland, Germany Belgium, and the Netherlands), Mediterranean (Spain, Italy and Greece) and Eastern (Poland and the Czech Republic) regions of Europe.

• Measure of well-being derived from several individual questions using a latent variable approach

Life satisfaction

Look forward to each day

Life has meaning

Look back with happiness

Full of opportunities

Future looks good

Optimistic about future

Feel prepared for my future

Happiness

Enjoyed life

Well-being

Factor 1

Factor 2

Factor 3

Factor 4

Best fitting/invariant well-being modelBest fitting/invariant well-being model

France

Italy

Spain

Sweden

Greece

Denmark

Austria

Germany

The Netherlands

-0.6

-0.4

-0.2

0

0.2

0.4

0.6

0.8

Country of residence

Well

-bein

g

Males

Females

Well-being country level comparisonWell-being country level comparison

Depression country level comparisonDepression country level comparison

Conclusion

• There is considerable between country variance in late-life mental health in Europe

• The Scandinavian countries seem to do best (low depression/high well being) followed by central European countries, while residents of Mediterranean countries report the worst mental health (NB Eastern Europe not considered)

• Results from analysis of ESS data also show older Northern Europeans happier than older Southern Europeans

• Social factors and social support important in all analyses – but countries with the most family based social support systems have worse rather than better indicators of well-being

• More innovatory approaches such as Ann Bowling’s needed to unravel meaning of measures and their implications.

Mr Paul Cann

Age UK Oxfordshire

Measuring Quality of Life

* What are the best measures of wellbeing and happiness across the lifecourse?* How does wellbeing differ across the lifecourse and do we need to take age into account when developing measures of happiness?* What factors predict quality of life amongst older people?* How can quality of life amongst older people be improved?* How will the new ONS measures of national well-being, including quality of life, impact on public policy?

Measuring Quality of Life

Measuring Quality of Life

Tuesday 10th May 2011www.ilcuk.org.uk