Yorkshire and Humberside Strategic Clinical Networks: Regional Leads Meeting

12 May 2015

Raphael WittenbergPersonal Social Services Research Unit

London School of Economics and Political Science

The evidence and economic case for post diagnostic support for people with dementia

The evidence and economic case for post diagnostic support for people with dementia

The work I am presenting was conducted by colleagues at the Personal Social Services Research Unit (PSSRU) at LSE

Some of it was supported by:•the Department of Health (DH) for England•NHS England •the National Institute for Health Research (NIHR) •the Economic and Social Research Council•the Alzheimer’s Society.

All views expressed in this presentation are those of the presenter, and are not necessarily those of the DH, NHSE, NIHR, ESRC or Alzheimer’s Society.

Acknowledgements and Disclaimer Acknowledgements and Disclaimer

•Dementia scenarios study for G7 Dementia Event, June 2014•Cost of Illness Study for Dementia UK second edition report, funded by the Alzheimer’s Society, September 2014•Alzheimer’s disease scenarios modelling for the Office for Life Sciences, November 2014•Economic analyses for a range of trials of different interventions•Economic case for dementia care, funded by NHS England, started this month•Comprehensive approach to modelling outcome and cost impacts of interventions for dementia(Modem), funded by the ESRC and NIHR, 2014 to 2018

PSSRU Studies of DementiaPSSRU Studies of Dementia

Estimates by PSSRU for Dementia UK: 2nd edition published by the Alzheimer’s Society Sept 2014

Annual cost of dementia in the UKAnnual cost of dementia in the UK

Total cost = £26.3 billionAverage cost per person (for the 816,000 people

with dementia) = £32,250

This is a 24% real terms increase in just 7 years

Average annual cost per person - depends on severity and care setting

Average annual cost per person - depends on severity and care setting

£25,723

£42,841£55,197

£36,738

Prince, Knapp et al Dementia UK 2nd Edition, 2014

These are just aggregate costs (not cost-effectiveness findings) – some are good and some bad:

‘Good costs’ – appropriate (evidence-based) treatment and care responses to assessed needs; and responding to individual preferences.

‘Bad costs’ – result from late or no diagnosis, unavailability of effective care, crisis admissions to hospital, unnecessarily long inpatient stays etc.

We must shift the balance from ‘bad’ to ‘good’ – by making evidence-based treatment & care much more widely available.

Good and bad costsGood and bad costs

o Risk reduction

o Screening & diagnosis

o Carer support

o Staff skills training

o Medications

o Psychosocial treatments

o Home-based care

o Case management

o Awareness and attitudes

Responding to the challenge: what works?

Responding to the challenge: what works?

For each area there is (some) international

evidence on effectiveness; not

much on cost-effectiveness.

+++++++++++Many trials underway

+++++++++++Review of evidence

now underway (MODEM)

Knapp et al Int J Geriatric Psychiatry 2013; Lombard et al In prep’n

Individual programme (8 sessions over 8-14 weeks, delivered by psychology graduates + manual); carers given techniques to:

ounderstand behaviours of person they care for

omanage behaviour

ochange unhelpful thoughts

opromote acceptance

oimprove communication

oplan for the future

orelax

oengage in meaningful, enjoyable activities.

START: a manual-based coping strategy

START: a manual-based coping strategy

Livingston et al BMJ 2013; Knapp et al BMJ 2013; Livingston et al submitted

Pragmatic, multicentre RCT – START vs usual support.

n=260 family carers of people with dementia, North London area.

Analyses 8 & 24 months after end of intervention

Carers with usual support were 4 times more likely to have clinically significant depression than carers with START; HADS-total = 2.10 (95% CI 0.51 to 3.75).Small incremental QALY gain for START group; mean 0.042 (95% CI 0.015 to 0.071). (QALY = quality-adjusted life year)

Livingston et al BMJ 2013

START improved carer mental health and

health-related quality of life over 8 months.

START: outcomes at 8 months

START: outcomes at 8 months

Carers getting START had slightly but not significantly higher costs (£252; 95% CI -28 to +565), adjusting for baseline.Cost-effectiveness: £118 (€201) per 1-point change on HADS-total; and £6000 (€7620) per additional QALY (quality-adjusted life year) … measuring carer service use only.

Cost of START was offset by reduced use of

other services by carers over 8

months. START is cost-effective.

Cost-effectiveness at 8 months

Cost-effectiveness at 8 months

Knapp et al BMJ 2013

Effects on carers:oBetter mental health: carers with usual support were 7 times more likely to have clinically significant depressionoSignificantly better quality of life

Outcomes & cost-effectiveness at 24 months

Outcomes & cost-effectiveness at 24 months

Livingston et al Lancet Psych 2014

Effects on people with dementia:oNo differences in health status or quality of lifeoSome delay to care home admission (not (yet?) significant)

Service costs go up in both groups over time; but care home costs go up more for people in the usual care group.Cost-effectiveness: START has better outcomes and doesn’t cost any more … It is clearly cost-effective.

CST is a group intervention in care homes & day centres for people with mild-to-moderate dementia: themed activities to stimulate cognitive function.Effective and cost-effective if delivered bi-weekly over 7 weeks.Maintenance CST (weekly for 24 weeks) improves QOL; in combination with ACHEI meds it improves cognition.Also cost-effective over 24 weeks, especially with ACHEIs.

Cognitive stimulation therapy (CST)Cognitive stimulation therapy (CST)

Woods et al Cochrane 2012; Orrell et al BJPsychiatry 2014; D’Amico et al, submitted

Home-based care Home-based care

Worringly little evidence on what works in home care.Patterns of home support provided to people with dementia and their carers - study led by David Challis (reporting 2015)Reablement home care – no direct evidence for people with dementia, but Glendinning et al (2010) report some success.

Glendinning et al SPRU/PSSRU report 2010; Hirani et al Age & Ageing 2014; Henderson et al Age & Ageing 2014

What evidence there is suggests quality of care is highly variable, and often very poor (e.g. see recent report from group chaired by Paul Burstow).

Home-based care Home-based care

Surprisingly little evidence on what works in home care.Patterns of home support provided to people with dementia and their carers - study led by David Challis (reporting 2015)Reablement home care – no direct evidence for people with dementia, but Glendinning et al (2010) report some success.

Glendinning et al SPRU/PSSRU report 2010; Hirani et al Age & Ageing 2014; Henderson et al Age & Ageing 2014

Telecare is widely seen as long-term solution. However, today’s evidence is not encouraging: oWSD trial telecare for (all) older people offers ‘small relative benefits’ over usual care, but is not cost-effective (cost per QALY = £297,000).

So, are robots the future?

• Summarise and update the existing evidence base on the cost-effectiveness of treatments and care arrangements;

• Identify those interventions that are consistent with both national policy frameworks and practice guidance, and for which an economic case might be developed;

• Summarise, update or generate (e.g. through simulation modelling) a corpus of up-to-date economic evidence on those interventions that is relevant to the English context;

• Prepare an accessible summary of that evidence that could potentially be useful to commissioners as they make their decisions.

Dementia treatment and care: making the economic case: new study

Dementia treatment and care: making the economic case: new study

o How many people with dementia between now and 2040?

o What will be the costs and outcomes of their treatment, care and support under present arrangements?

o How do these costs and outcomes vary with individual characteristics and circumstances?

o How could costs and cost-effectiveness change if better interventions were more widely available and accessed?

Methods – data-heavy modelling:

o Micro-simulation, macro-simulation, care pathways

MODEM: a projections study (2014-18)MODEM: a projections study (2014-18)

Team: Martin Knapp, Mauricio Avendano, Sally-Marie Bamford, Sube Banerjee, Ann Bowling, Adelina Comas, Margaret Dangoor, Josie Dixon, Emily Grundy, Bo Hu, Carol Jagger, Maria Karagiannidou, Derek King, Daniel Lombard, David McDaid, Jitka Pikhartova, Amritpal Rehill, Raphael Wittenberg,

QuestionsQuestions

Thank you for your attentionQuestions welcome

r.wittenberg@lse.ac.uk