Martin Knapp & Paul McCronePSSRU, London School of Economics

King’s College London, Institute of Psychiatry

Best Practice in Mental Health GP Led Commissioning

25 March 2011

Investing in Mental Health: the

Economic Case

N of people by disorder, England 2007 & 2026

1.24

2.28

0.21

0.580.61

2.47

0.117

1.14

1.45

2.56

0.940.69

2.64

0.1220.24

1.23

0

1

2

3

DEP ANX SCH BPD EAT PER CHI DEM

Num

ber

of p

eopl

e (m

illio

n)

McCrone, Dhanasiri, Patel, Knapp, Lawton-Smith, Paying the Price, King’s Fund, 2008

Current & projected future prevalence

Context: current & projected future costs

7.58.9

4

14.9

0.1

7.9

0.1

5.2

12.214.2

34.8

0.2

12.3

0.1

6.58.2

0

5

10

15

20

25

30

35

DEP ANX SCH BPD EAT PER CHI DEM

Tota

l cos

ts (

£ bi

llion

)

Cost by disorder, England 2007 & 2026

McCrone, Dhanasiri, Patel, Knapp, Lawton-Smith, Paying the Price, King’s Fund, 2008

Aim - model the costs and economic pay-offs of initiatives to prevent mental illness and promote mental well-being.

o Look at evidence-based mental health interventions (incl. non-NHS) – must have well-established outcomes

o Looked at 16 different areas and interventionso Use simple decision analytic modelling o Close liaison with DH officials; consultation with experts As far as the robust evidence base allows:o Include promotion, primary & secondary preventiono Look at widest range of economic impactso Estimate impacts over long time periodso If in doubt, adopt conservative estimates

Our approach

o Examine interventions from 2 perspectives:- pay-offs to society as a whole and- cash savings to the public sector

o The wider impacts are important, given the high ‘external’ costs of many MH problems … but are they considered?

o Over and above the economic pay-offs estimated here there are health and QOL benefits to individuals (‘patients’ etc)

Please be aware that …a. The findings are not definitive – they provide platform for

discussionb. These are simple, partial and incomplete modelsc. The interventions modelled are not necessarily the only

ones that are economically attractive

Please note …

What economic case needs to be made?

Costs are higher

Costs are lower

Outcomes are worse

Outcomes are better

What economic case needs to be made?

Costs are higher

Costs are lower

Outcomes are worse

A non-starter

Outcomes are better

What economic case needs to be made?

Costs are higher

Costs are lower

Outcomes are worse

A non-starter

Outcomes are better

A winner – but check the

timing and spread of impacts

What economic case needs to be made?

Costs are higher

Costs are lower

Outcomes are worse

A non-starter

A delight to penny-pinchers; a nightmare to everyone else

Outcomes are better

A winner – but check the

timing and spread of impacts

What economic case needs to be made?

Costs are higher

Costs are lower

Outcomes are worse

A non-starter

A delight to penny-pinchers; a nightmare to everyone else

Outcomes are better

?? Do the outcomes justify the

higher costs?

A winner – but check the

timing and spread of impacts

Many causes; widespread impacts

Mental health

Health care

Each of these links is evidence-based

Social careHousing Education Crim justice

NHS

LAsCLG

DfEMoJ

BenefitsEmployment

DWPFirms

Vol sectorIncome

CVOs

AllMortalityIndiv

Genes Family IncomeEmply’tResilience

TraumaPhys env

EventsChance

Parenting for conduct disorderTarget Prevalence of conduct disorder = 4.9% among

children aged 5-10Inter-vention

Mix of individual and group-based parenting programmes at age 5; average cost = £1,177 per family

Outcome evidence

Based on data from 20 RCTs, effectiveness = 33% (but low take-up, high drop-out)

Economic pay-offs

Reduced use of NHS, social care and special education services and reduced crime, from age 5-30. Excluded: employment / earnings, social security, adulthood MH, mortality

Findings Total return of £7.89 for every £1 invested, including savings in public expenditure of £2.86

Early intervention teams for psychosisTarget Young people aged 15-35 in general population with

first-episode psychosis. Estimated number per year = 6900.

Inter-vention

Multidisciplinary team intervention including medical and non-medical professionals. Emphasis on assertive approach to maintaining contact and heavy emphasis on vocational recovery.

Outcome evidence

Reduction in relapse rate (Craig et al, 2004), improvement in vocational recovery and quality of life (Garety et al, 2006).

Economic pay-offs

Reduction in: readmission rates, costs of homicide and suicide, and lost employment.

Findings Average cost savings in the short-term of £5777 pa, medium-term £4774 pa and long-term £2600 pa. Return for £1 spent: £5.82 short term, £7.69 medium term, £4.47 long term.

Early detection of psychosisTarget Young people aged 15-35 in general population with

prodromal symptoms of psychosis. Estimated number per year = 15,763.

Inter-vention

Early detection service (based on OASIS in South London; Valmaggia et al 2009). Consists of psychological and pharmacological treatment.

Outcome evidence

Reduced rate of transition to full psychosis and reduced duration of untreated psychosis for those who do develop it.

Economic pay-offs

Reduction in inpatient costs and lost employment, reduction in homicide rate, reduction in suicide rate.

Findings Short-term cost increase of £2228 per person pa, medium term cost saving of £3022 pa and long-term saving of £2604 pa. Annual return for £1 spent: £5.87 short term, £7.42 medium term, £5.05 long term.

Workplace well-being programmesTarget Working-age adult population accessed through their

place of employment Inter-vention

Multi-component health promoting programme, including a health risk appraisal and information and advice tailored to the employee’s readiness to change health-related behaviours. Cost = £80 per year employee per year

Outcome evidence

Quasi-experimental evaluation in UK company reported significantly reduced stress levels, reduced absenteeism and improved productivity (Mills et al 2007).

Economic pay-offs

Reductions in sickness absence and presenteeism; reduced costs of avoidable mental health problems to NHS

Findings Total savings = £9.69 for every £1 invested

Other interventions examined Post-natal depression –

health visitors (universal or targeted)

School-based social and emotional learning programmes

School-based anti-bullying initiative

Workplace screening for depression risk, then CBT

Debt counselling Alcohol misuse - GP screen

and advice

Suicide – population awareness scheme + CBT for people at risk

Suicide – ‘hotspots’ - e.g. safety barriers on bridges

Co-morbid diabetes and depression – collab. care

Medically unexplained symptoms – CBT

Older people – befriending schemes (various)

Dementia - physical exercise programmes

[previously done] Anti-stigma campaigns

NHSOther public sector

Non-public sector

Total

Early identification and intervention as soon as mental disorder arises Early intervention for conduct disorder 1.08 1.78 5.03 7.89Health visitor interventions to reduce postnatal depression 0.40 - 0.40 0.80

Early intervention for depression in diabetes 0.19 0 0.14 0.33Early intervention for medically unexplained symptoms b 1.01 0 0.74 1.75

Early diagnosis and treatment of depression at work 0.51 - 4.52 5.03

Early detection of psychosis 2.62 0.79 6.85 10.27Early intervention in psychosis 9.68 0.27 8.02 17.97Screening for alcohol misuse 2.24 0.93 8.57 11.75Suicide training courses provided to all GPs 0.08 0.05 43.86 43.99Suicide prevention through bridge safety barriers 1.75 1.31 51.39 54.45

Promotion of mental health and prevention of mental disorderPrevention of conduct disorder through social and emotional learning programmes 9.42 17.02 57.29 83.73

School-based interventions to reduce bullying 0 0 14.35 14.35Workplace health promotion programmes - - 9.69 9.69Addressing social determinants and consequences of mental disorder Debt advice services 0.34 0.58 2.63 3.55Befriending for older adults 0.44 - - 0.44

Summary of findings

o Very conservative models even so, many interventions look good value for money.

o Some are self-financing from NHS perspectiveo Some are very low cost: a small shift in expenditure from

treatment to prevention/promotion could generate efficiency gainso Many have broad pay-offs - both within public sector, and more

widely (educational performance, employment/earnings, crime).o Pay-offs may span many years; need to invest for the longer termo Process parameters - targeting, take-up rates, drop-out rates –

are important. It may be most cost-effective to increase take-up among high-risk groups, or improve ‘completion’ rates.

o Each modelled intervention is evidence-based – each has been shown to be effective (to achieve good outcomes) …

o … which means that there are health and QOL gains to individuals over and above the economic pay-offs here

Conclusions

Rebeea’h Aslam 1

Florence Baingana 1

Annette Bauer 1

Jennifer Beecham 1,4

Eva-Maria Bonin 1

Sarah Byford 2

Adelina Comas 1

Sara Evans-Lacko 2

Chris Fitch 5

Nika Fuchkan 1

Derek King 1

Martin Knapp 1,2

Canny Kwok 1

Paul McCrone 2

David McDaid 1

Team; and further informationIris Molosankwe 2

Gerald Mullally 1

A-La Park 1

Michael Parsonage 3

Margaret Perkins 1

Andres Roman 1

Marya Saidi 1

Azuso Sato 1

Madeleine Stevens 1

Jamie Vela 1

1 PSSRU, LSE2 KCL, IOP3 Centre for Mental Health4 PSSRU, Univ of Kent5 Royal Coll. Psychiatrists

Further details of work to date:

Report published by the DH, Jan 2011

More work likely to be undertaken in 2011

Contact:m.knapp@lse.ac.ukpaul.mccrone@kcl.ac.uk