Behavioural Insights in Action Forum Tuesday 13 August 2013 Presented by the Institute for Public Administration Australia (IPAA) and the NSW Department of Premier and Cabinet
Welcome Peter Achterstraat President, IPAA NSW NSW Auditor General
Welcome Professor Mary O’Kane NSW Chief Scientist and Engineer
Facilitator Stephen Brady Deputy Director General Strategic Initiatives and Performance NSW Department of Premier and Cabinet
Visit our Behavioural Insights Community of Practice website: http:/ / policytoolkit.nsw.gov.au/ bi
Keynote Speaker Professor Peter John Professor of Political Science and Public Policy at University College, London
Behavioural Insights: an International setting Peter John University College London
How interventions can find out how to get citizens to act pro- socially • Citizens are not passive - often interested in civic action and doing more for society • But often they find it hard to translate interest and intentions into reality • The behavioural economics take on this is that they tend to use short-cuts and go for easy options that can involve doing nothing - equivalent to leaving the letter on the mantelpiece
Influencing citizen behaviour • A line of experimental work shows that light-touch contacts with citizens can stimulate them to carry out more civic acts • Door knocks, requests (‘asks’), leaflets, telephone calls have effects on turnout (see Green and Gerber 2008), recycling (Cotterill et al 2009) – effect sizes of 4-7 per cent • Content of the message does not matter • M ore controversially – the mode does not matter (see John and Brannan 2008, Fieldhouse et al 2011) • Citizens respond to the cue - they want to do these things, but they need reminding - cuts into a behavioural economics take on civic action - people need shortcuts to do good
Second generation experiments • A new move to more targeted and manipulative experiments, using insights from social psychology • Basic idea is that the citizen sees their actions in the light of what other citizens are doing • Issue is that there are different ways of manipulating this information: a) social pressure b) recognition c) social information d) personalisation
M echanism: recognition • Cotterill et al. (2010) test the whether the numbers of books citizens donate to charity depend on the manner in which they were asked. • The research team randomly allocated 11,812 households in two electoral wards: – a control group that were just asked to donate books to Africa – a pledge group which were asked to pledge – a pledge-plus-publicity group, which got the pledge but who were told that if they donated their names would be put up in a public place.
Book Donations Control Group Pledge Group Pledge & Publicity Group Book donation 282 (7.2%) 320 (8.1%) 346 (8.8%) No book donation 3665 (92.8%) 3617 (91.9%) 3592 (91.2%) Total no. of 3937 3937 3938 households
The wider context of citizen- responsive services • Highly centralised traditions of public management, systems designed by experts, then path dependence • Consumerist trends – especially changes in the private sector, with large companies collecting micro information about consumers cf. Google • M ore vocal dissatisfaction with services > citizen action, complaints • M ove to bring the citizen back in put at the heart of services
The politics of nudge • Not to do with fiscal austerity: – interest pre-dates current round, e.g. under New Labour – Obama employed Cass Sunstein, new White House Unit • But more attractive in age of fiscal austerity - alternatives to money and laws (but note that laws and finance may be improved with behavioural techniques) • Coalition government on its own with Behavioural Insights team (set up for two years in July 2010, now extended)
How academics and policy-makers work together • Successful element of BIT - academics willing to help, flexible input • Need to find the right academics! • Work to their incentives: – interest in study design – interest in being close to policy – not too costly in terms of time – good to get a publication outcome – money is less important (but we do need some reward!)
The limits to nudge • How large are the effect sizes? How sustainable? (Depends on changing habits) • How many defaults are there to change? • Diverts from traditional instruments: see House of Lords Report, Behaviour Change • Natural life-cycle of new fads? • Depends on key people as supporters, e.g. Head of Civil Service Gus O’Donnell (but successor just as enthusiastic)
Limits to nudge (continued) • Does it really challenge the economic model? • Paternalistic? – see work of Bob Sugden • T op down: how to reconcile with decentralisation: see Nudging Citizens Towards Localism
The ‘Think’ agenda • T op down, manipulative > needs a citizen directed complement > Think agenda But think effects weak Or long term?
The impact of discussion versus information on organ donation • The project compared the impact of information alone versus information and discussion on organ donor registration levels. • 180 students were randomly assigned to three groups to receive one of the following interventions: a four page information booklet encouraging registration, the booklet followed by a 15 minute deliberation, information booklet about swine flu (placebo control) • After the intervention, students were invited to join the organ donor register. • Registration increased across all 3 groups • The information booklet had the greatest effect, and raised the number of students on the organ donor register from 23 to 57 per cent, an increase of 17 per cent whereas nudge+think increased from 26 to 41 per cent (control from 34 to 64!).
Conclusions • Behavioural science offers a powerful set of insights for policy makers • Part of citizen-orientated public management – making the citizen at the centre of decisions • Does not get rid of incentives and the economic model: often behavioural insights and incentives work together • How much evidence to change policies? Now moving beyond the quick wins. M ore RCTs • Political support for nudge? Y es
From the Beach to the Bed: Lessons for the recognition and management of the deteriorating patient Professor Cliff Hughes AO Clinical Excellence Commission 13th August 2013
Keynote Speaker Professor Clifford Hughes Chief Executive of the Clinical Excellence Commission
From the Beach to the Bed: Lessons for the recognition and management of the deteriorating patient Professor Cliff Hughes AO Clinical Excellence Commission
Safe Driving - NSW ? 1 ? 2 376 (435) Q 1 What happened here? Q 2 What caused this fall? CF Hughes: 26 September 2012
IIM S enthusiasm
1 August 2012 C F Hughes CF Hughes AO
The Problem • Unrecognised deterioration is a significant problem for patients* in all health systems despite ‘hallmark’ clinical signs of deterioration. • • *Vanessa Anderson (aged 16) 24 th October 2012 C F Hughes
Programme Aim Aim: To improve early recognition and response to clinical deterioration and thereby reduce potentially preventable deaths and serious adverse events in patients who receive their care in NSW public hospitals.
Physiological Parameter Docmented (%) 100% 10% 20% 30% 40% 50% 60% 70% 80% 90% 0% Reliability of Observations Patient Label Date and Time Frequency of Observations Observations graphed, not written Pulse Rate April Blood Pressure Respiratory Rate May Temperature June Pain Score July Bowels O2 Saturations O2 - How? How much? Weight C F Hughes Observations outside normal range Completed
The Solution Prevention Clinical Review Rapid Response Patient Condition Intervention on the Slippery Slope ALS Time C F Hughes
The Solution • Identify missed opportunities to: • prevent • recognise • escalate • respond
Standard Calling Criteria and Charts • Simple to use- single trigger • Most sensitive indicator of deterioration first • Graphed vs. written observations • Clinical usefulness and relevance • Consolidation of observations for a ‘global’ view. • Ordered A-G to support patient assessment • National standards • ‘Photocopiable’ (including patient details)
Adult Chart
Human factors principles • Reduce cognitive load and improve functionality – Top left hand corner is processed first – Font size and type – No overlap of parameters – Colour choice (emphasis) – Colour choice (colour blindness) – Consistency in formatting – Clear and descriptive labels – Low light legibility
Setting Vital Sign Thresholds: Getting the Balance Right • Patient Risk • Sensitivity • Specificity • Resource availability
• D etecting Deterioration, E valuation, T reatment, E scalation, and C ommunication in T eams • Manual • E-learning modules • Clinical skills workshop • Focus on improving the ability of clinicians to identify and respond to clinical deterioration at the ward level
Recommend
More recommend
