Falls detection and prevention using smartphone technology Dr Helen - - PowerPoint PPT Presentation

Falls detection and prevention using smartphone technology

Dr Helen Hawley-Hague NIHR Research Fellow

http://www.preventit.eu/ www.profound.eu.com www.farseeingresearch.eu www.fallsprevention.eu

www.iofbonehealth.org

30-40% community dwelling >65yrs fall in year

40-60% no injury 30-50% minor injury 5-6% major injury (excluding fracture) 5% fractures 1% hip fractures

Falls most serious frequent home accident 50% hospital admissions for injury due to fall History of falls a major predictor future fall

Masud, Morris Age & Ageing 2001; 30-S4 3-7

• Rubenstein. Age & Ageing; 2006; 35-S2; ii37-41

Falls can be prevented!

• Multiple-component group exercise

– RaR 0.71 [0.63-0.82] RR 0.85 [0.76-0.96]

• Multiple-component home-based

exercise – RaR 0.68 [0.58-0.80] RR 0.78 [0.64-0.94]

• Tai Chi

– RaR 0.72 [0.52-1.00] RR 0.71 [0.57-0.87]

RR=0.83 (95%CI 0.75-0.91)

(High Dose & Challenging RR=0.58 (95%CI0.48–0.69) Sherrington et al JAGS 2008 44 trials 9,603 participants

Gillespie et al 2012 159 trials 79193 participants

Two problems

• Identification

We do not always know when someone falls or what happens beforehand.

• Adherence

We are not always good at implementing the evidence base, older adults do not adhere!

FARSEEING EC FP7 project 10 partners I Bologna Florence Milan x 2 UK Manchester D Stuttgart Cologne NO Trondheim x2 CH Lausanne

1.Universita di Bologna 2.BTCINO 3.Servico Sanitario della Toscana 4.NoemaLife 5.Norges Teknisk-Naturvitenskapelige Universitet 6.SINTEF 7.Robert Bosch Krankenhaus 8.Deutsche Sporthochschule Koln 9.EPFL 10.University of Manchester

Real-World Falls Database

• Defined a data format and build up a meta-database to collect a

reasonable number of falls (more than 200).

• Collect data and signals through the monitoring of high-risk subjects

and fit elderly people.

• Build up a network of experts in this field willing to share

knowledge and data.

• Develop signal processing methods and novel algorithms for the

assessment of daily living activities and of health status.

• Defined an evidence-based fall risk model using advanced data

mining and reasoning techniques.

What does it mean for practice?

• More accurate detection
• Reduction in false alarms
• Outside detection
• Further understanding of what happens just

before someone falls?

• Already tested in Italy and Norway.

How smartphone technology can/could help adherence to evidence based exercise

• Estimated population changes will result in

fracture rates beyond the reach of current intervention(Benzinger et al., 2015), need innovative ideas!

• Soon everyone will have a smartphone! 50% of

people aged 55-64 now have one, this has more than doubled since 2012 (Ofcom, 2015)

• Older adults prefer existing technology rather

than new equipment (Hawley-Hague et al, 2014).

To assist in delivering exercise

• Clinical outcomes were found to be as good if not better

(Kairy et al. 2009).

• Adherence to telerehabilitation is good (Kairy et al. 2009).
• Social networks and friendship identified as important

aspects of group telerehabilitation programmes (Lai et al., 2004).

Smartphone Technology Exercise in the Home Feasibility trial

Can smartphone and teleconferencing technology be used to deliver an effective home exercise intervention to prevent falls amongst community dwelling older people?

School of Nursing, Midwifery & Social Work

The Technology

The technology will increase:

Exercise Safety Social support Motivation

Waist case belt used for wearing the smartphone

School of Nursing, Midwifery & Social Work

School of Nursing, Midwifery & Social Work

Study 1: Usability Testing

Acceptability and usability of technology with older adults/health professionals. Three weeks testing period with patients and staff.

• Recording of issues
• Interviews with

patients

• Focus groups with 3

falls services

Setup Phase

Steering group Creating smartphone applications Initial testing Working with health professionals PPI workshops

Revision Phase

Reflection

• n findings

from usability testing Revisions to applications and set-up Preparation for Study 2

Study 2: Feasibility RCT

• 1. Assess feasibility
• f interventions

delivered as full alternative to standard service.

• 2. Assess

feasibility/ acceptability of design and procedures.

• 3. Determine effect

sizes for sample- size calculations for definitive large scale RCT.

Project outline

App development

• Motivate me
• My activity

programme

With special thanks to Later Life Training for permission to use pictures and name ‘Motivate me’.

Recruitment so far

• 7 patients, 4 men and 3 women.
• 3 existing wifi, 1 provided wifi, 3 with 4G.

Key lessons so far

Patients

• They generally like the teleconferencing-

varied technical support required.

• They like self-reporting and the feedback that

they get.

• They generally find self-reporting and

feedback more motivating than messages on

• utcomes.
• Apps simple to use/some issues with

touchscreen.

Key lessons so far

Health professionals

• Their app could be made more intuitive and

more flexible.

• Some patients it is just not suitable for
• 1. Point programme/motivation
• 2. Cognition
• 3. Safety
• Introducing new exercises.

What next?

• More testing
• Focus groups with staff
• Co-design of feasibility RCT- more than 1 site.
• Full multi-site trial.
• Motivational and self-report apps could be used

within different patient cohorts.

• Could be used after group exercise interventions.
• What happens after intervention period?

Early risk detection and prevention in ageing people by self-administered ICT-supported assessment and a behavioural change intervention delivered by use of smartphones and smartwatches

PreventIT

• Will use smartphone and smartwatches to

deliver motivation.

• Will deliver an adapted version of Clemson et

al’s LiFE programme (Clemson et al, 2012 BMJ). Integrating activities as part of everyday life.

• Multi-site feasibility RCT

Ongoing work

• Work with Phillips Healthcare exploring the

acceptability and potential impact of the use

• f predictive analytics in the prevention of

readmission to hospital for falls.

• Work with Tiyga Health and mHealth

technologies to assess how self-report data alongside sensor data can aid assessment and intervention for falls.

Acknowledgements

Research Team: Health/Social care:

School of Nursing, Midwifery & Social Work

Prof Chris Todd Central Manchester Falls team: Prof.David French Bernie O’Dowd Dr Lis Boulton Ellen Martinez Prof Jorunn Helbostad Caroline Birch

• Prof. Lorenzo Chiari North Manchester Falls Team

Dr Sabato Malone Trafford Community Services Dr Carlo Tacconi AgeUK Tameside Dr Helen Hosker Julie Jerram

http://profound.eu.com/ Farseeingresearch.eu Preventit.eu Email: Helen.Hawley-Hague@manchester.ac.uk