– A US study by Sager et al (1996) found that a third of elderly medical inpatients experienced a decline in function whilst in hospital, therefore efficient use must be made of the limited time in the acute care setting.

– Brief but intensive rehabilitation programmes should be targeted at improving muscle strength, functional mobility and endurance in order to widen the patient’s safe functional abilities.

– Adequate practice opportunities throughout the day are crucial to achieving effective acute rehabilitation, e.g. delegating to physiotherapy assistants, working with nursing staff and health care assistants.

– The importance of exercise provision is highlighted by a systematic review completed by de Morton et al (2007) that found multidisciplinary intervention including exercise improved patient and hospital outcomes for hospitalised older medical patients.

– Regular liaison with the multidisciplinary team (MDT) regarding the estimated date of discharge (EDD) is essential so that the necessary assessments are carried out in a timely manner.

– A useful question to consider is: ‘Are the person’s goals likely to be met before the EDD?’

– If the individual is making progress on a daily basis and is likely to reach their pre-morbid function at the same time as the EDD, consider referral to community services for further rehabilitation if indicated, e.g. day hospital or community physiotherapy.

– If the individual has experienced a significant change in their function, discuss the options for further rehabilitation in the community, e.g. bed-based community hospital or intermediate care with the patient and in collaboration with the MDT.

Table 7.1 Evaluating therapy interventions – balance

Outcome measure	Reasoning	Testing tips
Timed Unsupported Steady Standing (TUSS)	This is a simple test of static steady balance for frail older people (Simpson et al 1996).	The end point is 60 seconds. It can be made more challenging by assessing with feet together (TUSSTOG) and feet in tandem (TUSSTAN).
TURN 180	This is a staff-rated performance based test of dynamic postural stability when turning 180° (Simpson et al 2002, Fitzpatrick et al 2005).	It can also be used as a screening test for risk of falling. Taking 5 steps or more to TURN 180 increases the relative risk of falling in the following year by 1.9 in community dwelling older people (Nevitt et al 1989). The older person needs to be able to stand unsupported for one minute in order to perform the test. It is not appropriate for older people with weight bearing restrictions, pain, anxiety or severe confusion.
Berg Balance Scale	This is a scale designed to measure balance by assessing the performance of 14 functional tasks in older people with balance impairment (Berg et al 1989).	It can be used in a variety of settings. It can be used as a screening test for risk of falling. A score of less than 45/56 has been shown to be predictive of falls (Berg et al 1992, Gillespie et al 2000). It can be quite tiring for frail older people or those recovering from an acute illness.
Functional reach	This test measures the maximum distance a person can reach forward beyond arms length with a fixed base of support in standing (Duncan et al 1990).	It can be used in many clinical settings; however the older person must be able to stand unaided and the test performance may be affected by fear of falling. It is not appropriate in patients with significant spinal deformities and may be difficult in those with marked dementia. It can also be used as a screening test for risk of falling (Duncan et al 1992).
Lateral reach test	This is a test of ability to reach directly left and right as far as possible from a fixed base of support in standing (Brauer et al 1999).	Lateral reach assesses a distinct component of postural stability. The older person must be able to stand unaided and the test performance may be affected by fear of falling.
One leg stance	This test measures the time in seconds of ability to balance while standing on one leg (Bohannon 2006a).	The test can also be used as a predictor of injurious falls in older people (Vellas et al 1997a). Inability to stand for 5 seconds on one leg is suggested as a marker of risk (Jonsson et al 2004). Impaired one leg stance is also a marker of frailty (Vellas et al 1997b).
Four step square test	This test is a timed performance based test that assesses dynamic standing balance (rapid stepping and obstacle avoidance) in active older community dwelling older people (Dite & Temple 2002)	The physiotherapist should have a clear view of the patient as they step. Have a second person to closely supervise the patient as they perform the test. This test is cognitively challenging as the older person has to understand and incorporate the stepping sequence. A cut-off time of greater than 15 seconds is associated with increased risk of recurrent falls.

Table 7.2 Evaluating therapy interventions – gait

Outcome measure	Reasoning	Testing tips
Get up and go test	This is a test of sit to stand, walking and turning ability for frail older people (Mathias et al 1986).	This is a staff-rated performance based test rated on a five point scale where 1 = normal and 5 = severely abnormal. A score of 3 or more is at risk of falling.
Timed Up and Go (TUAG)	This is a test of timed sit to stand, walking and turning ability for frail older people (Podsiadlo and Richardson 1991)	The test correlates well with every day function. The chair used should have a seat height of 44-47cms and arm rests as a lower chair may affect the validity of the test (Siggeirsdóttir et al 2002). Try to avoid talking to the person during the test as this may distract them from the task. The test can be used as a screening tool to see if further in-depth assessment of mobility is required. For community dwelling older people a TUAG of 12 seconds is regarded as normal. There is some variation according to age (Bohannon 2006b). TUAG can be used as a falls screening tool but its predictive validity and sensitivity has been variable upon the populations studied, setting and the research methodology (Shumway-Cook et al 2000, Chui et al 2003, Thrane et al 2007, Lindsay et al 2004, Large et al 2006, Kristensen et al, 2007).
6 metre timed walk	Distances limited timed walking tests are useful indicators of functional mobility of older people. Gait speed is important for safe community mobility, e.g.: crossing a road.	A 2 metre distance before and after the course minimises the affects of acceleration and deceleration. Normative data are available (Butler et al 2009)

Table 7.3 Evaluating therapy interventions – combined functional performance

Outcome measure	Reasoning	Testing tips
Elderly Mobility Scale	This is a standardised scale for assessment of mobility, gait, balance and key position changes in frail older people in an acute hospital or day hospital environment (Smith 1994, Prosser et al 1997).	The test correlates well with function. The test has a ceiling effect for older people who are more able. The functional reach scale differs from the original research paper.
Lindop Parkinson’s Assessment Scale	This is a functional assessment scale designed to measure bed and gait mobility in patients with Parkinson’s disease (Pearson et al 2009).	The bed mobility section should be performed without shoes on.

Table 7.4 Evaluating therapy interventions – endurance

Outcome measure	Reasoning	Testing tips
6 minute walk test	This is a useful measure of exercise capacity in older people (Mangan & Judge 1994).	Ideally the test should be conducted in a quiet hallway with cones placed at the beginning and end of 30 metres. The goal is for the individual to walk as far as possible in 6 minutes. The individual is allowed to self-pace and rest as needed as they traverse back and forth along a marked walkway. Encouragement increases the distance walked (Harada et al 1999)

Table 7.5 Evaluating therapy interventions – quality of life

Outcome measure	Reasoning	Testing tips
SF12/36	These generic scales measure functional health and well-being from the patient’s point of view (http://www.sf-36.org/tools/SF36.shtml).	They can be self-administered or completed by interview
EQ-D5	This generic scale measures health related quality of life consisting of five dimensions (mobility, self-care, usual activities, pain and anxiety) plus the individual’s rating on a Visual analogue scale of their current health status (Szende A et al 2007)	Self-administered or completed by interview.
PDQ-39	The Parkinson’s Disease Questionnaire is designed to address aspects of functioning and well-being for those affected by Parkinson’s disease (PDQ-39 – Isis Innovation Ltd, Ewert House, Ewert Place, Summertown, Oxford OX2 7SG UKT +44 (0)1865 280830 F +44 (0)1865 280831 E innovation@isis.ox.ac.uk)	Self-administered.

Table 7.6 Evaluating therapy interventions – fear of falling

Outcome measure	Reasoning	Testing tips
Falls Efficacy Scale-International (FES-I).	This is a test to measure level of concern of falling during a range of physical and social activities that is suitable for use across a range of cultures and languages (Yardley et al 2005, Kempen et al 2008).	To obtain a total score for the FES -I add the scores on all the items together, to give a total that will range from 16 (no concern about falling) to 64 (severe concern about falling).
Activities Balance Confidence Scale (ABC)	This is a test to measure confidence in doing a range of activities on a scale from 0-100% (Powell et al 1995).	If the older person normally uses a walking aid, ask them to rate themselves as if they were using the aid.

Exercise interventions to prevent falls in older people at high risk or who have fallen

• The following risk factors for falls can be improved by physiotherapy intervention:

– Muscle weakness

– Impaired balance

– Impaired gait

– Reduced reaction times.

• Exercise prescription will differ depending upon the individual’s history of falls, medical conditions and functional capacity.

• In older people with poor balance, some preparatory strength, co-ordination and flexibility training may be required before unsupported dynamic balance exercise starts.

• Safety is paramount, less challenging balance exercises may need to be prescribed initially, particularly if the older person is exercising at home unsupervised.

• Exercise programmes to prevent falls should include dynamic balance, strength and functional floor activities. They should also aim to include bone loading, power, flexibility, posture, gait training, supported endurance work and tasks to improve visual, vestibular and sensory input (DOH 2009).

• Evidence:

– Recent reviews and guidance recommend that falls intervention, including exercise, should be considered alongside osteoporosis management to reduce both the number of falls and falls-related injuries (NICE 2004, DOH 2001b, DOH 2007, Skelton and Todd 2004, Sherrington et al 2008).

– Studies of strength and specific balance training have shown falls reduction in intervention groups when evaluated at the end of one year (Tinetti et al 1994, Campbell et al 1999, Province et al 1995).

– A randomised controlled trial (RCT) of group based exercise over one year found that the intervention group (community dwelling older people n = 551) had a 22% lower rate of falls than the control group and 31% fewer falls in those who had fallen in the past year (Lord et al 2003).

– A UK study of community dwelling older people with a history of falls, halved their risk of falls with 9 months of weekly group balance and strength exercise combined with twice weekly home exercises (Skelton et al 1995). In this study the exercise intervention consisted of progressive resistance, gait, balance, functional activity, floor work, endurance and flexibility training. The exercise was individually tailored in type and intensity with most exercise in weight bearing positions, reducing upper limb support.

– In New Zealand, an RCT of home-based individually tailored exercise (OTAGO) including a warm up, muscle strength and balance exercises plus a walking programme three times per week for one year was found to be effective in reducing falls and injuries in community dwelling older women over 80 years old (Campbell et al 1999). During the study the physiotherapist visited each individual in the intervention group 4 times in the first 2 months followed by regular telephone contact.

– There have been several studies investigating the effects of Tai Chi on falls. There is some evidence that Tai Chi is effective as a preventative group exercise for older people with impaired balance and strength who have not fallen (Wolf et al 1996). However modified Tai Chi has not been found effective in frail older people (Wolf et al 2003). A review paper by Harmer et al (2008) recommended that longitudinal studies with consistent intervention parameters and clinically meaningful outcomes were required in order to establish the role of Tai Chi in falls prevention.

Long-term physical activity and exercise opportunities

• A major challenge is to ensure a continuum of exercise provision and to make a successful transition from a health care setting (one to one or small group basis) to a community based setting.

• Sherrington et al (2008) showed that the greatest relative effects of exercise on falls rates is seen in programmes that include a combination of a higher total dose of exercise (>50 hours over the trial period) and challenging balance exercises.

• It is important to provide a choice of exercise opportunities to ensure individual need and preference are met, as this is more likely to improve participation and uptake.

• Effectiveness will be determined by how receptive the individual is to the recommendations and by how capable they are when carrying out the exercises independently, safely and effectively (Dinan 2001).

• Evidence suggests that older people will be more receptive and more likely to undertake an exercise programme if the information provided discusses the wider benefits of exercise to well-being and maintenance of independence, rather than just to prevent falls (Skelton & Todd 2004, Yardley et al 2007a, Yardley et al 2007b).

Progressive Resistance training

• Resistance training aims to increase the ability of a muscle or group of muscles to generate force.

• To increase strength, a resistance should be used that allows 6–8 repetitions for each exercise. Aim for a resistance of 65–75% of 1RM. Make sure that the older person has a 1–3 second rest between repetitions and a 90–120 second rest between sets. Aim for 1–3 sets of each exercise using the major muscle groups 2–3 days per week with 48 hours between sessions. The level of effort should be moderate to high.

• On a 10 point scale, where no effort is 0 and maximal effort of a muscle group is 10, moderate intensity is 5–6.

• With frail older people, using body weight is often a sufficient training stimulus initially. Exercise intensity can be altered by adjusting the performance of the exercise, e.g. reducing the use of the upper limbs when practicing sit to stand or increasing the repetitions.

• Using lighter weights and a higher number of repetitions, particularly if the patient has musculoskeletal disease, will assist the development of endurance and power required for functional activity.

• Gradual progression in weight and repetitions should be made on a regular basis to maintain overload.

• Overuse injuries can occur during resistance training. To reduce the risk, precise teaching instructions and skilled demonstration together with observation of the person and feedback are required.

• Evidence:

– The greatest strength gains occur in programmes of moderate to high intensity where exercises are performed at 50% or more of the patient’s one repetition maximum.

– Some studies have shown that progressive resistance training programmes improve performance in functional tasks (Jette et al 1999, Fiatarone et al 1990, Vincent K 2002).

– Other studies have found large strength gains, but no or small improvements in physical performance or function (Jette et al 1996, Judge et al 1994, Skelton et al 1995).

– A Cochrane review by Liu and Latham (2009) found that resistance training conducted two to three times a week, at moderate to high intensity, with free weights, exercise machines or elastic bands increased strength in older people and had a positive effect on performance of activities such as standing up from a chair more quickly, walking and stair climbing. In addition progressive resistance training improved older peoples’ ability to perform more complex functional tasks such as preparing a meal. However the reviewers advised caution when transferring the research into practice as adverse events were not adequately reported in the studies.

– An RCT of home-based exercise (n = 29) that included progressive resistance training, balance and general physical activity over a 6 month period found that functional performance, dynamic balance, muscle strength and aerobic fitness improved significantly in the intervention group (Nelson et al 2004).

– An RCT of 3 months low intensity exercise programme versus unsupervised home based flexibility activities in frail older people (n = 84) found significant improvement in the exercise group on frailty markers such as strength, gait speed and balance (Brown et al 2000).

– Weighted vest exercise has been used safely in older people and has been found to be effective in improving balance, strength and bone mass (Shaw & Snow 1998, Bean et al 2004, Snow et al 2000).

– Research has also studied frail older people living in care homes although the sample numbers have been small. Fiatarone et al (1990) found that an 8 week high intensity strength training (n = 10) improved muscle strength and tandem gait speed in nonagenarians in care homes. Ikezoe et al (2005) found that a 12-month low-intensity exercise programme was effective in increasing strength and maintaining balance and mobility in frail older people residing in care homes (n = 28). Sauvage et al (1992) conducted an RCT of a 12 week (3 times per week) programme of PRT and stationary cycling (n = 14) on older men living in a care home. This resulted in significant although limited improvements for clinical mobility scores, strength, muscular endurance and gait velocity.