Physical activity in older people living in residential aged-care: A study of its measurement and a physical activity intervention using Active Video Games

Abstract

Regular physical activity for older people is essential for optimal physical and mental health; specifically, a reduction of falls risk and injuries from falls, prevention or mitigation of functional limitations, and also as an effective therapy for many chronic diseases and decreases mortality risk. Accurate measurement of physical activity of this population is important for both surveillance purposes and evaluation of intervention effectiveness. This thesis addresses both measurement and intervention topics in older people living in long-term residential aged-care. The physical activity levels and activity patterns of community-dwelling older people has been characterised using self-report and objectively using body-worn motion sensor devices. This information has been used to determine the relationships between physical activity and health status indices. However, quantifying and describing the habitual active and sedentary behaviours of aged-care residents has received much less attention. On this basis, the first aim of this thesis was to characterise the habitual activity and sedentary behaviour patterns of older people living in long-term residential aged-care. Secondly, there is evidence that physical activity programmes can maintain or improve the physical function of aged-care residents. This contributes to maintenance of a level of independence. However, aged-care residents have physical and cognitive impairments, which makes devising suitable activity programmes for residents challenging. Active video games (AVGs) might address this need. These are games that require the person to move to play the game. Their benefit is they are engaging and fun to play. Whether they are suitable for aged care residents has not been explored. On that basis, the second aim of this thesis was to evaluate the use of an AVG programme for the purpose of improving mobility and physical activity levels in aged-care residents. Thesis structure: The aims of the thesis were addressed using four studies. Studies one and two addressed the first aim of the thesis: (i) to characterise the habitual activity and sedentary behaviour patterns of older people living in long-term residential aged-care. The third and fourth studies addressed the second aim of the thesis (ii) to evaluate the use of an AVG programme for the purpose of improving mobility and free living activity in aged-care residents. Study One: Validation of a body-worn accelerometer Study aims: To determine the validity of a triaxial body-worn accelerometer for detection of gait and postures in people aged over 80 years. Methods: Community dwelling and aged care residents (n= 22, mean age 88.1± 5 y) performed a range of activities (sitting, lying, walking and standing) in both a fixed and free sequence while wearing the accelerometer. Analysed accelerometer data were compared against video observation as the reference measure. Results: The median absolute percentage errors between video observation and accelerometry were <1 % for locomotion and lying. The absolute percentage errors were higher for sitting (median, IQR, -22.3%, -62.8 to 10.7%) and standing (median, IQR, 24.7%, -7.3 to 39.6%). A second by second analysis found an overall agreement of ≥85% for all activities except standing (median, IQR 56.1%, 34.8 to 81.2%). Conclusion: This accelerometer provided a valid measure of lying and locomotion in people aged over 80 yrs. There was an error of approximately 25% when discriminating sitting from standing postures, which needs to be taken into account when monitoring longer term habitual activity in this age group. Study 2: Habitual activity patterns of older people living in residential aged-care Study aims: To describe the habitual activity levels and activity patterns of older people living in residential aged-care using body-worn accelerometry Methods: Accelerometer data collected at three time periods over eight weeks from aged-care residents (n=30, mean age 84.90 ± 6.16 yr) were used to calculate the duration of time spent upright and sedentary. Cognitive status (Abbreviated Mental Test Status), mobility (Timed Up and Go) and balance (the de Morton Mobility Index) were also assessed. Results: Participants spent most of their waking day sitting or lying down (8 hr 31 min ± 40 min per 10 hr measured). Over 50% of all upright (standing and walking) bouts were less than 10 minutes in duration. Increasing age was inversely related to upright time (β = -0.44, p<0.05). No correlation was found between upright time and cognition and mobility measures. Conclusions: Residents’ activity levels were low; and residents spent long periods of time sedentary. The lack of association between habitual activity levels and measures of cognition and physical performance, suggested that aged-care resident’s habitual activity levels were influenced by factors other than physical capacity or cognitive function. Study 3: Activity and energy expenditure in older people playing active video games Study aims: To quantify energy expenditure in older adults playing Xbox 360 Kinect and Nintendo Wii video games, while standing and seated. The secondary aim was to determine whether participants’ balance status influenced the energy cost associated with active video game play. Methods: Community-dwelling adults (n=19, mean age 70.7 ± 6.4 yrs. played nine active video games, each for five minutes, in random order. Two games (boxing and bowling) were played in both seated and standing positions. Energy expenditure was assessed using indirect calorimetry while at rest and during game play. Balance was assessed using the Mini-BESTest, the Activities-specific Balance Confidence Scale, and the Timed Up and Go. Movement was quantified using two dual-axial accelerometers worn on the participants’ right hip and dominant wrist. Results: Energy expenditure during the games ranged from 1.46 ± 0.41 METs to 2.97 ± 1.16 METs. There was no significant difference in energy expenditure, activity counts, or perceived exertion between equivalent games played while standing and seated. No significant correlations were observed between energy expenditure or activity counts and balance status. Conclusions: AVGs provided light intensity exercise, whether played while seated or standing. People who are unable to stand may derive equivalent benefits from active video games played while seated. Study 4: Keeping active using active video games: a cluster randomised controlled trial in residential aged-care facilities Aims: To investigate the use of active video games (AVGs) for aged-care residents for the purpose of improving mobility. Methods: Aged care residents from nine aged-care residential facilities in Auckland, New Zealand was conducted. Four residential aged-care facilities (n=29, mean age 84.7±7.4 yrs.) were randomised to AVGs held twice weekly for eight weeks. The remaining five facilities (n=36, mean age 85.8±7.2 yrs.) were randomised to usual activities (control). The AVGs were a supervised group exercise programme offered twice weekly for approximately 30 minutes per session. Exercises were done in standing whenever possible, with a walking frame or chair to hold onto if needed. The supervising physiotherapist modified the activities as required to suit the residents Outcome Measures: Mobility measures conducted at baseline (pre-intervention), and eight weeks post-randomisation were the Timed Up and Go (TUG) and the de Morton Mobility Index (DEMMI). Percentage of time spent upright (standing or walking), measured using a body-worn accelerometer was assessed as a secondary outcome measure. Cognition was assessed at baseline using the Abbreviated Mental Test Score. Adherence to the AVG programme, adverse events and retention to the study were also documented. Results: Participants attended a mean of 9.3±4.9 of 16 (58%) of the AVG sessions offered. No statistically significant differences in mobility measures (DEMMI or TUG) or upright time were found between intervention and control groups at eight weeks. A sensitivity analysis revealed an improvement in DEMMI scores (p=0.03) and a trend towards improvement in TUG scores (p=0.05) in AVG participants without cognitive impairment, when compared to those with cognitive impairment. No adverse events were reported. Conclusions: An eight week supervised AVG programme did not improve mobility measures in aged-care residents. Nevertheless, the comparatively good attendance rate and lack of adverse events showed the programme to be safe and acceptable for residents. A programme of longer than eight weeks may be needed to demonstrate significant improvements in residents’ mobility. Discussion Older people living in aged-care spend very brief periods of time upright (less than 10 minutes) and long periods of time sedentary. Increasing age is associated with a decline in habitual activity. However, the lack of association found between habitual activity levels, cognition and physical capacity measures (TUG and DEMMI) suggests that aged-care resident’s activity levels are influenced by factors other than cognition and physical capacity. These may include environmental, organisational, or resident-related i.e. health or attitudinal barriers to activity. This suggests that a variety of approaches maybe needed to accommodate these factors when devising physical activity programmes for residents. While no statistically significant improvements in mobility measures were found in favour of AVGs in the trial, there was a trend towards improved mobility scores in AVG participants without cognitive impairment. The comparatively good attendance rate of participants at AVG sessions suggest AVGs for the purpose of increasing activity hold promise as an approach to encourage activity. However, they require modification to make them more suited to aged-care residents. Specifically, games need to be slowed down; and activities simplified to accommodate older people with cognitive and physical impairments. There is the opportunity for suitability qualified health professionals, in conjunction with game designers to develop appropriate AVGs that could be supervised by residential care staff, with little additional training. Any future research needs to consider the sustainability of any improvements gained from using AVGs, the effect of the AVG on quality of life and care staff satisfaction; and whether individual differences (e.g. age, gender, frailty, cognitive and mood status) affect outcomes.