Strategies to enhance routine physical activity in care home residents: the REACH research programme including a cluster feasibility RCT

Accelerometer wear time

Extended wear of accelerometers by residents was high: 41 (89.1%) of the 46 recruited residents undertook extended wear of an accelerometer. Thirty (73.2%) wore the accelerometer on the hip, 10 (24.4%) wore the accelerometer on the wrist and one resident was asked to wear a commercially available device so that we could explore the data collected by this device. Unfortunately, there are no published cut-off points validated in older adults for wrist monitor wear; thus, the data for wrist accelerometers were not subsequently analysed. In addition, the outcomes of the commercially available device were based on distance moved; therefore, we decided that these data were unlikely to be sufficiently accurate for our purposes in this population.

Twenty-two (73.3%) of the 30 residents wearing a hip accelerometer had valid data (≥ 8 hours on ≥ 4 days of the week) and wore the accelerometers for a mean of 6 days (range 4–9 days). The mean wear time per day was 12 hours 24 minutes [standard deviation (SD) 2 hours 36 minutes] (Table 2).

The proportion of residents meeting the criteria for valid wear time (73.3%) was lower than that found in the development work (84.8%). This may, however, be attributed to differences in participant characteristics, for example inclusion of those deemed to lack capacity in this study. Enabling residents with cognitive impairment to wear monitors may have required greater support from staff; this may not have been prioritised by staff, given their heavy workloads. Furthermore, over half of the potential participants (10/19, 52.6%) in the development work declined to take part in the study because they did not wish to wear an accelerometer for an extended time (5 days were needed for data to be considered valid). It may have been that those who did wear an accelerometer were not representative of a CH population.

Compared with more recent studies conducted in long-term care settings that have employed similar criteria to define valid wear time, a similar or greater proportion of residents in the current study had valid data. 23,58,59

Activity logs

Correct identification of accelerometer wear time is imperative to ensure that accelerometer data are interpreted and analysed correctly. 60 Although different methods of identifying accelerometer non-wear are used in the literature, automated algorithms and activity logs are the most common. However, both methods present limitations. Automated algorithms have not been developed with this population specifically, and, on examination of the first sets of data collected, it was apparent that sedentary time may be mistakenly categorised as non-wear time, thus underestimating wear time and incorrectly categorising sedentary time as non-wear time. Consequently, although it was acknowledged that completion of activity logs may be viewed as burdensome, much emphasis was placed on attempting to collect accurate information about accelerometer wear using an activity log.

Accordingly, activity logs were administered in a way in which the CH managers felt would result in the greatest success. Nevertheless, completion was poor and appeared onerous for staff, even after they were asked principally to record the times of administration and removal of the monitor, and advised that recording reasons for removal were optional.

Eight (19.5%) of the activity logs were fully completed for the time accelerometers were in the residents’ possession, 20 (48.8%) were partially completed and 13 (31.7%) were not completed at all. Only three (10.7%) of the logs were completed by residents; 12 (42.9%) were completed by CH staff with the assistance of the researcher and 13 (46.4%) were completed by the reseacher alone.

Given the importance of collecting accurate wear time information in this population, the activity log was amended. The amended activity log, in addition to strategies proposed to encourage completion, was to be trialled and further refined, if necessary, in WS4 prior to their use in WS5, the feasibility trial.

Levels of physical activity and sedentary behaviour

Mean daily time spent in engaging in PA was 1 hour 13 minutes (SD 1 hour 8 minutes) (Table 3), which equated to 9.5% (SD 8.3%) of accelerometer wear time (Figure 3).

FIGURE 3.

Proportion of daily wear time spent sedentary and in combined low, light and moderate to vigorous PA levels in each individual CH and, on average, across all CHs (n = 22). SB, sedentary behaviour.

The majority of the time spent in PA was spent in low-intensity PA [1 hour 7 minutes (SD 1 hour 4 minutes); 8.7% (SD 7.7%) of wear time].

List 1: explanatory determinants

1. Knowledge.

2. Environmental context and resources.

3. Motivation and goals (intention).

4. Beliefs about capabilities (self-efficacy).

5. Emotion.

6. Social influences (norms).

7. Skills.

8. Beliefs about consequences (anticipated outcomes).

9. Action-planning.

10. Memory, attention and decision processes.

11. Social and professional role and identity.

The result of this step of the IM process consisted of a matrix of cells that included the intersection of proximal performance objectives (rows of the table) with specified determinants (columns of the table) (see Appendix 5). From the intersecting cells, a statement or a learning or change objective was identified that would support the achievement of the proximal performance objective. This was achieved by scrutinising each performance objective individually to identify what barriers were associated with achieving each desired outcome. Our change objectives outlined specific challenges that staff, residents and relatives face in increasing regular movement, and the objectives could be clearly mapped to psychological determinants of behaviour (e.g. ‘intention’, ‘social influences’ and ‘skills’). For example, one of the determinants selected to be targeted in the intervention was knowledge. This was in recognition of the fact that baseline level of knowledge about what PA is, in the context of increasing everyday movement, rather than structured PA in exercise classes, and how staff might engage residents in increasing movement a little every day, was necessary.

Via researcher interview

• Cognitive impairment (6-CIT). 47

• Mood [Geriatric Depression Scale (GDS)]. 70

• Perceived health [EuroQol-5 Dimensions, five-level version (EQ-5D-5L)]. 71

• Quality of life [Dementia Quality of Life (DEMQOL)]. 72

• Three questions from the World Health Organization Quality of Life-OLD (WHOQOL-OLD). 73

Via staff member

• Physical function and mobility [FAC50 and Elderly Mobility Scale (EMS)]. 74

• Medical history (based on the Charlson Comorbidity Index). 75

• Health-care resource use (for the health economic study).

Accelerometry

To allow for the objective measurement of PA, residents were asked at each data collection time point to wear an ActiGraph wGT3X-BT accelerometer (ActiGraph, LLC) on the hip, during waking hours, over the course of 7 days. A detailed protocol, refined through earlier WSs, was followed for administration of the accelerometers, which took place after the data collection defined above to ensure that accelerometer wear did not interfere with residents’ routine PA levels and influence their questionnaire outcome assessments. Researchers assisted with the administration of the accelerometers on the initial visit and instructed care staff in their daily application. Researchers routinely checked compliance by regular telephone calls and visits to the CHs (including a weekend telephone call or visit). Residents and CH staff were also encouraged to complete a purposely designed log to record daily wear time.

Care homes

Care homes in the MoveMore arm were, on average, smaller than the UC CHs, with a smaller number of beds, a smaller number of beds taking part and a smaller number of permanent residents, although there was greater variation in resident numbers across UC homes during the trial. UC homes were, however, less likely to provide rehabilitation or intermediate care and have telemedicine facilities (see Appendix 7, Table 23).

See Appendix 7, Table 24, for follow-up data of CH characteristics.

Care home managers

Information on CH manager demographics is provided in Appendix 7, Table 25.

The CH managers in the MoveMore homes had slightly more experience working in CHs, and more experience as a home or unit manager. However, the managers in the UC homes had been in their current post for, on average, 3 years longer than the managers in the MoveMore homes, although this was skewed by one particular home whose manager had been in the post for 22 years. There was a change in manager at one CH in the UC arm at both the 3- and 9-month follow-ups.

Information on staff demographics (see Appendix 7, Table 26) was obtained via completion of staff booklets, which were distributed to all staff who had face-to-face contact with residents, with the exception of those acting as a NC. Overall return rates were low, ranging between 20 and 39%, with a higher proportion of staff in the UC CHs completing the booklets at each time point. However, as the characteristics of those who completed the booklet at each time point were reasonably similar between the arms, any differences are likely to reflect the profile of the staff at the CHs, rather than any systematic differences in those completing the booklet. Thirteen members of staff across all CHs completed the staff booklet at every time point.

Resident screening and recruitment rates

Three hundred residents in the 12 CHs were screened for participation. Of these, 278 (92.7%) were eligible, 159 consented/had consultee agreement (57.2% of eligible; 53.0% of those screened) and 153 were registered to take part in the study (55.0% of eligible; 96.2% of those consenting/consultee agreement) (Figure 10).

FIGURE 10.

Flow of residents through screening, consent, baseline assessment and registration. a, Reasons not mutually exclusive.

Forty-three (59.7%) of 72 residents with mental capacity were recruited. Of the 203 residents with doubtful or who were lacking mental capacity, 187 had a potential PC; 73 (57.9%) of the 126 PCs who agreed to act agreed for the resident to take part in the study. A NC was used for 76 residents, 43 (56.6%) of whom agreed for the resident to take part in the study.

There was variation across CHs in the proportion of eligible residents who were registered, but the overall proportions were similar between the arms: 62 out of 113 (54.9%) eligible residents were registered in the MoveMore arm, compared with 91 out of 165 (55.2%) in the UC arm (see Appendix 7, Table 27). On average, 12.4 residents per home were registered in the MoveMore arm, and 13 residents per home were registered in the UC arm, although, in one CH, only six residents were registered in each arm (see Appendix 7, Table 27).

The average time for recruitment at the CHs was 64 days; the shortest time was 50 days and the longest was almost twice this (97 days), although these times also included approximately 3 weeks per CH for baseline data collection (depending on the number of participants in each CH).

Follow-up rates

Care home and resident retention during the study period was high, with no CH or resident protocol violations or withdrawals, although one CH in the UC arm temporarily withdrew from researcher visits at 3 months because of renovations in the home (Figure 11).

FIGURE 11.

The CONSORT diagram of the flow of CHs and residents through follow-up by randomised arm.

There was some variation in the proportion of residents followed up across the CHs (50.0–90.9% at 9 months) (see Appendix 7, Table 27), with a lower overall proportion in the MoveMore arm (69.4%) than in the UC arm (76.9%) (difference –2.65, 95% CI –17.50 to 12.20) (see Appendix 7, Table 28).

Losses to follow-up were due to deaths and movement of residents out of the CH (see Appendix 7, Table 29). Deaths and resident movement were also similar between arms at the CH level (see Figure 11 and Appendix 7, Table 28).

Physical activity

At baseline, residents in the MoveMore arm spent less time engaging in PA of any intensity (mean 1 hour 7 minutes, 8.5% of accelerometer wear time) than residents in the UC arm (mean 1 hour 53 minutes, 13.4% of accelerometer wear time) (see Appendix 8, Tables 60 and 61).

At each follow-up, the mean daily time spent engaging in PA of any intensity increased in both arms. The mean daily time that residents in the MoveMore and UC arms spent engaging in PA of any intensity at 9 months was 1 hour 25 minutes (SD 47 minutes) and 2 hours (SD 2 hours 16 minutes), respectively. This equated to 10.9% (SD 5.5%) of accelerometer wear time in the MoveMore arm and 12.6% (SD 10.8%) in the UC arm. The increase in PA of any intensity from baseline to 9 months was greater in the MoveMore arm than in the UC arm (see Appendix 8, Tables 60 and 61).

However, in both arms, all residents engaged in some level of light and moderate activity, although this was for < 2% of the day (see Appendix 8, Table 61). The increased proportions of time spent in PA were reflected mainly in periods of low intensity (see Appendix 8, Table 61).

In the MoveMore arm, there was a gradual increase in the number of days that residents engaged in low-intensity bouts of 5 minutes’ duration and in the number of 5-minute bouts per day (see Appendix 8, Table 63).

Sedentary behaviour

At baseline, the mean daily time that residents spent sedentary was similar between the arms (MoveMore arm: 11 hours 38 minutes; UC arm: 11 hours 41 minutes) (see Appendix 8, Table 60). This equated to a considerable proportion of accelerometer wear time (MoveMore arm: 91.4%; UC arm: 86.6%) (see Appendix 8, Table 61). As the CIs for these figures did not overlap, there was a suggestion that the proportion of time in sedentary behaviour was greater in the MoveMore arm. This is consistent with residents in the MoveMore arm having lower physical function and more comorbidities than those in the UC arm.

At 9 months, there was no change in the average time residents in the MoveMore arm spent sedentary (11 hours 31 minutes), compared with baseline (see Appendix 8, Table 60); however, this did account for a smaller proportion of accelerometer wear time (89.1%) (see Appendix 8, Table 61). Conversely, in the UC arm, an increase in the time spent sedentary was observed (12 hours 33 minutes) (see Appendix 8, Table 60), such that there was little difference in terms of the proportion of wear time (87.4%) (see Appendix 8, Table 61). Thus, at 9 months, the difference between the arms in terms of percentage of accelerometer wear time spent sedentary narrowed (MoveMore arm: 89.1%; UC arm: 87.4%) (see Appendix 8, Table 61).

Physical activity bouts

At baseline, residents in the MoveMore arm engaged in bouts of low-intensity PA of at least 5 minutes’ duration on fewer days, with less frequency per day and for shorter duration (see Appendix 8, Table 63) than residents in the UC arm, and they engaged in bouts of low-intensity PA of at least 10 minutes on fewer days and with less frequency per day than residents in the UC arm (see Appendix 8, Table 64). However, the proportion of residents engaging in low-intensity bouts of at least 10 minutes’ duration was lower in the MoveMore arm (19/49; 38.8%) than in the UC arm (49/77; 63.6%); hence, comparisons should be interpreted with caution (see Appendix 8, Table 64).

There were fluctuations during follow-up among residents in the UC arm in low-level intensity bouts of PA of ≥ 5 minutes’ and ≥ 10 minutes’ duration in the number of days they engaged in at least one bout, the number of bouts per day and the average duration of the bouts. However, by 9 months, there was a suggestion that residents engaged in low-intensity bouts of 5 minutes’ duration on fewer days than they did at baseline (see Appendix 8, Table 63). But, on the days they did engage, they had more bouts (see Appendix 8, Table 63), although the average duration of the bouts did not appear to change (see Appendix 8, Table 63).

At 9 months, there was no suggestion of a difference between the arms in the 5-minute low-intensity bout outcomes (see Appendix 8, Table 63), but the difference in the number of days and number of bouts per day for the 10-minute low-intensity bouts that was observed at baseline remained (see Appendix 8, Table 64). There were, however, few residents engaging in low-intensity bouts of 5 minutes’ and 10 minutes’ duration by 9 months; therefore, it is difficult to draw robust conclusions on such small sample sizes (see Appendix 8, Tables 63 and 64).

Breaks in sedentary behaviour

At baseline, the number of breaks in sedentary behaviour that residents had per day were fewer among residents in the MoveMore arm than among residents in the UC arm (see Appendix 8, Table 62).

At each follow-up, the number of breaks in sedentary behaviour increased in both arms, but by a greater amount in the MoveMore arm (see Appendix 8, Table 62), such that, at 9 months, there was no suggestion of any differences in the number of breaks in sedentary behaviour between the arms (see Appendix 8, Table 62).

Sedentary behaviour bouts

At baseline, all residents engaged in sedentary bouts of at least 30 minutes’ duration. On average, residents engaged in at least one 30-minute sedentary bout on 6 of the days that they wore the accelerometer (see Appendix 8, Table 65) [the average number of days’ wear was 6.9; on each of these days residents had, on average, a mean of 6.4 (SD 1.6) sedentary bouts in the MoveMore arm and 5.9 (SD 1.9) sedentary bouts in the UC arm (see Appendix 8, Table 65)]. The duration of the sedentary bouts was similar between arms, lasting, on average, 1.5 hours before the bout was broken by a period of PA, even if this was for only 1 minute.

The number and duration of 60-minute sedentary bouts was also similar between the arms at baseline (see Appendix 8, Table 66), although residents in the MoveMore arm appeared to engage in 60-minute sedentary bouts on more days than residents in the UC arm (see Appendix 8, Table 66). Therefore, although residents in the MoveMore arm appeared to spend a greater percentage of their day sedentary at baseline, there was a suggestion that the difference was in shorter bouts.

Changes in bouts of sedentary behaviour were observed in both arms during follow-up. At 9 months, residents in the MoveMore arm engaged in 30-minute sedentary bouts on the same number of days and had the same number of bouts per day as at baseline, but the duration of these bouts was shorter (see Appendix 8, Table 65). Residents in the UC arm engaged in 30-minute sedentary bouts on fewer days at 9 months than at baseline, but had more bouts per day, although the duration of these bouts was also shorter (see Appendix 8, Table 65). For the 60-minute sedentary bouts, a similar pattern was observed. Residents in the MoveMore arm also engaged in these bouts on fewer days at 9 months than at baseline, and had fewer bouts per day at 9 months than at baseline (see Appendix 8, Table 66).

Variation in outcomes between care homes

The intracluster correlation coefficients (ICCs) (see Appendix 8, Tables 56–59) provide an estimate of the amount of variation in each outcome, explained by the variation between CHs. These data inform sample size calculations for a definitive trial. In this feasibility study, we had a small number of CHs (clusters) with varying cluster size, with a maximum of between three and 20 residents per home, which led to an unbalanced design, for which there is limited research on the performance of ICC estimates.

Individual-level profiles

Displaying the profile of individual residents’ activity in CHs allowed us to examine whether there were any observed changes that may or may not be reflected at the level of the CH. It also allowed us to see the contribution of each resident to analysis at the different time points. We chose sedentary behaviour variables, as these had a larger sample in the population.

There was a large amount of variation between residents in all homes in both arms when examining the percentage of time per day spent sedentary (see Appendix 8, Figures 14 and 15). In particular, in the MoveMore arm there was marked fluctuation in individuals’ sedentary behaviour over the study duration. However, there did appear to be a number of residents in CHs in the MoveMore arm who showed a decrease in the percentage of time spent sedentary, most notably in CH3 and CH4 up to the 6-month follow-up (after which sedentary behaviour increased again) and in CH5 between the 6- and 9-month follow-ups. In the UC CHs, there was a clear increase in the percentage of time spent sedentary per day for all residents in CH7, CH8 and CH11, whereas, in the other CHs, the percentage of time spent sedentary stayed at a similar level or increased slightly. The presence of an activity co-ordinator did not appear to affect residents’ profiles.

The number of breaks in sedentary behaviour per day also showed variation between residents within homes and for individual residents across the study duration (see Appendix 8, Figures 16 and 17). The most notable change in the MoveMore CHs was a general decreasing trend for the majority of residents in CH5 up to 6 months, after which time the number of breaks increased, in line with the observed decrease in the percentage of time spent sedentary.

Setting

Rowntree Nursing Home is a fairly modern two-storey nursing home situated on the outskirts of a small, rural village. Local resources are very near the nursing home. Village shopping areas are diverse in character, and include cafes, pubs and other amenities, such as the local library; these were perhaps one-quarter of a mile from the CH. A green space, incorporating a field with seating areas, was roughly 100 m from the nursing home. The CH is owned, managed and run by a family who had set it up as a limited company. There were approximately 44 residents in total, roughly eight of whom were male. Rowntree Nursing Home provided mainly permanent care; however, on occasions, respite care was offered if a bed was available. Approximately half of the residents were self-funded, with the remaining half funded by the council. In addition, four beds were available that were funded by the NHS.

The routine of the organisation is concerned with carrying out the tasks of providing meals and dispensing medicines, as well as meeting residents’ needs in terms of general care. In addition to assisting residents with moving to different areas of the CH, such as the dining areas, encouraging them to finish their meals and assisting them with toileting, a more holistic approach to satisfying residents’ social and emotional needs allowed for the provision of regular daily activities, which usually took place in the afternoons.

Organisation, management and delivery of care

The managerial hierarchy in Rowntree Nursing Home centred on the family. Key responsibilities for overall management and administration were held by the general manager and his stepson; the general manager’s wife was the nurse manager.

Hands-on care was provided by a group of experienced care workers, most of whom had been employed in Rowntree Nursing Home for a number of years; this created an atmosphere of stability and regularity around procedures and engagement with residents. Most of the care workers were female and generally aged in their 40s, with a few others aged in their 20s or 30s. Some of the care workers displayed particularly strong empathetic skills in emotionally engaging with residents and were able to spend time talking to residents, holding their hands and listening to residents for longer periods than other careworkers, who undertook care in a more perfunctory manner (Table 7 presents further details of the staff body).

In general terms, routine procedures around mealtimes, dispensing medicines, and assisting residents with toileting and other aspects of personal care reflected a more holistic ethos to caring for residents. This approach involved socially interacting with residents in ways that were mindful of residents’ personal characteristics and likes/dislikes; this often produced richer, higher-quality social interactions, which seemed to support a strong sense of well-being among residents. This holistic ethos of care was in marked contrast to the more managerial approach to delivering care in other homes, driven by the need to perform specific tasks, such as gathering residents to the dining table at mealtimes, within a limited time frame, in which the quality of social interactions seemed poorer.

Care staff worked varying shift patterns according to the number of hours they worked each week. Irrespective of the shift patterns, there was always a combination of 10 care and nursing staff to cover the morning periods, nine to cover the afternoons, eight to cover the evenings and four to cover the night-time.

An experienced, creative and motivated activities co-ordinator planned and delivered various social activities and PAs, such as arts and crafts, musical performance, bingo, outdoor trips and quizzes. A physiotherapist assisted the activities co-ordinator in delivering armchair exercises on a fortnightly basis. Unusually for a CH, a complementary therapist provided various therapies, such as reiki, homeopathy and Indian head massage. In addition, a reminiscence therapist worked on a one-to-one basis with residents. Additional support was provided by speech and language therapists, nutritionists, district nurses, mental health nurses and medical consultants through the local NHS.

Some relatives took a fairly active role at the nursing home, occasionally accompanying residents and staff on day trips to provide additional support. During times when they visited the CHs, there was considerable banter between relatives and other residents, as well as with members of staff when in the communal spaces, and this conveyed a sense of warmth in the environment.

Physical environment

The approach to Rowntree Nursing Home comprised a large garden space with a lawn, flower beds and raised beds, as well as window boxes beneath residents’ windows on both upper and lower floors; the overall impact during the spring and summer period was the production of a vivid and colourful floral display.

Rowntree Nursing Home is a two-storey building. It used to be a former mill owner’s home and has been adapted and expanded in various ways that have allowed for additional rooms to become available for residents, and has enabled the conservatory space to be built. The top floor houses residents’ rooms, a dining area, the kitchen and the conservatory, which was used for delivering activities and was also a lounge space. The lower floor houses more residents’ rooms, a lounge/dining area and an administration office. Most of the residents’ rooms have en suite facilities.

In general terms, the home is decorated in warm bright colours, has modern decor and there are various pictures adorning the walls. The ground-floor lounge space is smaller than the upper lounge space and can seat approximately 10 residents. Residents would generally prefer to sit in the same seats rather than move from lounge to lounge, and so lower-floor residents would tend to remain there rather than venture to the upper floor, even to participate in activities. Seating on the ground floor was arranged around the perimeter of the lounge, with residents facing inward, and a television set was positioned in the corner of the room. A collapsible dining table rested against a wall and could seat about eight residents.

The upper-floor lounge space was much larger and there were a number of small dining tables that could each accommodate four residents. An open doorway linked the upper lounge to the conservatory space, which also acted as a lounge space. Comfortable seating was available in both the upper lounge and the conservatory space and residents tended to sit in corner spaces in both areas, which facilitated some communication.

In the lounge spaces on both floors, residents tended to do very little beyond the mid-afternoon period, usually 14.00–15.00, during which activities were provided for residents in the conservatory or lounge space on the upper floor. Residents were generally lethargic, spending time napping in their chairs, reading newspapers or else asleep. There was some conversation between residents, although this varied considerably. Sedentary behaviour was prevalent, with most residents seated for the majority of the time and shuffling infrequently between lounge space, bathroom and dining areas. However, three or four residents were extremely physically active, wandering about the CH for much of the time in a fairly aimless fashion.

Individual environment

Roughly 25 of the 40 residents were observed over the course of the fieldwork, as a large proportion spent most of their time in their rooms. Some of these residents were in exceptionally poor health.

The majority of the residents observed were aged between 80 and 100 years. Physical mobility was generally poor. Although 10 of the residents were fully able to self-mobilise, eight required a walking frame and two required the use of a wheelchair. Conversations and observations with both residents and staff indicated that most of the residents had lived for a number of years in the vicinity of Rowntree, and a number had known each other during earlier parts of their lives.

Based on the observations of the researcher, residents’ cognitive abilities were also limited, with approximately 12 residents deemed to have severe cognitive impairment and five deemed to have a mild cognitive impairment. Some of those with severe cognitive impairment were unable to speak or make themselves understood. Residents with mild cognitive impairment were able to hold conversations for short periods and to volunteer thoughts and opinions on various subjects.

The pen portraits in Box 1 summarise the key characteristics for three residents living at Rowntree Nursing Home. The first pen portrait presented describes a resident who is relatively able and active, the second is someone who is fairly average in terms of physical and cognitive ability, and the third is someone who is physically and cognitively frailer than average.

Daily routine

Morning routines centred on getting residents ready for breakfast, and helping them with washing and toileting. Most residents had breakfast in the communal spaces, either on the upper or the lower floor. There was some flexibility regarding the timing of breakfast, and in the time prior to lunch residents mostly sat about idle in the lounges. Yet this was the time of the day when they seemed most alert and energised, and it seemed to represent a good opportunity for delivering PA.

Generally, residents were extremely sedentary, spending most of their time in a seated position. In terms of routine life, getting ready for mealtimes, toileting and moving into the conservatory space to take part in activities were the peak times for physical movement (beyond actual participation in the few PA-orientated sessions provided).

Lunchtimes followed a similar pattern, with residents being served at their tables, and the activities that took place in the afternoons were a highlight for residents, giving them something to do.

The activities co-ordinator organised and facilitated a range of physical and social activities on a daily basis that mainly took place during the week. Activities took place mainly in the post-lunch period from 14.00 to 15.00, although occasionally there were late-morning activities. Although the activities co-ordinator worked from 10.00 to 15.00, most of her time in the morning was occupied with providing care support to residents, such as serving lunchtime meals, and comparatively little time was spent on activities.

Most activities were social rather than physical, and were usually orientated around music. A male guitarist and singer visited once or twice a week during the post-lunch period, performing songs from the 1940s and 1950s. This person also delivered painting, as well as arts and craft sessions on a fortnightly basis. Another male guitarist and singer also performed once per month. During Saturday afternoons, female singers as well as a male saxophonist also entertained residents. Musical activities also stimulated some physical movement through foot-tapping and hand-clapping.

Other social activities included bingo sessions, quizzes on topical issues, dominoes, board games and film showings on Sunday afternoons. Occasional visits to the CH included a children’s Christmas choir, and an organisation facilitating engagement with small farm animals. Pampering sessions involving hand massage and foot spa took place on a monthly basis. Various outdoor trips to local places of interest were also planned, but most were abandoned because of inclement weather.

A reminiscence therapist also visited the CH twice weekly. This support involved one-to-one therapy in residents’ rooms and so was not subject to observation. Similarly, a complementary therapist visited twice weekly and offered reiki, head massage and reflexology, but these sessions also largely took place in residents’ rooms.

Physical activities were supported by the activities co-ordinator as well as a physiotherapist. Chair exercises were organised on a fortnightly basis for 1 hour after lunch, and involved mainly stretching movements and playing with balloons, all of which was demonstrated and participation encouraged by the activities co-ordinator and physiotherapist. Occasionally, specific PA sessions were organised, such as hoopla, bread-making and parachute games. On a monthly basis, Music for Health sessions took place. These involved residents listening to pieces of music and performing directed physical actions, as well as music-based quiz questions. They were immensely popular, as were all music-related events, with residents enjoying the singing and social interaction.

At the conclusion of the set activity period, there was often a return to boredom and physical inertia, which were regular features of the CH routine, with residents in communal spaces spending long periods of the day doing very little at all. During the lead-up to teatime, the prevailing boredom combined with residents having to share communal spaces with others who were quite confined, and this created ongoing antagonism among some residents, which sometimes caused heated disputes that seemed partially a consequence of boredom and spatial confinement. Residents generally returned to their rooms by 19.00.

Changes at Rowntree Nursing Home over the course of the fieldwork

The process of conducting interviews with residents, relatives and staff involved spending further time at Rowntree Nursing Home, which allowed for a general sense of change as the observations were concluded. The rapid physical and mental decline of some of the residents was particularly striking, with some residents who had been in relatively robust health becoming very poorly. Some decorative changes were also evident in the ground-floor lounge, which was redecorated in brighter colours with new wooden flooring fitted together, with new doors leading to the garden area at the front of the building. A flat-screen television set had also been fitted, and these changes created the visual impression of being in a large space.

Setting

Owned by a small local company, Eden Park is part of a larger complex in the north of England that includes a nursing home and a development of houses and apartments for those aged > 55 years, all of whom have available to them a range of leisure facilities.

The home’s core objectives are to provide accommodation, meals, personal care and medications for older people who are no longer able to manage in their own homes. Occasionally, the home offers respite care whenever a bed is available. The home has two twin rooms and the rest are single rooms: all but one of the rooms have en suite facilities. Over the course of the fieldwork, a lounge was converted into two en suite rooms.

Organisation, management and delivery of care

Eden Park has 35 staff members. At the top of the ladder, the senior manager reports to the owner of the CH, and the assistant manager reports to the manager. During this study, Eden Park’s manager (Sandra) was rarely on the premises. The assistant manager (Chloe) took care of most managerial responsibilities and also sometimes worked on the floor as a senior-in-charge. The senior-in-charge, senior care assistants and kitchen and domestic staff all reported to the assistant manager, and the care assistants reported to the senior care assistants (Table 8 presents further information about the staff profile).

Interviews with residents and senior staff revealed that there was a high turnover among the care assistants, as opposed to the senior staff who had all worked at Eden Park for > 3 years. Because different ranks wore different uniforms, the organisational hierarchy was always clear.

The senior-in-charge was required to have National Vocational Qualification (NVQ) Level 3 certification, plus additional training in the administration of medications. Their role was to deputise for the assistant manager; allocate staff duties; oversee medication delivery; and handle communications with GPs, hospital admissions and ambulances. The senior care assistant was required to have NVQ Level 2 certification. Senior care assistants allocated the care assistants’ duties. Care assistants, who took care of residents’ essential daily needs (bathing, toileting, dressing, wheeling, etc.) were mostly trained in-house by the CH, on which they received NVQ Level 2 qualifications. Kitchen staff prepared residents’ food, keeping in mind their various dietary restrictions and needs. During main meals, kitchen staff also frequently helped to serve the food. Over the course of the study, a few residents were visited privately by different therapists. These visits were arranged by both staff and relatives.

Each day was divided into three shifts: morning (07.00–14.30), afternoon (14.30–21.00) night (21.00–07.00). A total of eight staff members were on hand during morning and afternoon shifts (one senior-in-charge, two senior care assistants, four care assistants and the assistant manager, who was on site and stepped in to help with delivering care if required). Care staff spent a lot of time attending to the needs of those residents who stayed in their bedrooms; therefore, staff were absent from the communal areas for long periods of time during the day. Five staff members were on duty during the night shift (one senior-in-charge, one senior care assistant and three care assistants). Domestic staff worked from 07.00 to 13.00, and kitchen staff worked from 07.00 to 16.00. After the end of the shift, their workload was shared among care staff, who often complained about the effect this additional work had on their caring roles. In addition to taking care of residents, the night shift was required to do all the domestic work, which increased their workload.

The care assistants at Eden Park adopted a task-orientated approach to care delivery, meaning that the focus was on having the caring job done within a specific time frame. Amplified by the high turnover among this group, there was minimal interaction with residents, which may have been due to inadequate knowledge about various residents. The senior staff, all of whom had worked in the CH for several years, possessed greater knowledge of the residents, and sometimes took a more resident-focused approach to delivering care. The assistant manager talked about the importance of staff ensuring that all residents got out of bed and had at least three meals per day. Senior staff also noted the importance of informing residents about the day, date, menus and activities of the day during breakfast. Overall, at Eden Park, meeting the essential care needs of residents, reinforcing routines and ensuring that residents’ safety was maintained was crucial.

Physical environment

Eden Park is housed in a two-story Victorian house that retains many original features, plus two car parks and two beautiful and spacious gardens outside.

Eden Park has eight public spaces that are located both on the upper and lower level of the building. Of the eight public spaces, six were routinely used by specific residents throughout the time of the study.

On the lower level are a staff office, two dining rooms, a kitchen, two lounges, a conservatory with a patio door (opening out onto the garden) and two communal toilets. The lower-level small office is used by all the staff for administrative purposes and to store personal items. The first dining room is used by residents who are physically frail and require help with feeding. It has two large windows, a good view of the garden and sufficient natural light throughout the day. Its furnishings include a stereo for playing music during mealtimes, two cabinets for storing residents’ medicines, three tables and a few dining chairs. Space is left around tables for residents in wheelchairs. This dining room has an exit that opens into the corridor leading to the lift, which is used to take residents to bedrooms on the upper and lower levels. A second dining room is used by residents who do not require help with feeding. With four tables, 16 chairs and a wall unit, it is a bit crowded. Because it faces north and does not get enough natural light, lights are switched on during daily meals. Both dining rooms have wooden floors and walls painted in different pastel colours.

The first lounge on the lower level seats four and has seven doors that exit to the upper-level stairwell, the toilet, the kitchen, the dining rooms, the second lounge and the main entrance. On the wall, there is the menu of the day and a notice board for planned activities. Given its central location, staff, residents, visitors, trolleys and deliveries compete in this space.

The second lounge on the lower level, which seats five, is furnished with a television, a clock, a coffee table and five heavy armchairs arranged to face the television. All residents who use these public spaces prefer to sit in this lounge, and there is considerable competition for the five chairs. This lounge opens onto a small conservatory and also has an exit to the hall with residents’ bedrooms.

The conservatory has small windows and is furnished with four heavy, closely arranged armchairs. The proximity of the chairs allows residents to form new friendships and to converse with friends, but can also lead to tensions between residents.

Reached by a lift or a steep flight of stairs, the upper level contains three communal lounges, a toilet and a manager’s office. Each communal lounge is furnished with a television, a clock, a coffee table and heavy-to-move armchairs. One lounge and one dining room (devoted to the physically frail and least able residents) also contain a stereo system for playing music. Staff post notices of the daily menu and planned activities and events on a centrally located notice board.

The upper level contains three communal lounges, a toilet, a lift and a manager’s office. The first communal lounge on the upper level is devoted to physically frail residents. It is furnished with five armchairs arranged with enough space for a hoist machine to transfer residents from their wheelchairs to the armchairs. This lounge also has a music stereo, a television and an (unused) residents’ computer and desk.

The second upstairs lounge, which seats eight people, is used for organised activities. It contains eight chairs and a television; unless there are activities, it is empty most of the time.

The third upstairs lounge contains two armchairs, two cabinets and a table, and is used by staff. All three lounges exit into the hall with residents’ bedrooms. A manager’s office is on the upper level, used mainly for administrative purposes. All lounges have attractive blue and cream carpeting that is well maintained.

Residents’ private rooms are divided into five zones which are distributed on both floors. Doors to lounges, painted brown, are heavy enough to be difficult for some residents to open. Doors to toilets are painted the same colour as the adjoining walls, making them difficult to see.

At Eden Park, rather unusually, residents and staff frequently use the home’s public spaces together. Both staff and residents with mobility roamed freely through offices, lounges, dining rooms, the garden and the stairwells. During staff’s short breaks, for instance, a few staff members generally congregated together in communal areas rooms, where they were sometimes joined by residents.

The way members of staff used the CH space was, in part, shaped by their role. The seniors-in-charge were frequently in public spaces giving medications, answering telephone calls and meeting with relatives and visitors who had enquiries. The senior care assistants and care assistants who helped residents with ADL spent most of their time in residents’ private rooms and were most evident in the CH’s public spaces during routine hours, such as mealtimes. Ten to 15 residents described by the manager as ‘poorly’ remained in their rooms and were dependent on the care assistants to help with ADL.

Individual environment

Based on the observations of the researcher, the residents of Eden Park varied in terms of their cognitive ability: some residents were very cognitively able, whereas others were in the later stages of dementia. In terms of mobility, there was a broad range: some residents were independently mobile, whereas others required a hoist. Most residents, however, experienced problems with their short-term memory and many were also physically weak. Residents frequently came to Eden Park following a crisis, such as a fall or bereavement. Many of the residents at Eden Park had either lived in the retirement flats owned by the owner of the CH or came from the neighbourhood, with the exception of approximately five residents who had come from a different town. Senior staff told the researcher that many residents come to Eden Park when they can no longer manage at home and their intent is to stay there until they die. According to the assistant manager, unusually, the proportion of privately funded residents at Eden Park CH is larger than the proportion funded by the local authority. Over the course of the fieldwork, a number of residents suffered an abrupt decline in their physical abilities and there were several deaths.

The pen portraits in Box 2 summarise key characteristics for three residents living at Eden Park CH. The first pen portrait describes a resident who was relatively able and active, the second is someone who is fairly average in terms of physical and cognitive ability, and the third is someone who is physically and cognitively frailer than average.

Daily routine

Residents’ daily routines revolved around the three main meals of the day: breakfast (07.30–09.30), lunch (11.30–13.00) and dinner (16.00–17.30). Because breakfast extends over 2 hours, there are periods when the dining room is empty and periods when it is full. Observations and interviews with staff showed that many residents prefer to have their breakfast in their private rooms. Those who choose to have breakfast in the public spaces are not obliged to use the dining room, and so most residents prefer to have breakfast in the public lounges as they watch television.

Observation data, further supported by staff interviews, showed that, after main meals, some residents are taken to specific lounges and sit on the same seat daily, a social seating pattern that is reinforced by the CH staff:

Hmm, if we’re trying to place them always on the same chair because it’s easier for us as well and it’s, I mean for them it’s easier because they remember where they’re sitting and they know, we always, for us it’s like we’ll go to the dining room and we’ll look so we know who’s missing, if they’ll be sitting at every day different seat we’ll be twice longer for us to count them, who’s there and who’s not, yeah.

Laurie, senior care

According to the assistant manager, this routine seating patterns suit most residents fairly well and facilitates staff efforts to meet residents’ many and varied needs:

It is knowing your residents . . . Linda doesn’t like lots of noise, lots of business, she likes to be just in the quiet. The same to Alyssa and so they have a separate lounge. Constance, Millicent and Emma read a lot and that is why you will find them in the conservatory.

Chloe, assistant manager

During mid-morning, residents are served tea, coffee, juices, water and biscuits. Staff, who are extremely busy at this time of day, do not assist physically frail residents with eating as they do at main meals. As a result, many residents either spill their drinks or end up not drinking them at all.

On Tuesdays and Fridays, staff reminded some selected residents to attend the Music for Health class that runs from 10.30–11.30. Staff also encouraged female residents to take advantage of the hairdresser who visits Eden Park on Wednesdays from 09.00 to 16.00. Most residents made use of this facility and seemed to enjoy it. On the last Friday of the month, there was a music class in the afternoon, which many residents delightfully attended.

During lunchtime, care assistants would begin to seat those who required help with eating at 11.30 in a specific dining room. At around 11.45, staff walked through the lounges fetching, supporting and encouraging other residents to come in to lunch. During this period, some residents were assisted to the toilet and into the dining rooms. A few independently mobile residents walked to the dining room just before the food was served. Just as in the lounges, the seating patterns in the dining rooms were reinforced by the care staff.

After lunch, residents spend their time in the lounges, where they rest, watch television or read newspapers. Some of those who were able to walk independently tended to wander in and out of the lounges. During warm weather, a few able residents arranged trips to local shops, theatres and museums. On warm days, two residents, Ashby and Eda, spent much of the day outside gardening. Care staff sometimes brought other residents to sit outdoors.

Afternoon tea was served at 14.45. Beginning at around 15.30, staff followed the same routine as at lunch. In the period following dinner, when most residents had returned (or were returned) to their private rooms, there were long periods of time when staff members were absent from the lounge spaces. This often meant that the few residents who remained in the lower-level lounges were left entirely on their own, sometimes for > 1 hour. During the evening was often when tensions arose between residents.

Changes at Eden Park Care Home over the course of the fieldwork

Several changes have been noted at Eden Park CH over the course of the fieldwork, including: residents’ changing use of outdoor space, care staff making an increased effort to involve those residents who were very physically and cognitively frail in activities, the dramatically declining health of several residents and the increasing dissatisfaction of some residents with the service offered by Eden Park CH.

Fieldwork commenced in May 2013; during this time, many residents spent ample time in the garden area. During the cooler weather, in October and November, residents spent increasing amounts of time in their bedrooms. Even those who had spent time in the lounges typically returned to their own rooms as soon as it got dark. With the garden no longer appealing and most lounges empty for some of the time, Eden Park took on a lonely and forlorn air.

When fieldwork began, only residents who were regulars in the communal lounges took part in music classes or other organised activities. The number of residents participating in the PA class slowly declined over the summer. In winter, just before the study was concluded, staff brought physically frail residents, many of whom had remained in their rooms, to activity classes, not to participate, but rather to listen to music and meet others. Senior staff reported that they felt that it was important for residents to come out of their rooms whenever possible to meet others. Staff were also observed to encourage residents sitting in the lounges to take part, and Irene (senior care assistant) said that, after attending training on how to help people with dementia, she now encourages other staff members to help residents participate in activities. This is a change from the earlier institutional policy of leaving such decisions to residents as a matter of free choice.

Over summer and autumn, seven Eden Park residents died. In all cases, these deaths were unpreventable and sudden. Staff report that, with older residents, these sudden fatal events are not uncommon. In addition to deaths, a number of residents suffered an abrupt decline in their physical abilities, and even a minor illness was seen to affect even the most active residents adversely. When 94-year-old Beatrix came to Eden Park in July 2013, for instance, she was active and well connected with the CH community. Her days were filled with things to do and people to meet. By the end of the study, Beatrix suffered continual bouts of diarrhoea, could no longer walk and was virtually bedridden.

Over the course of the study, the researcher observed a shift in some residents’ feelings about Eden Park CH. By the end of the study, a group of long-term residents who are physically mobile and who had good cognitive abilities complained to the researcher about boredom and the lack of an activities co-ordinator, the need for a new manager who could be present throughout the week and the lack of staff response to buzzers.

Towards the completion of this study, residents also noted that the way junior staff interacted with them had changed:

The staff changed a lot, honestly. They are not friendly as they used to be. The good ones [mostly care assistants] have left.

Ashby, CH resident

These changes were noticed by other residents, who felt that it was due to high turnover among this group:

They don’t stay [care assistants]. Only bad ones are still here . . . I don’t know their names as they are always hurrying and when you know them, they are gone . . . so we have to put up with unfriendly new faces everyday . . . you don’t know who is coming to your room next.

Myra, CH resident

Most emphatically of all, these residents complained about their desire to be allowed out without being accompanied by care staff. As one disgruntled interviewee commented:

It is crazy, absolutely crazy. The law states you’ve got to have three members of staff on here. Weekend is worse, it [Eden Park] has gone downhill a lot. I don’t think Beatrix, Eda, Evie have a good word either. This place wants a very good old shake-up, I tell you.

Ashby, CH resident

These residents had spoken well of Eden Park at the beginning of this study.

In addition, there were four residents with good mobility and short-term memory loss who joined the CH when this study had already began. These residents complained that management had misled them into coming to the home by promising activities that are not in fact offered in the recruitment package.

Setting

Bourneville CH is a three-storey residential home that is owned, run and managed by a large international corporate provider of health and social care to older adults in the UK. It is home to roughly 25 residents, virtually all of whom are female. It is located in a town in a semirural district, with ready access to moorland, on the outskirts of a large city in the north of England. It has ready access to shops and local amenities. It offers a combination of mainly permanent care, as well as more limited respite and rehabilitative care.

Organisation, management and delivery of care

The CH has a defined managerial hierarchy, with a general manager overseeing a deputy manager and a senior care assistant, who were responsible for medicine management. These senior staff members dictate the daily care procedures to more junior care assistants, although care is, in some ways, governed by residents’ requests for assistance or as a consequence of crises, such as falls (Table 9).

Care staff worked varying shift patterns according to the number of hours they worked each week.

Shift patterns meant that there were four care assistants in the morning, three in the afternoon and three on the night shift. There is always a senior care assistant on duty. Some staff work 6-hour shifts (i.e. 08.00–14.00 or 14.00–20.00), but others work 12-hour shifts (i.e. 08.00–20.00 or 20.00–08.00). The breakdown of staff that cover these shift patterns is given in Table 10.

In addition, district nurses and physiotherapists provided additional health care, based on identified residents’ needs.

An activity co-ordinator was in post who was much less experienced than the one employed at Rowntree CH, and did not offer the same variety of activities for residents.

Most hands-on care is undertaken by care assistants, comprising mainly women aged over 40 years, with one or two younger women aged in their 20s. In addition, there is a chef, a gardener, an office-based administrative worker and a part-time activities co-ordinator, who was responsible for delivering activities for residents. Some musical entertainers came in to perform for residents, but not as frequently as at Rowntree Nursing Home. As a residential CH, there were no nursing staff on site, although there were regular visits by district nurses as well as by GPs.

Both senior care assistants were highly experienced, although the CH manager had been appointed more recently. The care assistants presented a more mixed group; some of them had worked at Bourneville CH for a number of years, or else seemed very experienced.

Care assistants attend to a variety of daily tasks: responding to the sound of buzzers in residents’ rooms; helping them with toileting; feeding, when necessary; and maintaining daily records in the personalised care plan documents, which was a source of irritation for them, as it was seen to represent a laborious process. Tensions between the general manager and junior staff were evident through the conduct of team meetings and staff demeanour in the lounges. There was a general sense of low morale among staff, who seemed harassed for the most part and overwhelmed by their workload.

Although there is relative stability within the staff group, there was a sense of disquiet between the manager and care staff. During the period of observation, one member of care staff had left. After the study, it came to light that the CH manager had departed the CH, and had apparently taken up a managerial post at another site in the same organisation.

The ethos of care provided at Bourneville CH might best be described as managerial, in that the primary focus was on the performance of set tasks, such as undertaking paperwork or assisting residents to and from the dining area, within a highly restricted time frame. Hence, care staff would scuttle briskly from resident to resident in the lounge space, and back and forth to residents’ rooms, constantly mindful of the urgent needs of other residents. This produced a sense of anxiety, such that social interactions with residents became characterised by brief, rushed verbal exchanges centred on the task at hand, rather than meaningful engagement. The quality of these exchanges were limited in scope and depth, and did not support the development of relationship-based interactions that were representative of a more holistic ethos, as observed in Rowntree Nursing Home.

Relatives and friends played a fairly neutral role at Bourneville, spending time with the residents, but there was not the same sense of conviviality with other residents or with members of staff. Relatives spent their time mainly with their loved ones, and not really interacting with others in communal spaces.

Physical environment

Bourneville CH is a three-storey building set in its own grounds that has been converted from being a private home. The internal decor of Bourneville CH seemed quite old-fashioned, with the hallway area marked by dark wooden panelling, which gave it a gloomy appearance, as no natural light could penetrate this area of the building.

The top floor contains office space and a small care workers’ rest room. The ground and first floors house residents’ rooms interspersed with some communal bathrooms, although most rooms are en suite. The communal spaces are all on the ground floor, where two lounges are provided.

The front lounge, which more residents tend to use, could be entered from the main doorway. On the left side of this lounge were a number of chairs positioned along a wall facing inward towards a television set. On the right side were two small dining tables that could each seat about six residents. A doorway linked to a second rear lounge. On the right side of this rear lounge was a large dining table that could seat about a dozen residents. This table also acted as a resource for care staff to complete resident care-related documentation. Seating in this lounge was arranged along the same plane as the main lounge. During the day, residents were generally evenly spaced between both of the ground floor lounges, although some residents very rarely, if ever, frequented communal areas. Staff seemed to oscillate fairly evenly across both areas, as well as the dining areas, although they would often leave these areas to attend to residents who had requested assistance by pressing their buzzers.

The CH is surrounded by copious garden space, with raised beds for planting, ample seating, tables and secure pathways all around. Yet residents were generally reluctant to venture outdoors, although a few were more than happy to do so, and to stroll around the outdoor space. The exit from the rear lounge to the garden had a sloping ramp that was a barrier to those using a walking frame, as the width of the ramp was narrower than the walking frame.

Over the course of the fieldwork, significant structural changes in the front lounge involved the creation of a care workers’ office for general administration. This meant that the dining space in the front lounge, incorporating some small tables, was removed so that all residents were now dining in the rear lounge. This seemed to allow for greater social interaction among residents through having greater numbers of residents in a small space.

Individual environment

Seventeen residents were observed. Their ages ranged from roughly 85 to 100 years. Generally, physical mobility was poor, with two residents requiring a hoist to transfer them from armchair to wheelchair, three wheelchair users, six requiring a walking frame and five requiring either a walking stick or a walking frame. One resident could walk without assistance. There was a strong sense that most of the residents had spent a number of years living in the local environment prior to moving to the CH, based on the kinds of conversations that ensued between residents. The residents’ lengths of stay varied from a few months to several years. The CH manager was unable to provide the research team with information regarding the number of residents who were funded by the local authority or self-funded, owing to matters of confidentiality.

There are a total of 26 beds at Bourneville CH, with one of the residents currently in hospital. Based on the researcher’s observations, four residents had full capacity, eight had mild cognitive impairment, two had greater impairment and three had severe impairment. Some of those with a severe impairment were unable to verbally communicate and required assistance with eating. The majority, however, could maintain a simple conversation.

Over the course of the study, the rapid physical and mental decline of some of the residents was particularly striking, with some residents who had been in relatively robust health becoming very poorly, or dying.

The pen portraits in Box 3 summarise key characteristics for three residents living at Bourneville CH. The first pen portrait presented describes a resident who was relatively able and active, the second is someone who is fairly average in terms of physical and cognitive ability, and the third is someone who is physically and cognitively frailer than average.

Daily routine

The routines of daily living are centred on delivering medicines and organising mealtimes. The morning routine involves care workers spending time in residents’ rooms preparing them for breakfast. Residents gather at the dining tables from the lounge spaces, if they have been awake for a while, or use the lift to reach the dining room from the second floor. The general sense is that this can take some time, depending on residents’ health and well-being, given that some residents have their breakfast when most of the others have finished. Medicines management is carried out by the senior care assistants, who are responsible for handing out medicines and maintaining paper records. This process is carried out in the mornings, afternoons and evenings.

During the morning period between breakfast and lunchtime (about 12.30), and then between lunchtime and teatime (about 17.00), the residents spend most of their time sitting idle and bored, and some of them are frustrated by the absence of things to do. Despite there being an activities co-ordinator, and a weekly programme of activities that are supposed to occur, in practice, very few formal, organised activities were provided, other than a single singing session and two quiz sessions. One outdoor canal trip has been arranged. The programme of weekly activities includes two armchair exercises, a quiz, bingo, a craft session and board games. Hence, there is a complete disconnect between what is supposed to occur and what actually happens.

Informal activities have involved some residents going for short walks outdoors during fine weather, or sitting outside on benches, but these are infrequent. Sometimes the activity co-ordinator engages with some residents on a one-to-one basis in their rooms, but the nature of this engagement is not known. The mood of lethargy culminates in residents snoozing for most of the day.

Changes at Bourneville Care Home over the course of the fieldwork

Over the course of the fieldwork, there was an increase in PA through the delivery of armchair exercises three times per week; the participation of some residents, who seemed uninterested in PA during the period of observation, was notable.

Setting

Hebble House is a local authority-run CH that specialises in caring for older adults with a diagnosis of dementia. It is situated in a large, post-war social housing estate on the outskirts of a large city in Yorkshire. The area surrounding the CH is one of the most deprived in the region, and, according to information obtained from the local council, has a poor reputation, with poor housing stock, low levels of economic activity and high rates of crime. Talking to people who live locally, however, there is a strong sense of community; this is echoed by the presence of the community garden that runs down one side of the CH. The CH is approximately 5 minutes’ walking distance from local amenities, including a chemist, local shops, a church and transport links to the city centre. Other amenities, such as pubs, cafes, restaurants and supermarkets are a short (10- to 15-minute) bus ride away in the city centre.

The home is registered to provide residential care for 35 adults, aged ≥ 65 years, and also acts as a day centre for those living in the local community. In discussions with staff members, some referred to the CH as an elderly mentally infirm (EMI) unit. Although an outdated term, EMI refers to the fact that people throughout all stages of dementia (including those in the later stages of dementia) and those who experience other mental health problems associated with their diagnosis are cared for in this unit. The CQC carried out a routine inspection of the CH during the period of fieldwork, and produced a positive report, with the CH meeting all of the standards against which it was assessed.

At the beginning of the fieldwork, Hebble House accepted both permanent and short-stay residents, with a specific wing of the CH allocated for those who were short-stay residents. Over the course of the fieldwork, however, the number of short-stay residents placed at Hebble House to undergo assessment increased significantly, as they stopped accepting new permanent residents. This resulted in short-stay residents residing in all four wings of the CH.

Organisation, management and delivery of care

The staff body consists of 65 members and is well established, with a low turnover. The CH manager, Laura, has worked in management roles for the previous 12 years. She moved to work at Hebble House 7 years ago, and has been the CH manager for 5 years. She is trained to NVQ level four in dementia studies and dementia care. In conversations with Laura, she communicates a desire to create an ethos of enabling residents at Hebble House, but acknowledges the challenges of achieving this, which included the busyness of staff and their fear of putting residents at risk.

The CH manager worked weekdays and had overall responsibility for running the CH; she had to ensure that care was delivered to a high standard, that care plans were accurate and implemented, that issues with her staff were quickly and appropriately resolved, and that any bad practice was identified and addressed. She conducted staff supervisions and appraisals and oversaw the recruitment of new staff members. She also dealt with complex cases. Laura was also involved with overseeing ‘disciplinaries’ at other local authority-run CHs. She reported to the divisional manager.

Laura was initially supported by a full-time deputy manager, a full-time assistant unit manager and an administrator. During the day (including at weekends), the deputy manager or the assistant unit manager was usually present. Although the staff body was fairly static, a significant loss was the deputy manager, who left her post mid-way through the fieldwork. This post remains vacant.

The assistant unit manager, Karen, assisted the CH manager with staff supervisions and appraisals, oversaw the staff rota and booked in the medications (with the support of the administrator). She also helped with the running of the staff office, which involved liaising with external health and social care professionals and responding to any issues that arose during the day. A significant amount of her time, however, was spent on carrying out residents’ assessments (short stay):

So I just prioritise what needs doing. So if I need to do me flexi-bed assessments, depending on how many I’ve got in, you see, I could have six to do in a week, because if all six have come in at once (. . .). So it’s very varied, like I say, very different, and I prioritise through, like I say, the flexi-beds and things like that because I have to be on a timeline with them, because the social work, they only have 4 weeks in a flexi-bed (. . .).

What about when you’re working on the office, what kind of, is your priorities then, is that . . .?

It’s all about the service users. Making sure the environment’s clean, tidy, making sure that everybody’s doing what they should be doing, making sure the doctors are called if they’re needed, any of our service users as are end-of-life care, ‘cause we’ve got a couple, I like to be involved in that, in their care, at least once a day, I like to be able to go in and assist with changing or checking pressure points and things like that, making sure that the skin’s not breaking down, and just making sure generally that they’re OK (. . .). So, I prioritise, my first thing when I come on duty, I do the handover, and then anybody that’s an end-of-life care, I like to make sure I go and see ‘em (. . .) before the office gets too busy, because that office gets absolutely manic.

The home employed four, and then later five, senior carers and 23 care assistants (who do the majority of the face-to-face care). The hours worked by members of the care team ranged from 37 hours per week (full time) to 22 hours per week. In addition to their NVQ care qualifications, the majority of care staff have undergone dementia training. The senior carers and care assistants work in shift patterns. There are three shifts to cover the care throughout each 24-hour period, and each shift overlaps by at least 15 minutes to allow for transfer of information. The morning shift runs from 07.15 to 14:45, the afternoon shift runs from 14:15 to 21:45 and the night shift runs from 21:30 to 07:30. The morning and afternoon shifts have a senior carer and four care assistants (one care assistant will be allocated to each of the four wings of the CH; the remaining two will ‘float’ between wings to assist with jobs that need more than one person, e.g. using the hoist). The night shift has one senior carer and two care assistants.

Senior carers were responsible for the running of their allocated shift and overseeing the work of the care assistants. This comprised tasks such as allocating the care assistants to particular areas of the CH, conducting the handover, arranging visits from health-care professionals, distributing medications (although some care assistants had also been trained to do this), keeping the diary and communication book up to date, helping out the care assistants and working with those residents who were distressed:

Could you take me through a typical day, what a typical shift would look like for you, from when you get in to when you finish?

Well, me, I’m always in half an hour before me shift. Normally, we do the allocation where people are going, where the staff are going on a morning, get the handover from night staff or if we’re in on lates, then it’s often the morning staff. If we’re on a mornings, we get the handover then we give the handover to the staff; they need to get a full handover before they go onto floor. They go onto the floor, we start looking in the diary, communication book to see what needs doing for that day, so 8 o’clock we start getting in touch with the doctors or the district nurses or community matron, social workers if we need to and deal with the stuff that we need to deal with. Then we go into medication, which that can take you quite a while and after you’ve done the medication, normally, it’s just seeing the running of the floor or whichever floor or the building if it comes to where you’re working on your own and just to make sure that everybody’s doing what they’re supposed to be doing. We have a lot of service users that are demanding, I won’t say demanding, that they need more your help than anybody else (. . .) on an afternoon, washing up, setting the tables, you know, seeing to laundry, so if you can take a little bit of that off ‘em [the care assistants], you know, then the shift runs a lot smoothly.

Care assistants were responsible for face-to-face care: their role comprised bath and body work (such as washing, dressing and toileting residents), serving meals and assisting residents to eat and drink, ensuring residents were well hydrated and filling out the necessary paperwork. They tended to prioritise residents’ health needs and were also attentive to whether or not residents were distressed. If they had time, care assistants would sit and talk with residents, sometimes engaging them in activities. Care assistants would often sit and talk with residents while completing their paperwork:

I’ll start at half past 7, I’ll come up [upstairs], try and give personal care and then I have to break off at 9 o’clock from personal care in order to start dispensing the medications so they’re getting it at the allotted times. So I have to do medication and then, once that’s done, it’s go back to my service users and then it’s any toileting or if there’s anything that they need to try and get that in; then I break off at 11 o’clock again to get the files out to write down what’s happened in the morning, and at that time it’s toileting, you know, taking them, assisting with toileting and then back into the dining room for lunch. Once they’ve had their lunch, you start your transfers and then I’m broken off again ‘cause I have to go back again and start the paperwork to write down what diet input they’ve had and how they’ve been, so it’s constantly breaking away from your service users to make sure the paperwork is done.

Maya, care assistant, Hebble House

There was a sense of collaborative working at the CH and staff members commented that they felt supported by others, including those in more senior and managerial positions. One complaint noted by the researcher over the period of fieldwork related to the increased workload created by the intake of more short-stay residents, who were there to be assessed; therefore, their care needs were unknown. Depending on the particular resident, such placements could result in care assistants spending a considerable amount of time helping the resident to settle in, and learning about the new resident’s needs, behaviours and preferences. This often meant that care assistants had less time for permanent residents. Care staff also sometimes mentioned that it would be beneficial to have more care staff on duty, but also acknowledged that this was always going to be the case in such settings.

Domestics, a chef, kitchen assistants and a handyman were also employed at Hebble House. Over the course of the research, the activity co-ordinator post was vacant (due to maternity leave). However, towards the end of the fieldwork, they had recruited a new activities co-ordinator who was due to start in the new year. There was also an exercise professional who visited the home once a fortnight to lead a seated exercise session. External people were also brought in from time to time to provide activities for residents, including ‘pat the dog’.

In addition to the paid staff, there was a volunteer who came to the home twice per week to provide one-to-one support to a particular resident. Over the course of the fieldwork, the activity co-ordinator, who was on maternity leave, arranged for a group of volunteers from a local business to come to the CH and decorate a sensory room and tidy one of the gardens. There was also a small, but active, group of relatives who organised special events, including a Halloween party and fundraising events, and tried to arrange trips out for the residents. Over the course of the fieldwork, this group of relatives were successful at arranging a range of events. However, they struggled to arrange a trip out for the residents because of a lack of volunteers and transport.

The staff at Hebble House had access to a wide range of external professionals from whom they regularly sought input and advice. Health-care professionals were called on regularly, including doctors, the district nurse and the community matron. They also liaised with social workers. In addition, the staff at Hebble House worked closely with the community psychiatric nurses and, if a particular resident was causing concern in terms of their behaviour, they worked together with this team and with the resident’s family members to produce a plan of how they could best manage the situation. External experts were brought in to assess the mobility of residents if staff members noted a change in residents’ mobility.

Physical environment

Hebble House is a fairly modern, T-shaped building with two floors. Inside, the CH is divided into four wings (two downstairs and two upstairs), as well as a large, open-plan day centre space on the ground floor. Each of the four wings comprises a lounge area, a kitchen/dining room area, bathrooms and toilets, and the residents’ bedrooms. Residents’ bedrooms were small, with just enough room for a single bed, a wardrobe, a chair and a few personal belongings. Residents rarely spent time in their rooms, unless they were unwell or receiving palliative care. Residents tended to spend time in the lounge on their wing, moving a short distance to the kitchen/dining area for meals.

Each floor of the CH also had a brightly coloured ‘sensory room’, which was rarely used by residents or staff, and an ‘activity room’ which, for the duration of the fieldwork, was used as storage space. The upper floor also had a small additional lounge, which was sometimes used by residents. Care assistants would also very occasionally sit Grace (see Box 4) in this room with another resident or, if she was on her own, they would sit with her. Grace was unable to speak, but she vocalised, sometimes loudly, which resulted in other residents shouting at her. The ground floor also had a ‘smoking room’, a room used by the hairdresser, the laundry room and the staff locker rooms. The seated exercise class and other formally organised activities took place in the day centre area.

Although each of the four wings of the CH were, in fact, long corridors, attempts had been made to make them more accessible for residents. For instance, the toilet doors were painted red and images were hung on all the doors to communicate the purpose of the room, such as a picture of a plate, knife and fork on the doors to the kitchen/dining rooms. Personal photographs had been placed on residents’ bedroom doors, to help them identify their room. There was also a ‘pit-stop’ chair on the lower level, so that residents could pause and have a rest if they tired when walking along the corridor. This chair was located outside the main staff office, and so Joy (see Box 4), who liked to be kept informed of goings-on, used to spend a lot of time sat in the pit-stop chair.

Despite attempts to make Hebble House a more accessible and homely environment for residents, some residents were unable to orientate themselves in the setting. Those residents with moderate and later-stage dementia (the majority) became disorientated if they left the lounge areas, and many were unable to locate the toilet or their bedrooms. Many residents were unaware that they were living in the CH. Staff frequently responded to a wandering resident by asking them where they were going and, if the resident replied that they did not know, they were routinely encouraged to return to their chair in the lounge. Some residents, who were physically able and happy to venture beyond the lounge area, were allowed to wander the corridors.

The floors were linked by two staircases, both of which were out of bounds for residents, and a lift. Although residents were allowed to use the lift, there was only one resident, James, who regularly did so to access the smoking room on the lower level. Mabel occasionally used the lift if she wished to speak with a senior member of staff, which usually occurred when she had become distressed or disgruntled with the care assistants or her fellow residents as a result of a perceived misdemeanour. The vast majority of residents, therefore, rarely ventured beyond the floor where their rooms were located. Residents from the upper floor were taken to the lower floor by care assistants so that they could visit the hairdresser, and, occasionally, so that they could sit in the day centre, participate in the seated exercise class or go for a walk in the garden. Care assistants reported that residents liked routine and the familiar space of their lounge area, and so often disliked it if taken to another area of the CH.

Although there were distinctive areas of the CH that were intended for staff use, boundaries between staff spaces and resident spaces were somewhat blurred. On both floors, there is office space for staff, including the CH manager’s office, assistant unit manager’s and administrator’s office, and general staff office on the ground floor, and a small office space for those working on the top floor. With the exception of the assistant unit manager’s and administrator’s office, the offices all had large glass windows that looked out onto lounge areas. The CH manager, Laura, regularly spent time in all three offices on the lower level, and also with residents in the communal areas of the CH. Over the course of the study, she began inviting a recently arrived resident, Julie (see Box 4), into her office while she was working. She encouraged Julie to occupy herself by organising piles of paper, which Laura felt Julie enjoyed. The door to the general staff office, where the majority of the day-to-day running of the CH took place, was often propped open. Some residents, in particular Joy and Polly, would sit outside this office interacting with staff members. Occasionally, one or two residents would sit in this office with staff members. The office on the upper floor was a thoroughfare between a lounge and kitchen/dining room and the residents’ bedrooms; therefore, residents passed through this office. Care assistants rarely spent much time in the office spaces. If they had time, they would also sit with residents in the lounge areas and engage them in conversation, reminiscence or other ad hoc activities such as singing, throwing a ball or playing a game. They would also complete their paperwork in the lounge areas or kitchen/dining rooms.

The CH is surrounded by three small gardens and a car park. One of the gardens had been developed into a community space, where local residents and school children grow vegetables. There was, however, little interaction observed between those who use the community garden and the CH residents. During the warm weather, some day-centre participants and CH residents sat outside in the gardens. It tended to be the case that those residents who lived on the lower level were those who were able (or were assisted) to access the garden spaces, with the exception of a male resident, James, who, despite living on the upper level, regularly used the lift, unaccompanied, to access the lower level.

The external doors of the CH were locked and residents rarely left the CH setting. Occasionally, one resident, Julie, would go along for a ride in the minibus with the day-centre participants when they were driven home. Residents were occasionally taken for a short walk and a cigarette. Towards the end of the fieldwork, a small group of residents were taken for a pub lunch. This had been requested by a particular resident, Caroline, and facilitated by several committed care assistants and the CH manager, Laura. A small group of relatives tried to organise a day trip for residents; however, this was indefinitely postponed. Relatives and visitors rarely took residents out of the CH. In her interview, the CH manager bemoaned the fact that relatives rarely took residents out:

. . . I feel sad that the families as well think that, once their loved one comes here, that that’s it.

Do you think that’s quite common?

I think that if the families came in and continued doing what they used to do with their loved one in their home environment, I think that person would be maintained at a steadier level for a longer period of time. So if they still continued to take them out to the pub of an evening or if they still continued to say ‘I’m, mum I’m popping to Morrison’s, do you want to come with me?’ But it’s like, as soon as they come through the door, that’s it, the family don’t want, think that’s it, there’s nothing more that they can do. (. . .) But they should just carry on the way that they used to do, if you know what I mean? Because that would help. It’s like none of them go out for Christmas but why wouldn’t you want to take them out and have them over at your house for Christmas? I don’t understand it. That, isn’t that what families do? (. . .) I know some of them, like Edith, couldn’t at the moment but Julie could, Joy could. There’s still quite a few upstairs that could, but it’s just like, that’s it, no they don’t. They don’t want to do it. And, see, that would keep them going, wouldn’t it? Stimulate their mind, you know, it’s a change of scenery going out for the day. It’s sad, I think.

Interviews with residents’ relatives, however, highlighted some of the difficulties of taking residents out of the CH, including that residents sometimes became distressed and disorientated by an unfamiliar environment.

Individual environment

The majority of residents are female and have moderate, or are in the later stages of, dementia. Many therefore have impaired memory. Some also experience visual disturbances, mobility problems, communication impairments, and/or impaired eating and appetite. In terms of mobility and physical abilities, there is great variation among the residents. The majority of residents require some form of support from a staff member (i.e. they usually require a staff member to physically support or encourage and instruct them when moving). Several are independently mobile and several residents require a hoist.

The pen portraits in Box 4 summarise key characteristics for three residents living at Hebble House. The first pen portrait describes a resident who was relatively able and active, the second is someone who is fairly average in terms of physical and cognitive ability and the third is someone who is physically and cognitively frailer than average.

Daily routine

Residents’ typical daily routine was structured around mealtimes. Breakfast was served at 09.00, lunch (the main meal of the day) was eaten at 13.00 and the evening meal between 16.30 and 17.00. During mealtimes, groups of four residents sat around small round tables while care assistants served the meals and assisted some to eat. There was usually some conversation between residents and also between residents and staff during mealtimes. Kitchen/dining room areas were used by residents primarily during mealtimes, although Mavis would sometimes help care assistants wash and put away dishes at the end of the meal.

With the exception of a small number of residents who walked around the CH (i.e. Julie, Evie and Emily) or occupied themselves by tidying the CH (Mavis), residents tended to spend the time between meals sitting quietly, watching television or sleeping in the lounge areas. The arrival of the tea trolley broke up the morning and afternoon. Those residents who smoked also occupied themselves by going outside or to the smoking room for a cigarette.

Generally, there was little conversation between residents. A small number of residents, however, regularly conversed with others. Ann, Joy and Mavis regularly struck up conversations with each other and with other residents, and Caroline and Jane happily engaged in lengthy conversations from time to time, although there was little shared understanding. Some of the more able residents, including Mabel, Ann and Joy, often demonstrated patience and thoughtfulness in their interactions with other residents. Residents sometimes helped others, for instance Mabel regularly fetched Evie’s forgotten walking stick. A few residents vocalised or talked to themselves, which sometimes prompted reactions from others including concern (‘are you OK?’) or disparaging comments (‘shut up’). There were times when tensions were high between residents; staff tried to manage these situations. Care assistants would also strike up conversations with residents when they were in the lounge areas. Residents enjoyed conversations with members of the care staff, which tended to be light-hearted and jovial. Care staff demonstrated good knowledge of individual residents’ histories, likes, dislikes and abilities. Many were extremely enabling in the approach they took to their interactions with residents, often encouraging them to do as much for themselves as they could.

Some residents had regular visits from relatives and friends, which were often welcomed, with some residents visibly delighted by the arrival of their visitors. Visits tended to consist of residents and visitors sitting together and talking in the lounge areas. Some visitors (who are familiar with the home) spent time talking with other residents and staff. For some residents, these visits formed a topic of conversation and they enjoyed telling others about their visitors. For others, however, they were soon forgotten.

Although activity timetables were displayed on the walls of the wings, very few of these formal activities actually took place over the course of the fieldwork. The organisation and delivery of such activities were considered to be the role of the activity co-ordinator, who was on maternity leave. When the care staff had spare time, however, they organised spontaneous activities. One member of the care staff, who used to work in a kitchen teaching adults with learning disabilities, organised baking sessions with small groups of residents during which she encouraged them to do as much as possible.

Changes at Hebble House over the course of the fieldwork

In addition to the changes detailed throughout the previous sections, over the course of the fieldwork, the physical and cognitive decline of several residents, including Ann, Mabel, Grace and Bella, was particularly apparent and difficult to witness. Other residents, such as Joy, James, Jane and Evie, however, experienced a much more gradual decline.

The ethos of care

There were different philosophies of care at the four participating CHs. These philosophies were shaped, in part, by the wider political economy of care, by the policies of the care company and by the managerial team. At Rowntree Nursing Home, Bourneville CH and Hebble House, the CH manager had a very strong influence over the ethos of care and how it was operationalised.

At Rowntree Nursing Home, the philosophy of care centred around the notion of the family and the provision of care was holistic and personally meaningful in its delivery. Rowntree Nursing Home invested significant resources in employing a broad range of professionals to support an array of activities and therapies for residents, thereby providing residents with more opportunities to participate in organised activities than the other CHs.

At Eden Park, the CH manager was largely absent over the course of the study, and the assistant manager was acting up to fill the gap left by the manager. The ethos of care was paternalistic in nature, in that the emphasis was on ‘caring for’ (doing for) residents. Although there were discussions of resident choice, residents were not often enabled to make choices or take action. They were, therefore, often constructed as passive recipients of care.

At Bourneville CH, the philosophy of care was closely aligned with the managerial and audit culture, in that the emphasis was on procedures, documentation and a task-focused approach to care delivery. There were tensions between the manager at Bourneville CH and the care staff, which was, in part, shaped by the manager’s focus on procedures and documentation.

At Hebble House, the ethos of care was constructed around enabling ‘service users’ through the provision of personally meaningful and enabling support. The CH manager, Laura, wished to foster an atmosphere of interdependency in which residents were supported to do certain everyday activities, rather than the care assistants doing such activities to or for them.

The ethos of care at each of the four CHs had implications for the care environment in relation to how care was managed, organised and delivered. Thus, it was one factor that had shaped and helped to (re)produce residents’ pattern of daily life.

Organisation, management and delivery of care

Individual environment

Physical environment

Daily routine: residents’ daily life in the participating care homes

The pattern of residents’ daily life was largely structured around regenerative activities and discretionary/leisure activities.

Physical function assessments

Cognitive assessment

ActiGraph accelerometer

Accelerometers are electronic sensors that measure the acceleration of an object along reference axes. 114 Acceleration is proportional to external force; thus, the accelerometer reflects the intensity and frequency of human movement. 115

The outputs generated from the ActiGraph accelerometer are referred to as ‘activity counts’. Activity counts are proprietary units that reflect raw accelerations that have undergone some data processing and have been summed over a user-defined discrete time period (epoch). 116 The sum of the activity counts in a given epoch is related to activity intensity and may be categorised based on validated activity count cut-off points. 117,118 Using ActiLife software, it is then possible to apply algorithms to classify how long individuals spend sedentary and engaging in PA of different intensities, and to calculate energy expenditure estimates.

Accelerometers were worn on the wrist (attached by a strap) or on an elasticated belt around the waist (positioned above the hip). We used the ActiGraph GT3X+ model accelerometer for the study.

Assessment of Physical Activity in Frail Older People

The APAFOP is a PA questionnaire that has been purposely developed for and tested for reliability with older people with and without cognitive impairment. 46 It is administered verbally by a researcher with the interviewee, who is asked to recall activities undertaken in the previous 24 hours, with a focus on ADL.

Observational study: exploration of the criterion validity of estimates of physical activity and sedentary time derived from hip- and wrist-worn accelerometers

Accelerometer wear location is an important methodological consideration when using these devices, as the activity counts recorded are likely to differ markedly dependent on where the accelerometer is worn on the body. 119 Yet guidance on how to interpret activity counts recorded by an accelerometer worn on a location other than the hip in a meaningful way is limited. Nonetheless, wrist-worn accelerometers are increasingly being used in large-scale epidemiological studies in the light of reports of improved compliance, compared with a hip-worn monitor. 120 Furthermore, there is some evidence to suggest that an accelerometer worn on the wrist rather than the hip may be better suited to measuring the PA of frail older adults, as it can capture upper body movements typical of non-ambulatory activities (e.g. ADL, such as dressing) and the influence of atypical gait patterns is likely to be minimised. 60 However, to date, there is a paucity of studies comparing the outputs derived from hip- and wrist-worn accelerometers among older adults. Furthermore, no studies have assessed agreement with a criterion measure. Therefore, it is not possible to make inferences about which wear location results in the most valid measure of PA in this population.

The aims of this observational study were to:

• use observations of residents’ activities as a reference to compare with data captured by accelerometers to assess the validity of using accelerometers in a frail, elderly and largely inactive population

• identify whether an accelerometer worn on the hip or an accelerometer worn on the wrist was the most accurate in recording the types of activities in which frail, older adult residents in CHs typically engage.

Residents who had been recruited to WS2 were approached in the communal areas of the CHs and asked whether or not they would be prepared to participate in the observational study. Residents were invited to wear a hip-worn and/or wrist-worn accelerometer over a 2-hour ‘free-living’ period while being observed by the researcher.

The researcher used a predetermined list of activities to record systematically every 15 seconds what they observed participants doing over the monitoring period. The list of activities included both sedentary behaviours and PAs that were thought to be typical of CH residents. Any additional details that would facilitate the accurate classification of activities during analysis were also noted. For example, if a participant was observed walking, a note was made regarding walking speed. The time the 2-hour monitoring period started was documented to ensure appropriate data comparisons could be made across the two recording methods.

Feasibility study: use of accelerometers to assess physical activity over extended periods

Participants in habitual PA studies are typically asked to wear an accelerometer during all waking hours over a 7-day period. 121 However, compliance with this requirement is variable. Consequently, studies typically require participants to have met a prespecified wear-time criterion in an effort to ensure that estimates of PA and sedentary behaviour are reliable. While numerous wear-time criteria are reported in the literature, these are typically derived from studies among younger adults and empirical evidence supporting the superiority of a specific criterion is absent. 122 Nevertheless, in studies involving community-dwelling older adults, a wear-time criterion of at least 8 hours per day on at least 4 days is often required for inclusion in analysis. 52,53

In addition, given that the use of questionnaires remains an attractive method to measure PA, recent efforts have also been directed at developing questionnaires appropriate to old-old CH residents (i.e. those aged ≥ 85 years).

Thus, the aims of this feasibility study were to:

• investigate both the practical feasibility of using accelerometers and the appropriateness of the aforementioned wear criteria in older CH residents

• describe the levels of PA and sedentary behaviour

• explore the use of a PA questionnaire (APAFOP).

All recruited residents in the participating CHs were invited to wear an accelerometer around the hip for a period of up to 10 days, while they continued with their usual routine.

During the wear period, it was requested that a daily log of wear time (i.e. the time the monitor was put on and removed) was kept. The ‘activity log’ also included a column for recording the reason(s) the monitor was removed (e.g. for showering). Residents capable of completing the log and putting the accelerometer on themselves were encouraged to do so, although the process of completing the activity log was also explained to staff and they were asked to offer support with this, when appropriate. The arrangements for completing the activity logs were discussed with the CH managers in each home prior to the monitors being administered.

Following the collection of the accelerometers, the APAFOP questionnaire was trialled.

Observational study: exploration of the criterion validity of estimates of physical activity and sedentary time derived from hip- and wrist-worn accelerometers in older care home residents

To enable direct comparison between counts recorded by the hip and wrist accelerometers, all four counts recorded by the wrist accelerometer were adjusted according to the recommendations of the manufacturer (Table 12). These adjusted wrist counts were used in all analyses.

To align with the observational data, each of the accelerometer data sets was reviewed alongside the observational tool. Equivalent epochs for which participants were unavailable for observation (either because the participant was out of a communal area or because the observer was unavailable) were manually removed.

On the predetermined list of activities used systematically to record what participants were doing, activities were categorised as either sedentary behaviour or PA, based on metabolic equivalent of task (MET) values assigned to the same (or similar) activity in the Compendium of Physical Activities124 (Table 13). Thus, each observation was coded as either sedentary behaviour or PA.

Next, to enable direct comparison with observational data, accelerometer counts were coded as either sedentary behaviour (i.e. < 10 counts) or PA (≥ 10 counts). These cut-off points were developed in a sample of 37 community-dwelling older adults aged ≥ 65 years. 125

The agreement between the categorisation of sedentary behaviour and PA derived from the observational data and the accelerometer data was evaluated. Specifically, sensitivity, PV, overall agreement (Equations 1–3) and the kappa statistic were calculated:

With regard to sensitivity, PV and overall agreement, values close to 100% were deemed indicative of good agreement between the observational and accelerometer data, whereas a value close to zero suggested that the estimated time spent engaging in PA and sedentary behaviour, as categorised by the accelerometer data, was very different from that derived from the observational data. In line with recommendations from Landis and Koch,57 the kappa statistic was interpreted as follows: a value of between 0 and 0.2 was indicative of slight agreement, between 0.2 and 0.4 indicated fair agreement, between 0.4 and 0.6 indicated moderate agreement, between 0.6 and 0.8 indicated substantial agreement and between 0.8 and 1.0 indicated almost perfect agreement.

Feasibility study: use of accelerometers to assess physical activity over extended periods

Raw activity count data collected by the accelerometers were downloaded, processed with the normal filter and aggregated over 60-second epochs using the ActiLife software (version 6.8.0). Activity counts recorded in the vertical axis were used for analysis.

To reduce the risk of distorting data provided by the least active residents,126 all data were manually screened alongside the activity logs, and periods of non-wear time were removed. Daily wear time was determined by subtracting non-wear time from total possible minutes in a day (1440 minutes). Valid accelerometer wear time for this sample was identified as ≥ 8 hours on ≥ 4 days of the week.

Next, cut-off points, which had previously been applied to data collected in a sample of CH residents,24 were applied to the vertical axis accelerometer cpm to identify time spent engaging in sedentary behaviour (< 100 cpm),76,127 low-intensity PA (100–759 cpm), light-intensity PA (760–2019 cpm) and moderate to vigorous intensity PA (≥ 2020 cpm). 76,128 The data for each of these separate PA outcomes were also converted into a percentage of daily wear time to account for intraindividual and interindividual differences in daily wear time. 129

To facilitate discussion around the patterns of sedentary behaviour and PA, the number and length of ‘bouts’ of sedentary behaviour and PA were calculated. Specifically, a bout of sedentary behaviour was defined as a period of at least 30 minutes up to 59 minutes, or at least 60 consecutive minutes, and a bout of PA was defined as a period of at least 10 consecutive minutes of activity. The total amount of sedentary and PA time accumulated in bouts was also calculated.

Differences in the demographic characteristics and physical function of the following groups of participants were compared using a series of one-way ANOVAs or Kruskal–Wallis tests, as appropriate. First, those residents who agreed to wear a hip-worn accelerometer were compared with those who did not. Second, those participants wearing a hip-worn accelerometer and who provided valid data were compared with those participants wearing a hip-worn accelerometer who did not provide valid data.

Descriptive statistics (medians, IQRs and ranges) were used to describe the levels of sedentary behaviour and PA. Next, to explore the effect of personal characteristics on levels of PA and sedentary behaviour, participants were grouped according to age (< 85 years or ≥ 85 years), sex (male or female), functional walking ability (score of 0–1 or 2–3), physical function [BI score of ≤ 11 (dependent) or > 11 (independent)] and level of cognitive impairment (6-CIT score of < 8 or ≥ 8). Differences between the groups were assessed using a series of one-way ANOVAs or Kruskal–Wallis tests, as appropriate.

Practical feasibility of using accelerometers

Forty-one (89.1%) of the 46 recruited residents available for the phase 2 study undertook extended wear of an accelerometer. Four residents were deemed too unwell to participate and one resident refused to wear the accelerometer.

A hip-worn accelerometer was administered to 30 (73.2%) participants and, at the request of the research leads in two of the CHs, a wrist-worn accelerometer was administered to 10 (24.4%) participants. The reason for requesting residents to wear wrist-worn accelerometers rather than hip-worn accelerometers was that the CH managers felt that wrist-worn accelerometers would be more appropriate for their frail residents. In addition, one resident was asked to wear a commercially available device so that we could explore the data collected by this device.

There were no significant differences in the personal characteristics of those participants who were fitted with a hip-worn accelerometer and those who were fitted with a wrist-worn accelerometer (p > 0.05). However, those participants who wore a hip-worn accelerometer had higher BI scores than those participants who wore a wrist-worn accelerometer [mean BI scores: 13 (SD 5) and 9 (SD 6) respectively; p < 0.05].

Although wrist-worn accelerometers are increasingly being used in large-scale studies as a result of reports of improved compliance of wrist-worn accelerometers, compared with hip-worn accelerometers, there is a paucity of studies comparing the outputs derived from hip- and wrist-worn accelerometers among older adults. The results of the observation study outlined above suggested that data collected from a hip-worn accelerometer had better validity than wrist data in this population. Furthermore, there is limited guidance available on how best to interpret data collected from a wrist-worn accelerometer. Consequently, the decision was taken not to further analyse the wrist data collected in this study.

Twenty-two of the 30 participants who wore an accelerometer on the hip met the criteria for valid accelerometer data: two monitors were lost and six residents did not meet the criteria for valid wear time (≥ 8 hours on ≥ 4 days of the week). All of these six residents wore the monitor for < 4 days. Participant distress and/or confusion and the accelerometer interfering with the provision of care were the most common reasons provided to explain the removal of accelerometers over the monitoring period. There were no differences in the personal characteristics (sex, age, length of residence, height, weight, mental capacity, total BI score) of the 22 participants who met the criteria for valid accelerometer data and the eight who did not (p > 0.05).

Completion of the activity logs

Eight (19.5%) of the 41 activity logs distributed with the accelerometers were fully completed for the period of time the activity monitors were in the residents’ possession (range 3–9 days); 20 (48.8%) were partially completed and 13 (31.7%) were not completed at all.

A designated individual in most of the homes, usually the activity co-ordinator, was asked by the CH manager to assist in the completion of the activity logs. However, in practice, the researcher completed 25 (89.3%) of the logs, either alone [n = 13 (46.4%)] or with the assistance of CH staff [n = 12 (42.9%)]. Only three (10.7%) of the logs were completed by the residents (Table 19).

Accelerometer wear

Of the 22 participants with valid accelerometer data, accelerometers were worn for a mean of 6 days. Eight (36%) residents had 7 valid days, two (9%) residents had 6 valid days, three (14%) residents had 5 valid days and nine (41%) residents had 4 valid days. Mean daily wear time was 12 hours 24 minutes (SD 2 hours 36 minutes). Details regarding the daily wear time at each of the five CHs is displayed in Table 2.

Levels of sedentary behaviour and physical activity

Patterns of sedentary behaviour and physical activity

As discussed previously, residents spent the majority of their time engaging in sedentary behaviours and the little PA they did engage in was predominantly of low intensity. Figures 5 and 6 provide further information on the patterns of sedentary behaviour and PA across a ‘typical day’. Although Figures 5 and 6 represent the patterns of movement of two residents from a single CH (CH5), these two patterns of activity were ‘typical’ of the residents wearing the accelerometers across all of the CHs.

As can be seen in Figure 5, this resident spent the majority of his time sedentary, and the little PA that he did engage in was associated with self-care activities (getting out of bed and dressed in the morning and getting ready for bed in the evening). Conversely, the resident in Figure 6 engaged in more PA throughout the day and actually managed to accumulate continuous ‘bouts’ of at least 10 minutes of low-intensity PA. Based on observational data, the main form of regular movement observed in all the CHs generally involved accessing the toilet or moving to a different location for meals/organised activities.

Relationships between sedentary behaviour and physical activity and age, sex, physical function and cognitive ability

The differences between groups (e.g. age, sex) in the time and percentage of time spent in sedentary behaviour and PA are shown in Table 21. As can be seen from the table, there were no significant differences in the time spent engaging in sedentary behaviour or PA of any intensity between those aged < 85 and those aged ≥ 85 years, or between males and females (p > 0.05). Similarly, no differences in sedentary behaviour or PA were observed between those participants categorised as having a significant score on the 6-CIT (≥ 8) (i.e. higher level of cognitive impairment) and those who did not (6-CIT score of < 8).

However, those residents with better functional walking ability and better physical function, as assessed by the BI, spent a smaller proportion of accelerometer wear time engaging in sedentary behaviours than those with poorer walking ability and poorer physical function (p < 0.01). In addition, those residents with better functional walking ability and physical function also spent more time, and a greater percentage of accelerometer wear time, engaging in low-intensity PA than those with poorer walking ability and physical function (p < 0.01) (see Table 21).

Assessment of Physical Activity in Frail Older People questionnaire

The APAFOP46 was trialled with two purposively selected residents with relatively strong cognitive ability in CH1 (Table 22).

Figure 13 shows the activities of the two residents over a 24-hour period, as recorded by the APAFOP questionnaire. Little walking or activities of any form were recorded. Both residents spent the majority of their time engaged in typically sedentary activities (sitting or lying). One resident who participated in completing the APAFOP was immobile and displayed no bouts of activity. The other resident was most active at 08.00, 12.00 and 19.00, coinciding with getting up, having lunch and going to bed.

FIGURE 13.

Activities of two residents in a 24-hour period, as recorded by the APAFOP. Cumulative energy expenditure (kcal) is shown for each category of activity undertaken.