Few trials have included such a comprehensive process evaluation carried out independently of the main trial. The OPERA process evaluation has given us a unique insight into both workings of the care home sector for older residents and the implementation of a large complex cluster randomised controlled trial in this setting. It is important to try and understand why an intervention is effective or not. Our study is one of a growing number of published randomised controlled trial process evaluations [
22‐
26]. Here, we explore two possible explanations for the negative result of the OPERA trial: whether the intervention changed the culture of the homes to promote more physical activity among residents, and whether the residents engaged with the exercise classes.
Did the intervention change the culture in the homes?
One of the most striking findings of this evaluation is the difference between care homes within the OPERA study. However, all the homes in OPERA volunteered to be in the study so we might imagine that the differences across the entire care home sector would be even more extreme. The activity sweeps show patterns of activity that appear to be relatively consistent across time within the homes but very different between homes. Differences in the mix of residents alone seem unlikely to explain why the culture around physical activity appears to differ between homes. It is unclear how the very different physical environments we observed might impact on the residents and care staff and further work exploring these environmental issues is ongoing [
27]. The process evaluation shows that it was possible to run standardised, evidence-based exercise groups requiring special equipment in all the homes in the OPERA study. The stability of patterns of activity across 12 months within the case study homes suggests that changing the culture of homes, at least with regard to patterns of activity among residents, might be more difficult.
While the physiotherapists developed a very good relationship with the homes and residents, the planned provision of a ‘home champion’ did not happen. Possibly the best placed member of staff to take on this role would have been the activities coordinator. However, while a number of activity coordinators were observed just fulfilling this role many were also allocated other tasks, taking them away from the ‘activity’ role. Where the role was recognised and exclusive residents benefited from having someone who was dedicated to increasing activity, but this was rare and in general the identification of ‘home champions’ was not successful. It seems that the role of home champion was unrealistically demanding for most staff in most homes.
Access to a health professional was something home managers and staff found beneficial. Most of the managers and staff interviewed reported problems with obtaining basic NHS services for residents such as physiotherapy, or even to obtain provision of simple mobility aids. The British Geriatrics Society has highlighted the issue of providing seamless healthcare provision for patients who are resident in homes in the independent sector [
28]. Our process evaluation identified that OPERA physiotherapists succeeded in obtaining simple, relatively low cost mobility aids (from NHS sources) that care homes had been either unaware of, or unable to access.
The OPERA physiotherapists assessed residents and provided aids, equipment and advice which should have made it easier for homes to increase the activity of residents when the OPERA staff were not there. There was little evidence that this happened. Each resident had an individual exercise prescription yet there was little or no evidence that homes made use of these.
Care staff were positive about the staff training sessions, when they attended. However, it is doubtful if this one-off session with only just over half the staff attending was sufficient to bring about the necessary changes; particularly when the training promoted the safe mobility of residents as a mood enhancer.
The OPERA study took place across a period of increasing economic uncertainty and, latterly, in an evolving economic recession. The majority of UK care homes are part of the independent sector and one of the largest groups in the UK, Southern Cross, announced it was closing down just after the end of the study [
29]. Even at baseline, mean occupancy was 87% and the field team reported that vacant places increased across the duration of the study. Observation and interviews in the homes suggested that care staff often have to work very hard and some resource-stretched homes may have little spare capacity to engage in cultural shifts that consume carer time, such as promoting physical activity among frail, older residents.
Did the residents engage with the exercise?
The exercise groups were delivered as planned and information from the focus groups and interviews identified many positive effects of the groups and almost universal enthusiasm from the home staff and residents for the groups, which only appeared to increase across the duration of the study. Moreover there was little evidence of fall-off in the numbers attending of the classes across time, suggesting that an ongoing exercise-class-based intervention in care home settings is viable.
On the surface it seems that attendances at group sessions were good, with some attending every session, but a closer look at the data reveal a number of issues. There were a large number of residents coded as ‘unwilling’ to attend. It is difficult to know if this was real unwillingness or because staff or the physiotherapists did not have time to gather residents together for the session. The physiotherapists reported that, when they arrived for a group session, the care staff were often too busy to help get people to the group. Less than half of study participants (that is, those who may have contributed to the primary outcome) attended an average of one session per week (the predefined definition of an ‘adequate dose’) or more. Thus, overall exposure to the group sessions was poor.
The frailty of the residents and high levels of cognitive impairment may be a contributing factor. The SPPB results revealed that many residents had poor physical function (particularly in the lower limbs), contributing to the fact that the exercise groups were mostly seated. Residents enjoyed the sessions and exhibited outward signs of joy and happiness. However, perhaps they did not reach an intensity of physical activity sufficient to bring about long-term physiological or psychological mood enhancing changes.
Secondary analysis of the impact of a number of baseline variables on group attendance reveals that neither age, poor physical function (SPPB) nor greater cognitive impairment (MMSE) seemed to stop residents attending the groups. What we did find is that those who were not depressed attended significantly more groups than those who were depressed. This suggests that our exercise intervention was missing those who may have benefited most from it. In future, it may be useful to explore how these depressed residents could be encouraged to join in the group activities. We also found that those with lower MMSE scores attended more groups, which is somewhat counterintuitive. What it does suggest is that cognition is not a barrier to attendance at groups such as these but we are cautious at overinterpreting this result that is based on a post hoc analysis.
Two other similar but smaller trials have tested interventions to increase physical activity in older people with depressive symptoms. Neither of them found any effect of the intervention on the prevalence of depressive symptoms. One of these trials was delivered to older people in residential care but did not report any process data that would allow further exploration of possible reasons for the negative result [
30]. A trial with community-dwelling older people tested individually prescribed moderate intensity exercise combined with walking, each for 30 minutes three times a week [
31]. The researchers reported good uptake in the intervention group, with 84% of randomised individuals receiving all the planned visits, although the participation in the prescribed activity was less. At 6 months just 29% were carrying out the exercises three times a week, and 37% were walking three times a week. Another trial of physical activity in older people provided detailed process evaluation of a negative result, but was not targeted at reducing depression but on reducing falls [
22]. The researchers tested providing t’ai chi classes twice a week for 13 weeks. Both instructors and participants were very positive about the classes, but nevertheless, only 64% of the interventions group completed the course and only 47% were rated as having achieved an adequate attendance of at least 21 of the 26 classes. It seems that low uptake and adherence may be a common issue although neither of the two trials reporting uptake took place in a residential facility. In addition, the fact that the participants were living independently implies that they were on average considerably less frail than the participants in the OPERA trial. A recent Cochrane review of exercise for depression found that exercise may have a positive effect on depression, but there is considerable uncertainty about the optimum duration and frequency of the exercise interventions to achieve a positive outcome [
32]. Few of the trials included frail care home residents similar to those in the OPERA trial; as noted earlier, most were less frail and more mobile.
This process evaluation had some limitations. The number and spread of interviews carried out was smaller than planned, particularly with regard to home residents. Relatives were also difficult to access as we had no access to their contact information. Our detailed observations were limited to 8 of the 78 homes in the trial. Nevertheless, even in this small sample we observed a wide diversity in levels of activity within the homes. We are cautious in our interpretation of the post hoc analyses, as these were unplanned, unadjusted and lack power. We have therefore provided a full set of descriptive statistics to enable the reader to draw their own conclusions.
We have explored in detail two possible reasons why the OPERA trial was negative. Eliciting a cultural shift in the care home sector is problematic and the OPERA intervention failed to bring about enough of a change. We have also found that the care home population is older and frailer than anticipated when the OPERA trial was planned, perhaps as a result of a prevailing ethical and financial imperative to maintain frail older people in their own homes with support for as long as possible. While the staff and residents were positive about the exercise groups it appears that the intervention did not encourage enough engagement or intensity of exercise to address low mood more than momentarily, and that those with most potential to benefit from an intervention ameliorating depression (that is, the depressed) were least likely to engage in it. Indeed, from many hours observing the intervention, we appeared to make a difference. These differences may have just been transient or, alternatively, the measures we were using may not have been sensitive enough to capture them. It is clear that life for our aging population in a care home setting can be very sedentary and this increases the possibility of low mood and a poor quality of life. Alternative approaches are needed to both relieve the burden of depression in care home residents and to improve their overall quality of life. Researchers in the future need to consider that the instruments we use to measure depression and quality of life may need to be tailored to the population being studied, and any promising approaches should be tested empirically before implementation.