This current systematic review summarised evidence based on process, resource, management, and scientific feasibility of physical activity intervention for older adults living with and beyond carcinomas. The results of these studies suggest that physical activity interventions are feasible based on adherence, acceptability, retention, and safety, however, recruitment was identified as a challenge. Resource and management feasibility continue to have limited to no reporting, even though these metrics provides important information for planning and implementation like the need for adequate recruitment timeframes and using suitable settings for study and intervention delivery. Finally, a key limitation to most studies within this review was the little to no reporting of pre-specified criteria of success, which is important in assessing the studies feasibility.
Process Feasibility
Recruitment
A major challenge for most studies in this current review was recruitment, as seen by large exclusion and low retention rates, which is common within the literature (105). Three key barriers to recruiting this population included a lack of interest, high frailty, and transport barriers. A lack of interest in physical activity is common among older adults and might be attributed to underlying barriers or beliefs that physical activity is not relevant, beneficial, or suitable (106, 55, 107, 108). Whilst the exclusion of frailty is likely due to concerns around safety (109, 110), our review suggests the inclusion of frail participants is feasible (85, 111, 86, 97, 95). Finally, transport barriers include lack of public transport, poor parking, reliance on family or friends to provide transport, proximity, poor weather conditions, and cost. This not only affected recruitment but also had implications for acceptability and retention (53, 112, 106, 113). As a result, studies may still be failing to recruit a typically older carcinoma population who may present as physically inactive and frail.
A possible solution to improving recruitment is engaging with relevant stakeholders (i.e., people living with and beyond cancer) in the process of designing the intervention and study (54, 114). For example, stakeholders may be able to provide feedback on solutions to transport barriers like home-based programs, including social support, and reimbursing travel expenses, or offer alternate solutions (54, 55, 113, 114). This should support the design of a relevant and meaningful study with appropriate data collection methods (i.e., the balance between research needs and participant burden). Stakeholder involvement has been shown to improve the acceptability, appropriateness, and relevance of interventions and study designs, and consequently increase recruitment rates (115–117).
Adherence
In general markers of adherence to the intervention were good across all studies. An explanation for the good adherence may be the acceptability of the interventions by participants (scientific feasibility). An alternative explanation could be the supervisory component, as systematic reviews reveal that supervision is superior for adherence to interventions and improving health outcomes compared to non-supervision (118, 119). Nevertheless, there were discrepancies in reporting adherence (e.g., collective or individual adherence and different %’s of adherence) and minimal reporting of intensity and duration adherence, which can be attributed to a lack of consensus on the definitions and appropriate use of adherence measures (120). As seen in our review, the attainment of the prescribed intensity was not feasible for some participants which is valuable information for efficacy trials which can make modifications based on findings. Therefore, going forward retention, attendance, and intensity and duration adherence should be used with the same definitions to allow for comparison (120).
Resource and management feasibility
Feasibility metrics pertaining to resource and management feasibility were not well reported. The limited reporting may be explained by many resource and management outcomes (e.g., provider's capacity and willingness to complete research tasks, researchers’ time to complete research activities) not naturally fitting within the structure of typical scientific article, and the importance of these features in terms of their feasibility not being widely appreciated. Therefore, the importance of reporting these outcomes needs to be better underlined, perhaps by adding instructions to reporting guidelines, and signposting where this detail could be incorporated (e.g., as supplementary files). Indeed, lessons learnt from resource and management feasibility could help future researchers better plan and implement studies in relation to time frames, resources needed, and in preparing for challenges that previous studies have encountered (74). For example, a key learning from this current review is the need for adequate recruitment timeframes. In addition, future studies should aim to measure the fidelity of intervention delivery. This can be achieved through directly observing if sessions are being delivered as planned, allowing for any deviations to be explored and resolved (121, 122). Arguably, it is important for feasibility studies to measure if interventions and study designs can be implemented as intended to understand the effect on the study outcomes (e.g., prespecified criteria of success).
Scientific feasibility
Adverse events
Exercise may be safe for older adults living with carcinomas based on a few minor adverse events reported, however, few studies reported events with varying methods of recording incidents and severity. There is still uncertainty about the safety of exercise for cancer populations in different stages of treatment, and compounded by the limited representation of older adults within the literature this indicates the need for further exploration (12, 21, 22). To understand the safety of exercise for older cancer populations, adverse events should be consistently reported in line with guidelines (123).
Acceptability
Participants found the interventions acceptable with several studies reporting supervision and self-monitoring as key motivating elements within the interventions. In addition to being reported as motivating, supervision and self-monitoring were also reported as educational and increasing awareness, which in accordance with previous research has a positive effect on physical activity interventions for cancer populations (35, 124, 118, 125). Thus, the inclusion of supervision and self-monitoring within interventions may be important for acceptability and behaviour change.
The reporting of acceptability could however be more robust. Firstly, studies reported acceptability retrospectively. Future studies may consider measuring acceptability throughout the study (start, during, and end) to actively adapt the intervention or study design and test the feasibility of such adaptions (33). Secondly, only one study reported acceptability from providers, which is important to report, as this may identify unanticipated barriers and facilitators to implementing (intervention fidelity) the intended intervention, for example, the need for further competence training or resources, such as protected time to deliver the intervention (117, 75, 33). Finally, few study reported on the acceptability of the study design which could also contributing factor to participation, retention (i.e., study burden), and adherence to data collection (i.e., completing physical activity diaries). The regular measurement of participant and provider acceptability may contribute towards enhanced feasibility outcomes like recruitment, intervention fidelity, adherence, and retention.
Estimated effects
Most studies appeared to have a positive effect on their secondary outcomes which included physical activity, physical function, and QoL. Previous studies have found little to no benefits of physical activity interventions when older adults are included, especially those with poor physical function or no exercise history (30, 31, 126), indicating the need for targeted intervention for older adults living with cancer. The feasibility studies showed promise towards improving secondary outcomes, therefore, indicating the need to target age-related factors (e.g., poor physical function, comorbidities, negative beliefs about age and cancer).
The reporting quality of feasibility studies
In general, the studies were reported adequately, meeting most of the criteria set out in extended CONSORT guidelines for pilot and feasibility studies (72). The poorest areas of reporting were pre-specified criteria of success and the sample size rationale (56, 127). Pre-specified criteria for success are important to determine if the feasibility study should proceed or not to full trial (with or without design modification). Whilst the correct sample size is needed for useful information to be drawn from the feasibility study (69). However, the under-reporting might be explained by no clear methodological guidance on how to determine these criteria (56). Interestingly, one method for determining sample size is linked to estimating and hypothesising pre-specified criteria of success (e.g., recruitment, fidelity, retention (128)). In addition, few studies were pre-registered. Pre-registration allows transparency of the intended study plan (research question, data collection, analysis) and any reasons for deviations made, therefore increasing the credibility of findings (129). This may create more ‘trust’ that a study was an intended feasibility trial and not a small underpowered study (56).
Limitations
There are several limitations to this review. The widely used age criteria of ≥ 65 to define an older adult may have resulted in some studies being missed (48, 49). At this age, comorbidities and functional limitations are assumed and mortality risk is greater. However, it is acknowledged that defining ‘older adults’ by chronological age is debated and arguably limited, with biological age (frailty) more precise but less used within the literature (130, 131). A second limitation is the inclusion of different intervention (e.g., exercise type, intensity, supervision) and population characteristics (e.g., carcinomas, cancer stages), plus the small number of studies included made it difficult to determine which factors may have influenced feasibility. Nevertheless, the focus of this review was on physical activity interventions for older adults living with carcinomas and we were able to demonstrate feasibility outcomes for this population. Finally, there were two minor deviations from the pre-registered protocol. First, we described we would do sub-group analyses of feasibility outcomes, but we did not clearly state this as a secondary aim. Secondly, we included stage 0 cancer participants, however, this only made up 10% of the study population in two studies (88, 103).