Table 3.

Study results

Source	Outcome(s)	Feasibility results	Primary outcome results	QoL results	Follow-up results	Modifications/changes to intervention
Alibhai 2019 [47]	Feasibility Physical fitness Cost-effectiveness General and prostate-specific QoL: FACT-G plus FACT-P Fatigue: FACT-F	Recruitment rate: 25.4% (59/232 eligible participants) Retention rate at 6 months: 76.3% QoL outcome captur0:e 80% Satisfaction of at least 4 out of 5: 88%	See feasibility	FACT-P HOME group poorer QoL at 6 months than PT difference in Δ = 4.3, (95% CrI − 8.1 to − 0.5, probability of inferiority = 74%) GROUP Δ = − 1.4 (95% CrI − 5.4 to 2.6, probability of inferiority = 21%). FACT-G the change from baseline to 6 months was 2.9 points worse for HOME and 1.7 points worse for GROUP than PT, with the probability of inferiority being 38 and 26%, respectively. Changes in FACT-F were similar between arms.	Not reported
Bourke 2011 [48]	Feasibility Diet and exercise behaviour Prostate specific QoL; FACT-G and FACT-P Fatigue; FACT-F	Retention 12 weeks intervention = 84% control = 88% 6 months intervention = 60% control = 52% Attendance (for 21/25 men) 360/378 sessions (95%) Compliance in sessions 329/378 (87%) at least 25–30 min recorded in logs.	Total PA behaviour higher in intervention group post intervention Godin LSI points 33.8 vs 17.4 (mean diff Δ = 16.3, 95% CI 8.8–23.8; p < .001) and 6 months (25.9 vs. 15.6 Godin LSI points, mean diff Δ = 11.3, 5.0–17.5; p = .001) Diet macronutrient intake reductions in total energy intake (mean diff Δ = − 285.5 kcal, − 32.5 to − 484.5; p = .005), total fat (mean diff Δ = − 19.8 g, − 7.3 to − 32.3; p < .001), saturated fat (mean diff Δ = − 8.6 g, − 3.7 to − 13.5; p < .001), and monounsaturated fat intake (mean diff Δ = − 6.6 g, − 2.0 to − 11.2; p < .001)	Fatigue (FACT-F) Improvement at 12 weeks in intervention group (mean diff Δ = 5.4, 95% CI = 0.8–10.0; adjusted p = .002) FACT-G No significant differences in groups at 12 weeks (p = .25) FACT-P No significant differences in groups at 12 weeks (p = .21)	6-month follow-up Total PA behaviour higher in intervention group post intervention Godin LSI points 6 months (25.9 vs. 15.6 Godin LSI points, mean diff Δ = 11.3, 5.0–17.5; p = .001) Fatigue improvements maintained at 6-month follow-up (mean diff Δ = 3.1, 95% CI = 0.3 to 6.4; adjusted p = .006) FACT-G No significant differences in groups at 6 months (p = .36) FACT-P No significant differences in groups at 6 months (p = .45)
Demark-Wahnefried 2006 [49]	Diet quality Physical activity behaviour General and cancer specific QoL; FACT-G plus FACT-B or FACT-P	3000 participants identified by cancer registries, 74% had sufficient data to enable contact 688/2010 contactable patients returned consent forms and screeners for 34% response rate Of these, 182 were enrolled and eligible (26%) 168/182 completed all telephone counselling sessions 160/182 completed 12 month follow-up data (cumulative dropout rate of 12.1%)	Behaviour change Improvement in diet quality (between group p < .003) for intervention	Enhanced physical functioning though non-significant (between group p = .23) Depression scores improved in both groups (between group p = .55) FACT-G QoL Both group improved at 6 months (between group p = .38)	All scores returned to near baseline levels except QoL which remained at 6 month levels for both groups.
Demark-Wahnefried 2018 [50]	Feasibility Absence of serious events General QoL; SF-36	Invitation letter mailed = 694 Contactable pool = 624 Enrolled and consented = 46; 24 intervention, 22 WL control Retention rates 22/24 (92%) intervention; 20/22 (91%) WL control	No changes in self-efficacy as it was high to begin with. Increase in intervention arm in social support to garden (p = .002) Increased F&V intake among intervention (within group p = .02) but not control (between group p = .06) Both arms had improvements in physical performance	Perceived stress stable within both arms Reassurance of worth increased in intervention arm whilst decreasing in WLC (between group p = .02) More positive results in SF-36 QoL measures for WLC versus intervention group; i.e. pain worsened (p = .02) in intervention and physical role (p = .01) and overall mental health (p = .01) improved in controls	Not reported
Desbiens 2017 [51]	Fatigue General and breast specific QoL; FACT-G and FACT-B	Not explicitly stated but the challenge of recruitment were a limitation. They were unable to perform two of their intended study objectives.	FACT-F non-significant improvement within both groups; no differences between groups No group difference mean that individual exercise was also potentially effective.	FACT-G non-significant trends towards improvement within both groups; no differences between groups FACT-B non-significant trends towards improvement within both groups; no differences between groups	N/A	Subsequent survey was administered to assess willingness to exercise because accrual was poor
Lai 2017 [52]	Cardiopulmonary intolerance Pulmonary function Cancer specific HRQoL; EORTC QLQ-C30 lung supplement	None explicitly stated	The mean postoperative length of stay (6.9 ± 4.4 vs. 10.7 ± 6.4 days, p = .010) and total in-hospital stay (16.0 ± 4.5 vs 19.7 ± 6.5 days, p = .012) were significantly reduced in the PR group 6MWT and peak expiratory flow increased significantly in PR group. Potentially due to fewer post pulmonary op complications.	EORTC-QLQC30 & EORTC-LC13_CN no difference was observed between the groups in terms of: global QoL (− 0.5; P = .785) physical function (− 0.67; P = .691) emotional function (− 2.2; P = .206) dyspnoea score (0.37; P = .808)	N/A
Loh 2019 [53]	Anxiety Mood Social and emotional well-being using subscales from FACT-G	N/A	Anxiety within group change for intervention (− 3.51, p = .003); between group difference favouring intervention for 75th (− 5.39, p = .001) and 95th (− 10.97, p < .001) percentiles Mood within group change for intervention (3.08, p = .046) and control (4.57, p = .002); between group difference favouring intervention for 75th (− 5.04, p = .032) and 95th (− 11.12, p = .007) percentiles	SWB no significant within group change for intervention or control; between group difference favouring intervention for 5th (3.90, p < .001) and 25th (1.39, p = .006) percentiles EWB within group change for intervention (1.04, p < .001) and control (0.80, p = .010); between group difference favouring intervention for 5th (1.82, p = .025) percentile	N/A
Miki 2014 [54]	Feasibility Cognitive function Cancer specific QoL; FACT-G	Deemed feasible due to no dropouts in either group and no adverse events in the intervention group. Highly accepted based on patient reports of how fun it was.	Cognitive function with Frontal Assessment Battery (FAB) found significant time effect (F = 24.39, p < 0.001, partial η2 = 0.247) and a significant group effect (F = 9.26, p = 0.003, partial η2 = 0.109) Also significant interaction between the two groups on the FAB score (F = 7.88, p = 0.006, partial η2 = 0.094)	FACT-G Baseline scores 75.29(15.76) vs. 74.30(14.27) Week 4 scores 77.47(14.01) vs. 75.42(15.42) Interaction p = .738; time effect p = 0.297; group effect p = .612	N/A
Monga 2007 [55]	Fatigue General and prostate-specific QoL; FACT-G plus FACT-P	None explicitly stated however Nine patients (4 after enrolment, 5 after randomisation) refused to participate, because they wanted to be in the intervention group. Three of the 5 patients who disenrolled after randomisation and initial baseline testing were from the control group	Fatigue with Piper Fatigue scale (higher scores = greater fatigue) Significant between group differences in favour of exercise group for fatigue (change − 4.3 ± 2.1; t = − 4.72, p ≤ .001)	Within intervention group changes: FACT-G Physical well-being (PWB: change 2.3 ± 1.8; t = 4.20, p = .002), social well-being (SWB: change 1.5 ± 1.9; t = 2.67, p = .02) FACT-P (change 7.4 ± 10.4; t = 2.36, p = .04) Within control group changes SWB (change − 1.7 ± 2.4; t = − 2.28, p = .05) Between group changes in favour of intervention PWB (change 3.6 ± 2.0; t = 4.19, p ≤ .001) SWB (change 3.2 ± 2.1; t = 3.47, p ≤ .002) Functional well-being (FWB: change 4.1 ± 4.2; t = 2.24, p = .04) FACT-P (change 13.8 ± 10.1; t = 3.12, p = .006)	N/A
Morey 2009 [56]	Functional status General QoL; SF-36	558/641 (87%) completed 12 month measures 488/641 (76%) completed 2-year measures.	Change in functional status using physical function subscale of SF-36 (higher score means better function) and Late Life Function and Disability Index (basic and advanced lower extremity function subscales) Control group experienced a mean score change of − 4.84 (95% CI, − 3.04 to − 6.63); more than double that of the intervention group (− 2.15 [95% CI, − 0.36 to − 3.93]); group difference p = .03. Significant difference between groups in basic lower extremity function as function changed negligibly in the intervention group (mean, 0.34 [95% CI, − 0.84 to 1.52]), but control group showed a decrease (− 1.89 [95% CI, − 0.70 to − 3.09]; group difference P = .005).	Full SF-36 Overall HRQoL decreased in every subscale in the control group. In the intervention group, decreases in subscale scores were of lower magnitude and were sustained for overall health and mental health Overall HrQoL score mean change between groups was 2.71 (95%CI, 0.58 to 4.84); p = .03	Not reported
Park 2012 [57]	Functional physical fitness General QoL; SF-36	No side effects or safety issues arose from exercise programme. 26/33 completed trial and were analysed from exercise group 25/33 completed trial from control group; 2 were excluded from analyses due to missing data	Functional physical fitness Intervention group had greater improvements in fitness (p < .001), flexibility (p = .027), and balance (p = .015).	SF-36 Physical composite score of SF-36 decreased about the same in both groups after surgery (p < .001). Physical score recovered to preoperative level in exercise group (p < .001) but not in control group (p = .225) after 12 weeks. Mental composite score improved after 12 weeks in exercise group (p = .017) but not control group (p = .773).	N/A
Porserud 2014 [58]	Feasibility Physical function General QoL; SF-36	Deemed not feasible due to unknown number who were not interested, unknown number not invited to participate, large number of dropouts. 5/9 intervention; 8/9 control completed week 14 measures 4/9 intervention; 6/9 control completed 1 year measures attended 76% of groups exercise sessions and taken daily walks on 87% of the days in the 12 week period	Both groups improved 6MWD	SF-36 Role physical domain in intervention group improved more than control (p = .031) after intervention (14 weeks). No other differences were observed at 14 weeks.	No differences remained or were observed at 1-year follow-up.
Sprod 2015 [59]	Cancer-related fatigue Global side-effect burden; Clinical Symptom Inventory	attendance in sessions averaged 6.2/8 sessions Original study had 410 randomised (206;204) and retained 361 (174;187). Unknown what retention rates were for specific age group of interest in the current study.	Cancer related fatigue (CRF) with Multidimensional Fatigue Symptom Inventory — Short Form (MFSI-SF) Yocas group reported significantly lower CRF than WLC (total score; p = .03), physical fatigue (p < .01) and mental fatigue (p < .01).	Global side effect burden with Clinical Symptom inventory Yocas group had significantly lower level of global side effect burden (p < .01)	N/A
Winters-Stone 2016 [60]	Feasibility General Physical and Mental QoL; Physical and Mental summary scores from SF-36	22% enrolment rate No dropouts in intervention group but 5 couples dropped out of WLC Median attendance to exercise sessions was 78% for PCS, 76% for spouses, and 75% for couples 94% were fully compliant with training	Feasibility primary outcome	Physical and mental summary components of SF-36 No significant group differences among men. Among spouses, mental health increased in intervention while WLC decreased slightly; however, non-significant (p = .06). No significant group differences in physical function and vitality subscales of SF-36 in either PCS or spouses.	N/A

QoL quality of life, PT personal training, FACT-G functional assessment of cancer therapy-general, FACT-P FACT-prostate, FACT-B FACT-breast, FACT-F FACT-fatigue, SF-36 Short form-36, EORTC QLQ-C30 European Organisation for Research and Treatment of Cancer QoL Questionnaire 30, LSI leisure score index, F&V fruits & vegetables, WLC wait list control, SWB social welling, PWB physical well-being, EWB emotional well-being, CI confidence interval, 6MWD 6-min walk distance, CRF cancer-related fatigue, HrQoL health-related quality of life, PCS prostate cancer survivors