Sage Journals: Discover world-class research

Abstract

Background: Although exercise benefits female cancer survivors, clinical decision-making regarding timing, frequency, duration, and intensity is lacking. Optimizing exercise interventions in this population is necessary. This study aimed to describe existing digital home-based exercises and to assess their effectiveness at improving physical health in female cancer survivors upon completion of therapy. Design: We conducted a systematic review using articles from Web of Science, Embase and Medline (Ovid). We included intervention studies examining the effects of digital home-based exercise programs on post-treatment recovery in female cancer survivors. Rob2 and ROBIN I were used to assess quality of studies. Quality-of-life, fatigue score, and physical performance were assessed using meta-analysis. Results: This study involved 1578 female cancer survivors in 21 interventions. Following guidelines and supervised exercise with coaches led to better outcomes than interventions without guidelines, programs without coaches, or lower intensity exercise. Exercise led to significant improvement in some physical performance outcomes. Significant improvements were seen in physical performance outcomes, including the 6-min walk test, metabolic equivalent task, and number of steps per day. Conclusion: Providing cancer survivors with standard guidelines for home-based, coach-supervised, vigorous exercise on digital platforms could improve their physical function, health, and quality-of-life.

Keywords

cancer digital-based exercise home-based exercise post-treatment intervention women

Introduction

Cancer has caused a high burden of diseases among women worldwide¹ because not only the high mortality and incidence rate but also the high number of disability-adjusted life year (DALY) compared to men.² Female cancer survivors tend to have higher incidence rates³ and live longer than male cancer survivors.⁴ Females have a significantly higher incidence rate of cancer than males between the age of 18 and 60, according to an official report from the United Kingdom.³ In 2012, there were an estimated 6.7 million new cases and 3.5 million deaths attributed to cancer among women globally.⁵ Fortunately, average 5-year survival rates of almost all cancer types have increased significantly over the last three decades, particularly for breast cancer (74.8% in 1970 vs 89.7% in 2007).⁶ This is most likely attributable, at least in part, to the progress that has been made in the diagnosis and treatment of cancer among women.

Indeed, since the widespread introduction of radiotherapy in cancer treatment in the 1990s, 60% of cancer patients have received therapeutic radiation alongside chemotherapy, immunotherapy, hormone therapy, and surgery to shrink tumors or alleviate pain in patients with metastases.⁷ Worldwide, as a result of radiation therapy 3.5 million people may have a chance of cure, with a similar number receiving post-treatment palliative care.⁸ Cancer treatment extends the lives of half of those diagnosed with the disease for 10 years or more.³ While the benefits are evident, one must also acknowledge potential downsides. The quality-of-life can be adversely affected by side-effects of cancer treatment, including fatigue, sleeplessness, and cancer-related cognitive impairment.⁹ Therefore, post-treatment care for cancer survivors is essential to ease the burden of treatment-related side-effects and hence to reduce the time to recovery.

Exercise interventions are increasingly used to help cancer survivors to improve their health and functionality following recovery.¹⁰ Exercise therapies are safe, have the potential to minimize the risk of mortality, and to produce significant improvements in clinical, functional, and quality of health outcomes in cancer-surviving patients.¹¹ Exercises are also proven to improve in quality-of-life, muscular and aerobic fitness, and overall health¹² and improve management of post-therapy symptoms, especially fatigue and sleep disturbance.^13,14 With cutting-edge digital tech, home-based exercise programs offer a promising post-treatment strategy. They efficiently manage interventions, motivate home-based physical activity, and maintain contact with healthcare providers after hospital discharge.^15–17 A digital home-based program is an intervention that an individual can undertake from home for any duration of follow-up using a smartphone, tablet, laptop, or home computer.^16,17 Hence, digital home-based exercise treatments should be developed and put into practice with a view to improving health outcomes of cancer patients. The purpose of this systematic review was to provide a synthesis of evidence regarding the characteristics, implementation approach, and effectiveness of digital home-based exercise treatments on health outcomes. Specifically, the following research questions are addressed:

1. What are the characteristics of digital home-based exercise interventions for female cancer survivors after completing cancer treatments?

2. What effects have these interventions had on health outcomes of interest for cancer-surviving women following cancer treatments?

Methods

This systematic review was conducted following the “The Preferred Reporting Items for Systematic Reviews and Meta-Analyses” (PRISMA) guideline.¹⁸ The detailed protocol has been registered on PROSPERO (#CRD42022348049).

Eligibility criteria

Inclusion and exclusion criteria for literature selection were defined using the Population, Intervention, Control, Outcomes, Study Design (PICOS) method.¹⁹

Types of population

The following inclusion criteria were all required to be met: (1) women (18 years of age or older) with a diagnosis of any cancer type; and (2) completed an episode of main and adjuvant cancer treatment, such as radiation alongside chemotherapy, immunotherapy, or hormone therapy. Following treatment, each woman could have hospital outpatient examinations according to their treatment schedule. Exclusions were for: (1) patients receiving acute or inpatient hospital care; and/or (2) metastatic cancer patients (more than 80% of participants).

Types of intervention

Digital home-based exercise interventions were considered for: (1) implementation at home; (2) patients who had completed cancer treatment; and (3) when exercise interventions needed to include at least one type of physical exercise, such as aerobic exercise (AE), resistance exercise (RE), Yoga, or tai chi¹¹; and (4) when the participant possessed a mobile device, handheld tablet, or laptop computer to offer the intervention via smartphone apps, online platforms (websites, Zoom, Skype, WhatsApp, social media channels, etc.), video gaming, and DVD videos.²⁰ The review excluded hospital/clinic-based or home exercise interventions like cooking, cleaning, and gardening without supervision or intensity.

Types of control and study design

Intervention studies included randomized controlled trials (RCTs), quasi‐experimental studies, and multiple intervention arms trials with no control groups. Non-experimental studies such as case reports, cross-sectional and observational studies were excluded. Before/after studies with no comparison groups were also deemed as ineligible for inclusion.

Types of outcomes

This review focused on self-reported quality-of-life (QoL) outcomes and selected physical outcomes:

- Fatigue: the outcome could be assessed or self-reported using validated questionnaires, such as Functional Assessment of Chronic Illness Therapy-Fatigue (FACIT-F).

- Health-related quality-of-life (cancer-specific or, cancer site-specific): using validated questionnaires, such as Functional Assessment of Cancer Therapy-General (FACT-G), European Organization for Research and Treatment of Cancer Quality-of-life Questionnaire-Core 36 (EORTC QLQ-C30), and Functional Assessment of Cancer Therapy-Breast (FACT-B).

- Physical activity: maximum rate of oxygen consumption during physical exertion (VO₂ max), distance walked per unit time (MET – hour/week), 6-min walk test (6MWT), and metabolic equivalent task in minutes per week (MET – min/week).

Language of publication

No restrictions were set on the language of publication, although all search terms used were in English.

Search methods

On 25 July 2022 a search was performed for published interventions in four different bibliographic databases: Web of Science; Embase; Medline; The Cochrane Central Register of Controlled Trials. The search strategy was presented in Appendix 1, Table S1. Our search included a combination of Medical Subject Headings (MeSH) and key words describing the population and intervention. An initial limited search of Medline (Ovid) was undertaken to identify articles on the topic and key terms used to describe the articles. Those key terms were then matched to search for MeSH terms and key words were used to search for titles, abstracts and keywords in Medline.

Study selection

Search results were imported into EndNote 20, where duplicates were eliminated before a title screening. The two-part study selection process involved a title and abstract review, followed by a full-text review. Two reviewers (KHN and LB) independently assessed titles/abstracts, reaching a consensus on full texts. Potentially eligible articles were assessed by the same reviewers, with a third reviewer (HHN) available for consultation if needed.

Data extraction

Data were extracted from selected articles, encompassing general information (country, authors, publication details), participant details, intervention specifics, data analysis methodology, outcome measures, and results. Continuous outcomes, such as physical fitness, fatigue, and quality-of-life scores, were extracted with means, standard deviations, and participant numbers assessed. Two researchers (KHN and LB) independently extracted data, verified by a third researcher (HHN) for accuracy. Disagreements could be resolved by consulting a fourth researcher (HDT).

Assessment of risk of bias

Two reviewers independently conducted a risk of bias assessment. According to a recommendation from the National Institutes of Health (NIH),²¹ the Cochrane Handbook for Systematic Reviews of Interventions (Risk of Bias 2; Rob 2) tool and the Risk of Bias In Non-randomized Studies of Interventions (ROBINS-I) tool were used.^22,23

Data synthesis

The meta-analyses were conducted using R software (version 4.2.0) using meta package. Given methodological heterogeneity in target population, intervention implementation, and outcome measurement, the random-effects model was employed to address variations across studies.²⁴ Statistical heterogeneity was assessed using the I² statistic, which quantifies the inconsistency between separate studies. An I² value of up to 25% was considered as predictive of low heterogeneity, an I² value of around 50% as moderate heterogeneity, and an I² value of 75% or more as considerable heterogeneity.²⁵

Results

Study selection

Figure 1 depicts the flow diagram of this study’s selection process. The bibliometric search found 5406 studies. After removing 1881 duplicates, 3525 articles underwent title and abstract screening. Of these, 3476 were excluded for not meeting criteria, leaving 49 for full-text screening. Twenty-six studies were excluded for not meeting eligibility criteria. In the end, the review included 23 experimental studies reporting outcomes from 21 interventions. There were two pairs of articles reporting the same interventions.^7,26–30

Figure 1.

Flow diagram of this study’s selection process.

Characteristics of participants

The main characteristics of the 1578 female cancer survivors (age ≥ 18 years) recruited in the 23 included articles (21 interventions)^{26–29,31–49} are presented in Table 1. Study participants averaged 46–66 years old. The most frequently reported cancers were those of the breast (15 interventions), followed by endometrium^33,47 and ovaries^35,36 (2 interventions each). There was one intervention focused on lung cancer,³¹ while one intervention included unspecified cancer survivors.³⁷ Participants received different regimens of cancer treatment,such as surgery,^{26,32–34,39,40,46,49} chemotherapy,^{25–27,31–33,35,36,39,40,44,46,49} radiation,^{26,32–34,39,40,44,46,48,49} or hormonal therapies.³⁹ Participants completed cancer therapy 5 weeks to 44 months before starting an exercise intervention. The drop-out rates associated with included interventions varied from 0 to 44.6%, with 7 studies having excellent attendance^{31,35,38,41,43,47} and,⁴⁹ while 6 studies had drop-out rates under 10%.^{26,28,32,34,39,45} The reasons for the lost-to-follow-up included health problems or poor physical conditions, time constraints to continue the study, and unspecified reasons. The characteristics of the control groups are detailed in Table 2. Randomization selected control groups in RCTs. Control group individuals were encouraged to keep their usual physical activity or follow the routine guidelines given to all breast cancer patients treated at their treatment facility. Importantly, they did not participate in physical activity treatments. In single-arm studies, participants’ outcomes were examined before and after the intervention.

Table 1.

Characteristics of participants among 21 selected interventions.

	Paper name	Study design	Cancer types	No. of participants	No of participants at the end of study	Dropout rate (%)	Mean age	BMI	% White	Can er sta ge^a	Time since treatment^b	Time since treatment^c	Chemotherapy	Radiation	Surgery
1	A. J. Hoffman, 2014	Single arm	Lung cancer	7	7	0	65	NA	NA	1	1.25	1	5
2	M. Baruth 2015	RCT	Breast cancer	32	30	6	57.4	28.15	75	1	5.125	2	23.7	25	32
3	N. Galiano-Castillo 2016	Single arm	Endometrial cancer	63			58.2	34.4	88.8889	1	27.6	1	7	21	35
4	N. Galiano-Castillo 2016	RCT	Breast cancer	81	76	6	48.3	NA	NA	1	NA	2	4	4	81
5	N. Galiano-Castillo 2016	RCT	Breast cancer	81	76	6	48.3	NA	NA	1	NA		73	NA	NA
6	I. M. Lahart 2016	RCT	Breast cancer	80	70	12.5	53.6	Cat	97.5	1	9.5	2	NA	NA	NA
7	I. M. Lahart 2016	RCT (32 in 80 participants)	Breast cancer	80	29	9.4	52.3	NA	NA	1	2.725	1	NA	NA	NA
8	K. E. Uhm 2018	Quasi-randomized controlled trial	Breast cancer	356	339	5	50.3	NA	NA	Stage 4	19.5	3		242	532
9	Z. Xiaochen 2017	Single arm	Ovarian cancer	10	10	0	63	24	100	State 3–4	44	3	5	NA	NA
10	Y. Zhou 2017	RCT	Ovarian cancer	144	113	21.5	57.3	29	95.1389	State 3–4	20.4	3	134	NA	NA
11	L. J. Frensham 2018	Quasi-randomized controlled trial	Cancer survivors	102	91	10.8	65.6	NA	95.6044	NA	NA	NA	NA	NA	NA
12	S. J. Foulkes 2019	Non-randomized trial	Breast cancer	17	17	0	51.6	25.9	NA	NA	NA	NA	NA	NA	NA
13	J. McNeil 2019	RCT	Breast cancer	45	41	8.9	58.7	NA	80	1	35.3	3	36	36	45
14	Z. Pope 2019	Single arm	Breast cancer	10	8	20	45.8	NA	90	1	NA	3 months and 5 years	3	1	10
15	E. Grazioli 2020	Case study	Breast cancer	2	2	0	48.5	NA	NA	NA	NA		NA	NA	NA
16	F. Holtdirk 2021	RCT	Breast cancer	363	306	16	49.93	NA	NA	NA	NA	1	NA	NA	NA
17	V. Natalucci 2021	RCT	Breast cancer	30	30	0	52.3	26	NA	1	10.4	2	NA	NA	NA
18	Y. Qiao 2021	RCT	Breast cancer	56	41	26.8	54.9	NA	80.4878	NA	22.8	3	30	27	NA
19	Z. Gao 2022	RCT	Breast cancer	36	35	3	56.17	27.14	88.5714	NA	NA	1	NA	NA	NA
20	S. Gomaa 2022	Single arm	Breast cancer	39	21	44.6	58.18	NA	76.9231	1	NA	1	23	27	39
21	J. S. Gorzelitz 2022	RCT	Endometrial cancer	40	40	0	60.9	NA	NA	1	NA	2	NA	NA	NA
22	E. Ochi 2022	Single arm	Breast cancer	50	44	22	48.5	20.95	NA	Stage 1–2	5.5	2	NA	23	NA
23	L. Sagarra-Romero 2022	Quasi-randomized controlled trial	Breast cancer	15	15	0	55.53	NA	NA	1		> 6 months	10	10	15

^aCancer stage: sta ge I, II, III.

^bmonths.

^cTime since treatment, 1: < 6 month, 2: < 12 month, 3: over 1 year).

Table 2.

Characteristics of control group.

	Paper name	Study design	Control group
1	A. J. Hoffman 2014	Single arm	Before-After
2	M. Baruth 2015	RCT	Maintain their usual physical activity levels
3	N. Galiano-Castillo 2016	Single arm	Before-After
4	N. Galiano-Castillo 2016	RCT	Basic recommendations (written format) for exercise
5	N. Galiano-Castillo 2016	RCT	Basic recommendations (written format) for exercise
6	I. M. Lahart 2016	RCT	Standard information regarding PA (i.e. current recommended PA guidelines), as provided to all breast cancer patients treated at the site
7		RCT (32 in 80 participants)
8	K. E. Uhm 2018	Quasi-randomized controlled trial	Conventional program
9	Z. Xiaochen 2017	Single arm	Before-After
10	Y. Zhou 2017	RCT	Conventional program
11	L. J. Frensham 2018	Quasi-randomized controlled trial	Conventional program
12	S. J. Foulkes 2019	Non-randomized trial	Standard medical care throughout the trial
13	J. McNeil 2019	RCT	Maintain their baseline PA levels and did not receive any aspect of the PA interventions
14	Z. Pope 2019	Single arm	Before-After
15	E. Grazioli 2020	Case study	Before-After
16	F. Holtdirk 2021	RCT	Conventional program
17	V. Natalucci 2021	RCT	Conventional program
18	Y. Qiao 2021	RCT	Conventional program
19	Z. Gao 2022	RCT	Conventional program
20	S. Gomaa 2022	Single arm	Before-After
21	J. S. Gorzelitz 2022	RCT	Conventional program
22	E. Ochi 2022	Single arm	Before-After
23	L. Sagarra-Romero 2022	Quasi-randomized controlled trial	Conventional program

Characteristics of interventions

The main characteristics of the interventions are presented in Table 3. Notably, no interventions were conducted in low- and middle-income countries. They were conducted mostly in North America; the United States with 10 interventions,^{31–33,35,36,40,44–47} Canada with one³⁹ and Europe: Spain,^26,27,49 Italy^41,43 and the United Kingdom,²⁸ and also in South Korea³⁴ and Japan.⁴⁸

Table 3.

Characteristics of 21 selected interventions.

	Paper name	Delivery method	Unsupervised	Supervised	Developing guideline
1	A. J. Hoffman 2014	Video game: Nintendo Wii Fit Plus	x		Theory of Symptom Self-management
2	M. Baruth 2015	In-person training and calling follow-up	x		Active Choices model developed and refined by King and colleagues (Stanford Health Promotion Resource Center, 2005)
3	N. Galiano-Castillo 2016	Participants used a handheld computer (Hewlett-Packard iPAQ RX1950) to complete the EMA. Telephone counseling, print materials, and a pedometer were provided	x		ACSM guidelines for cancer survivors (American College of Sports Medicine, 2006)
4	N. Galiano-Castillo 2016	This program was implemented using the e-CUIDATE system (www.cuidateconnosotros.com)		x	American College of Sports Medicine for cancer survivors
5	N. Galiano-Castillo 2016			x	American College of Sports Medicine for cancer survivors
6	I. M. Lahart 2016	Participants were given a PA pack consisting of an information booklet and a DVD (previously developed by Breast Cancer Care)	x		Breast Cancer Care Guidelines for exercise testing and prescription. 9th ed. Philadelphia
7	I. M. Lahart 2016		x
8	K. E. Uhm 2018	Group 1: pedometer and brochure groups	x		NA
8	K. E. Uhm 2018	Group 2: newly developed smartphone exercise application called Smart After Care (BIT Computer Co., Ltd., Seoul, South Korea) and an InBodyBand pedometer (InBody Co., Ltd., Seoul, South Korea)	x		NA
9	Z. Xiaochen 2017	Walking program DVD (Sansone, 2014) and a physical activity tracker. The Fitbit Zip is a wireless physical activity tracker that synchronizes data to smartphones, as well as to a web-based data platform		x	NA
10	Y. Zhou 2017	Using the weekly 26-chapter book that we developed, the trainer provided weekly individualized counseling via telephone to motivate participants to exercise.	x		ACSM and the American Cancer Society
10	Y. Zhou 2017	The attention control arm received weekly phone calls from a WALC staff member, along with a 26-chapter book that contained exclusively ovarian cancer survivorship–related information	x		ACSM and the American Cancer Society
11	L. J. Frensham 2018	STRIDE website	NA		STRIDE is designed upon social cognitive theory which posits that physical activity interventions are most effective when they include intrapersonal mediators
12	S. J. Foulkes 2019	Participants were provided digital exercise equipment to perform exercise at home	x	x	NA
13	J. McNeil 2019	After baseline, participants were provided digital exercise equipment to perform exercise at home	x		American College of Sports Medicine position stand.
14	Z. Pope 2019	MapMyFitness and the study’s Facebook page, with breast cancer survivors encouraged to ask questions during the training session and at any time throughout the course of the intervention			Social Cognitive Theory
15	E. Grazioli 2020	Smartphone with WhatsApp video-calling			NA
16	F. Holtdirk 2021	Optimum was developed with a responsive web-design approach and can be accessed via a password-protected, secure (https-encrypted) website (https://optimune.broca.io) with any computer or smartphone equipped with a contemporary web browser			Cognitive behavior theory
17	V. Natalucci 2021	Both in-person and online supervision training		x	NA
18	Y. Qiao 2021	Both in-person and online supervision training		x	American College of Sports Medicine (ACSM)’s recommendation
19	Z. Gao 2022	7 Minute Chi app.	x		NA
19	Z. Gao 2022	Private Facebook group	x		NA
20	S. Gomaa 2022	Zoom platform.		x	NA
20	S. Gomaa 2022	Facebook group: consists of instructional videos matching the progress of weekly classes for at-home practice and promotes peer support in tai chi engagement		x	NA
21	J. S. Gorzelitz 2022	In-person training and providing materials	x		NA
21	J. S. Gorzelitz 2022	Facebook group	x		NA
22	E. Ochi 2022	Smartphone app	x		NA
23	L. Sagarra-Romero 2022	Zoom platform.	x	x	WHO physical activity recommendations
23	L. Sagarra-Romero 2022	Exercises were developed with body weight, elastic bands, and home-available materials (such as small plastic bottles of water or milk, or shopping bags)	x	x	WHO physical activity recommendations

Guidelines

Six interventions^{26,28,33,36,39,44} used the American College of Sports Medicine recommendations.⁵⁰ In which, participants were first evaluated pre-exercise, which comprised taking their medical history, performing a physical examination and fitness testing. Based on an individual’s pre-exercise evaluation results, an appropriate exercise program was prepared to include components (warm-up, stimulus or conditioning phase, recreational activities, and cool-down) and exercise prescription (mode, duration and intensity of exercise). An intervention for lung cancer survivors applied World Health Organization guidelines,⁵¹ which provide details for the intensity of physical activities among different age groups and specific population groups.³¹

Supervision

Exercise classes were delivered by a coach to participants and were conducted either through in-person home visits or using online meeting platforms such as Zoom or Skype.^46,49 In contrast, for other interventions,^{26,28,31,33,34,37,40–42,45,46,48} the coach only guided exercise at the baseline appointment, then participants did it at home without program staff monitoring. Smartphone apps and websites were employed in this case. These networks offered peer support, health education, and weekly class recordings for at-home performance.^{33,37,40,41,45,46,48} Other digital supplementary materials such as video games or DVD videos were provided for participants to support them to practice exercises at home.^{26,28,31,34,42} For example, the Nintendo Wii Fit Plus video game was used to provide a variety of video exercise routines such as yoga, strength training, aerobics, and balance in a mini games format, which participants could practice at home.³¹

Interventions using standard guidelines and programs supervised by trained fitness coaches improved participant results more than those without. The study of Galiano et al.^26,27 (using the American College of Sport Medicine for cancer survivor guidelines) reported a two-fold increase in quality-of-life score following intervention compared to a study that did not apply any standard physical exercise guidelines.³² Furthermore, the mean difference in 6MWT results between treatment and control groups in a supervised intervention²⁶ was higher than that for either of two comparable but unsupervised interventions^37,48 (Figure 4).

Digital technology was also utilized to supplement the measurement of physical performance. For instance, the following digital devices measured anthropometric indices of participants: (1) Tanita UM-108 to measure body weight, body mass index (BMI), body fat percentage, water level of body, heart rate^37,40; (2) Pedometer to measure the number of steps and provide an estimate of distance walked^37,44; (3) digital dynamometer to measure handgrip strength⁴¹; and (4) the Actigraph GT3X + accelerometer to measure overall level and intensity-specific level of physical activity.³⁹

Characteristics of exercises

Table 3 summarizes exercise programs that were included in this study. In terms of physical exercise type, programs can be of a single type or a combination of aerobic exercises, resistance exercises, and tai chi. The majority of included studies combined exercise types, with a mix of aerobic and resistance exercises being the most common format.^26,34,38,47 Only 4 studies involved walking^26,32,37,40 and 2 interventions offered tai chi exercise as a single type of exercise.^45,46 As an intervention, tai chi appears to achieve less significant improvement to physical functioning of female cancer survivors than do either aerobic exercise or resistance training. Pope et al.⁴⁰ (aerobic intervention) reported greater impact on physical functioning among the intervention group compared to the result of Gao et al.⁴⁵ (tai chi intervention), wherein both studies used the same questionnaire (Patient Reported Outcome Measurement Information System).

Duration, frequency, and intensity

Various duration, frequency, and intensity levels were used across exercise types and studies. Specifically, 12 interventions were delivered from 8–16 weeks, with a frequency of 2–7 days per week and an intensity of 7–150 min each day.^{26,31,32,34,38,40,41,43,45–49} In addition, 6 interventions were implemented for longer than 16 weeks, with a frequency of 3–5 days per week and at a moderate intensity (activities ranging between 3 and 6 MET⁵⁰)^{28,29,33,35,36,44} (see Table 4). Normally, exercise programs included several sessions, for each of which the intensity gradually increased from start to finish. For example, an online home-based exercise intervention developed for breast cancer survivors during COVID-19 lockdown spent the first 2 weeks focusing on exercise familiarization. Thereafter, each session was divided into three successive periods: warm-up, exercise, cool-down.^26,49 Exercise undertaken at vigorous intensity showed more significant improvement to participant outcomes in comparison to interventions with lower intensity exercise. The study of Hoffman et al.,⁵² conducted over 16 weeks with daily sessions lasting 30 min each, reported a higher daily step count compared to studies of shorter duration, less frequency, and lower intensity^40,53 (Table 4).

Table 4.

Characteristics of exercises that were used in 21 selected interventions.

	Paper	Aerobic exercises	Resistance exercises	Tai chi	Type of exercise	Individual based designed	Duration	Frequency	Intensity	Physical fitness	Fatigue	Quality of life	Outcome
1	A. J. Hoffman 2014	1	0	0	Walking and balance exercises		Phase 1: 6 weeks	Everyday	30 mins/day	0	1		Feasibility, acceptability, and preliminary efficacy of the intervention. Cancer-related Fatigue
1	A. J. Hoffman 2014	1	0	0	Walking and balance exercises		Phase 2: 10 weeks	Everyday	30 mins/day	0	1
2	M. Baruth 2015	1	0	0	Walking		12 weeks	Week 1: 3 days/week	Week 1: 20 mins (moderate)	1	1	1	QoL, Fatigue, Physical activities (MET)
2	M. Baruth 2015	1	0	0	Walking		12 weeks	Week 5: 5 days/week	Week 5: 30-40 mins (high)	1	1	1	QoL, Fatigue, Physical activities (MET)
3	N. Galiano-Castillo 2016	1	0	0	Walking		6 months	5 days or more per week	Moderate intensity walking	1	0	1	Physical activity, QoL, sexual health, psychological distress
3	N. Galiano-Castillo 2016	1	0	0	Walking		6 months	5 days or more per week	30 mins per day	1	0	1
4	N. Galiano-Castillo 2016	1	1	0	Resistance and aerobic exercise training		8 weeks	3 sessions/week	90 mins/session	1	1	1	QoL, Pain, Isometric handgrip strength, Isometric abdominal strength, Isometric back strength, lower body strength, fatigue
5	N. Galiano-Castillo 2016	1	1	0	Resistance and aerobic exercise training		8 weeks	3 sessions/week	90 mins/session	1	1	1	6-min walk test
6	I. M. Lahart 2016	1	0	0	General physical activities such as walking		6 months	3 to 5 days	30 mins of moderate intensity PA	1	0	1	PA (MET-min∙wk−1 ). Body composition (body fat %) HRQoL
6	I. M. Lahart 2016	1	0	0	General physical activities such as walking		6 months	per week	30 mins of moderate intensity PA	1	0	1	PA (MET-min∙wk−1 ). Body composition (body fat %) HRQoL
7			0	0	Walking and aerobic					0	0	0	, cardiorespiratory fitness (V_O2max). Secondary outcomes included exercise tolerance test variables, time to exhaustion (TTE) and peak heart rate (HR peak); resting cardiovascular measures, resting HR (RHR), systolic and diastolic blood pressure (BP), and mean arterial pressure (MAP); self-reported physical activity outcomes via International PA Questionnaire
8	K. E. Uhm 2018	1	1	0	Aerobic and resistance exercises		12 weeks		150 mins of moderate intensity aerobic exercise every week	1	0	0	Body mass index (BMI), systolic blood pressure (SBP), diastolic blood pressure (DBP), pulse rate, arm circumference, the presence of hand edema, upper, and lower extremity strength, and cardiovascular respiratory endurance. Satisfaction questionnaire survey
									90 mins of exercise per week
									Resistance exercise: two sets of 10 repetitions for
									each exercise was instructed twice a week.
9	Z. Xiaochen 2017	1	0	0	Aerobic		26 weeks		Moderate intensity	1	0	0	Physical activity
					exercise:				225 mins/week
					walking; bicycling, exercise on cardio-
					vascular training equipment
10	Y. Zhou 2017	1	0	0	Aerobic		6 months		Moderate	1	1	1	Intervention Adherence HRQOL and CRF Scores
10	Y. Zhou 2017	1	0	0	Aerobic		6 months		150 mins/week	1	1	1	Physical activity
11	L. J. Frensham 2018	1	0	0	Walking		NA	NA	NA	1	0	1	Anthropometry, physiological measures, diet quality, mediators of physical activity change, functional status and quality of life
12	S. J. Foulkes 2019	1	1	0	Aerobic and resistance exercises	1	12 weeks	2 sessions/week	60 mins/session (30 mins aerobic and 30 mins resistance training)	1	0	0	VO2 max, Echocardiography, CMR imaging, …
13	J. McNeil 2019	1	0	0	Walking and casual activity	1	12 weeks	NA	1 group: moderate	1	0	0	Sleep time, physical activity
13	J. McNeil 2019	1	0	0	Walking and casual activity	1	12 weeks	NA	1 group: high intensity	1	0	0	Sleep time, physical activity
14	Z. Pope 2019	1	0	0	Walking	0	10 weeks	5 days/week	Three 10-mins bouts of PA	1	0	0	Physical activity, weight or body composition, cardiovascular fitness, psychosocial constructs, and quality of life
15	E. Grazioli 2020	1	0	0	Aerobic exercises	0	16 weeks	2 lessons/week	60 mins/session The work intensity: 50-70% of the maximum heart rate (HRmax),	0	1	1	Quality of Life, Psychological and Fatigue Assessments
15	E. Grazioli 2020	1	0	0	Aerobic exercises	0	16 weeks	2 lessons/week	evaluated by a heart rate monitor provided to the patients by the training staff.	0	1	1	Functional Evaluations
16	F. Holtdirk 2021	1	0	0		1			NA	1	0	1	Quality of Life, Physical activity
17	V. Natalucci 2021	1	0	0	Aerobic:	0	3 months	3 days/week	Aerobic exercise training	1	0	0	Anthropometrics, Mediterranean diet adherence, physical activity level (PAL), cardiorespiratory fitness (VO2max), echocardiographic parameters, heart rate variability (average standard deviation of NN intervals (ASDNN/5 mins) and 24 h very- (24 hVLF) and low-frequency (24 hLF)), and metabolic, endocrine, and inflammatory serum biomarkers (glycemia, insulin resistance, progesterone, testosterone, and high-sensitivity C-reactive protein (hs-CRP)
					walking, running, or cycling				program having progressive increases in exercise intensity (from 40% to 70% of heart
					walking, running, or cycling				rate reserve) and duration (from 20 to 60 mins).
18	Y. Qiao 2021	1	0	0		1	30 weeks		150 mins per week of moderate intensity or 75 mins per week of vigorous intensity	0	1	0	Perceived physical fatigability, self-administered Pittsburgh Fatigability Scale (PFS)
									aerobic activity and at least 2 days/week of moderate
									intensity resistance training for every major muscle group
19	Z. Gao 2022	0	0	1	Tai chi	0	12 weeks	3 times/day & 5 days/week	7 mins tai chi/time	1	1	1	Quality of Life, Physical activity
20	S. Gomaa 2022	0	0	1	Tai chi	0	12 weeks	3 times/week	40 - 60 mins mins/each	1	0	0	Quality of Life, fatigue
21	J. S. Gorzelitz 2022	0	1	0	Resistance exercise training	0	10 weeks	2 sessions/week	Participants reported 19.2 sessions over the 10-week period out of the expected 20 sessions	1	0	1	Lean mass, functional fitness assessments,
21	J. S. Gorzelitz 2022	0	1	0	Resistance exercise training	0	10 weeks	2 sessions/week		1	0	1	blood biomarkers, and quality of life outcomes
22	E. Ochi 2022	1	0	0	Body weight exercises		12 weeks	3 times/week	Intensity of the exercise performed (18 ± 2)	1	0	0	Body weight
22	E. Ochi 2022	1	0	0	Body weight exercises		12 weeks	3 times/week	Intensity of the exercise performed (18 ± 2)	1	0	0	exercises, fatigue
23	L. Sagarra-Romero 2022	1	0	0	Aerobic	0	16 weeks		60 mins for each day & 4 days per week = 2 days functional supervised training sessions + 2 days aerobic unsupervised training sessions	1	0	0	Cardiorespiratory fitness
23	L. Sagarra-Romero 2022	1	0	0	Aerobic	0	= 2 weeks for familiarization and 14 weeks of exercises			1	0	0	walking speed (brisk walking test), and agility (8-foot up-and-go test)

Risk of bias assessment

“Risk of bias” assessments for RCTs and non-RCTs are depicted in separate Figures 2 and 3, respectively. Baruth et al.’s and Lahart et al.’s RCTs showed a high risk of bias due to selective reporting and incomplete outcome, respectively.^28,32 Lack of allocation concealment, random sequence generation, and participant and staff blinding raised problems in seven research.^{36,39,42–45,47} ROBINS-I showed that Grazioli et al’s study had a high risk of bias due to recruiting issues and potential intervention variations.⁴¹ Pope et al. had a serious risk of bias due to important concerns regarding confounding.⁴⁰ The remaining nine studies were rated as having moderate risk of bias.^{31,33–35,37,38,46,48,49}

Figure 2.

Risk of bias assessment for RCTs.

Figure 3.

Risk of bias assessment for non RCTs.

Effects of interventions

Fatigue

Table 5 summarizes six studies that reported a reduction in cancer-related fatigue,^{26,31,32,36,42,47} three of which showed a statistically significant improvement. Interventions associated with significant decrease of fatigue were both provided via internet, including an 8-week combined exercise training program available online/via internet and a website named Optimum.^26,42 Overall, self-reported fatigue decreased after intervention with various questionnaires. First, Baruth et al.³² and Zhou et al.³⁶ assessed their participants’ fatigue using the Functional Assessment of Cancer Therapies – Fatigue (40 items) (FACT) questionnaire.⁵⁴ Both studies found that intervention groups experienced a 4- to 32-point reduction in fatigue.^32,36 In the study by Gorzelitz and colleagues,⁴⁷ a modified version of the FACT questionnaire, Functional Assessment of Cancer Therapy – Endometrial (FACT-En), was used, which added three endometrial cancer-related items to the original one. While the mean fatigue score decreased from 50.5 (SD = 8.9) to 48.1 (SD = 2.9) after 10 weeks of exercise in the intervention group, this difference was not statistically significant.⁴⁷ Other studies using cancer-related fatigue severity scores, Piper Fatigue Scale – revised 22 items, and cancer-related fatigue scores corroborated a reduction in fatigue following intervention: a 0.4- to 2-point reduction.^26,32,42

Table 5.

Description cancer-related fatigue of participants among selected interventions.

	Study	Tool	Effect size	Statistically significant
1	A. J. Hoffman 2014	Cancer-related fatigue severity scores	Before: n = 7, mean = 3.3 (SD = 1.7)	NA
1	A. J. Hoffman 2014	Cancer-related fatigue severity scores	After 16 weeks: n = 5, mean = 1.32 (SD = 1.71)	NA
2	M. Baruth 2015	Functional Assessment	Intervention	NA
			of Cancer Therapy: Fatigue (FACT-Fatigue)
			Before: n = 20, mean = 14.5 (SD is missing)
			After 12 weeks: n = 20, mean = 10.0 (SD is missing)
			Control
			Before: n = 12, mean = 15.3 (SD is missing)
			After 16 weeks: n = 12, mean = 14.1 (SD is missing)
4	N. Galiano-Castillo 2016	Piper Fatigue Scale-revised 22 items	Cohen’s d baseline 4 score, after 6 weeks: 2.1 score (Only in Figure).	p < 0.001
4	N. Galiano-Castillo 2016	Piper Fatigue Scale-revised 22 items	The ES was large after intervention (d =-0.89 CI95% - 1.30; -0.48) and moderate after follow-up (d =-0.74 CI95% -1.19; -0.29)	p < 0.001
9	Y. Zhou 2017	Functional Assessment of Cancer Therapy–Fatigue (FACT-F)	Intervention	p = 0.06
			Before: n = 74, mean = 36.7 (11.1)
			After 6 months: n = NA, mean = 4.0 (1.8 to 6.2)
			Control
			Before: n = 70, mean = 35.8 (10.8)
			After 6 months: n = NA, mean = 1.2 (–1.1 to 3.5) 95%CI
16	F. Holtdirk 2021	Cancer-related fatigue (Brief Fatigue Inventory, 9 items, BFI-9	Intervention	0.23 (0.02–0.44) (p < 0.05)
			Before: n = 181, mean = 4.41 (2.28)
			After 3 months: n = 141, mean = 4.09 (2.16) cohen-d
			Control
			Before: n = 182, mean = 4.71(2.24)
			After 3 months: n = 165, mean = 4.61 (2.28)
21	J. S. Gorzelitz 2022	Functional Assessment of Cancer Therapies (FACT-En) is a 43-item scale	Intervention	p > 0.05
			Before: n = 20, mean = 50.5 (8.9)
			After 10 weeks: n = 20, mean = 48.1 (2.9)
			Control
			Before: n = 20, mean = 50.9 (3.1)
			After 10 weeks: n = 20, mean = 48.7 (2.9)

Quality-of-life

The quality-of-life outcome was reported in eight interventions,^{26,32,34,37,40,42,45} with two studies indicating a statistically significant improvement, as presented in Table 6. A resistance and aerobic training program provided by the e-CUIDATE system showed significant increase of quality-of-life in the intervention group, from 59.19 to 72.86 after 6 months.²⁶ The European Organization for Research and Treatment of Cancer Quality-of-life Questionnaire Core 30 (EORTC QLQ-C30) was completed by participants in the studies of Galiano-Castillo et al.²⁶ and Uhm et al.³⁴ The former reported a 13- point increase in quality-of-life score [global health domain] post-intervention,²⁶ which is double the index attained in the latter study.³⁴ Due to the difference in study design, meta-analysis was not applied. Several other questionnaires to measure the quality-of-life for a general population were used. These included the Medical Outcomes 36-item Short Form Health Survey (SF-36),^32,37 the Patient Reported Outcome Measurement Information System,^40,45 and the WHO Quality-of-life Questionnaire.⁴² All these studies reveal, to a greater or lesser extent, an improved quality-of-life among cancer survivors following treatment. Using the WHOQOLBREF questionnaire, an intervention delivered via a responsive website in the study of Holtdirk and colleagues significantly improved quality-of-life among participants in the intervention group after 3 months, from 66.11 before intervention to 69.44 afterwards.⁴² Baruth et al. and Frensham et al. reported an increase of quality-of-life score of 6.7 and 0.5, respectively, using the SF36 questionnaire.^32,37

Table 6.

Description Quality of life of participants among selected interventions.

	Study	Tool	Effect size	Statistically significant
2	M. Baruth 2015	The Medical	Intervention
		Outcomes 36-item Short Form Health Survey (SF-36)	Before: n = 20, mean = 68.6 (SD is missing)
		(physical functioning)	After 12 weeks: n = 20, mean = 75.3
			(SD is missing)
			Control
			Before: n = 12, mean = 76.6 (SD is missing)
			After 16 weeks: n = 12, mean = 73.5
			(SD is missing)	p = .06
4	N. Galiano-Castillo 2016	EORTC QLQ-C30	Intervention	p < .05
		Global health status	Before: n = 39, mean = 59.19 ± 21.01
			After 6 months: n=, mean = 72.86 ± 19.93
			Control
			Before: n = 37, mean = 54.73 ± 22.53
			After 6 months: n =, mean = 57.21 ± 21.71
7	K. E. Uhm 2018	European Organization for Research and Treatment	Intervention	p-value after intervention < 0.05,
		of Cancer Quality of Life Questionnaire Core 30 (EORTC QLQ-C30)	Before: n = 179, mean = 62.8 ± 20.1	but p-value between 2 groups > 0.05
		Global health status	After 12 weeks: n = 167, mean = 68.0 ± 20.9
			Control
			Before: n = 177, mean = 58.7 ± 22.6
			After 12 weeks: n = 173, mean = 64.0 ± 23.0
10	L. J. Frensham 2018	Short Form	Intervention	p > .05
		(36) Health Survey’ (SF-36v2)	Before: n = 74, mean = 79.5 (15.8)
		Physical functioning	After 24 weeks: n =, mean = 80.9 (17.5)
			Control
			Before: n = 70, mean = 74.5 (20.5)
			After 24 weeks: n=, mean = 75.1 (17.5)
14	Z. Pope 2019	The Patient Reported Outcome Measurement	Before: n = 10, mean = 1.22 ± 0.58	NA
		Information System Physical functioning	After 10 weeks: n=, mean = 1.06 ± 0.12
		valuated on a 5-point Likert-type scale	After 10 weeks: n=, mean = 1.06 ± 0.12
16	F. Holtdirk 2021	World Health Organization Quality of Life Questionnaire (WHOQOLBREF, 26 items	Intervention	p < .05
			Before: n = 181, mean = 66.11 (14.22)
			After 3 months: n = 141, mean = 69.44 (13.63) Cohen-d
			Control
			Before: n = 182, mean = 65.05 (12.86)
			After 3 months: n = 165, mean = 65.7 (13.56)
19	Z. Gao 2022	Patient Reported Outcome Measurement Information	Intervention	NA
		System	Before: n = 20, mean = 1.51 ± 0.36
		Physical health	After 3 months: n = 20, mean = 1.38 ± 0.34
			Control
			Before: n = 16, mean = 1.64 ± 0.65 After 3 months: n = 15, mean = 1.47 ± 0.49

Physical activity

Table 7 shows each physical activity performance before and after intervention for included interventions. The effects of physical activity were measured by various methods, including 6MWT,^27,37,48 MET–min/week,^28,34,42,43 minutes spent doing exercise per week,^{31,33,35,36,38–40} and number of steps walked per week.^31,35,40 These methods could be assessed by self-reporting,³¹ or by using a digital measure such as an Actigraph GT3X + accelerometer,³⁵ or a pedometer.³⁷ Overall, all physical performance of participants was improved across different exercise interventions. Three trials found six-minute walking distance increases in the intervention group compared to the control^27,37,48 (MD = 30.13 meters, 95% CI -102.41 – 162.67; Tau² = 1977.4026; Chi² = 8.32, df = 2 (p = 0.02); I² = 76%; substantial heterogeneity; Figure 4)). Four studies measured the number of MET minutes per week^28,34,42,43 and indicated increases in number of the MET minutes per week in the intervention group versus the control (MD = 375.81 MET minutes, 95% CI -678.43 – 1430.04; Tau² = 108106.6544; Chi² = 5.00, df = 2 (p = 0.08); I² = 60%; substantial heterogeneity; Figure 5). Three studies measured the number of steps walked per week^31,35,40 and showed decreases in the number of steps walked per week in the intervention group versus the control (MD 1734.45, 95% CI -2301.19 – 5770.08; Tau² = 0; Chi² = 0.32, df = 1 (p = 0.57); I² = 0%; no heterogeneity; Figure 6). Five studies measured cardiorespiratory fitness (VO₂ max; ml. kg/min) as a marker of physical health attainment^{29,38,39,43,48} and showed increases in maximal oxygen consumption for the intervention group versus the control (MD = 1.29 VO₂ max (ml. kg/min), 95% CI -1.80 – 4.37; Tau² = 0; Chi² = 1.50, df = 2 (p = 0.47); I² = 0%; no heterogeneity; Figure 7).

Table 7.

Description Physical activity of participants among selected interventions.

	Study	Tool	Effect size
	Walking minute
6	N. Galiano-Castillo 2016	6-mins walk test (6MWT)	Intervention
			Before: n = 39, mean = 312.71 ± 207.91 (245.31 to 380.10)
			After 6 month: n = 38, mean = 419.62 ± 216.59 (349.41 to 489.83)
			Control
			Before: n = 37, mean = 281.73 ± 144.22 (233.64 to 329.81)
			After 6 month: n = 37, mean = 322.41 ± 145.61 (273.86 to 370.96)
10	L. J. Frensham 2018	6-mins walk test (6MWT)	Intervention
			Before: n = 46, mean = 530,4 (66.8)
			After 24 weeks: n = 46, mean = 565.3 (81.3)
			Control
			Before: n = 45, mean = 515.2 (81.3)
			After 24 weeks: n = 45, mean = 528.9 (102.6)
22	E. Ochi 2022	6-mins walk test (6MWT)	Intervention
			Before: n = 21, mean = 586 (43)
			After 12 weeks: n = 21, mean = 615 (42)
			Control
			Before: n = 23, mean = 603 (59)
			After 12 weeks: n = 23, mean = 631 (68)
5	I. M. Lahart 2016	International Physical Activity Questionnaire (IPAQ, 27 items) MET (metabolic equivalent task, in minutes per week)	Intervention
			Before: n = 40, mean = 1354.74 (1073.44)
			After 6 month: n = 37, mean = 1899.01 (985.84)
			Control
			Before: n = 40, mean = 1974.60 (1478.57)
			After 6 months: n = 33, mean = 1968.19 (1133.87)
7	K. E. Uhm 2020	MET (metabolic equivalent task, in minutes per week)	Intervention
			Before: n = 179, mean = 2050.6 ± 2182.2
			After 12 weeks: n = 167, mean = 3026.9 ± 2489.5*
			Control
			Before: n =177, mean = 2091.5 ± 1811.2
			After 12 weeks: n = 173, mean = 2560.4 ± 2354.9*
16	F. Holtdirk 2021	MET (metabolic equivalent task, in minutes per week)	Intervention
		International Physical Activity Questionnaire (IPAQ, 27 items)	Before: n = 181, mean = 3749 (2867)
			After 3 months: n = 141, mean = 3967 (2456)
			Control
			Before: n = 182, mean = 3223 (2437)
			After 3 months: n = 165, mean = 3198 (2580)
17	V. Natalucci 2021	MET (metabolic equivalent task, in minutes per week)	Before: n = 30, mean = 647 ± 547
			After 9 weeks: n=, mean = 1043 ± 564
			p-value < 0.001
8	Z. Xiaochen 2018	Steps per day	Before: n = 10, mean = 948, IOR = 592–1074
8	Z. Xiaochen 2018	Steps per day	After 26 weeks: n = 8, mean = 1162 IOR = 652–1338 Cohen d = 0.417
1	A. J. Hoffman 2015	The mean number of steps taken per day	Before: n = 7, mean = 4650 [SD, 3105]
1	A. J. Hoffman 2015	The mean number of steps taken per day	After 16 weeks: n = 5, week 14 at 7687 (SD, 5127) steps
14	Z. Pope 2020	Average daily steps	Before: n = 10, mean = 4,930 ± 1,376
14	Z. Pope 2020	Average daily steps	After 10 weeks: n = 10, mean = 6,587 ± 1,22
1	A. J. Hoffman 2014	PERCEIVED SELF-EFFICACY FOR WALKING DURATION	Before: n = 7, mean = 96.4%
		The mean PSE for walking 30 minutes (not specific to walking with the Wii)	(SD, 6.1%)
			After 16 weeks: n = 5, mean = 99.4% (SD, 1.36%)
8	Z. Xiaochen 2017	— moderate- to vigorous-intensity physical activity in minutes per day	During the 26 weeks of exercise intervention, according to the accelerometer, moderate-intensity activity increased by 15 minutes per day (p-value = 0.05) Before: n = 10, mean = Average MVPA mean = 18, IOR = 9–30 After 26 weeks: n = 8, mean = mean = 38 IOR= 12–41 d = 0.491
3	N. Galiano-Castillo 2016	Exercise duration:	0.8- hour increase in their total hours of activity per week (SD = 10.9) from baseline to 6 months.
		(1) ecological momentary assessment (EMA)
		(2) Community Health Activities Model Program for Seniors (CHAMPS)
9	Y. Zhou 2017	Mean exercise/week	Before: n = 74, mean = 26.0 (644.2)
9	Y. Zhou 2017	Mean exercise/week	After 6 months: n = NA, mean = 166.0 6 66.1
12	S. J. Foulkes 2020	Active Australia physical activity survey	Exercise Qc was lower at 16 months compared with baseline and 4 months (p-value < 0.001), which was due to a blunted augmentation of SV during exercise (p-value = 0.032; a 14% reduction in peak SV), with no changes in heart rate response
13	J. McNeil 2019	Minute do physical activity per week (min/week)	Intervention: Higher intensity
			Before: n = 15, mean = 324 (96)
			After 24 weeks: n = 12, mean = 348 (84)
			Intervention: Lower intensity
			Before: n = 15, mean = 348 (90)
			After 24 weeks: n =,14 mean = 372 (84)
			Control
			Before: n = 15, mean = 306 (102)
			After 24 weeks: n = 12, mean = 318 (96)
14	Z. Pope 2019	Average daily MVPA duration (mins)	Before: n = 10, mean = 26.8 ± 13.8
14	Z. Pope 2019	Average daily MVPA duration (mins)	After 10 weeks: n = 10, mean = 29.4 ± 22.5
	Cardiorespiratory fitness (VO2 max)
11	I. M. Lahart 2016	Cardiorespiratory fitness (VO2 max) - (ml. kg/min)	Intervention
			Before: n = 16, mean = 24.1 ± 4.8
			After 6 month: n = 15, mean = 26.2 ± 5.3
			Control
			Before: n = 16, mean = 26.6 ± 4.4
			After 6 months: n = 14, mean = 26.6 ± 3.3
12	S. J. Foulkes 2019	Cardiorespiratory fitness (VO2 max) - (ml. kg/min)	At 4 months, AC was associated with 18% and 6% reductions in VO2peak in the UC and ET groups, respectively, which persisted at 16 months (UC, −16%; ET, −7%) and was not attenuated by ET (interaction, p = 0.10).
13	J. McNeil 2021	Cardiorespiratory fitness (VO2 max) - (ml. kg/min)	Intervention: Higher intensity
			Before: n = 15, mean = 16.6 (7.6)
			After 24 weeks: n = 12, mean = 20.4 (7.6)
			Intervention: Lower intensity
			Before: n = 15, mean = 19.8 (8.4)
			After 24 weeks: n =14, mean = 23.9 (8.0)
			Control
			Before: n = 15, mean = 16.0 (4.8)
			After 24 weeks: n = 12, mean = 18.0 (5.1)
17	V. Natalucci 2022	Cardiorespiratory fitness (VO2 max) - (ml. kg/min)	Before: n = 30, mean = 30.5 ± 5.8
			After 9 weeks: n = 30, mean = 33.4 ± 6.8
			p < 0.002
22	E. Ochi 2022	Cardiorespiratory fitness (VO2 max) - (ml. kg/min)	Intervention
			Before: n = 21, mean = 25.0 (3.0)
			After 12 weeks: n = 21, mean = 25.9 (2.8)
			Control
			Before: n = 23, mean = 24.9 (4.6)
			After 12 weeks: n = 23, mean = 24.1 (4.0)
	BMI
10	L. J. Frensham 2020	Body mass index (BMI)	Intervention Before: n = 46, mean = 30.4 (4.9). After 24 weeks: n = 45, mean = 31.1 (4.9)
10	L. J. Frensham 2020	Body mass index (BMI)	Control Before: n = 46, mean = 28.6 (4.5). After 24 weeks: n = 45, mean = 28.6 (4.4)
13	J. McNeil 2020	Body mass index (BMI)	Intervention: Higher intensity
			Before: n = 15, mean = 30.6 (7.9)
			After 24 weeks: n = 12, mean = 30.5 (7.6)
			Intervention : Lower intensity
			Before: n = 15, mean = 28.6 (5.0)
			After 24 weeks: n=14, mean = 28.3 (4.9)
			Control
			Before: n = 15, mean = 25.4 (3.5)
			After: 24 weeks: n = 12, mean = 25.9 (3.8)
17	V. Natalucci 2022	Body mass index (BMI)	Before: n = 30, mean = 26.0 ± 5.0
17	V. Natalucci 2022	Body mass index (BMI)	After 9 weeks: n= 30, mean = 25.5 ± 4.7 p-value = 0.035
23	L. Sagarra-Romero 2022	Body mass index (BMI)	Before: n = 21, mean = 25.5 (3.8)
23	L. Sagarra-Romero 2022	Body mass index (BMI)	After 9 weeks: n = 15, mean = 25.4 (3.9)
	Strength of the body
4	N. Galiano-Castillo 2017	Isometric abdominal strength (number of seconds)	F = 9.43; p < 0.001 (d = 0.66 CI95% -5.14; 6.45)
4	N. Galiano-Castillo 2018	Isometric back strength (number of seconds)	F = 18.78; p < 0.001 (d = 0.58 CI95% -1.78; 2.95)
4	N. Galiano-Castillo 2019	Lower body strength (number of seconds)	F = 6.87; p = 0.002 (d= -0.56 CI95% - 1.62; 0.50).
22	E. Ochi 2024	Grip strength (kgw)	Intervention
			Before: n = 21, mean = 22.4 (3.8)
			After 12 weeks: n = NA, mean = 23.2 (4.2)
			Control
			Before: n = 23, mean = 23.3 (4.2)
			After 12 weeks: n = NA, mean = 23.7 (3.4)
22	E. Ochi 2025	Chair stand test (s)	Intervention
			Before: n = 21, mean = 14.4 (3.2)
			After 12 weeks: n = NA, mean = 13.5 (2.9)
			Control
			Before: n = 23, mean = 13.7 (2.5)
			After 12 weeks: n = NA, mean = 13.3 (2.5)
23	L. Sagarra-Romero 2022	Chair stand test (s)	Before: n = 21, mean = 16.2 (4.7)
23	L. Sagarra-Romero 2022	Chair stand test (s)	After 9 weeks: n = 15, mean = 19.1 (3.5) p-value = 0.018
	Other measurement
7	K. E. Uhm 2018	Difference in upper arm circumference (mm)	Intervention
			Before: n = 179, mean = 3.5 ± 13.1
			After 12 weeks: n = 167, mean = 3.6 ± 11.6
			Control
			Before: n = 177, mean = 2.0 ± 11.1
			After 12 weeks: n = 173, mean = 1.6 ± 9.7
7	K. E. Uhm 2019	Difference in forearm circumference (mm)	Intervention
			Before: n = 179, mean = 1.3 ± 13.2
			After 12 weeks: n = 167, mean = 2.0 ± 11.0
			Control
			Before: n = 177, mean = 0.4 ± 9.8
			After 12 weeks: n = 173, mean = 1.0 ± 10.0
10	L. J. Frensham 2019	Waist girth (cm)	Intervention Before: n = 46, mean = 99.7 (13.1). After 24 weeks: n = 45, mean = 99.8 (13.5)
10	L. J. Frensham 2019	Waist girth (cm)	Control Before: n = 46, mean = 97.6 (13.3). After 24 weeks: n = 45, mean = 97.8 (13.1)
17	V. Natalucci 2024	Fat mass (%)	Before: n = 30, mean = 31.1 ± 6.3
17	V. Natalucci 2024	Fat mass (%)	After 9 weeks: n = 30, mean = 30.7 ± 5.9
23	L. Sagarra-Romero 2024	Fat mass (kg)	Before: n = 21, mean = 25.0 (6.7)
23	L. Sagarra-Romero 2024	Fat mass (kg)	After 9 weeks: n = 15, mean = 24.6 (6.6)
23	L. Sagarra-Romero 2025	Waist circumference (cm)	Before: n = 21, mean = 83.8 (9.5)
23	L. Sagarra-Romero 2025	Waist circumference (cm)	After 9 weeks: n = 15, mean = 82.4 (9.1)

Figure 4.

Meta-analysis results for the outcome of 6-min walk test.

Figure 5.

Meta-analysis results for the outcome of MET minutes per week.

Figure 6.

Meta-analysis results for the outcome of Steps walked per week.

Figure 7.

Meta-analysis results for the outcome of VO₂ max.

Discussion

Our review included 21 digital home-based exercise interventions among a combined total of 1578 participants. This review found that digital home-based exercise interventions improve quality-of-life (7 studies), physical function (20 studies), and cancer-related fatigue (6 studies) in female cancer survivors post treatment. Interventions involving standard guidelines, programs monitored by qualified exercise coaches, and intense exercise improved participant outcomes more than those without standard guidelines, programs without exercise coaches, and lower intensity exercise.

Our study suggested that following standard recommendations with trained coaches and engaging in intensive activity improved participant outcomes compared to programs without guidelines, without coach supervision, or lower-intensity workouts. This finding corroborates observations from previous studies.^55,56 Corres et al. found that a 16-week hypocaloric diet intervention improved cardiometabolic profiles, moving people from Metabolically Unhealthy Obesity (MUO) to MHO. Unfortunately, after 6 months without observation, participants’ biochemical profiles returned to MUO.⁵⁵ The importance of guidelines and coaches was previously shown by Gunderson et al. and Church et al.^57,58 Following standard guidelines with trained coaches provides better structured, safer workouts that enhance participant compliance and reduce confusion. This expert guidance to vigorous exercise stimulates greater physiological adaptations, fostering improved outcomes compared to when following less structured or lower-intensity programs.

The current study included three types of exercise intervention, namely aerobic exercise, resistance training and tai chi. Tai chi appears to improve physical functioning of female cancer survivors less than aerobic exercise or resistance training. The relative intensity of these two forms of exercise may help to explain this difference. While aerobic exercise and training often demand moderate to rigorous intensity, in the study of Gao et al.⁴⁵ tai chi required only minor intensity (7 min of balance exercises). Due to the marked heterogeneity of exercise interventions, the difference in impact on quality-of-life and fatigue were not comparable between the 21 reviewed interventions. Hence, there is a pressing need for further research to focus on creating exercise guidelines that support recommendations at different stages of the cancer care state. This should consider both current and anticipated side-effects of treatment, as well as the health status of an individual.

All 21 programs in this review were in high-income countries. While residents of low- and middle-income countries (LMIC) account for > 70% of all cancer-related deaths, there is a scarcity of supportive care interventions as well as a general lack of research to evaluate their efficacy in these resource-poor settings.⁵⁹ Therefore, more exercise interventions are required to be implemented in LMIC, which should also consider targeting women living with types of cancer other than breast cancer.

Limitations

This study has some inherent limitations. Many included papers were excluded from the meta-analysis due to outcome measurement and reporting heterogeneity. However, the interventions’ findings were narratively presented and compared. In addition, none of the RCTs included in this review had high-quality evidence due to a poor blinding process. The risk of bias for allocation concealment and blinding of participants and personnel is high in intervention groups because both patient and coach know about the intervention. Finally, interventions were often conducted in small trials. A consequence of this is that statistical power was compromised in several studies and also negatively impacted on the generalizability of the study.

Conclusion

Digital home-based exercise interventions improved side-effects of cancer treatment. Standard guidelines, exercise coaches, and intense exercise should be considered when designing exercise interventions for cancer survivors.

Supplemental Material

Supplemental Material - Digital home-based post-treatment exercise interventions for female cancer survivors: A systematic review and meta-analysis

Supplemental Material for Digital home-based post-treatment exercise interventions for female cancer survivors: A systematic review and meta-analysis in Huyen Thi Hoa Nguyen, Nguyen Thi Khanh Huyen, Linh Khanh Bui, Ha Thi Thuy Dinh, Andrew W Taylor-Robinson in Health Informatics Journal

Footnotes

Acknowledgements

This study received financial support from the VinUniversity 2022 Fast Track Funding Scheme. We are grateful to the College of Health Sciences, VinUniversity, Hanoi, Vietnam, for providing in-kind support.

Author contributions

HT, LK, TT, and AWT-R generated study aims and collaborated to conduct the study. HN designed the study, organized the search methodology, and data analysis. HT and HN performed data screening and data extracting. HT wrote the first draft of the manuscript. HN wrote the methodology sections of the manuscript. All authors contributed to the manuscript revision, and read, and approved the submitted version. AWT-Rrevised the writing style for the final version of the manuscript and HN is responsible to submit the manuscript.

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This work is funded by Vin University 2022 Fast Track Funding No. 22-0011-P0001.

ORCID iDs

Nguyen Thi Khanh Huyen

Andrew W Taylor-Robinson

Supplemental Material

Supplemental material for this article is available online.

References

Soerjomataram

Lortet-Tieulent

Parkin

, et al. Global burden of cancer in 2008: a systematic analysis of disability-adjusted life-years in 12 world regions. Lancet 2012; 380(9856): 1840.

American cancer Society . Global burden of cancer in women. Available from: https://www.cancer.org/content/dam/cancer-org/research/cancer-facts-and-statistics/global-cancer-facts-and-figures/global-burden-of-cancer-in-women.pdf

Cancer Research UK . UK cancer survival statistics: British medical Journal publishing group; 2022. Available from: https://www.cancerresearchuk.org/health-professional/cancer-statistics/incidence/age#heading-Zero

Jung

K-W

Park

Shin

, et al. Do female cancer patients display better survival rates compared with males? Analysis of the Korean National Registry data, 2005–2009. PLoS One 2012; 7(12): e52457.

Ferlay

Soerjomataram

Dikshit

, et al. Cancer incidence and mortality worldwide: sources, methods and major patterns in GLOBOCAN 2012. Int J Cancer 2015; 136(5): E359–E386.

Our world in Data . Cancer death rates are falling; five-year survival rates are rising 2019. Available from: https://ourworldindata.org/cancer-death-rates-are-falling-five-year-survival-rates-are-rising

Majeed

Gupta

. Adverse effects of radiation therapy. St. Petersburg, FL: StatPearls [Internet]: StatPearls Publishing, 2022.

Gelband

Jha

Sankaranarayanan

, et al.

Cancer: disease control priorities

(volume 3). 2015.

Palesh

Scheiber

Kesler

, et al. Management of side effects during and post-treatment in breast cancer survivors. Breast J 2018; 24(2): 167–175.

10.

Midtgaard

Hammer

Andersen

, et al. Cancer survivors’ experience of exercise-based cancer rehabilitation - a meta-synthesis of qualitative research. Acta Oncol 2015; 54(5): 609–617.

11.

Stout

Baima

Swisher

, et al. A systematic review of exercise systematic reviews in the cancer literature (2005-2017). PM&R. 2017;9(9):S347–S384.

12.

Segal

Zwaal

Green

, et al. Exercise for people with cancer: a systematic review. Curr Oncol 2017; 24(4): 290–315.

13.

Mercier

Savard

Bernard

. Exercise interventions to improve sleep in cancer patients: a systematic review and meta-analysis. Sleep Med Rev 2017; 36: 43–56.

14.

Takemura

Cheung

DST

Fong

DYT

, et al. Effectiveness of aerobic exercise and Tai Chi interventions on sleep quality in patients with advanced lung cancer: a randomized clinical trial. JAMA Oncol 2023; 10: 176–184.

15.

Amedisys. Benefits of Home Health Care. Available from: https://resources.amedisys.com/benefits-home-health-care

16.

Argent

Daly

Caulfield

. Patient involvement with home-based exercise programs: can connected health interventions influence adherence? JMIR Mhealth Uhealth 2018; 6(3): e47.

17.

Van Den Helder

Van Dronkelaar

Tieland

, et al. A digitally supported home-based exercise training program and dietary protein intervention for community dwelling older adults: protocol of the cluster randomised controlled VITAMIN trial. BMC Geriatr 2018; 18(1): 183.

18.

Moher

Shamseer

Clarke

, et al. Preferred reporting items for systematic review and meta-analysis protocols (PRISMA-P) 2015 statement. Syst Rev 2015; 4(1): 1–9.

19.

Eldawlatly

Alshehri

Alqahtani

, et al. Appearance of Population, Intervention, Comparison, and Outcome as research question in the title of articles of three different anesthesia journals: a pilot study. Saudi J Anaesth 2018; 12(2): 283–286.

20.

Soobiah

Cooper

Kishimoto

, et al. Identifying optimal frameworks to implement or evaluate digital health interventions: A Scoping Review Protocol. BMJ Open 2020; 10(8): e037643.

21.

NIH . Available from: https://www.nihlibrary.nih.gov/services/systematic-review-service/tools-resources?fbclid=IwAR04X3OEyFHPLSnu-nPCnnzAsZjU8D4ITHaFzR6-WJE9Osjq3DTfoLzMWK8

22.

Sterne

Savović

Page

, et al. RoB 2: a revised tool for assessing risk of bias in randomised trials. BMJ 2019; 366: l4898.

23.

Sterne

Hernán

Reeves

, et al. ROBINS-I: a tool for assessing risk of bias in non-randomised studies of interventions. BMJ 2016; 355: i4919.

24.

Harrer

. Doing Meta-Analysis in R: A hands-on guide. Available from: https://bookdown.org/MathiasHarrer/Doing_Meta_Analysis_in_R/pooling-es.html#rem

25.

Higgins

Thomas

Chandler

, et al. Cochrane handbook for systematic reviews of interventions. Hoboken, NJ: John Wiley & Sons, 2019.

26.

Galiano-Castillo

Cantarero-Villanueva

Fernandez-Lao

, et al. Telehealth system: a randomized controlled trial evaluating the impact of an internet-based exercise intervention on quality of life, pain, muscle strength, and fatigue in breast cancer survivors. Cancer 2016; 122(20): 3166–3174.

27.

Galiano-Castillo

Arroyo-Morales

Lozano-Lozano

, et al. Effect of an Internet-based telehealth system on functional capacity and cognition in breast cancer survivors: a secondary analysis of a randomized controlled trial. Support Care Cancer 2017; 25(11): 3551–3559.

28.

Lahart

Metsios

Nevill

, et al. Randomised controlled trial of a home-based physical activity intervention in breast cancer survivors. BMC Cancer 2016; 16: 1–14.

29.

Lahart

Carmichael

Nevill

, et al. The effects of a home-based physical activity intervention on cardiorespiratory fitness in breast cancer survivors; a randomised controlled trial. J Sports Sci 2018; 36(10): 1077–1086.

30.

Stein

Colditz

. Modifiable risk factors for cancer. Br J Cancer 2004; 90(2): 299–303.

31.

Hoffman

Brintnall

Brown

, et al. Virtual reality bringing a new reality to postthoracotomy lung cancer patients via a home-based exercise intervention targeting fatigue while undergoing adjuvant treatment. Cancer Nurs 2014; 37(1): 23–33.

32.

Baruth

Wilcox

Der Ananian

, et al. Effects of home-based walking on quality of life and fatigue outcomes in early stage breast cancer survivors: a 12-week pilot study. J Phys Act Health 2015; 12(Suppl 1): S110–S118.

33.

Armbruster

Song

Bradford

, et al. Sexual health of endometrial cancer survivors before and after a physical activity intervention: a retrospective cohort analysis. Gynecol Oncol 2016; 143(3): 589–595.

34.

Uhm

Yoo

Chung

, et al.

7. Effects of exercise intervention in breast cancer patients: is mobile health (mHealth) with pedometer more effective than conventional program using brochure?

Breast Cancer Res Treat 2017; 161(3): 443–452.

35.

Xiaochen

McClean

Morgan

, et al. Exercise among women with ovarian cancer: a feasibility and pre-/post-test exploratory pilot study. Oncol Nurs Forum 2017; 44(3): 366–374.

36.

Zhou

Cartmel

Gottlieb

, et al. Randomized trial of exercise on quality of life in women with ovarian cancer: women’s activity and lifestyle study in connecticut (WALC). JNCI: Journal of the National Cancer Institute 2017: 109(12): djx072.

37.

Frensham

Parfitt

Dollman

. Effect of a 12-week online walking intervention on health and quality of life in cancer survivors: a quasi-randomized controlled trial. Int J Environ Res Publ Health 2018; 15(10): 2081.

38.

Foulkes

Howden

Bigaran

, et al. Persistent impairment in cardiopulmonary fitness after breast cancer chemotherapy. Med Sci Sports Exerc 2019; 51(8): 1573–1581.

39.

McNeil

Brenner

Stone

, et al. Activity tracker to prescribe various exercise intensities in breast cancer survivors. Med Sci Sports Exerc 2019; 51(5): 930–940.

40.

Pope

Lee

Zeng

, et al. Feasibility of smartphone application and social media intervention on breast cancer survivors’ health outcomes. Transl Behav Med 2019; 9(1): 11–22.

41.

Grazioli

Cerulli

Dimauro

, et al. New strategy of home-based exercise during pandemic COVID-19 in breast cancer patients: a case study. Sustainability. 2020; 12(17): 6940.

42.

Holtdirk

Mehnert

Weiss

, et al. 16. Results of the Optimune trial: a randomized controlled trial evaluating a novel Internet intervention for breast cancer survivors. PLoS One 2021: 16(5): e0251276.

43.

Natalucci

Marini

Flori

, et al. Effects of a home-based lifestyle intervention program on cardiometabolic health in breast cancer survivors during the covid-19 lockdown. J Clin Med 2021: 10(12): 2678.

44.

Qiao

Van Londen

Brufsky

, et al. Perceived physical fatigability improves after an exercise intervention among breast cancer survivors: a pilot randomized clinical trial. J Clin Oncol. 2021; 39(15 SUPPL): 30–37.

45.

Gao

Ryu

Chen

. Effects of Tai Chi App and Facebook health education programs on breast cancer survivors’ stress and quality of life in the Era of pandemic. Compl Ther Clin Pract 2022; 48: 101621.

46.

Gomaa

West

Lopez

, et al. A telehealth-delivered tai chi intervention (TaiChi4Joint) for managing aromatase inhibitor-induced arthralgia in patients with breast cancer during COVID-19: longitudinal pilot study. JMIR Form Res 2022; 6(6): e34995.

47.

Gorzelitz

Stoller

Costanzo

, et al. Improvements in strength and agility measures of functional fitness following a telehealth-delivered home-based exercise intervention in endometrial cancer survivors. Support Care Cancer 2022; 30(1): 447–455.

48.

Ochi

Tsuji

Narisawa

, et al. Cardiorespiratory fitness in breast cancer survivors: a randomised controlled trial of home-based smartphone supported high intensity interval training. BMJ Support Palliat Care 2022; 12(1): 33–37.

49.

Sagarra-Romero

Butragueño

Gomez-Bruton

, et al. Effects of an online home-based exercise intervention on breast cancer survivors during COVID-19 lockdown: a feasibility study. Support Care Cancer 2022; 30(7): 6287–6297.

50.

Ferguson

. ACSM’s guidelines for exercise testing and prescription 9th Ed. 2014. The Journal of the Canadian Chiropractic Association 2014; 58(3): 328.

51.

WHO . Physical activity, 2020. Available from: https://www.who.int/news-room/fact-sheets/detail/physical-activity

52.

Hoffman

. Treatment of the adult growth hormone deficiency syndrome: directions for future research. Growth Horm IGF Res 2005; 15(Suppl A): S48–S52.

53.

Xiaosheng

Xiangren

Shuyuan

, et al. The effects of combined exercise intervention based on Internet and social media software for postoperative patients with breast cancer: study protocol for a randomized controlled trial. Trials 2018: 19(1): 477.

54.

Ishikawa

Thuler

LCS

Giglio

, et al. Validation of the Portuguese version of functional assessment of cancer therapy-fatigue (FACT-F) in Brazilian cancer patients. Support Care Cancer 2010; 18(4): 481–490.

55.

Corres

MartinezAguirre-Betolaza

Fryer

, et al. Long-term effects in the EXERDIET-HTA study: supervised exercise training vs. physical activity advice. Res Q Exerc Sport 2020; 91(2): 209–218.

56.

Hayes

Singh

Reul-Hirche

, et al. The effect of exercise for the prevention and treatment of cancer-related lymphedema: a systematic review with meta-analysis. Med Sci Sports Exerc. 2022; 54: 1389–1399.

57.

Gunderson

Willging

Trott Jaramillo

, et al. The good coach: implementation and sustainment factors that affect coaching as evidence-based intervention fidelity support. J Child Serv 2018; 13(1): 1–17.

58.

Church

Ali

Smith

, et al. Examining clinical practice guidelines for exercise and physical activity as part of rehabilitation for people with stroke: a systematic review. Int J Environ Res Public Health 2022; 19(3): 1707.

59.

Bray

Ferlay

Soerjomataram

, et al. Global cancer statistics 2018: GLOBOCAN estimates of incidence and mortality worldwide for 36 cancers in 185 countries. CA Cancer J Clin 2018; 68(6): 394–424.

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.00 MB

0.01 MB