Secondary Logo

Journal Logo

ARTICLES: ONLINE ONLY

The Effect of Nonpharmacological Interventions on Managing Symptom Clusters Among Cancer Patients

A Systematic Review

So, Winnie K. W. PhD, RN; Law, Bernard M. H. PhD; Chan, Dorothy N. S. PhD, RN; Xing, Weijie PhD, RN; Chan, Carmen W. H. PhD, RN; McCarthy, Alexandra L. PhD, RN

Author Information
doi: 10.1097/NCC.0000000000000730
  • Free

Abstract

Cancer is currently one of the most common life-limiting diseases, and its global incidence is expected to increase in the coming years.1 Recent advances in cancer treatment have undoubtedly contributed to better survival rates,2 but the disease and its treatments often result in multiple undesirable symptoms.3,4 Some of these symptoms, which appear to be related to each other, may co-occur in patients, thereby forming a symptom cluster.5 Dodd et al5 first defined a symptom cluster as 3 or more concurrent symptoms related to each other, and Molassiotis et al6 later revised the definition to 2 or more symptoms that are clinically meaningful together, related to each other and with a significant variance in the cluster.6,7

Although the pathophysiology of these clusters is still not fully understood, current evidence suggests that inflammation could be one of the key events contributing to the occurrence of cancer-associated symptom clusters. For example, the psychoneurological cluster, comprising symptoms that include pain, fatigue, sleep disturbance, cognition, and depression,8 appears to be caused by proinflammatory events. Cancer treatment has also been shown to induce changes in the white matter of the brain, which subsequently causes cognitive impairment,9,10 and this effect has been suggested as attributable to the action of proinflammatory cytokines.11,12 Furthermore, inflammation would induce an increase in the expression of serotonin transporters in the brain, decreasing the availability of the neurotransmitter serotonin, and this was suggested as linked to the development of depression.13 Overall, cancer-associated symptom clusters appear to have complex etiologies, where symptoms in these clusters may exhibit an exacerbating effect on each other. This effect in turn leads to a drastic reduction in patients’ physical and social functioning, resulting in a decrease in their health status and quality of life (QOL).14–16

With the increasing numbers of cancer patients worldwide and the detrimental effects of cancer- and treatment-associated symptom clusters, it is of paramount importance to develop effective strategies for cancer symptom management, in order to reduce the healthcare burden of such patients in societies worldwide. Nonpharmacological intervention, as a complement to pharmacological interventions, has the potential to be effective in managing various individual symptoms in cancer patients.17,18 Nonpharmacological interventions comprise the physical based (eg, body massage and auricular point acupressure), cognitive-behavioral (eg, mindfulness-based stress reduction [MBSR], guided imagery, relaxation and meditation), and educational (eg, counseling, supportive instruction).

Two recent systematic reviews19,20 have evaluated the effectiveness of some of these interventions on certain symptom clusters in cancer patients during and/or after treatment. However, the review by Kwekkeboom19 focused only on interventions targeting a specific cluster. Moreover, the author focused only on symptom cluster severity as an outcome and did not report the effects of the interventions on QOL or functional ability. Although Xiao et al20 did carry out a meta-analysis of the effects of interventions on symptom clusters among cancer patients during and/or after treatment, the scope of the review was limited to the psychoeducational type. Other types of nonpharmacological intervention developed for cluster management were excluded.

There is thus no systematic review that provides a general overview of the effectiveness of nonpharmacological interventions in addressing the severity of symptom clusters and QOL in individuals being treated for cancer. The objective of this article is therefore to provide an overview of the nonpharmacological interventions currently tested and to evaluate their effect on outcomes among cancer patients at various stages of the disease trajectory.

Methods

The preparation of this review adheres to the PRISMA (Preferred Reporting Items for Systematic Reviews and Meta-analyses) statement for reporting systematic reviews.21 This review was guided by the Revised Model for Symptom Management (the Model) developed by Dodd et al5 in order to have a better understanding of the effect of nonpharmacological interventions on managing symptom clusters among cancer patients. The Model is composed of 3 interrelated dimensions—the symptom experience, symptom management strategies, and outcomes. The dimension of symptom experience consists of an individual’s perception of a symptom, evaluation of the meaning of a symptom, and response to a symptom. The dimension of symptom management strategies is focused on preventing or delaying a negative outcome through biomedical, professional, and self-care strategies. Seven specifications of “what,” “when,” “where,” “why,” “how much,” “to whom,” and “how” questions are used to guide development of an effective intervention. The outcomes consist of 8 multidimensional indicators—functional statuses, self-care, costs, QOL, morbidity and comorbidity, mortality, and emotional status. Also, person, environment, health, and illness are variables in influencing all 3 dimensions.

In our review, we focused on investigating the effect of the nonpharmacological intervention on cancer patients’ experience of symptom clusters and other outcomes. Experiences of symptom clusters were measured by occurrence, frequency, intensity, or distress of symptom clusters reported by cancer patients. The 8 multidimensional indicators were used as keyword search of other patient outcomes. The 7 specifications were used to summarize essential components of nonpharmacological interventions.

Eligibility Criteria

Types of Studies

Original studies reporting randomized controlled trials (RCTs) were included in this review as providing high-quality evidence. Acceptable studies were full-text articles in English, but those in Chinese were also included because a large number of complementary therapy studies had been published in Chinese journals. Since the concept of symptom clusters was first introduced in 2001, there is justification for limiting the studies to be included to those published after January of that year.

Types of Participants

Studies involving adult patients with any type of cancer, at any stage of the disease, and under any stage of treatment were included.

Type of Interventions

Included studies had to evaluate the effectiveness of at least 1 type of nonpharmacological intervention in managing symptom clusters.

Types of Outcome

The primary outcomes were symptom clusters, identified by occurrence, frequency, intensity, or distress, and measured by multiple-dimension symptom or individual symptom questionnaires. Studies that reported interventions addressing individual symptoms or multiple symptoms unrelated to each other were excluded. The secondary outcomes were the 8 multidimensional indicators identified by the Model.5 For details of the outcome-related keywords, refer to Table 1.

Table 1 - The Search Strategy
“Cancer” OR “Neoplasm” OR “Carcinoma” OR “Tumor” OR “Malignancy”
AND
Symptom cluster” OR “Symptom clusters” OR “Syndrome” OR “Syndromes”
AND
“Intervention” OR “Therapy” OR “Programs” OR “Treatment” OR “Therapeutics”
AND
“Quality of life” OR “Quality of life satisfaction”
OR
“Symptom burden” OR “Symptom” OR “Symptoms”
OR
“Symptom severity” OR “Symptom severity level” OR “Level of symptom severity”
OR
“Symptom occurrence” OR “Symptom” OR “Symptoms”
OR
“Symptom distress”
OR
“Symptom limitation” OR “Symptom” OR “Symptoms” OR “Limitation” OR “Limitations”
OR
“Financial burden” OR “Financial” OR “Finances” OR “Costs”
OR
“Emotional status” OR “Emotional” OR” Emotion” OR “Feeling” OR “Feelings” OR “Mood” OR “Moods” OR “Affective state” OR “Status” OR “Statuses” OR “Stat”
OR
Co-morbidity OR “Co morbidity OR “Comorbidities” OR “Comorbidity” OR “Morbidity”
OR
“Mortality” OR “Death rate” OR “Death rates” OR “Mortalities” OR “Fatality rate” OR “Fatality rates” OR “Rate death” OR “Rates death” OR Mortality statistics” OR “Statistics mortality” OR “Statistics death” OR “Mortality vital statistics” OR “Mortality rate” OR “Mortality rates”
OR
“Functional status”
OR
“Self-care” OR “self care”

Search Strategy

A 3-step search strategy was undertaken in January 2018. First, an initial limited search of MEDLINE and CINAHL was conducted to identify relevant keywords that could be used in the comprehensive search, which was then performed in 7 databases: OVID MEDLINE (since 1946), PubMed, EMBASE, PsycINFO (since 1806), CINAHL, CNKI (Chinese), and Wanfang (Chinese). The search strategy used in PubMed is presented in Table 1. The keywords covering patient outcomes included in the search strategy were based on the symptom management model described by Dodd et al.5 The combination of keywords used for the search was agreed by 3 authors (W.K.W.S., B.M.H.L., and D.N.S.C.). Finally, a snowball search was carried out to find additional texts by screening the reference lists of all related studies.

Data Extraction and Summary

Data were extracted from the studies by 2 independent reviewers (B.M.H.L. and D.N.S.C. for English articles and B.M.H.L. and W.X. for Chinese articles) and included study settings, patients’ clinical characteristics, sample size, intervention, time of data collection, symptom clusters examined, instruments used in cluster assessment, within-group variations, and between-group differences in various patient outcomes guided by Dodd’s model (ie, occurrence/frequency/intensity of symptom clusters, the 8 multidimensional indicators). Any disagreements in the study selection and data extraction processes were resolved by discussion between the reviewers to reach a consensus.

Quality Assessment of Included Studies

The quality of the RCTs in the studies was assessed by 2 investigators independently (B.M.H.L. and D.N.S.C. for English articles and B.M.H.L. and W.X. for Chinese articles), using a 27-item checklist from the Checklist for Both Randomized and Nonrandomized Studies developed by Downs and Black.22 The scoring methods for the quality of each study have been previously described.23 In short, the quality of the included studies was assessed on the following 5 subscales: (1) reporting, (2) external validity, (3) bias, (4) internal validity, and (5) statistical power. Points were allocated to each of the items in these areas if the criteria were observed, as instructed in the checklist.23 The maximum total score in this assessment was 28. Studies scoring 9 or less (<33%) were considered of low quality and excluded from the review.23

Data Analysis

Given the heterogeneous nature of the interventions involved and the outcome measures investigated across selected studies, meta-analysis was not undertaken. Instead, extracted data were used to perform narrative and descriptive analysis with studies grouped according to types of intervention and outcomes reported. Within each specific outcome (eg, frequency/occurrence/intensity of symptom clusters, QOL, functional ability), the findings were grouped according to types of intervention and then narratively summarized by the investigators (B.M.H.L., D.N.S.C., and W.K.W.S).

The strength of evidence (the level of certainty) that indicates the effect of reported interventions in the studies included on the above patient outcomes was described according to the guidelines of the US Preventive Services Task Force, under which the criteria for assessment are adapted and defined as follows24:

HIGH LEVEL OF CERTAINTY

The available evidence is generated from consistent findings from well-conducted studies among representative populations with appropriate study designs. These studies should also involve an assessment on the effects of the reported interventions on the patient outcomes involved in this review.

MODERATE LEVEL OF CERTAINTY

The available evidence is sufficient to evaluate whether the reported interventions may have a beneficial effect on the patient outcomes involved in this review. However, the quality of the evidence is restricted by the limited number of studies reporting such outcomes, the small sample size, the quality of methodologies, the inconsistency between the findings across studies, limited generalizability of the findings on oncology nursing practice, or the lack of coherence in the available evidence for the patient outcomes involved.

LOW LEVEL OF CERTAINTY

The available evidence is not sufficient to evaluate whether the reported interventions may have a beneficial effect on the patient outcomes involved in this review. The insufficiency of evidence could be attributed to the small number of studies reporting such outcomes, small sample size, serious flaws in study design and methodologies, serious inconsistency between findings across studies, appearance of critical gaps in the chain of evidence generated, lack of generalizability of the generated evidence on oncology nursing practice, or lack of information on the reported patient outcomes.

Results

Search Results

A total of 93 054 citations were identified in the initial search. One thousand three hundred forty-seven articles remained after removal of duplicates and those not reporting RCTs. After screening titles and abstracts, 338 full-text articles were then subjected to screening to identify RCTs that described nonpharmacological interventions and reported their effect on symptom clusters. A total of 325 original articles that did not meet these criteria were excluded, with 13 studies left to be included in the final review. The PRISMA diagram of the flow of the selection process is presented in the Figure.

Figure
Figure:
The PRISMA diagram.

Methodological Quality of Included Studies

The quality of the studies ranged from moderate to high, with overall scores ranging between 14 and 23 out of 28. Of the 13 articles, 10 (76.9%) attained a score of 8 or above for the reporting subscale. Nevertheless, most of the studies (77%) did not explicitly state whether adverse events had occurred among the participants during the intervention. The external validity score of the 13 studies was generally fair, as it was unclear whether participants approached during subject recruitment and enrolled in the study accurately reflected the sampling frame. All studies achieved a moderately high score (≥4 out of 7) for the bias subscale. To address bias, some studies (31%) reported attempts to blind the participants and/or data collectors, although blinding was not feasible in some of the studies because of the nature of the intervention. In addition, to reduce information bias, most studies (85%) used previously validated instruments for the purpose of data collection in all outcome measures. Most of the studies (92%) achieved a moderately high score (≥4 out of 6) on the internal validity subscale. To ensure the internal validity of the studies, true randomization procedures were used during the allocation of participants to treatment groups, including a lucky draw,25 minimization,26,27 block randomization,28 and computer-assisted random numbering of subjects.29–32 Nevertheless, only 2 studies (15%) indicated that group allocation was concealed during the study.29,31 Power analysis was also carried out in 5 studies for sample size estimation.25,26,28,29,31 The ratings of the studies’ quality in conducting and reporting RCTs are presented in Table 2.

Table 2 - The Quality Ratings of the Included Studies
References Reporting (Max, 11) External Validity (Max, 3) Bias (Max, 7) Internal Validity (Max, 6) Power (Max, 1) TOTAL Score (Max, 28)
Chan et al,25 2011 8 1 5 4 0 18
Charalambous et al,26 2016 9 2 6 5 1 23
Fleming et al,29 2014 7 0 4 5 1 17
Jarden et al,30 2009 9 2 4 5 0 20
Jiao et al,33 2015 9 1 4 5 0 19
Kwekkeboom et al,31 2012 10 0 5 6 1 22
Lengacher et al,34 2012 8 1 4 4 0 17
Mao et al,35 2018 9 1 4 5 0 19
Miladinia et al,28 2017 9 1 7 5 1 23
Reich et al,36 2017 7 0 4 3 0 14
Yeh et al,32 2016 9 1 6 4 0 20
Yorke et al,27 2015 9 1 5 4 0 19
Zhang and Zhang,37 2016 7 1 4 5 0 17

Having achieved a total score of 10 or above, all these studies were eligible to be included in this review.

Description of Included Studies

The publication years of the 13 studies fell between 2009 and 2018. Four were conducted in the United States,31,32,34,36 3 in mainland China,33,35,37 2 in the United Kingdom,27,29 and the remaining 4 in Cyprus,26 Denmark,30 Hong Kong Special Administrative Region,25 and Iran.28 The sample size of the studies ranged from 31 to 299, with a mean of 115.

PARTICIPANTS

The participants in the 13 included studies were diagnosed with thoracic/lung cancer,25,27,35 leukemia,28 or breast cancer32,37; individuals previously treated for breast cancer34,36; or patients with various forms of blood cancer.30 In addition, 4 studies involved various forms of cancer within the groups of participants, including a combination of breast and prostate cancer26; breast, prostate, bowel, and gynecological cancer patients who had completed treatment29; lung, prostate, colorectal, and gynecological cancers31; and advanced gastrointestinal, lung, and breast cancer.33 Patients included in these studies were either receiving or had received chemotherapy,26,28,37 radiotherapy,25 chemotherapy or radiotherapy,31,32 chemotherapy and radiotherapy,35 or allogeneic hematopoietic stem cell transplantation.30 Additionally, 5 studies involved cancer patients who had completed first-line treatment.27,29,33,34,36

INTERVENTIONS

The majority (69%) were carried out in outpatient clinics at either local hospitals or oncology centers, except 4 studies,30,33,35,37 where the intervention took place in an inpatient setting. As the interventions reported in the studies exhibited a high degree of heterogeneity, all the interventions were categorized into 3 groups—physical-based interventions, cognitive-behavioral interventions, and educational interventions.17,18

A detailed description of the nature of these interventions is provided in Table 3—educational intervention,25,27 physical-based intervention (ie, slow-stroke back massage, auricular point acupressure),28,32,33,35 cognitive-behavioral interventions (ie, cognitive-behavioral therapy, MBSR),26,29,31,34,36 or multimodal interventions (ie, interventions comprising physical-based activities combined with psychoeducational or relaxation components).30,37 Five studies involved the provision of educational materials to participants as an integral part of the intervention, alongside its core components.25,27,31,34,36 These materials contained information on the causes of symptoms experienced by patients,25,31 guides to suggested strategies for symptom management,25,27,31 or advice on relaxation techniques and healthy lifestyles.34,36 The interventions lasted from 2 to 6 weeks, and were delivered face-to-face, either on a one-to-one basis25–28,30–33,35,37 or in small groups.29,34,36 They were delivered by nurses,25,28,29,31,33 psychologists,34,36 or the study investigators.30,32,37 In addition, 1 study27 used a multidisciplinary team consisting of nurses, physiotherapists, and complementary therapists. Two studies26,35 did not specify who delivered the reported intervention. Four studies developed their interventions based on a theory or particular framework.25,27,28,32 Chan et al25 developed an analytical model to depict the relationship between the intervention and its outcomes. Yorke et al27 applied the Medical Research Council framework for the development and evaluation of complex interventions. Miladinia et al28 used symptom management theory as the theoretical framework. Yeh et al32 developed an auricular point acupressure intervention based on reflex theory. Nevertheless, none of the included studies stated how much the reported intervention costs.

Table 3 - A Summary of the Characteristics of the Included Studies
Authors, Year; Country Study Design/Setting Patient Characteristics/Sample Size N Interventions Data Collection Periods Symptom Clusters Measured and Other Outcome Measures Assessment Tools Results
Chan et al,25 2011;
Hong Kong
Randomized controlled trial (RCT); outpatient radiotherapy unit of a public hospital Lung cancer patients receiving radiotherapy;
N = 140 (intervention: 70; control: 70)
Intervention group:
Psychoeducational intervention
 - Within 1 wk before start of radiotherapy, a 40-min educational package and training on progressive muscle relaxation (PMR) techniques were delivered.
 - Educational package includes leaflets containing information on the 3 symptoms, factors that ameliorate and exacerbate them, and advice on self-management strategies of these symptoms.
 - Audiotape and educational leaflets were provided to patients.
 - Patients were encouraged to practice PMR daily.
 - Telephone reminder was given at the end of the second week.
 - The intervention was delivered by a registered nurse.
Control group: Usual care: Consisting of individual briefing of radiotherapy procedure, discussion of side effects, especially on skin care, and attendance of a health talk focusing on general care.
Baseline
Week 3
Week 6
3 mo after intervention
 - Anxiety-breathlessness-fatigue
- Functional ability
 - Piper Fatigue Scale
- State-Trait Anxiety Inventory
- Visual analog scale (VAS)
- SF-36 Health Survey
Between baseline and 6 wk after start of intervention
- Significant difference was observed in the pattern of change between the intervention and control groups with reduced breathlessness, fatigue, and anxiety and better functional ability
Breathlessness: P = .002 Fatigue: P = .011 Anxiety: P = .001 Functional ability: P = < .001
Between baseline and 12 wk after start of intervention
- Significant difference was observed in the pattern of change in only breathlessness, anxiety, and functional ability between the intervention and control groups but not fatigue: Breathlessness: P = .001 Fatigue: P = .034 Anxiety: P = .005 Functional ability: P = .002
Charalambous et al,26 2016; Cyprus RCT;
Outpatient clinics of 3 cancer care centers
Breast or prostate cancer patients undergoing chemotherapy;
N = 208 (intervention:104; control: 104)
Intervention group:
Guided Imagery and Progressive Muscle Relaxation
- Four weekly supervised and daily unsupervised sessions of guided imagery and PMR
- The intervention included a 2-min breathing exercise, 10-min PMR exercise and 15-min pleasant guided imagery session
Control group: Usual standardized treatment of symptoms described in international guidelines
Baseline
Week 4
 - Pain
- Cancer- related fatigue
- Anxiety
- Depression
- Nausea, vomiting, and retching
- Health-related quality of life
 - 10-point numeric pain scale
- Cancer Fatigue Scale
- The Revised Rhodes Index of Nausea, Vomiting, and Retching
- Zung Self-rating Anxiety Scale
- Beck Depression Inventory- II
European Organization for the Research and Treatment of Cancer Quality of Life Questionnaire (EORTC-QLQ-C30)
- European Organization for the Research and Treatment of Cancer Quality of Life Questionnaire - Breast Cancer Module (EORTC-QLQ-BR23)
European Organization for the Research and Treatment of Cancer Quality of Life Questionnaire - Prostate Cancer Module (EORTC-QLQ-PR25)
Within-group changes: - Statistically significant reduction in the severity of investigated symptoms in intervention group over time
Pain: P < .0001 Fatigue: P < .02
Nausea, vomiting, and retching:
P < .0001
Anxiety: P < .0001
Depression: P < .0001
Health-related quality of life (HRQOL): increased, P < .0001
Between intervention and control groups:
Differential effect observed for intervention and control groups before and after intervention
Pain levels: I: decreased (P < .0001) C: increased (P = .0004) Fatigue levels: I: decreased (P < .0001) C: increased (P < .0001) Nausea and vomiting
I: decreased (P < .0001)
C: increased (P < .0001)
Anxiety levels: I: decreased (P < .0001) C: increased (P < .0001) Depression: I: decreased (P < .0001) C: increased (P < .0001)
HRQOL:
I: increased (P < .0001) C: decreased (P < .0001)
Fleming et al,29 2014;
UK
Secondary analysis of RCT;
oncology clinics at Glasgow Beaton Oncology Centre or Aberdeen Royal Infirmary
Breast, prostate, bowel or gynecological cancer patients; N= 113
(intervention: 73; control: 40)
Intervention group:
Cognitive Behavior Therapy Five weekly 50-min sessions of CBT with content including stimulus control, sleep restriction, and cognitive therapy strategies
Control group: Usual care
Baseline
Post-treatment
6 mo after therapy
Fatigue-anxiety-depression  - Hospital Anxiety and Depression Scale
- Fatigue Symptom Inventory
Within-group changes
% change of patients experiencing clinical-level fatigue preintervention-postintervention
Intervention: -10.9%
Control: +2.5%
% change of patients experiencing clinical-level anxiety preintervention-postintervention
Intervention: no change
Control: +5%
% change of patients experiencing clinical-level depression preintervention-postintervention
Intervention: -5.5%;
Control: +5%
Between intervention and control groups:
Significant association exists between treatment group allocation and postintervention change in some of the symptoms investigated: Fatigue: P = .03
Anxiety: P = 1.00 (no association)
Depression: P = .09 (no association)
Jarden et al,30 2009; Denmark RCT;
Copenhagen University Hospital
Patients with hematologic malignancy (eg, leukemia) who receive allogeneic hematopoietic stem cell transplantation; N = 42 (intervention: 21; control: 21) Intervention group:
Multimodal Intervention One-hour sessions on 5 d per week of multimodal intervention, consisting of activities including stationary cycling, dynamic and stretching exercises, resistance training, and progressive relaxation and psychoeducation sessions.
Control group: Convention treatment and care, with the hospitals' standard care for physical activity
Baseline
Week 1–6
3 and 6 mo after intervention
Symptom clusters identified from study participants at baseline:
Gastrointestinal cluster: nausea, vomiting, stomach pain, loss of appetite, diarrhea Cognitive cluster: Lowered concentration, memory problems, fatigue Affective cluster: Nervousness, anxiety, stress Functional cluster: Muscle ache, joint ache Mucositis cluster: Mouth and throat pain, difficulty swallowing
- Karnofsky Performance Status
- Stem Cell Transplantation Symptom Assessment Scale
Between intervention and control groups:
Significantly lower symptom severity scores were observed for intervention group over time compared to control group for most of the symptom clusters: Gastrointestinal cluster: P = .017
Cognitive cluster: P = .002
Affective cluster: P = .605
Functional cluster: P = .009 Mucositis cluster: P = .019
Jiao et al, 2015;35 China RCT;
inpatient integrative medical unit at Cancer Hospital of Tianjin Medical University
Patients with advanced gastrointestinal, lung, breast cancer;
N=100
(intervention: 50; control: 50)
Intervention group:
Acupressure
- A 4-wk acupressure intervention for Neiguan, Hegu, Zusanli, 3–5 min a time, 4–6 times per day
- An education about acupressure from nurse
Control group: Convention treatment hospitals' standard care
Baseline
Week 4
- Pain-fatigue-short of breath-sleepiness
- Nausea-poor appetite-dry mouth-vomiting
- Disturbed sleep-distress-forgetfulness-sadness
- M. D. Anderson Symptom Inventory Between intervention and control groups:
Significantly lower symptom cluster scores were observed for intervention group over time compared to control group for 2 symptom clusters:
- Pain-fatigue-short of breath-sleepiness: P = .027
- Nausea-poor-appetite-dry mouth-vomiting: P = .001
- Disturbed sleep-distress-forgetfulness-sadness Affective cluster: P = .196
Kwekkeboom et al,31 2012; USA RCT:
Outpatient chemotherapy or radiation therapy clinics at a National Cancer Institute–designated comprehensive cancer center
Patients with advanced (metastatic or recurrent) colorectal, lung, prostate or gynecological cancers receiving chemotherapy or radiotherapy;
N= 78 (intervention: 36; control: 42)
Intervention group:
Patient-controlled cognitive behavior intervention
- Participants received a single one-on-one training session from a research nurse.
- A 2-wk intervention with components that include (1) information on pain, fatigue, and sleep disturbance during cancer treatment and their causes, (2) rationale for how that cognitive-behavioral strategies affected symptoms, (3) overview of the 12 cognitive-behavioral strategies, (4) individualized recommendations for practice based on patients’ usual symptom patterns.
- Participants were encouraged to use the strategies at least once per day. These cognitive behavioral strategies were presented in 4 categories including relaxation exercises, presentation of nature imagery, symptom-focused imagery and nature sounds.
- Educational booklet was given to participants to guide training.
Wait-list control group: Usual care
Baseline
Week 2
 - Pain-fatigue-sleep disturbance
- Symptom interference
- Mood
- Brief Pain Inventory
- Brief Fatigue
Inventory
- Pittsburgh Sleep Quality Index
- M. D. Anderson Symptom Inventory
- Profile of Mood States-Short form
Between intervention and control groups:
- Intervention group had significantly lower symptom cluster scores 2 wk after intervention
- Intervention group had significant less pain (P = .006) and less fatigue (P = .049) but not sleep disturbance and symptom interference, compared to control group.
Lengacher et al,33 2012;
USA
RCT;
H. Lee Moffitt Cancer Center and Research Institute
Breast cancer survivors up to 18 mo after end of chemotherapy and/or radiotherapy
N= 84 (intervention: 41; control: 43)
Intervention group:
Mindfulness-based stress reduction intervention Six weekly 2-h sessions of intervention comprising
(1) educational material on relaxation, meditation and mind-body connection;
(2) meditation practice in weekly group sessions and home assignments;
(3) group discussion of barriers to the practice of meditation and application of mindfulness in daily situation;
(4) supportive interaction between group members;
(5) group support sessions on symptom management
- taught by a licensed clinical psychologist
Wait-list control group: Standard posttreatment visits to practitioners
Baseline
6 wk after intervention
Symptom clusters identified from participants at baseline:
Gastrointestinal cluster: nausea, vomiting, lack of appetite, shortness of breath, dry mouth, numbness Cognitive/psychological cluster: distress, sadness, remembering, pain Fatigue cluster: Fatigue, sleep disturbance, drowsiness
 - M.D. Anderson Symptom Inventory Within-group changes
- In intervention group, a significant reduction of symptom cluster score was observed for 2 of the 3 clusters: Gastrointestinal cluster: P = .005 Cognitive/psychological cluster: P = .01 (not sig.)
Fatigue cluster: P = .001
- In control group, a significant reduction of symptom cluster score was observed for 2 of the 3 clusters:
Gastrointestinal cluster: P = .005
Cognitive/psychological cluster: P = .002
Fatigue cluster: P = .024 (not sig)
(Significance level was set at P < .01 in this study)
Between intervention and control groups:
No significant difference was observed in the changes in symptom cluster score for all 3 clusters between the intervention and control groups (P value not specified)
Mao et al,36 2018; China RCT;
inpatient radiotherapy unit at Hunan Cancer Hospital of Xiangya Medical College, Central South University
Patients with thoracic cancer undergoing concurrent chemotherapy and radiotherapy
N = 120
(intervention: 60; control: 60)
Intervention group:
Acupoints stimulation
A 2 wk acupoints stimulation intervention for Neiguan, Zusanli, and stomach 1 time per day, 20 min 1 time
Control group:
Convention treatment hospitals' standard care
Baseline
postintervention
4 wk after intervention
 - Loss of appetite-nausea-vomiting-constipation  - The Memorial Symptom Assessment Scale Between intervention and control groups:
Significantly lower symptom occurrence and severity observed at postintervention for intervention group over time compared to control group
- Loss of appetite: P < .05
- Nausea-vomiting: P < .05
- Constipation: P < .05
Significantly lower symptom occurrence and severity observed at 4 wk after intervention for intervention group over time compared to control group
- Loss of appetite: P < .01
- Nausea-vomiting: P < .01
- Constipation: P < .01
Miladinia et al,28 2017;
Iran
RCT; outpatient chemotherapy and hematology units at Shafa Hospital Leukemia patients undergoing chemotherapy
N = 60 (intervention: 30; control: 30)
Intervention group:
Slow-stroke back massage intervention Receive 10-min massage sessions delivered by trained oncology nurses 3 times a week for 4 wk in addition to routine care
Control group:
Routine care and medical interventions
Baseline
Week 1
Week 2
Week 3
Week 4
Week 5
30th day since start of intervention (sleep quality)
Pain-fatigue-sleep disturbance  - Brief Pain Inventory
- Brief Fatigue Inventory
- Pittsburgh Sleep Quality Index
Within-group changes - Intervention significantly reduced the intensity of pain, fatigue, and sleep disturbance over time in the intervention group (P values not shown) - Significant improvement observed between pretest and posttest sleep quality score in the intervention group (P = .001) - No significant difference in the intensity of the above 3 symptoms during study period, although a slight increase in such intensity was observed in the control group.
- No difference was observed between pretest and posttest sleep quality score in the control group (P = .118)
Between intervention and control groups:
- Significant difference was observed in the changes of intensity of the symptoms investigated over time (baseline vs week 5) between the intervention and control groups, with less pain, fatigue and sleep disturbance observed in intervention group. Pain: P = .001 Fatigue: P = .001 Sleep disturbance: P = .015
- Also, significant difference was observed in the changes of sleep quality score over time (30th day since start of intervention) between the 2 groups (improvement in intervention group) (P = .003)
Reich et al,34 2017;
USA
RCT;
Moffitt Cancer Center, Carol and Frank Morsani Centre for Advanced Healthcare, and the life Hope Medical Group
Breast cancer survivors, with a diagnosis of stage 0-III breast cancer and within 2 wk to 2 y off-treatment;
N = 299
(intervention: 152;
control: 147)
Intervention group:
Mindfulness-based stress reduction program
Consisted of 3 components:
- Educational materials related to relaxation, meditation, mind-body connection, and a healthy-lifestyle for survivors
- Practice of meditation in group meetings and homework assignments
- Group processes related to barriers to the practice of meditation and supportive group interaction
Intervenor: trained psychologist delivered the intervention in 6 weekly 2-h sessions
Control group:
Usual care (standard post treatment clinic visits and were asked to refrain from practicing meditation, yoga techniques)
Baseline
Week 6
Week 12
Symptom clusters identified from study participants at baseline:
Psychological cluster: Anxiety, depression, stress
Fatigue cluster: fatigue, sleep, drowsiness
Cognition cluster: memory, mindfulness
- Concerns about Recurrence Scale
- The Center for Epidemiological Studies Depression Scale
- State-Trait Anxiety Inventory
- The Perceived Stress Scale
- Cognitive and Affective Mindfulness Scale-Revised
- M.D. Anderson Symptom Inventory
- Pittsburgh Sleep Quality Index
- Fatigue Symptom Inventory
- Brief Pain Inventory
- Everyday Cognition scale
- Short-Form General Health Survey (SF-36)
- Significant difference between groups between baseline and 6 wk after intervention with improvement of severity of certain symptom clusters were observed in the intervention group: Psychological: P = .007 Fatigue: P < .001 But not for other clusters: Cognition: P = .78
- No significant difference between 6 wk and 12 wk after intervention observed in terms of improvement of severity of each of the symptom clusters investigated between groups: Psychological: P = .84 Fatigue: P = .80 Cognition: P = .26
Yeh et al,32 2016;
USA
RCT;
cancer outpatient clinics
Breast cancer patients;
N = 31 (intervention: 16; control: 15)
Intervention group:
Auricular Point Acupressure (APA) - Attachment of plant seeds onto designated acupoints for pain, fatigue and sleep using tape.
- Apply pressure to the acupoints with the thumb and index finger for moderate stimulation.
- Treatment once a week for 4 wk, where the seeds are attached to the acupoints for 5 d
Control group: Attachment of plant seeds onto acupoints that are unrelated to pain, fatigue and sleep using tape
Baseline
Week 1
Week 2
Week 3
Week 4
1 mo postintervention
- Pain-fatigue-sleep disturbance
- Quality of life
- M. D. Anderson Symptom Inventory
- World Health Organization Quality of Life
 - Clinically significant differences were observed in the changes of the mean score for a number of symptoms, including pain, fatigue and sleep, before and after the intervention.
Within-group changes
- Reduction in the mean score for pain (71%), fatigue (44%) and sleep disturbance (31%) and interference to daily activities (61%) after 4 wk of intervention for the intervention group.
- Control group reported a moderate reduction in the mean score of the above outcome measures, but numbers not specified.
Between intervention and control groups:
- No significant differences for most of the outcome measures (except pain) Pain (P = .0217) Fatigue (P = .2351)
Sleep (P = .0642)
Interference to daily activities (P = .2792)
- Better QOL and level of symptom interference with daily lives were reported for patients in intervention group than those in control, but the difference was not significant.
Yorke et al,27 2015;
UK
RCT:
11 participating centers: 7 secondary care teaching hospitals, 2 specialist cancer centers and 2 district general hospitals
Lung cancer patients;
N = 71 (intervention: 31; control: 40)
Intervention group:
Respiratory distress symptom intervention - Two 1-h training sessions of face-to-face education on controlled breathing techniques, cough easing techniques and acupressure techniques. These were accompanied by the provision of an information pack containing information on symptom experience and advices on symptom management.
- One follow-up telephone call 2 wk after the second training session.
- The intervention was delivered by specialist nurse, physiotherapist and complementary therapist.
Control group: Usual care
Baseline
Week 4
Week 12
- Respiratory distress symptom cluster: Breathlessness-cough-fatigue
- Health-related QOL
- Numerical rating scale and Dyspnea-12 scale for breathlessness
- Manchester Cough in Lung Cancer scale
- Functional Assessment of Cancer Therapy-Fatigue
- 3 level version of EQ-5D (EQ-5D-3L) (index and visual analogue scale (VAS) - for measurement of HRQOL)
Between baseline and 4 wk after start of intervention:
Between intervention and control groups:
- Nearly significant difference was observed in breathlessness (P = .05), cough (P = .07) between groups.
- No significant difference in fatigue and QOL score.
Between baseline and 12 wk after start of intervention Between intervention and control groups:
- Significant difference was observed in breathlessness (P = .026) and QOL (P = .009) between groups.
- No significant difference in fatigue and cough between groups.
Zhang and Zhang,37 2016; China RCT;
inpatient chemotherapy unit at Affiliated Fourth Hospital of Harbin Medical University
Breast cancer patients undergoing chemotherapy (≥2 chemotherapy cycles);
N = 144 (4 groups)
- Aerobic exercise group N = 36
- Relaxation therapy group N = 36
- Aerobic exercise combined with relaxation therapy group N = 36
- Control group N = 36
Intervention group:
Aerobic exercise and relaxation therapy
- Aerobic exercise group: individual exercise program, moderate intensity, targeted heart rate 55%–65% max heart rate; choosing walking, climbing stairs, cycling, jogging according to individual preference; 20–30 min each time, 3–5 times per week, lasting 6 wk
- Relaxation therapy group: self-led relaxation program based on provided audio, 10 min each day, before going sleep, lasting 6 wk
- Combined intervention group: aerobic exercise and relaxation therapy as described above
Control group: Usual care (education with leaflets)
Baseline
Week 6
 - Fatigue
- Loss of appetite
- Nausea
- Taste change
- Sleep disturbance
- Health-related QOL
- The Memorial Symptom Assessment Scale (MSAS)
- Functional Assessment of Cancer Therapy–Breast Cancer
Within-group changes: - Statistically significant reduction in the severity of some investigated symptoms in intervention group over time
-Aerobic exercise group
Fatigue: reduction, P < .05
Loss of appetite: reduction, P < .05
Nausea: P > .05
Taste change: P > .05
Sleep disturbance: reduction, P < .05
Health-related QOL: P > .05
-Relaxation therapy group
Fatigue: P > .05
Loss of appetite: p>.05
Nausea: P > .05
Taste change: P > .05
Sleep disturbance: reduction, P > .05
Health-related QOL: P > .05
-Aerobic exercise combined with relaxation therapy group
Fatigue: reduction, P < .05
Loss of appetite: P > .05
Nausea: P > .05
Taste change: P > .05
Sleep disturbance: reduction, P < .05
Health-related QOL: increased, P < .05
-Control group: no significant reduction of any symptom, and no significant improvement of QOL

OUTCOMES

All studies used a questionnaire survey approach to assess patient outcomes and to determine the effect of the interventions. The most commonly assessed individual symptoms in the 13 studies included fatigue (n = 11), pain (n = 6), sleep disturbance (n = 6), and anxiety (n = 4). Four studies specifically assessed the effect of the interventions on the severity of the pain–fatigue–sleep disturbance symptom cluster.28,31–33 Other common clusters investigated in more than 1 study included the gastrointestinal,30,33,34 the cognitive,30,34,36 and the fatigue clusters.34,36 In addition to symptom assessment, 6 studies reported the effect of interventions on other patient outcomes—4 on QOL,26,27,32,37 2 on symptom interference with daily lives,31,32 and 1 on functional ability.25

The assessment of outcomes was primarily performed on a preintervention and postintervention basis, with 5 only at baseline and postintervention.26,31,33,34,37 Four studies additionally conducted a follow-up assessment after the intervention in order to evaluate its long-term effects.27,29,35,36 Another 4 studies25,28,30,32 also performed outcome assessments 1 or more times during the course of the intervention, evaluating temporal changes in the outcomes. A summary is provided in Table 3 of the outcome measures used and the periods of follow-up outcome assessment in each of the studies.

EFFECT OF INTERVENTIONS ON SYMPTOM CLUSTER SEVERITY

A summary of the effect of the interventions described in the studies is provided in Tables 3 and 4. Despite the heterogeneous nature of the interventions and the outcome variables measured, the effects of the interventions on patient outcomes were fairly consistent. The majority of the studies (n = 12) reported postintervention improvement in the severity of at least one of the individual symptoms in the clusters in the intervention group compared with the control group. The exception was the MBSR intervention reported by Lengacher et al.34 Three symptom clusters were commonly involved in the assessment of the beneficial effects of nonpharmacological interventions: pain–fatigue–sleep disturbance, cognitive/psychological, and the gastrointestinal clusters. A detailed summary of the findings concerning the effects on these clusters is presented below.

Table 4 - Strength of Evidence of the Different Types of Nonpharmacological Interventions Included in the Review
Intervention Type Intervention Name Author/Year/Sample Size Significance of Differences in Patient
Outcomes Between-Group Comparison
Quality Score of Study (Maximum 28) Overall Consistency of Outcomes Between Studies Strength of Evidence
Pain – Fatigue – Sleep Disturbance Cognitive Cluster Gastrointestinal Cluster Fatigue Cluster Quality of Life Functional Ability Symptom Interference With Daily Lives Miscellaneous Symptom Clusters / Outcomes
Educational interventions Psychoeducational  intervention Chan et al,25  2011
(N = 140)
S Anxiety-breathlessness-fatigue
S (6 wk)
S (12 wk, anxiety and breathlessness only)
18 NAa Moderate
Respiratory distress  symptom intervention Yorke et al,27 2015
(N = 71)
NS (4 wk)
S (12 wk)
Breathlessness-cough-fatigue
NS (4 wk)
S (12 wk, breathlessness only)
19
Cognitive-behavioral interventions Guided Imagery  and Progressive Muscle Relaxation Charalambous et al,26 2016
(N = 208)
Differential effect between groups Pain, cancer-related fatigue, anxiety, depression, nausea, vomiting, and retching
Differential effect between  intervention and control  groups
23 Inconsistent Moderate
Cognitive Behavior Therapy Fleming et al,29 2014
(N = 113)
Fatigue-anxiety-depression
S (fatigue only)b
17
Patient-controlled  cognitive  behavior intervention Kwekkeboom et al,31 2012
(N = 78)
S (pain and fatigue only) NS 22
Mindfulness-based stress  reduction intervention Lengacher et al,33 2012
(N = 84)
NS NS NS 17
Mindfulness-based stress  reduction intervention Reich et al,34 2017
(N = 299)
NS
(Baseline to 6 wk, 6–12 wk)c,d
S (Baseline to 6 wk only)c Psychological cluster (Anxiety, depression, stress)
S (Baseline to 6 wk only)c
14
Physical-based interventions Acupressure intervention Jiao et al,35 2015
(N = 100)
S S Disturbed sleep-distress-forgetfulness-sadness
NS
19 Overall consistent Moderate
Acupoints stimulation Mao et al,36 2018
(N = 120)
S 19
Slow-stroke back  massage intervention Miladinia et al,28 2017
(N = 60)
S 23
Auricular point  acupressure intervention Yeh et al,32 2016
(N = 31)
S (pain only) NS NS 20
Multimodal interventions Aerobic exercise and  relaxation therapy Zhang and Zhang,37 2016g
(N = 144)
NS (aerobic exercise group)e
NS (relaxation therapy group)e
S (combined therapy group)f
NS (control group)e
Fatigue, loss of appetite, nausea, taste change, sleep disturbance
S (aerobic exercise  group – fatigue, loss of  appetite, and sleep  disturbance only)f
NS (relaxation therapy group)e
S (combined therapy  group - fatigue and sleep  disturbance only)f
NS (control group)e
17 NAa Low
Multimodal intervention  (combining  physical-based  activities and  psychoeducational  components) Jarden et al,30 2009
(N = 42)
S S Affective cluster (Nervousness,  anxiety, stress)
NS
Functional cluster
(Muscle ache, joint ache)
S
Mucositis cluster
(Mouth and throat pain,  difficulty swallowing)
S
20
aOutcome consistency is not applicable as the 2 studies reported the effectiveness of the group of interventions on different patient outcomes.
bChanges in the score of only fatigue was significantly associated with treatment allocation.
cComparison in the extent of symptom severity improvement between baseline and 6 weeks postintervention, between the 2 groups
dComparison in the extent of symptom severity improvement between 6 weeks postintervention and 12 weeks postintervention, between the 2 groups
eStatistically nonsignificant reduction in severity of all symptoms/quality of life between preintervention and postintervention in a within-group comparison.
fStatistically significant reduction in severity of all symptoms/quality of life between preintervention and postintervention in a within-group comparison.
gThe study reported only within-group changes in their study. No between-group comparisons were made.
Abbreviations: NA, not applicable; NS, nonsignificant; S, significant.

PAIN–FATIGUE–SLEEP DISTURBANCE

Four studies reported the effect of the interventions on the severity of the pain–fatigue–sleep disturbance cluster.28,31–33 These interventions include physical-based intervention (nurse-delivered massage intervention28 and acupressure interventions32,33) and cognitive-behavioral intervention.31 Three31–33 used the M. D. Anderson Symptom Inventory to assess the severity and/or degree of interference of the cluster in the participants’ daily activities. Two28,31 used the 10-point numeric rating scales from the Brief Pain Inventory, Brief Fatigue Inventory, and Pittsburgh Sleep Quality Index to measure the severity of the individual symptom in the clusters. In general, the 4 studies obtained consistent results for the effect of the interventions on this cluster. Kwekkeboom et al31 reported a statistically significant reduction in the severity of all 3 symptoms in the cluster after the cognitive-behavioral intervention (P < .001), and the same applied to the distress level caused by these symptoms (P < .001). Between-group analyses also revealed significant differences in the severity of pain and fatigue between the intervention and control groups (P < .049), although the difference in sleep disturbance failed to reach statistical significance. Similarly, Miladinia et al28 reported significant improvement in the 3 symptoms in the cluster (P < .015) and in sleep quality (P = .003) among those receiving the massage intervention, these effects not being observed in the control group. With respect to the acupressure intervention described by Yeh et al,32 a significant difference in the extent of improvement was found only in severity of pain, but not fatigue or sleep disturbance, between intervention and control groups postintervention. Nevertheless, they reported a clinically significant reduction, defined as more than 30%, in the severity of all 3 symptoms associated with the intervention. In another study focused on acupressure intervention,33 a significant difference was reported in pain–fatigue–sleep disturbance scores, but it did not report the change in the scores of individual symptom. Overall, the severity of the cluster reportedly decreased among those receiving the interventions, suggesting their effectiveness in the relief of the symptom burden induced by this cluster among cancer patients.

COGNITIVE SYMPTOM CLUSTER

Three studies30,34,36 investigated the effect of nonpharmacological interventions on the cognitive symptom cluster. These interventions include a multimodal intervention30 and a cognitive-behavioral intervention (MBSR intervention34,36). They used a range of different instruments in their assessments, including the Stem Cell Transplantation Symptom Assessment Scale,30 M. D. Anderson Symptom Inventory,34 Cognitive and Affective Mindfulness Scale,36 and Everyday Cognition Scale.36 Inconsistent results were reported by these 3 studies on the effect of interventions on the severity of the cluster. Jarden et al30 reported that the severity (reduced concentration, memory problems, and fatigue) among patients who received the multimodal intervention was significantly lower than that among control subjects over time (P = .002). Lengacher et al34 were able to demonstrate a significant decrease in the severity of the cluster (distress, sadness, pain, and memory problems) among the MBSR intervention subjects (P = .01), but the between-group differences (intervention vs control) were reported not to be statistically significant (P value not reported). Likewise, when Reich et al36 assessed the level of memory problems and mindfulness to evaluate the effectiveness of the MBSR intervention in reducing the severity of the cognitive symptom cluster, they found that the between-group difference in severity was nonsignificant (P = .78), even though improvement was observed postintervention.

GASTROINTESTINAL SYMPTOM CLUSTER

Four studies30,33–35 assessed the effect of their interventions on the gastrointestinal symptom cluster, using the same instruments described above. The interventions involved were a multimodal intervention,30 a cognitive-behavioral intervention (ie, an MBSR intervention,34) 2 physical-based interventions (ie, a massage intervention,33 and an intervention involving acupoint stimulation.35) Inconsistent results for the between-group differences in the cluster severity were also reported. Jarden et al,30 Jiao et al,33 and Mao et al35 found significant between-group differences in the severity of the gastrointestinal cluster (nausea–vomiting–stomach pain–loss of appetite–diarrhea, nausea–poor appetite–dry mouth–vomiting, and loss of appetite–nausea–vomiting–constipation) (P = .017, P = .001, and P < .01). However, with that cluster defined as nausea, vomiting, lack of appetite, breathlessness, dry mouth, and numbness, Lengacher et al34 failed to observe any significant between-group differences in its severity, although there was such a difference in the intervention group (P = .005). Therefore, despite within-group changes in the severity of gastrointestinal clusters, the variations in between-group differences of this parameter among the 4 studies30,33–35 suggest that the evidence for any benefit from the interventions is inconclusive.

FATIGUE SYMPTOM CLUSTER

Two studies34,36 assessed the effect of the cognitive-behavioral interventions on the fatigue cluster. The interventions involved were both MBSR intervention. The instruments involved in symptom assessment include M. D. Anderson Symptom Inventory,34 Pittsburgh Sleep Quality Index,36 and Fatigue Symptom Inventory.36 Interestingly, both studies yielded contradictory findings on the effect of the MBSR intervention in this symptom cluster. While Reich et al36 reported a statistically significant between-group difference (P < .001) in the extent of improvement in the symptoms in the fatigue cluster, Lengacher et al34 noted a nonsignificant between-group difference for this outcome parameter (P value not reported). It is thus difficult to draw a concrete conclusion as to the effect of MBSR intervention on the fatigue symptom cluster.

In summary, data in these studies demonstrate the potential efficacy of the interventions in alleviating the severity of some of the symptom clusters experienced by cancer patients during and after treatment, particularly the pain–fatigue–sleep disturbance cluster. Overall, physical-based interventions were found to be effective in alleviating symptoms or clusters of symptoms, including pain–fatigue–sleep disturbance and gastrointestinal symptoms, although there was no evidence to support that these interventions are effective against cognitive symptoms. Likewise, educational interventions, including those that involve dissemination of supportive instructions, appeared effective in the management of certain symptoms in the reported symptom clusters, notably breathlessness. Meanwhile, although cognitive-behavioral or mindfulness-based interventions were shown to be effective in managing symptoms in the pain–fatigue–sleep disturbance cluster, their effect on other clusters remains unclear because of conflicting data. Nevertheless, it is worth making a note on the difficulty in giving any recommendation due to the limited number of studies, fair overall quality, and heterogeneity of interventions. Further studies are recommended to examine which type of intervention or which element in it plays a beneficial part in symptom cluster management.

Effect of Interventions on Patients QOL, Symptom Interference With Daily Lives, and Functional Ability

Four studies26,27,32,37 investigated the effect of the 4 different types of interventions on QOL. The evaluation of QOL was measured by the European Organization for the Research and Treatment of Cancer Quality of Life Questionnaire (EORTC-QLQ-C30),26 the 3 level version of EQ-5D (EQ-5D-3L),27 the World Health Organization Quality of Life,32 and Functional Assessment of Cancer Therapy–Breast Cancer scales.37 Generally, data from these studies indicate the potential of the interventions to improve the QOL of cancer patients during or after treatment. Notably, Charalambous et al26 observed that after a Guided Imagery and Progressive Muscle Relaxation intervention the QOL score of that group was significantly increased (P < .0001), whereas that of the control was significantly reduced (P < .0001). Similarly, Yorke et al27 demonstrated a differential effect of their educational-based respiratory distress symptom intervention on QOL 12 weeks after the start of the intervention, although this effect was not observed at the first follow-up assessment of patient outcomes, 4 weeks after the start of the intervention. The acupressure intervention described by Yeh et al32 yielded greater improvement in the QOL score of the physical health domain in the intervention than in the control group; however, the difference in QOL scores between the 2 groups did not reach statistical significance (P = .208). The aerobic exercise combined with relaxation therapy described by Zhang and Zhang37 showed significant improvement in QOL (P < .05), especially in social (P < .05) and functional well-being (P < .05). In addition, 1 study25 investigated the effect of interventions on functional ability, using the relevant subscale of the SF-36 Health Survey, Chinese version. The authors demonstrated that the functional ability levels of the participants who underwent psychoeducational intervention were significantly different from that observed among those who did not (P < .001). Moreover, 2 studies reported the effect of interventions on the level of symptom interference with patients’ daily lives through M. D. Anderson Symptom Inventory. Interestingly, both studies reported no significant difference in the level of symptom interference between the intervention participants and the control counterparts postintervention. In total, these studies show that all the 4 types of nonpharmacological interventions generally have a positive effect on patients’ QOL and functional ability, although no effect was reported regarding symptom interference with daily lives.

Strength of Evidence Indicating Intervention Effects on Patient Outcomes

As indicated by the level of certainty of the findings, the strength of the evidence generated from the included studies on the effect of reported interventions on various patient outcomes is shown in Table 4. The level of certainty on the effect of various intervention types on various symptom clusters and level of symptom interference with daily lives, QOL, and functional ability range from low to moderate. Overall, when the reported interventions are grouped by types, generally consistent effects were observed among educational interventions and physical-based interventions on the aforementioned patient outcomes. Nevertheless, given the low number of studies reporting such intervention types or the overall low-quality ratings of these studies, the strength of evidence on a positive effect of educational interventions and physical-based interventions is considered moderate. Meanwhile, the included studies that reported cognitive-behavioral interventions yielded inconsistent results on patient outcomes. While studies generally reported a positive effect of cognitive-behavioral interventions on certain symptoms in various symptom clusters, a study on MBSR intervention reported no effect on any of the symptom clusters examined.34 However, with multiple studies providing evidence on the effect of cognitive-behavioral interventions on patient outcomes, the strength of evidence on this type of interventions is also considered moderate. Moreover, given only 2 selected studies with modest quality rating reported the effect of multimodal interventions on patient outcomes, the strength of evidence on this intervention type is rated low.

In addition, the consistency between the intervention effects on symptom clusters examined in multiple studies, level of symptom interference with daily life, QOL, and functional ability is also assessed. Notably, the reported intervention effects on pain–fatigue–sleep disturbance symptom cluster were generally consistent, with the quality rating of those studies being the highest. In contrast, the reported effects on the cognitive symptom cluster appeared to be inconsistent, with the quality of relevant studies being modestly low. Generally inconsistent effects were also reported from the gastrointestinal symptom cluster and QOL. Therefore, the intervention effect on these outcome parameters should be interpreted with caution.

Discussion

The included studies of the present review had reported a wide array of nonpharmacological interventions that could benefit cancer patients in symptom management and QOL improvement. Our review demonstrates that these interventions can generally be classified into educational, cognitive-behavioral, physical-based interventions, and multimodal interventions. Particularly, a number of these interventions involved the provision of education for patients on the nature of symptoms experienced and the techniques required to manage these symptoms. The review shows that a decrease in the severity of certain symptom clusters is generally reported as a result of the interventions described. Notably, the severity of the pain–fatigue–sleep disturbance, cognitive, and gastrointestinal symptom clusters was significantly decreased. Such findings provide further evidence on the potential utility and benefits of nonpharmacological interventions in addressing undesirable symptom experience among cancer patients during and after treatment.

A common feature of the reported interventions is that, although they primarily targeted 1 or 2 of the symptoms in the cluster, they alleviated the severity of all of the symptoms in that cluster. In particular, Charalambous et al26 demonstrated that the Guided Imagery and Progressive Muscle Relaxation intervention, which primarily targets pain, fatigue, and psychological symptoms, was able to ameliorate the severity of nausea and vomiting. Likewise, the MBSR intervention reported by Reich et al,36 with components primarily targeting psychological symptoms, reported a significant alleviation of the cluster, including fatigue and drowsiness. In the cognitive behavior therapy described by Fleming et al,29 the components primarily targeted sleep disturbance, suggesting its potential to alleviate fatigue, anxiety, and depression. These findings indicate that interventions targeting 1 or 2 symptoms within a cluster could alleviate multiple symptoms within that cluster.

There are some caveats to this review. A considerable degree of heterogeneity in the study findings was reported. First, the nature of the interventions was generally diverse. Different interventions were designed to address different sets of symptoms with different etiologies. It is therefore likely that the content of the interventions reported might not be able to address certain symptoms that the studies examined. Second, the heterogeneity of the participants could play a role in the discrepancies of the intervention effects reported. Participants had various cancer diagnoses, received different forms of treatment, and were at different stages of the disease trajectory—response shift might therefore have influenced their QOL ratings. All these factors might have resulted in participants having different symptom experiences and severity. Third, the 13 studies used a wide range of instruments for symptom severity assessment to evaluate the effectiveness of the interventions.

Moreover, the outcomes assessed among the included studies of this review are largely limited to symptom cluster severity, QOL, and functional ability of the intervention participants. None of the studies have investigated the effect of nonpharmacological interventions on other outcomes, such as patient morbidity and mortality, and competence of self-care, as indicated in the Revised Symptom Management Conceptual Model presented by Dodd et al.5 As these outcomes can also be considered as indicators on the effectiveness of nonpharmacological interventions on promoting patient well-being, future studies may consider the assessment of effectiveness of interventions based on these outcomes.

Recommendations for future research work and practice in symptom cluster interventions should be directed to the following areas. First, large-scale RCTs should be carried out, ensuring that the trials have sufficient power to detect between-group differences in outcomes. The interventions described in the 13 studies all lasted for a relatively short time, with the longest reported intervention being only 6 weeks. Evidence for the effectiveness of long-term nonpharmacological interventions is generally lacking. Future work on longer-term interventions with additional follow-up assessments of patient outcomes should therefore be undertaken and would potentially provide useful data on the optimal period a particular intervention should last in order to achieve the maximum beneficial effect.

Second, future studies are encouraged to use theory, model, or framework to develop their interventions, measurement, and outcomes and to clarify the mechanisms of how interventions can work. In the 13 studies included in this review, although there were 4 that adopted theories or a framework to develop or support their interventions,25,27,28,32 only 1 study explained the potential relationship between interventions and outcomes.25 Interventions for symptom clusters are usually comprehensive, and measurements of outcomes are also multidimensional. Specific theories or frameworks should therefore be adopted to guide the study design and implementation.

Third, despite the abundance of studies that describe interventions for cancer patients still under treatment, few have actually developed interventions that specifically address symptom clusters among people no longer under active treatment. As indicated, this latter group of individuals experience a catalogue of symptoms38,39 and may need nonpharmacological interventions that help them overcome such symptoms, particularly those that limit physical and psychological health and that are known to affect QOL. In this light, more symptom cluster interventions designed for people after active treatment for various cancer types should be developed, addressing their currently unmet health needs.

Limitations

This review has 3 major limitations. First, the findings were summarized rather than synthesized because of the heterogeneity of interventions and methodologies. Second, some of the studies included30,32 had small sample sizes, perhaps leading to the possibility of small-sample bias that results in misleading interpretations of intervention effects. Third, only studies reporting RCTs are included in this review. The findings of those that were quasi-randomized or of pre-post test design were not analyzed, and certain additional nonpharmacological symptom cluster interventions might therefore have been excluded.

Conclusions

To date, a number of nonpharmacological interventions addressing a variety of cancer-associated symptom clusters have been developed. This review reveals that these interventions generally exert a beneficial effect by reducing symptom severity among the patients and in enhancing their QOL. Nevertheless, inconsistencies in the study findings do exist, potentially due to the use of varied instruments for symptom assessment in evaluating the effectiveness of interventions and to the heterogeneity of both participants and interventions. Moreover, current data are still sparse on the effectiveness of cluster interventions in specific types of cancer patients receiving specific treatments. As different types of cancer and treatment are associated with different cancer-related clusters, the scarcity of such data can lead to limitations in the comparison of the beneficial effects of various interventions on a particular cluster. In light of these limitations, further research could provide useful additional data on the effectiveness of the various types of intervention on improving symptom experience among cancer patients during and/or after treatment.

References

1. Torre LA, Bray F, Siegel RL, Ferlay J, Lortet-Tieulent J, Jemal A. Global cancer statistics, 2012. CA Cancer J Clin. 2015;65(2):87–108.
2. Hashim D, Boffetta P, La Vecchia C, et alThe global decrease in cancer mortality: trends and disparities. Ann Oncol. 2016;27(5):926–933.
3. Giacalone A, Lleshi A, Zanet E, Tirelli U. Symptom burden in cancer survivors 1 year after diagnosis: a report from the American Cancer Society’s Studies of Cancer Survivors. Cancer. 2012;118(7):1955. author reply 1955-1956.
4. Gosain R, Miller K. Symptoms and symptom management in long-term cancer survivors. Cancer J. 2013;19(5):405–409.
5. Dodd M, Janson S, Facione N, et alAdvancing the science of symptom management. J Adv Nurs. 2001;33(5):668–676.
6. Molassiotis A, Wengstrom Y, Kearney N. Symptom cluster patterns during the first year after diagnosis with cancer. J Pain Symptom Manage. 2010;39(5):847–858.
7. Kim HJ, McGuire DB, Tulman L, Barsevick AM. Symptom clusters: concept analysis and clinical implications for cancer nursing. Cancer Nurs. 2005;28(4):270–282. quiz 283-284.
8. Kim HJ, McDermott PA, Barsevick AM. Comparison of groups with different patterns of symptom cluster intensity across the breast cancer treatment trajectory. Cancer Nurs. 2014;37(2):88–96.
9. Duan F, Cheng J, Jiang J, Chang J, Zhang Y, Qiu S. Whole-brain changes in white matter microstructure after radiotherapy for nasopharyngeal carcinoma: a diffusion tensor imaging study. Eur Arch Otorhinolaryngol. 2016;273(12):4453–4459.
10. Schagen SB, Wefel JS. Chemotherapy-related changes in cognitive functioning. EJC Suppl. 2013;11(2):225–232.
11. Cheung YT, Ng T, Shwe M, et alAssociation of proinflammatory cytokines and chemotherapy-associated cognitive impairment in breast cancer patients: a multi-centered, prospective, cohort study. Ann Oncol. 2015;26(7):1446–1451.
12. Greene-Schloesser D, Robbins ME. Radiation-induced cognitive impairment—from bench to bedside. Neuro Oncol. 2012;14(suppl 4):iv37–iv44.
13. Smith HR. Depression in cancer patients: pathogenesis, implications and treatment [review]. Oncol Lett. 2015;9(4):1509–1514.
14. Fan G, Filipczak L, Chow E. Symptom clusters in cancer patients: a review of the literature. Curr Oncol. 2007;14(5):173–179.
15. Lin S, Chen Y, Yang L, Zhou J. Pain, fatigue, disturbed sleep and distress comprised a symptom cluster that related to quality of life and functional status of lung cancer surgery patients. J Clin Nurs. 2013;22(9–10):1281–1290.
16. Dong ST, Costa DS, Butow PN, et alSymptom clusters in advanced cancer patients: an empirical comparison of statistical methods and the impact on quality of life. J Pain Symptom Manage. 2016;51(1):88–98.
17. Hökkä M, Kaakinen P, Pölkki T. A systematic review: non-pharmacological interventions in treating pain in patients with advanced cancer. J Adv Nurs. 2014;70(9):1954–1969.
18. Larkin D, Lopez V, Aromataris E. Managing cancer-related fatigue in men with prostate cancer: a systematic review of non-pharmacological interventions. Int J Nurs Pract. 2014;20(5):549–560.
19. Kwekkeboom KL. Cancer symptom cluster management. Semin Oncol Nurs. 2016;32(4):373–382.
20. Xiao W, Chow KM, So WK, Leung DY, Chan CW. The effectiveness of psychoeducational intervention on managing symptom clusters in patients with cancer: a systematic review of randomized controlled trials. Cancer Nurs. 2016;39(4):279–291.
21. Moher D, Liberati A, Tetzlaff J, Altman DG, PRISMA Group. Preferred Reporting Items for Systematic Reviews and Meta-analyses: the PRISMA statement. BMJ. 2009;339:b2535.
22. Downs SH, Black N. The feasibility of creating a checklist for the assessment of the methodological quality both of randomised and non-randomised studies of health care interventions. J Epidemiol Community Health. 1998;52(6):377–384.
23. Selman LE, Brighton LJ, Hawkins A, et alThe effect of communication skills training for generalist palliative care providers on patient-reported outcomes and clinician behaviors: a systematic review and meta-analysis. J Pain Symptom Manage. 2017;54(3):404–416.e5.
24. US Preventive Services Task Force. Procedure Manual web site. https://www.uspreventiveservicestaskforce.org/Page/Name/procedure-manual. Accessed October 31, 2018.
25. Chan CW, Richardson A, Richardson J. Managing symptoms in patients with advanced lung cancer during radiotherapy: results of a psychoeducational randomized controlled trial. J Pain Symptom Manage. 2011;41(2):347–357.
26. Charalambous A, Giannakopoulou M, Bozas E, Marcou Y, Kitsios P, Paikousis L. Guided Imagery and Progressive Muscle Relaxation as a cluster of symptoms management intervention in patients receiving chemotherapy: a randomized control trial. PLoS One. 2016;11(6):e0156911.
27. Yorke J, Lloyd-Williams M, Smith J, et alManagement of the respiratory distress symptom cluster in lung cancer: a randomised controlled feasibility trial. Support Care Cancer. 2015;23(11):3373–3384.
28. Miladinia M, Baraz S, Shariati A, Malehi AS. Effects of slow-stroke back massage on symptom cluster in adult patients with acute leukemia: supportive care in cancer nursing. Cancer Nurs. 2017;40(1):31–38.
29. Fleming L, Randell K, Harvey CJ, Espie CA. Does cognitive behaviour therapy for insomnia reduce clinical levels of fatigue, anxiety and depression in cancer patients?Psychooncology. 2014;23(6):679–684.
30. Jarden M, Nelausen K, Hovgaard D, Boesen E, Adamsen L. The effect of a multimodal intervention on treatment-related symptoms in patients undergoing hematopoietic stem cell transplantation: a randomized controlled trial. J Pain Symptom Manage. 2009;38(2):174–190.
31. Kwekkeboom KL, Abbott-Anderson K, Cherwin C, Roiland R, Serlin RC, Ward SE. Pilot randomized controlled trial of a patient-controlled cognitive-behavioral intervention for the pain, fatigue, and sleep disturbance symptom cluster in cancer. J Pain Symptom Manage. 2012;44(6):810–822.
32. Yeh CH, Chien LC, Lin WC, Bovbjerg DH, van Londen GJ. Pilot randomized controlled trial of auricular point acupressure to manage symptom clusters of pain, fatigue, and disturbed sleep in breast cancer patients. Cancer Nurs. 2016;39(5):402–410.
33. Jiao J, Luo Z, Qiang W. Study for the effects of acupressure on symptom clusters for patients with advanced cancer. J Nurses Train (Chinese). 2015;30(24):2221–2223.
34. Lengacher CA, Reich RR, Post-White J, et alMindfulness based stress reduction in post-treatment breast cancer patients: an examination of symptoms and symptom clusters. J Behav Med. 2012;35(1):86–94.
35. Mao T, Chen Y, Liu X, Wei D, Li H. Effects of acupoints stimulation combined with gastric pacing on gastrointestinal symptom clusters in thoracic cancer patients undergoing chemoradiotherapy. J Nurs Sci (Chinese). 2018;33(8):38–41.
36. Reich RR, Lengacher CA, Alinat CB, et alMindfulness-based stress reduction in post-treatment breast cancer patients: immediate and sustained effects across multiple symptom clusters. J Pain Symptom Manage. 2017;53(1):85–95.
37. Zhang Y, Zhang J. Influence of aerobic exercise combined with relaxation therapy on symptom clusters and quality of life of breast cancer patients with chemotherapy. Chin Nurs Res. 2016;30(8A):2764–2767.
38. So WK, Ling GC, Choi KC, et alComparison of health-related quality of life and emotional distress among Chinese cancer survivors. Int J Nurs Pract. 2013;19(3):306–317.
39. Ng HS, Roder D, Koczwara B, Vitry A. Comorbidity, physical and mental health among cancer patients and survivors: an Australian population-based study. Asia Pac J Clin Oncol. 2018;14(2):e181–e192.
Keywords:

Cancer; Nonpharmacological intervention; Quality of life; Symptom cluster

Copyright © 2019 Wolters Kluwer Health, Inc. All rights reserved.