Effect of cognitive behavioral therapy on cancer-related fatigue and psychological status in ovarian cancer patients: A meta-analysis

Abstract

BACKGROUND

Substantial clinical evidence supports the efficacy of cognitive behavioral therapy (CBT) for various diseases, particularly in oncology. However, the true impact of CBT interventions on cancer-related fatigue and mental health in patients with ovarian cancer remains unknown.

AIM

To evaluate the effects of CBT on fatigue, anxiety, depression and quality of life in patients with ovarian cancer.

METHODS

Randomized controlled trials (RCTs) on CBT for patients with ovarian cancer were searched in the PubMed, EMBASE, Web of Science and Cochrane Library databases. According to the preferred reporting items for systematic reviews and meta-analyses statement, we formulated the inclusion and exclusion criteria, strictly screened the literatures, extracted data and performed a meta-analysis.

RESULTS

Six RCTs with 332 ovarian cancer patients were included. Compared with the control group, cancer fatigue [mean difference (MD) = -0.98, 95% confidence interval (CI): -1.47 to -0.50], anxiety [standardized mean difference (SMD) = -0.64, 95%CI: -0.91 to -0.36] and depression levels (SMD = -0.41, 95%CI: -0.76 to -0.06) of the patients in the experimental group reduced after CBT intervention. Quality of life (MD = 1.28, 95%CI: 0.65 to 1.90) and sleep quality (MD = -0.49, 95%CI: -0.66 to -0.33) of the patients improved, and the differences between the groups were statistically significant (P < 0.01). The quality evaluation results suggested that the quality of the included RCTs was low. The meta-regression results showed that patient age and nurse guidance affected treatment outcomes, especially anxiety, whereas the specific method of CBT had a non-significant effect.

CONCLUSION

CBT effectively improves mental status and cancer-related fatigue in patients with ovarian cancer undergoing chemotherapy. Future research should prioritize adequately powered RCTs with standardized outcome measures and longitudinal designs to establish sustained efficacy.

Key Words: Ovarian cancer; Cognitive behavioral therapy; Cancer-related fatigue; Anxiety; Depression; Meta-analysis

Core Tip: Despite the role of cognitive behavioral therapy (CBT) in oncology, its specific impact on debilitating fatigue and mental health in patients with ovarian cancer remains unclear. This meta-analysis revealed that CBT simultaneously alleviates cancer-related fatigue, anxiety, depression, and sleep disturbances, while significantly improving quality of life in chemotherapy-treated patients. Nurse-guided delivery enhances these benefits. These findings suggest that CBT may be an essential non-pharmacological strategy for managing this population’s complex symptom burden.

INTRODUCTION

Ovarian cancer (OC) is the seventh most prevalent malignancy in women worldwide, accounting for over 324000 annual diagnoses and 200000 deaths worldwide[1]. Epidemiological analyses have revealed a 50-year surge in OC incidence, particularly in developed nations, where it is the fifth leading cause of cancer-related mortality among women[2]. Despite the declining incidence trends observed in certain regions (e.g., United States, 2000-2017), delayed diagnosis persists (> 70% of cases at advanced stages), contributing to suboptimal 5-year survival rates below 50%[3]. Current management relies on cytoreductive surgery with platinum-based chemotherapy, for which complete resection rates demonstrate strong prognostic associations with survival outcomes[4].

Patients with OC frequently experience multidimensional chronic symptoms, including anxiety, depression, sleep disturbances, somatic complaints, and impaired health-related quality of life (QoL). These manifestations not only disrupt daily functioning but may also synergize with psychosocial stressors (e.g., social isolation) to exacerbate tumor progression and survival outcomes[5]. Despite the documented high rates of psychological distress (> 60% prevalence), systematic psychological support remains in routine oncology care. For example, cancer-related fatigue, anxiety, and depression are common problems faced by cancer patients[6,7]. They seriously affect the QoL of patients, and multiple painful symptoms can persist for a long time even after curative treatment has ended. Therefore, it is necessary to develop effective psychological interventions to support patients with OC undergoing chemotherapy.

Cognitive behavioral therapy (CBT) is a psychological treatment method based on the cognitive-behavioral model, with the core assumption that psychological disorders and emotional disturbances are maintained by maladaptive cognitive patterns and avoidance behaviors. It aims to alleviate symptoms by changing cognitive and behavioral patterns. In recent years, CBT has led to the development of new therapies based on traditional cognitive concepts such as acceptance and commitment therapy (ACT), mindfulness-based cognitive therapy, dialectical behavior therapy, and metacognitive therapy. CBT has been widely proven to be effective for various psychological disorders such as anxiety, depression, and obsessive-compulsive disorder, and is recommended as a first-line therapy in clinical guidelines[8]. However, there is a lack of reliable evidence on the efficacy of CBT in the treatment of psychological problems in patients with OC. The true effects of CBT interventions in patients with OC remain unknown, and no published systematic reviews or meta-analyses have been conducted in this context. This study aimed to investigate the efficacy of CBT on mental health status and cancer-related fatigue in patients with OC by evaluating published controlled trials to obtain the most authentic research evidence.

MATERIALS AND METHODS

This meta-analysis was conducted in accordance with the preferred reporting items for systematic reviews and meta-analyses guidelines and recommendations of the Cochrane collaboration[9]. The study was registered in PROSPERO under the registration number of CRD42022327092.

Search strategies

The following databases were searched: PubMed/MEDLINE, EMBASE, Web of Science, and the Cochrane Library. The retrieval time for all databases was from the start of the databases to March 1, 2025, and the search was conducted without language restrictions. The search was conducted for “ovarian cancer”, “cognitive behavioral therapy” (including specific therapies from the third wave), and “randomized clinical trials” as keywords. The specific search strategy is shown in Supplementary Table 1. In addition, the reference track, ClinicalTrials.gov (https://www.clinicaltrials.gov/, accessed March 2025) were used as a supplementary search.

Selection criteria

The inclusion and exclusion criteria were pre-specified according to the PICOS principle (population, intervention, comparison, outcomes and study), as shown in Table 1.

Table 1 Inclusion and exclusion criteria.

PICOS	Inclusion criteria	Exclusion criteria
Population	Patients diagnosed with ovarian cancer who are undergoing chemotherapy after surgery	Patients with other concomitant diseases
Intervention	Using CBT for treatment, but not limited to specific schools of thought within CBT theory, including the third wave, such as ACT, MBCT, DBT, etc.	In addition to CBT, the experimental group also received other treatments
Comparison	CBT is not used, but specific treatment methods are not restricted	No controlled study
Outcome	Cancer-related fatigue, mental health status (such as anxiety and depression levels), and quality of life of patients; there are no restrictions on the scales used to assess the outcomes	Unreliable or incomplete data
Study design	Randomized controls trails	Case report, conference paper, review article and letter; duplicate literature or duplicate studies

PICOS: Population, intervention, comparison, outcomes and study; CBT: Cognitive behavioral therapy; ACT: Acceptance and commitment therapy; MBCT: Mindfulness-based cognitive therapy; DBT: Dialectical behavior therapy.

Data extraction

Detailed information of the included studies was collected independently by two authors (Zhao F and Bo Y) and included: (1) Name of the first author and year of publication; (2) Study design; (3) Patient baseline characteristics (age, sample size, tumor stage); (4) Information on treatment/control group, including the specific forms and details of CBT intervention, follow-up time, etc.; (5) Primary outcome indicators (mental health status, such as anxiety, depression; QoL; cancer-related fatigue); and (6) Secondary outcome indicators [quality of sleep (QoS)].

All data are presented as mean ± SD. If multiple groups of data meet the inclusion and exclusion criteria in the same literature, the data from different groups were extracted and marked. Based on internationally recognized criteria for psychological assessment scales (such as validity, reliability, and scope of use), select the set of data that best reflected the results of the analysis. In this meta-analysis, the comparison was the change in the value of the data pre- and post-intervention between the experimental group and the control group; therefore, the extracted data needed to be converted.

Change value = post-intervention assessment value-baseline (before intervention) assessment value. If the change value was negative, the estimated value after the intervention was lower than that before the intervention; otherwise, the result was positive, indicating an increase. The formula for calculating the mean of the change value is as follows: The formula for calculating SD value change as follow: (corr = 0.5)[10,11].

It should be noted that higher scores on depression scales indicate a more severe degree of depression, and this also holds true for anxiety scales. For both scales, a greater magnitude of pre- to post-intervention score change (consistent with the direction of either decrease or increase) reflected a more significant improvement in depression or anxiety symptoms. In contrast, higher scores on the QoL and QoS scales corresponded to better outcomes in these domains. Therefore, conclusions should be drawn based on different outcome measures.

Quality evaluation of included studies

Two authors (Zhao F and Bo Y) independently assessed the risk of bias of the included studies. Any disagreements during evaluation were resolved by a third author (Su XL). The randomized controlled trials (RCTs) used the revised Cochrane collaboration risk of bias tool (ROB 2.0)[10].

Statistical analysis

R software (version R x64 4.2.2; R Foundation for Statistical Computing, Vienna, Austria; https://www.r-project.org/) was used for data analysis; the mean difference (MD) or standardized mean difference (SMD) with a 95% confidence interval (CI) for continuous variables was calculated. Heterogeneity was assessed using I² and τ² statistical tests. An I² > 50% indicated high heterogeneity. A random-effects model was adopted when high heterogeneity was observed. Otherwise, a fixed effects model was used. Moreover, sources of heterogeneity were searched, and subgroup analyses according to detection time points and sensitivity analyses were performed. P < 0.05 was considered statistically significant.

RESULTS

Literature selection

After searching the following databases, we identified a total of 259 potential studies (PubMed, n = 13; EMBASE, n = 43; Web of Science, n = 88; the Cochrane Library, n = 110; supplementary searching, n = 3), of which 59 were duplicate studies (Figure 1). By reviewing titles and abstracts, 253 studies were excluded for failing to meet inclusion criteria or lacking full-text availability. Ultimately, 6 studies involving 332 patients were included (all English literature).

Figure 1  Literature screening flow chart.

Characteristics of the included studies

Table 2 shows that among the included studies, 4 studies originated from China and the United States, and the others were from the United Kingdom and Canada[12-17]. Two studies were multicenter, but only 2 studies had been registered. Baseline characteristics were consistent between the experimental and control groups in all studies (e.g., age, sex, and sample size). All studies assessed the mental health status of patients after the intervention, including levels of depression and anxiety; four studies assessed the relief of patients’ cancer fatigue; three studies assessed the improvement in patients’ QoL or QoS. However, the scales used for the assessment were somewhat different.

Table 2 Characteristics of the six included studies[12-17].

Ref.	Country	Study design	Multi-center	Registered	Record patient time	Ovarian cancer patients	Age (T)	Age (C)	Sample size		Intervention (T)	Intervention (C)	Outcome
									T (n)	C (n)
Frangou et al[12]	United Kingdom	RCT	Yes	Yes	November 2013 to January 2018	Primary or relapsed epithelial OC	58.1 ± 9.46	60.9 ± 10.2	34	33	CBT	Standard care	Depression, QoL
Zhou et al[13]	China	RCT	No	NR	January 2018 to March 2019	Meets the diagnostic criteria for OC	58.64 ± 13.82	60.23 ± 15.78	37	36	CBT + conventional nursing	Conventional nursing	Cancer-related fatigue, depression, anxiety, sleep quality, QoL
Zhang et al[14]	China	RCT	No	NR	November 2014 to November 2015	Ages of 18 and 80 with OC	NR	NR	33	34	CBT + exercise	Usual care	Cancer-related fatigue, depression, sleep quality
Petzel et al[15]	United States	RCT	Yes	Yes	September 2012 to February 2013	Stage III/IV or recurrent (any stage) epithelial OC	59.6 ± 10.0	55.5 ± 8.4	16	12	Website based on CBT	Control website	Distress, depression, anxiety
Moonsammy et al[16]	Canada	RCT	No	NR	March 2011 to July 2011	Newly diagnosed with OC	52.7 ± 12.1	57.8 ± 12.0	3	5	CBT	Surveillance	Cancer-related fatigue, depression, anxiety
Rost et al[17]	United States	RCT	No	NR	NR	Women with stage III or IV OC	56	56	15	16	ACT	Treatment as usual	Distress, depression, anxiety, QoL, acceptance

T: Treatment group; C: Control group; RCT: Randomized controlled trial; NR: Not report; OC: Ovarian cancer; CBT: Cognitive behavioral therapy; ACT: Acceptance and commitment therapy; QoL: Quality of life.

Table 3 presents treatment characteristics. The CBT intervention in three studies included exercise, that is, cognitive intervention combined with behavioral intervention. Of the remaining studies, two used conventional CBT and one used CBT-based ACT. Interventions were mostly provided through collaboration between oncology nurses or nurses with experience in cancer care (two studies), clinical psychologists (two studies), and CBT counsellors (two studies). All interventions were performed between or during patients’ chemotherapy cycles. The frequency of intervention ranged from 2-3 times per week to once every two weeks, with 12 consecutive sessions of intervention lasting for 2-3 months. Specific forms of CBT interventions included face-to-face sessions, telephone interviews, and website visits. In terms of follow-up time, three studies continued to observe for three months after the end of the intervention, one study observed within one month, and one study observed for 24 months.

Table 3 Characteristics of cognitive behavioral therapy treatment, n (%).

Ref.	Intervention (T) specific method	Intervention (C) specific method	CBT implementer	Intervention duration/frequency/total times or time	Intervention time point	Assessment time point	Follow-up time after treatment	Lost to follow-up
Frangou et al[12]	CBT (face to face sessions)	Standard care	Doctoral-level clinical/counselling psychologist	90 minutes/NR/3 months	The 6-12 weeks post-chemotherapy	Baseline/after treatment 3, 6, 9, 12, 15, 18, 24 months	24 months	40 (37.4)
Zhou et al[13]	At-home CBT + conventional nursing	Conventional nursing	Nurses	Half an hour/3-4 times per week/NR	In the light of chemotherapy	Baseline/after treatment 1, 2, 3 months	3 months	0 (0)
Zhang et al[14]	Nurse-led home-based exercise + CBT	Nurse-led usual care	CBT-trained nurses	1 hour/once weekly/12 consecutive weeks	Before the sixth chemotherapy treatment	Baseline/immediately after treatment/3 month	3 months	3 (4.3)
Petzel et al[15]	Visit website (social cognitive theory + CBT)	Visit website (usual care materials)	Medical and psychological providers	Unlimited time/2-3 times per week/60 days	Post-operative checkup or planned chemotherapy	Baseline/1 month	1 month	6 (17.1)
Moonsammy et al[16]	CBT (counselling sessions by phone)	Surveillance	CBT-trained nurses	1 hour/once every two weeks/12 weeks	Before completing two cycles of adjuvant chemotherapy	Baseline/immediately after intervention/3 months	3 months	5 (26.3)
Rost et al[17]	ACT (face to face meetings)	Treatment as usual	A PhD-level clinical psychologist	1 hour/4 months/12 occasions	In the chemotherapy treatments	Baseline/at the end of the 4th, 8th, and final (12th) session	End of treatment	16 (34)

T: Treatment group; C: Control group; CBT: Cognitive behavioral therapy; ACT: Acceptance and commitment therapy; NR: Not report; Baseline: Before treatment.

Quality evaluation of the included studies

Figure 2 shows the quality assessment of the included studies using the Cochrane RoB 2.0. This tool assesses bias risk from five perspectives: “Randomization process”, “deviations from intended interventions”, “missing outcome data”, “measurement of the outcome” and “selection of the reported result”, with results represented “high risk”, “low risk”, or “some concerns risk/uncertain risk”.

Figure 2 Risk of bias of the included studies by the Cochrane RoB 2.0 tool. A: Risk of bias graph; B: Risk of bias summary. CBT: Cognitive behavioral therapy.

Only 3 (50%) studies in the included RCTs mentioned “random” allocation methods, but no specific methods were detailed. So they were judged as “uncertain risks” in random sequence generation. The other studies were low-risk. Three studies described allocation concealment, three studies reported single-blind trials (blinding of participants, intervention implementers, or measurers), and only one study reported blinding of intervention implementers and outcome measures. The blinding of the other studies was unclear. This may be related to the difficulty of practicing blinding in clinical trials. Five studies reported loss-to-follow-up and incomplete data, ranging from 4.3% to 37.4%, which was related to the poor prognosis of patients with OC. The loss-to-follow-up rate of four studies was more than 15% due to disease progression, patient death, reluctance to continue to receive intervention, etc. Most studies were judged to be at uncertain risk of bias due to uncertainty about whether missing outcome data would have affected the results. Regarding selective reporting bias, only two studies had a low risk of bias, while the risk of bias for the remaining studies was uncertain. However, in terms of overall risk, the majority of studies were assessed as high risk.

Cancer-related fatigue

Three of the included studies used the revised piper fatigue scale to assess patient fatigue. Compared to the control group, the cancer-related fatigue degree in the experimental group was reduced after CBT intervention (MD = -0.98, 95%CI: -1.47 to -0.50, P < 0.01; heterogeneity I² = 0%), and the difference between the two groups was statistically significant. The results of the subgroup analysis showed that the reduction effect was more pronounced at three months after the intervention than at one month after the intervention (Figure 3A). In addition, a subgroup meta-analysis comparing the results of the subscales of cancer fatigue also showed that after three months of treatment, CBT reportedly reduced the fatigue of patients in terms of behavioral, sensory, and cognitive fatigue, and the difference between the groups was statistically significant. Given that heterogeneity was > 50%, a random effects model was used; the MD value was -2.66 and the 95%CI was -4.07 to -1.25, P < 0.01 (Figure 3B). Further results from the Hedges’ g test showed Hedges’ g = -0.286, 95%CI: -0.577 to 0.005, suggesting that the experimental intervention had no significant effect on improving cancer-related fatigue.

Figure 3 Meta-analysis of cancer-related fatigue of ovarian cancer patients after cognitive behavioral therapy treatment. A: Total cancer-related fatigue; B: Subscales of cancer-related fatigue (negative values indicate a reduction in cancer-related fatigue). CBT: Cognitive behavioral therapy; CI: Confidence interval.

Depression

Depression is the most common psychological problem among patients with cancer. Three studies used the Zung self-rating depression scale, one used the patient health questionnaire-9 depression test, one used the center for epidemiologic studies-depression scale, and one used the hospital anxiety and depression scale to assess patients’ depression level; therefore, the pooled effect size was analyzed using SMD. The results of the meta-analysis showed that, compared with the control group, after the CBT intervention, patients’ depression scores were reduced, and the difference between the groups was statistically significant (SMD = -0.41, 95%CI: -0.76 to -0.06, P < 0.01; heterogeneity I² = 61.5%). The results of the subgroup analysis showed that the depression level of patients in the experimental group was not significantly reduced at one month (SMD = -0.31, 95%CI: -0.81 to 0.20; heterogeneity I² = 60.4%) or three months (SMD = -0.52, 95%CI: -1.08 to 0.04; heterogeneity I² = 71.2%) post-intervention (Figure 4A). However, the heterogeneity of both the overall and subgroup meta-analyses was high (> 50%), which may have been related to the use of different depression assessment scales. Further results from the Hedges’ g test showed Hedges’ g = -0.324, 95%CI: -0.600 to -0.048, indicating that the experimental intervention had a significant effect on improving depression.

Figure 4 Meta-analysis of depression and anxiety of patients with ovarian cancer after cognitive behavioral therapy treatment. A: Depression; B: Anxiety (negative values indicate a reduction in depression or anxiety). CBT: Cognitive behavioral therapy; CI: Confidence interval.

Anxiety

The meta-analysis of anxiety included eight independent data groups at two time points from four studies. Among the four included studies, one used the self-rating anxiety scale, one used the hospital anxiety depression scale, one used the beck anxiety inventory, and one used the state trait anxiety inventory (STAI-Y) scale to assess anxiety in patients. The pooled effect size was analyzed using SMD. In Moonsammy et al[16], the STAI-Y scale was used to evaluate total and trait anxiety, resulting in two anxiety values; thus, data analysis of the two groups. The meta-analysis results showed that after CBT intervention, the anxiety level of patients in the experimental group was significantly declined (SMD = -0.64, 95%CI: -0.91 to -0.36, P < 0.01; heterogeneity I² = 21.7%); the difference between the groups was statistically significant. The results of the subgroup analysis showed that the anxiety level of patients in the experimental group was significantly reduced at one month after treatment (SMD = -0.57, 95%CI: -0.91 to -0.23), but there was no statistically significant difference at three months (SMD = -0.34, 95%CI: -1.94 to 1.03) (Figure 4B). Further results from the Hedges’ g test showed Hedges’ g = -0.526, 95%CI: -0.887 to -0.165, indicating that the experimental intervention had a significant moderating effect on anxiety improvement.

QoS and QoL

Two studies reported the QoS and QoL. The Pittsburgh sleep quality index was used to evaluate the incidence and type of sleep disorders. The QoL questionnaire-core 30 scale, which includes six single items, three cardinal symptom modules, five functional modules, and one general health module, was used to evaluate QoL. As shown in Figure 5, the QoS and QoL of patients in the experimental group improved after the CBT intervention; the difference between the groups was statistically significant (MD = -0.49, 95%CI: -0.66 to -0.33; MD = 1.28, 95%CI: 0.65 to 1.90, P < 0.01).

Figure 5 Meta-analysis of quality of life and quality of sleep. A: Quality of life-general health; B: Quality of sleep; C: Subscales of quality of sleep (positive value indicates an improvement in quality of life or quality of sleep). CBT: Cognitive behavioral therapy; CI: Confidence interval.

Sensitivity analysis

The impact of each study on the pooled results was assessed by sequentially excluding individual studies (Figure 6). In the meta-analysis of cancer-related fatigue, heterogeneity remained at 0% regardless of which study was removed. However, in the meta-analysis of anxiety, heterogeneity decreased to 0% after excluding a specific study (Moonsammy et al[16]), suggesting that this study may be the primary source of heterogeneity. However, in the meta-analysis of depression, excluding any of the included studies did not significantly alter the heterogeneity of the effect size, which remained high. This suggests that individual studies were not the primary source of heterogeneity, and the stability of the meta-analysis results did not undergo significant changes, thereby validating the rationality and reliability of our analysis.

Figure 6 Sensitivity analysis of cancer-related fatigue, depression, and anxiety of patients with ovarian cancer after cognitive behavioral therapy treatment. A: Cancer-related fatigue; B: Depression; C: Anxiety.

Meta-regression analysis

Univariate meta-regression analyses of the relationships among depression, anxiety, patient characteristics, and specific interventions were performed. As shown in Table 4, there was no correlation between patients’ depression levels and patients’ age, nationality, specific interventions of CBT, or depression assessment scale (all P > 0.05). However, whether the implementation process was nurse-guided had a significant effect on depression (P < 0.05). In contrast, nationality was associated with an improvement in anxiety levels after the intervention. Additionally, whether nurses provided guidance was associated with patients’ anxiety and depression levels post-intervention (P < 0.05), suggesting that nurses’ involvement may influence intervention outcomes. However, it should be noted that nurse involvement may help alleviate patients’ depression (with a negative estimate), but the presence of nurses may increase anxiety levels (with a positive estimate). Other than this, we did not find any evidence that any other moderating factor influenced the strength of the depression-anxiety association.

Table 4 Meta-regression analysis of depression and anxiety.

Analysis	Depression				Anxiety
	Estimate (SE)	95%CI	P value		Estimate (SE)	95%CI	P value
Patients characteristics
Age	0.02 (0.14)	-0.26 to 0.29	0.8838		-0.06 (0.04)	-0.14 to 0.03	0.1733
China	-0.91 (0.59)	-2.06 to 0.24	0.1229
United States	-0.22 (0.65)	-1.48 to 1.05	0.7388		-1.28 (0.56)	-2.36 to -0.19	0.0217a
United Kingdom	-0.16 (0.67)	-1.46 to 1.15	0.8127		-0.85 (0.59)	-2.01 to 0.31	0.1497
Intervention methods
CBT/exercise	-0.49 (0.43)	-1.33 to 0.34	0.2485
I-CBT	0.98 (0.55)	-0.10 to 2.05	0.0767		0.06 (0.53)	-0.98 to 1.10	0.9092
Traditional-CBT	0.23 (0.41)	-0.57 to 1.02	0.5727		-0.27 (0.40)	-1.06 to 0.51	0.4966
Assessment scales
CES-D	0.68 (0.69)	-0.66 to 2.03	0.3179	STAI-Y	0.88 (0.64)	-0.37 to 2.14	0.1704
HADS	0.98 (0.61)	-0.21 to 2.16	0.1073	HADS	0.06 (0.53)	-0.98 to 1.10	0.9092
PHQ-9	0.53 (0.53)	-0.50 to 1.56	0.3157	SAS	-0.40 (0.40)	-1.19 to 0.40	0.3266
SDS	-0.22 (0.45)	-1.10 to 0.66	0.6292
Nurse-led	-0.56 (0.28)	-1.12 to -0.00	0.0486a		1.15 (0.55)	0.08 to 2.22	0.0359a

^aP < 0.05.

CI: Confidence interval; CBT: Cognitive behavioral therapy; I-CBT: Internet-based cognitive behavioral therapy; CES-D: Center for epidemiologic studies-depression scales; HADS: Hospital anxiety-depression scale; PHQ-9: Patient health questionnaire-9; SDS: Self-rating depression scale; STAI-Y: State trait anxiety inventory scale; SAS: Self-rating anxiety scale.

DISCUSSION

Cancer-related fatigue is a persistent subjective feeling of physical, emotional, and/or cognitive tiredness or weakness resulting from cancer treatment; this is a problem faced by many patients with cancer[18,19]. Women with OC have been reported to experience fatigue over time, which has also been reported as a pre-diagnostic symptom. Treatments for cancer-related fatigue include pharmacological treatments, such as psychostimulants[20], and non-pharmacological treatments, including CBT[21] and physical exercise[22]; these have been reported to significantly reduce fatigue in patients. Similar results were observed in the present meta-analysis. Regardless of whether it was one month or three months after CBT intervention, compared with the control group, the cancer fatigue of the patients was reduced, and the intergroup differences were statistically significant (MD = -0.98, 95%CI: -1.47 to -0.50, P < 0.01). In addition, the results of the subgroup meta-analysis suggested that CBT treatment was effective in improving behavioral, sensory, and cognitive fatigue but had no significant effect on affective fatigue.

Psychological problems such as anxiety and depression are more prevalent in patients with OC, especially in newly diagnosed patients. The occurrence of these symptoms was significantly associated with factors such as marital status, presence of pain, and chemotherapy history. Therefore, identifying and managing these psychological problems is important for improving the overall health of patients with OC[23,24]. Depressive symptoms have been reported in more than 50% of patients with OC after the completion of chemotherapy for primary or recurrent disease. Depression is a toxicity caused by chemotherapy itself, a hypothesis supported by in vivo experiments in which mice showed signs of depression after cancer chemotherapy[25]. One study reported that depressive symptoms improved spontaneously within three months in patients receiving chemotherapy and were not affected by CBT-based intervention[12]. However, the results of our meta-analysis suggest that CBT can also improve negative mood in patients with OC (depression: SMD = -0.41, 95%CI: -0.76 to -0.06; anxiety: SMD = -0.57, 95%CI: -0.86 to -0.27, P < 0.01). CBT was more effective in reducing anxiety than depressive symptoms. This may be related to the fact that CBT is a treatment method used to correct cognitively incorrect beliefs, such as catastrophizing, which often causes people to feel tense, thus producing anxiety.

However, given the very small sample size of one included study (Moonsammy et al[16]), which may have an impact on the authenticity of the results, we attempted to exclude this study. After exclusion, depression = -0.42 (-0.79 to -0.05) and anxiety = -0.57 (-0.86 to -0.27) (Figure 6), which was consistent with the results without exclusion. However, this suggests that future large-scale studies are required to validate these results.

These disease states are mutually causal and affect each other. Cancer-related fatigue is more prominent in OC, causing physical and mental distress, and even severe negative emotions such as anxiety and depression[26,27]. At the same time, sleep disturbance is considered to be an important cause of cancer-related fatigue, which in turn affects QoS and forms a vicious circle[28]. Medysky et al[29] found that sleep disturbances are directly proportional to the severity of fatigue in patients. Therefore, for patients with OC, active nursing intervention is needed to reduce cancer-related fatigue, thereby improving emotional function, QoS, and, ultimately, QoL.

A variety of CBT-based techniques were used in the included studies, including traditional CBT and ACT. CBT focuses on identifying and reorganizing the cycles of negative emotions and related behaviors, such as social withdrawal, while ACT therapy has the effect of making people aware of their entanglement with negative emotions while teaching them the skills to participate in and experience meaningful activities. Both can effectively solve psychological problems in patients with cancer[30]. Moreover, the personnel who implement CBT differ. Nevertheless, the results of this study show that regardless of whether CBT-based treatment is nurse-led, online, via psychologists, or home-based, it can improve negative emotions such as anxiety and depression and relieve cancer fatigue. It is worth mentioning that nurses seem to play a larger role in psychological interventions. In the six included studies, nurses participated in four intervention trials, all of which successfully improved patients’ mental status. Studies on high-quality nursing care in the perioperative period showed that nurse-led psychological interventions significantly reduced anxiety, depression, and cancer-related fatigue in patients with OC[31,32]. This may be due to the fact that nurses are closer to patients, easily accessible and providing intervention. Overall, cognitive-behavioral care can help patients with OC regain hope in life.

Limitations and implications

The quality of the included studies is not high enough. Among the six RCTs, half of the studies did not report the “random sequence generation” and “allocation concealment”, which has the risk of selection bias. Furthermore, the blinding method is not good enough; just one study reported double-blind, and no study reported triple-blinding, which may be related to the difficulty in implementing blinding due to the particularity of clinical trials. Additionally, the loss-to-follow-up rate of 5 studies exceeded 15%, which had the risk of reporting bias.

Also, the sample sizes of the included studies ranged from 8 to 73. The inclusion of small-sample studies may have affected the authenticity of the results. However, owing to the limited number of relevant published studies and the minimal impact on the analysis results after inclusion, these studies were ultimately included. However, if large-sample data becomes available in the future, such small-sample studies may be excluded.

Finally, in the meta-analysis results, heterogeneity for depression and cancer fatigue (subscale) remained relatively high (both > 50%) even after applying random-effects models and subgroup analyses; however, sensitivity analyses indicated no significant change in the stability of the meta-analysis findings. These findings collectively indicate that heterogeneity stems from clinical variability, potentially related to the following factors: (1) Differences in depression assessment scales; (2) Inconsistencies in the specific methods and duration of CBT interventions; and (3) Individual patient differences.

There are limited RCTs on CBT for OC, and the quality of the published studies is generally low. Currently, it is difficult to obtain high-quality evidence for analyses in this research field. This is a limitation of the present review. More large-sample, high-quality RCTs are needed in the future to provide more real research evidence.

CONCLUSION

To our knowledge, this is the first meta-analysis to evaluate the effectiveness of CBT on mental status and cancer-related fatigue in patients with OC undergoing chemotherapy. The present review suggests that CBT interventions may help relieve cancer-related fatigue, alleviat symptoms of anxiety and depression, and contribut to improvements in QoL and QoS in this patient population. These findings may assist physicians, nurses, and patients in recognizing the potential role of CBT as part of comprehensive cancer care, with a view to potentially enhancing mental, physical, and social health outcomes, thereby supporting more informed treatment decisions. However, larger samples, a more rigorous methodology, and more standardized outcomes in RCTs are needed to verify the sustained effects in the future.