Psychological interventions to alleviate death anxiety among patients with cancer: a systematic review and meta-analysis of randomized controlled studies

Abstract

Objective

Death anxiety is a common psychological distress in cancer patients and is associated with reduced quality of life. However, the efficacy of current psychological interventions for relieving this symptom remains unclear. This study aims to assess the effectiveness of psychological interventions in alleviating death anxiety among cancer patients and explore key effective intervention components.

Methods

A systematic review of randomized controlledstudies.

Data sources

Eight databases (PubMed, Embase, Cochrane Library, Web of Science, PsycINFO, CNKI, Wanfang, and VIP) were searched from inception to February 28, 2025. The Cochrane Risk of Bias 2 (RoB 2.0) tool was used to assess the risk of bias in the included studies.

Results

Nine papers involving 1,024 cancer patients were included. According to intervention components, the studies were divided into two categories: (1) Exploration of Life Meaning Interventions (ELMI) and (2) Cognitive Restructuring Interventions (CRI). Meta-analysis showed ELMI had a statistically significant but modest effects on death anxiety (SMD: − 0.17, 95% CI: (–0.32, − 0.01), P = 0.03). Narrative synthesis suggested that CRI may alleviate death anxiety among cancer patients. GRADE assessment rated the evidence for death anxiety as very low.

Conclusions

The effect of psychological interventions on alleviating death anxiety in cancer patients remains inconclusive. ELMI showed a statistically significant but modest effect, while evidence for CRI was inconsistent. Further high-quality studies with larger samples, standardized protocols, and longer follow-up are needed to explore combined or personalized intervention strategies.

Trial registration

The study protocol for this systematic review has been registered in PROSPERO(CRD420251169425).

Supplementary Information

The online version contains supplementary material available at 10.1186/s40359-026-04901-y.

Background

The Global Burden of Disease Study reported approximately 18.5 million new cancer cases globally in 2023, with around 10.4 million cancer-related deaths [1]. Against this backdrop, cancer patients experience psychological distress at a rate nearly three times higher than the general population [2]. Such distress is not transient but persists throughout the treatment trajectory: from initial diagnostic shock and fear, to anxiety induced by chemotherapy-related pain [3], disease progression, and inadequate health information during treatment [4]. Even during survivorship, many patients endure persistent fears of cancer recurrence [5–8]. Beneath these overt emotional concerns lies a fundamental fear of death, commonly defined as “death anxiety.”

Death anxiety is a complex emotional state marked by fear, unease, and worry, triggered by individuals’ awareness of death’s inevitability, unpredictability, and the finitude of human life [9]. To clearly delineate the research scope, this study distinguishes death anxiety from existential anxiety. Though both share core features of psychological distress and reflections on life value, existential anxiety mainly arises from confusion over one’s existential status and is typically diffuse and abstract [10]. By contrast, death anxiety centres on “the end of life”, with clear directionality and definite objects. Owing to such conceptual specificity, the present review focuses exclusively on the well-defined psychological construct of death anxiety.

Death anxiety exerts marked adverse effects on cancer patients’ physical and mental health, as well as the overall medical intervention process. It may impair treatment adherence, prompting patients to avoid disease-related information or discontinue standard therapeutic regimens [11]. In the terminal stage, severe death fear frequently hinders communication between patients and medical staff, thereby obstructing shared clinical decision-making [12]. Such barriers disrupt the delivery of palliative and end-of-life care, ultimately compromising the efficacy of clinical interventions [9, 13]. Furthermore, death anxiety may drive overtreatment tendencies, including frequent hospital attendance and redundant examinations, which aggravate patients’ psychological and economic burdens and waste limited personal and public healthcare resources [14, 15]. Considering these multiple detrimental outcomes, targeted and effective interventions to mitigate death anxiety are clinically imperative.

Currently, psychological interventions for death anxiety have evolved to include programs such as Managing Cancer and Living Meaningfully (CALM) and Cognitive Behavioral Therapy (CBT) [16, 17]. These interventions aim to alleviate anxiety at the cognitive and existential levels by helping terminally ill cancer patients reflect on life experiences, identify and challenge cognitive misconceptions [18, 19]. However, despite sound theoretical rationales for these interventions, their actual efficacy shows high heterogeneity in empirical research, with unclear overall effectiveness. Taking CBT as an example, Togluk et al. demonstrated that it significantly reduced death anxiety, whereas other studies reported negative findings, showing no significant improvement in this symptom [17, 18] Furthermore, clinical evidence for other mainstream therapies remains inconclusive, with inconsistent conclusions across studies. Such uncertainty in efficacy highlights the urgent need for a systematic evaluation of death anxiety interventions in the current field.

Reviewing previous research, a prior systematic review [20] summarizing interventions for advanced cancer patients still has significant limitations. First, it included a small number of randomized controlledstudies (RCTs), leading to a weak evidence base. Second, and most crucially, its conceptual definitions did not rigorously distinguish between “death anxiety” and “existential anxiety”. As few included studies adopted “death anxiety” as a core outcome measure, the conclusions were mostly based on inferences from existential anxiety-related or non-specific psychological outcomes (such as anxiety or depression), which may result in misjudgments of intervention efficacy. Furthermore, the latest study included in that review was published in 2016. Since then, several high-quality studies have emerged over the past nine years [16–18, 21–23]. Therefore, it is essential to conduct a systematic review incorporating high-quality literature over the past decade, especially RCTs, to summarize existing interventions and analyze their efficacy, to guide clinical practice. This review has two objectives: 1) to evaluate the effectiveness of psychological interventions for death anxiety in cancer patients; and 2) to identify common intervention components. By accomplishing these goals, we aim to propose targeted strategies to alleviate death anxiety, support future research and clinical practice, and ultimately improve patients’ psychological well-being and quality of life.

Methods

Design

This systematic review was conducted and reported in accordance with the 2020 Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) statement, ensuring rigorous methodological and reporting standards [24]. The review protocol was preregistered on PROSPERO (CRD420251169425).

Search strategies

A comprehensive search was conducted across five English databases (PubMed, Embase, Cochrane Library, Web of Science, and PsycINFO) and three Chinese databases (CNKI, Wanfang, and VIP). The search spanned from the inception of each database to February 28, 2025. Search terms including “neoplasms”, “death anxiety”, and “interventions” were combined via free-text words and Medical Subject Headings (MeSH) terms in each database. The full search strategy is provided in Appedix file.

Inclusion criteria

To ensure the relevance of evidence, we first clearly defined and differentiated death anxiety from existential anxiety. Secondly, given that outcomes including anxiety, depression, anger and quality of life are not specific to death anxiety, they may represent accompanying symptoms or derivative outcomes but fail to reflect its core mechanisms accurately. Therefore, studies solely reporting these non-specific outcomes were excluded from this review. Nevertheless, studies with both specific death anxiety measures and these non-specific assessments were included to comprehensively extract and analyze core data pertaining to death anxiety. The inclusion criteria are summarized as follows:

• Participants: All adult cancer patients aged 18 years and older. For studies that recruited both patients and caregivers, only patient-specific outcome data were extracted for analysis.

• Interventions: Any theory-based psychological intervention systematically delivered by professional personnel (e.g., clinical psychologists, counselors) and designed to reduce death anxiety.

• Controls: Eligible control groups consisted of usual, placebo, or wait-list controls.

• Outcomes: Data extracted in this study included two categories of assessment outcomes: (1) Primary outcome: Death anxiety, measured by validated death anxiety scales, such as Templer’s Death Anxiety Scale (DAS), Death and Dying Distress Scale (DADDS), and Death Anxiety Questionnaire (DAQ). (2) Secondary outcomes: General psychological distress (e.g., anxiety, depression) and quality of life as secondary endpoints.

• Study designs: RCTs were included in this review.

Exclusion criteria

Studies meeting any of the following criteria were excluded:

• Publications written in languages other than Chinese or English.

• Studies without available full texts.

• Studies using non-validated, unstandardized, or self-designed assessment tools were excluded.

Literature screening

After the literature search, EndNote X9 was used to identify and remove duplicate records. Literature screening was performed independently by two researchers, Qi Y and Li J. Full-text articles of potentially eligible records were then assessed for eligibility. Any disagreements during screening were resolved via discussion with reviewer Xiao J to reach a consensus.

Data extraction

Data extraction was conducted independently by two reviewers, Li J and Qi Y. One reviewer extracted relevant information from eligible studies, while the other performed cross-verification. Extracted contents covered authorship, study country, participant demographics (age, cancer type, and disease stage), intervention characteristics (theoretical basis and specific measures for both experimental and control groups), as well as outcome indicators and primary research findings.

Risk of bias assessment

The included RCTs were assessed using the Cochrane Risk of Bias 2 (RoB 2.0) tool [25]. Domains evaluated included bias arising from the randomization process, bias due to deviations from intended interventions, bias from missing outcome data, bias in the measurement of the outcome, and bias in the selection of the reported result. RoB 2.0 judgements for each domain were categorized as low risk, some concerns, or high risk. Two reviewers independently performed the assessments. Any disagreements were resolved by discussion, with consensus reached through consultation with a senior expert when needed.

Data analysis

Review Manager 5.3 software was used for the meta-analysis. Because different validated scales measured the same outcomes, the standardized mean difference (SMD) with 95% confidence interval (CI) was calculated for continuous data. Heterogeneity was assessed using the I² statistic, with values of 25%, 50%, and 75% indicating low, moderate, and high heterogeneity, respectively [26]. The random-effects model was applied if significant heterogeneity was present (e.g., P < 0.1 for the Chi² test or I² > 50%); otherwise, the fixed-effects model was used. Where heterogeneity was substantial (I² > 75%), studies were considered too heterogeneous to pool statistically and were analysed narratively. Pre-planned sensitivity analyses were performed by sequentially removing individual studies to assess their influence on pooled effect estimates and heterogeneity. Studies identified as influential outliers were examined for potential clinical or methodological sources of divergence and were removed from the primary analysis, with their impact discussed narratively.

Results

Search results

The initial search yielded 2,557 records. After removing 840 duplicates, the remaining 1,717 citations were screened based on titles and abstracts. Of these, 1,540 were excluded, and 177 full-text reports were assessed for eligibility. The study selection process is presented in the PRISMA flowchart (Fig. 1).

Fig. 1.

Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) flow diagram

Study characteristics

The characteristics of the included interventions are presented in Table 1.

Table 1.

Study characteristics of included studies

Authors(s), (year), country	Sample characteristics	Interventions	Outcome measurement/Assessment time	Attrition rate (%)	Analytic approach for missing data	Description of results
Kissane et al. [21], (2023), Australia	Various types of cancer, 107 patients were selected, no cancer staging report, with mean ages of 63 years in Usual Care and 60 years in Meaning and purpose (MaP) therapy	Type: MaP therapy delivered face-to-face Intervention group: session intervention with MaP therapy (each lasting over 1 h) Intervenors: Psychologists and social workers	The questionnaire includes DADDS, MQOL, PHQ-9, and GAD-7/baseline, 12 weeks after intervention, and after 12 weeks(every two weeks)	12 weeks: 37.38 After 12 weeks: 53.27	ITT used; PP analysis also reported	It failed to alleviate patients’ death anxiety, with no significant improvements in depression, anxiety, or quality of life.
Heidary et al. [23], (2023), Iran	Diagnosed with advanced cancer, 64 patients were selected, stages III–IV, with a projected survival of at least 6 months; the mean age of the logotherapy group was 53.93 (SD, 11.55) and the control group was 52.56, (SD, 9.23)	Type: Logointervention delivered face-to-face Intervention group: session group logotherapy course(2 h per session) Wait-list control group: After measurement were provided opportunity of intervention Intervenors: psychologist Intervenors: psychologist	The questionnaire includes DAS, /pre- and post-treatment (10 weeks, weekly)	1.56	NR	Compared with baseline levels, cancer patients in the intervention group reported significantly lower death anxiety, while no significant reduction was seen in the waiting-list control group.
Amini et al. [22], (2020), Iran	Gastrointestinal Cancer, 162 patients were selected, no cancer staging report, mean age of the intervention group was 53.12 (15.58) years, and control group was 60.5 (14.62) years	Type: Spiritual care delivered face-to-face Intervention group: session spiritual care(30–45 min per session) Intervenors: Nurses and clergyman	The questionnaire includes DAS/pre-and post-treatment (3 days, per day)	10.49	NR	A statistically significant difference was observed in death anxiety.
Rodin et al. [27], (2018), Canada	Various types of cancer, 305 patients were selected, stage III or IV, with mean age of the intervention group at 59.05 (10.55) years, and control group at 59.10 (11.48) years	Type: CALM delivered face-to-face Intervention group: 3 to 6 CALM psychotherapy sessions(40–60 min per session) Intervenors: Master’s degree-level social workers and psychiatrists	The questionnaire includes DADDS, PHQ-9, GAD-7/Baseline, 3 months and 6 months(monthly)	3 months:18.69 6 months: 31.48	ITT analysis only; PP not reported	It did not alleviate cancer patients' death anxiety, however, it improved end‑stage quality of life and reduced depressive symptoms.
Mehnert et al. [28], (2020), Germany	Various types of cancer, 206 patients were selected, stage III or IV, with mean age of the intervention group being 59.5 years, and the control group was 56.5 years	Type: CALM delivered face-to-face Intervention group: 3 to 6 CALM psychotherapy sessions Control group: non-manualized supportive psycho-oncological counselling intervention Intervenors: Psycho- -oncologists	The questionnaire includes DADDS, BDI, PHQ-9, DT, QUAL-EC, GAD-7/Baseline, 3 months and 6 months(monthly)	3 months: 26.21 6 months: 33.01	ITT analysis only; PP not reported	It failed to alleviate participants’ death anxiety, anxiety, or depression, nor did it improve quality of life.
Zhang et al. [16], (2024), China	Breast cancer, 36 patients were selected, stage IV, with mean age of the intervention group of 45.11 (8.89) years, and the control group was 49.65 (10.94) years	Type: CALM WeChat- delivered Intervention group: 3 to 6 CALM psychotherapy sessions(40–60 min per session) Wait-list control group: following assessment, CALM therapy was provided. Intervenors: Psycho-oncologists and GIPPEC-certified CALM therapists	The questionnaire includes DADDS, PHQ-9, DT, QUAL-EC/Baseline, 3 months, and 6 months(monthly)	5.56	NR	It failed to alleviate patients’ death anxiety, yet it relieved depressive and anxious symptoms, significantly improving their quality of life in the terminal stage.
Jelvehzadeh et al.[17], (2022), Iran	Breast cancer, 32 patients were selected, no cancer staging report.	Type: CBT delivered face-to-face Intervention group: 8 sessions of group CBT(120 min per session)	The questionnaire includes DAS, MOQL, and DASS/pre-, posttreatment and follow up (three months)	33.33	NR	It failed to alleviate death anxiety among breast cancer patients, yet it improved symptoms including depression, anxiety, as well as overall quality of life.
Togluk et al.[18], (2024), Turkey	Various types of cancer, 80 patients were selected, no cancer staging report.	Type: CBT delivered face-to-face Intervention group: 8 sessions of CBT(50 min per session) Intervenors: Researchers with CBT training certification	The questionnaire includes DAS, CAR/pre-and Posttreatment(4weeks, twice a week)	9.09	NR	It can significantly alleviate patients’ death anxiety and anger.
Onyechi et al.[29], (2016), Nigeria	Thirty-two cancer patients consisting of 20 with breast cancer, 8 with cervical cancer, and 4 with prostate cancer, no cancer staging report, with mean age of the intervention group being 48.33(6.17) years, and the control group was 48.32(6.43) years	Type: Rational-emotive hospice care therapy delivered face-to-face Intervention group: 10 sessions of Rational-Emotive Hospice Care Therapy (REHCT)(45 min per session) REBT Hospice Care Manual Intervenors: Researchers trained and proficient in REBT theory	The questionnaire includes DAQ, K10/Baseline, 10 weeks and follow up (14 weeks, weekly)	0	NR	Only patient-level death anxiety data were extracted and analysed; caregiver data were not included in the present meta-analysis. Death anxiety, depression, and anxiety were significantly mitigated in study participants.

NR No report, BDI  Beck Depression Inventory, CALM  Managing Cancer and Living Meaningfully, CAR Controlled Anger Scale, CBT  Cognitive Behavioral Therapy, CRI  Cognitive Restructuring Interventions, DADDS Death and Dying Distress Scale, DAQ Death Anxiety Questionnaire, DASS Depression Anxiety Stress Scales, DAS Templer’s Death Anxiety Scale, DT  Distress Thermometer, ELMI  Explorations of Life Meaning Interventions, GAD-7 Generalized Anxiety Disorder-7, GIPPEC Global Institute of Psychosocial Palliative & End-of-Life Care, ITT  Intention-to-Treat, K10 Kessler Psychological Distress Scale, MaP Meaning and Purpose, MQOL  McGill Quality of Life Questionnaire, PHQ-9  Patient Health Questionnaire-9, PP Per-Protocol, QUAL-EC Quality of Life at the End of Life-Cancer

Study location

Among the nine studies, three were conducted in Iran [17, 22, 23], and one each in Australia [21], Canada [27], Germany [28], China [16], Turkey [18], and Nigeria [29].

Characteristics of the participants

The nine studies encompassed 1,024 patients diagnosed with various types of cancer. Of note, one study [29] enrolled both patients and family caregivers; from this study, only patient-level data on death anxiety were extracted and analyzed. Study sample sizes ranged from 32 to 305 participants. The participants were predominantly middle-aged and older adults, with the mean ages across studies ranging from 45.1 to 65.9 years. Notably, one study [16] focusing on patients with metastatic breast cancer had a considerably younger sample with a mean age of 45–49 years. Most were diagnosed with breast or gastrointestinal cancer. Cancer staging ranged from stage I to stage IV, with the majority of cases being advanced-stage (stages III–IV). Attrition rates varied across studies, from 0% to 53.27%. The main reasons for dropout included disease progression, death, loss to follow-up, scheduling conflicts, and other personal factors. We extracted data from the most rigorously reported analysis: for the three studies that explicitly adopted an ITT approach [21, 27, 28], those results were used; for the remaining six studies, the analytic approach for missing data was not reported (NR) [16–18, 22, 23, 29]; we therefore extracted the results as presented in the original reports, which likely reflect non-ITT analyses such as available-case or per-protocol approaches (Table 1). Given this lack of clarity, the potential for attrition bias in these studies cannot be ruled out.

Theoretical bases of the interventions

The six intervention types offer clear theoretical rationales for their design, with each study grounded in one to three theoretical frameworks. These include meaning management theory, narrative theory, and cognitive behavioral theory. Supplementary Table S1 summarizes these theoretical models, highlights their key differences, and outlines their respective strengths and limitations.

Intervention content

Based on intervention components, relevant approaches can be categorized into two types: (1) Explorations of Life Meaning Interventions (ELMI) and (2) Cognitive Restructuring Interventions (CRI). These two categories cover seven therapeutic modalities, as detailed in Table 2.

Table 2.

Summary of intervention contents

Categories	Intervention types	Description
ELMI	Meaning and purpose therapy	Establish a trusting relationship with cancer patients. Through carefully designed questions, guide patients to organize their life stories and identify consistent personal strengths, values, and past sources of accomplishment. Building on this foundation, use independent homework assignments to help cancer patients weave scattered fragments of life experiences into a coherent, positive self-narrative. This process enables them to deepen their awareness of life’s meaning and self-worth when confronting mortality, ultimately achieving inner peace and fulfillment.
	Logointervention	Establish trust with cancer patients and create a safe therapeutic environment. Guiding patients to revisit meaningful life activities and emotional connections enables them to reflect on life value beyond their illness. Assist patients in developing value-driven goals that link them to the external world, other people, and the future, and translate these into actionable “meaningful action” plans.
	Spiritual care	Establish trust-based communication to evaluate patients’ psychological distress. Encourage cancer patients to express their emotions, and actively assist them in achieving meaningful communication and reconnection with family members and friends. Provide simplified religious rituals (especially prayer sessions) and spiritual guidance (such as reading religious texts) to help patients develop a profound spiritual connection with God, thereby fostering insights into death, inner strength, and the meaning of life.
	CALM	Build trust among cancer patients to facilitate disease management and doctor–patient communication. Address shifts in interpersonal relationships with family and friends by exploring illness-related impacts on self-identity, thereby alleviating social isolation. Guide patients to recount their life narratives and reflect on unresolved issues, helping them clarify and reconstruct personal life values and goals. Through open discussion of diverse death-related concerns, patients regain a sense of control and life meaning when confronting life’s ultimate limitations.
CRI	CBT	Establish a trusting relationship with cancer patients. Using interactive techniques such as case studies and role-playing, this approach guides patients to analyze the intrinsic connections between their thoughts, emotions, and behaviors. It aims to correct negative thought patterns and beliefs—such as catastrophizing death—while providing education on cancer and associated psychological concerns. Foster adaptive coping skills and stress management competencies, and promote emotional regulation and relaxation via breathing and relaxation therapies.
	Rational-emotive hospice care therapy	Build trust with research participants and conduct a comprehensive assessment of physical, mental, social, and spiritual needs. Guide patients with cancer to analyze the intrinsic connections between their thoughts, emotional distress, and behaviors; identify core irrational beliefs; clarify responsibility orientation; and help them grasp the fundamental distinction between the inevitability of death and personal control over life. Through guided discussions, questioning, and cognitive challenging of irrational beliefs, patients are encouraged to accept reality. Simultaneously, behavioral activation strategies help them regain a sense of control over their daily lives.

CALM  Managing Cancer and Living Meaningfully, CBT  Cognitive-Behavioral Therapy, CRI  Cognitive Restructuring Interventions, ELMI  Explorations of Life Meaning Interventions

The therapies within both categories enhance cancer patients’ awareness of and reflection on inner experiences. Simultaneously, both emphasize the establishment of a trusting relationship between cancer patients and healthcare professionals and rely on professionals’ guidance to facilitate psychological transformation. Interventions focused on meaning in life [16, 21–23, 27, 28] aim to help cancer patients uncover positive values and deeper significance embedded in past events through life-review or life-narrative approaches. This process fosters psychological growth and self-transcendence by guiding cancer patients to acknowledge death’s inevitability, thereby enabling a renewed sense of life’s meaning. Interventions based on cognitive restructuring [17, 18, 29], however, target correcting cancer patients’ irrational beliefs about death by assisting them in identifying and challenging cognitive distortions such as “death equals loss of value.” Through combining rational substitution with behavioral interventions, these approaches facilitate the development and internalization of new cognitive frameworks, ultimately achieving positive transformation in thought patterns.

Intervention delivery methods

Among the nine included studies, eight adopted face-to-face intervention delivery [17, 18, 21–23, 27–29], and one implemented the intervention via WeChat, a widely used Chinese social media platform [16]. Full details are presented in Table 1.

Intervention dosage

The number of intervention sessions across the included studies ranged from six to ten. In one study, interventions were delivered twice daily for three consecutive days [22], whereas frequencies in the remaining studies varied from once or twice weekly to once monthly. Session duration correspondingly ranged from 30 to 120 min, with details provided in Table 1.

Intervention facilitators

Eight studies reported intervention facilitators. Six noted providers received tailored training (content not specified [16, 18, 21, 27–29]). Four of these identified providers as psychiatrists, psychologists, or oncology psychologists [21, 23, 27, 28]; two listed social workers (one including clergy [22]). Two studies specified professionally trained, certified providers [16, 21]. Multi-provider delivery was noted in three studies [21, 22, 27] (Table 1). Despite adequate provider reporting, three studies lacked standardized protocols and detailed procedures, limiting clinical replicability [17, 22, 29].

Risk of bias

Figures 2 and 3 summarize RoB 2.0 assessments across domains for the included studies. One study exhibited concerns regarding randomization bias, with deficiencies in sequence generation or allocation concealment [18]. Two studies presented high risk of bias due to intervention deviations, including lack of standardized protocols and absence of intention-to-treat analysis, potentially compromising result reliability [17, 22]. One study had high risk of bias from missing data driven by high dropout rates, reducing effect estimate accuracy [17]. Four studies showed high risk of outcome measurement bias linked to unblinded assessors and subjective self-reports [16–18, 22]. Two studies faced potential selective reporting risks owing to lack of prospective registration [18, 23]. Overall, the methodological quality of the included studies was heterogeneous, with a substantial proportion exhibiting a moderate to high risk of bias, necessitating cautious interpretation of the pooled results.

Fig. 2.

Risk of bias summary

Fig. 3.

Risk of bias graph

Effectiveness of interventions

Intervention effectiveness across included studies was assessed using 11 scales or questionnaires, grouped into primary outcomes (death anxiety) and secondary outcomes (general psychological distress, quality of life). Given the small number of eligible studies and variability in intervention delivery, dosage, and measurement tools, the data available for meta-analysis in this review were relatively limited.

Death anxiety

All nine included studies assessed death anxiety. In the overall meta-analysis of the six ELMI studies, the pooled effect was (SMD = − 0.46, 95% CI: −0.90 to − 0.20, P = 0.04) with high heterogeneity (I² = 87%). Sensitivity analysis identified Heidary 2023 [23] as an outlier (see Supplementary Table S2). After removing this study, heterogeneity dropped to 0%, and the pooled effect for death anxiety remained significant (SMD = − 0.17, 95% CI: −0.32 to − 0.01, P = 0.03; Fig. 4), In contrast, three CRI studies [17, 18, 29] were not pooled for quantitative synthesis due to substantial heterogeneity in interventions, outcome measures, and effect sizes (I²= 96%; see Supplementary Figure S1). Narrative analysis revealed significant reductions in two CRI studies [18, 29] but no effect in one [17].

Fig. 4.

Effects of ELMI on death anxiety

GRADE assessment rated the evidence for death anxiety as very low, primarily due to serious risk of bias, inconsistency, imprecision, and indirectness (Table 3). Publication bias was not assessed as the number of included studies was fewer than ten.

Table 3.

GRADE Summary of Findings for the primary outcome (Death anxiety)

Outcomes	No. of participants (studies)	Effect estimate (95% CI)	Certainty assessment	GRADE	Comments
Death anxiety Follow-up: Short–medium term (3 days to 6 months)	9 RCTs, 1024 patients ELMI: 5 RCTs (meta-analyzed) CRI: 3 RCTs (narrative synthesis)	ELMI interventions: SMD = -0.17 95% CI: -0.32 to -0.01 P = 0.03 Small beneficial effect CRI interventions: High heterogeneity; not pooled (due to high methodological and clinical heterogeneity)  2/3 studies showed significant reduction	-1: Risk of bias (unclear/random sequence generation, allocation concealment, missing data) -1: Inconsistency (heterogeneity across studies) -1: Imprecision (small-moderate sample; wide CIs) -1: Indirectness (different interventions/scales) Publication bias: Not assessed due to the small number of included studies	Very low	Evidence is very uncertain. ELMI has the potential to reduce death anxiety. CRI effects are inconsistent.

ELMI Explorations of Life Meaning Interventions, CRI Cognitive Restructuring Interventions, SMD standardized mean difference, CI confidence interval

General psychological distress

Four ELMI studies examined intervention effects on anxiety and depression [16, 21, 27, 28]. Three found no significant improvements in these symptoms (P > 0.05) [21, 28], whereas one reported significant reductions (P < 0.05) [16, 27]. Among the three CRI studies, significant favourable changes were identified in anxiety, depression [17, 29], and anger [18] (P < 0.05).

Quality of life

Four ELMI studies assessed quality of life outcomes among patients with cancer [16, 21, 27, 28]. Two of these reported significant improvements specifically in end-of-life quality of life (P < 0.05) [16, 27], while no significant changes in overall quality of life were detected across the included studies (P > 0.05). Regarding CRI‑based interventions, only one study evaluated quality of life in cancer patients; this CBT-related study showed significant enhancements in overall quality of life (P = 0.03) [17].

Discussion

This systematic review included nine eligible studies, categorizing the interventions into ELMI and CRI. Meta-analysis showed that ELMI produced statistically significant but small effect sizes for death anxiety; its clinical efficacy therefore remains inconclusive. For CRI, meta-analysis was not feasible, and narrative synthesis yielded inconsistent findings. Only marginal improvements in secondary outcomes were observed across both approaches. Given the modest ELMI effect, the inconsistent CRI findings, the reliance on narrative synthesis, the small number of included studies, and the very low certainty of evidence, the current evidence does not yet support definitive clinical recommendations. Large-scale, rigorously designed prospective studies are urgently needed to verify and strengthen this evidence base.

A pooled analysis of five ELMI studies revealed a statistically significant but small reduction in death anxiety, indicating limited clinical relevance. This modest effect likely stems from several issues. Death anxiety arises from interrelated factors such as symptom burden, depression, and social support [30, 31]; although these are theoretically addressable within ELMI’s multidimensional framework (incorporating meaning therapy, spiritual care, and CALM), in practice, spiritual and psychosocial components were often delivered at a general supportive level. Although some interventions included death-related discussions, the lack of fidelity assessment makes it unclear whether these components were delivered with sufficient intensity or focus to directly challenge the core fear of death. Participant engagement was also suboptimal: high attrition, driven largely by disease progression, death, and scheduling conflicts, led to poor session attendance; the fatigue and frailty characteristic of advanced cancer likely compounded this, limiting the potential efficacy of brief, low-intensity programs. Moreover, the presence of an outlying study in sensitivity analyses [23], likely due to its use of group logotherapy versus individual therapy in other ELMI studies, together with the broader heterogeneity in intervention delivery, outcome measurement timing, and other methodological aspects across ELMI studies, underscores the need for standardized protocols. Spiritual care and CALM components are suggested to be implemented as active, structured, and dose-controlled elements, with dedicated sessions focused on death-related concerns, legacy, and end-of-life preparation [27, 32]. For patients with longer survival, tailored booster sessions may help sustain treatment gains [33]. Flexible delivery formats (e.g., telehealth) and initiating the intervention earlier in the disease trajectory can further improve retention and engagement [34].

The impact of CRI on death anxiety is unclear. Comparable interventions have produced similarly inconclusive findings. For instance, a meta-analysis of third-wave CBT reported significant improvements only in quality of life, while evidence on death anxiety remained insufficient and its efficacy uncertain [35]. These mixed findings may reflect methodological inconsistencies in death anxiety measures, session durations, and intervention components. Attrition was similarly high in some CRI studies [17]. CRI was categorized separately in this review precisely because its theoretical mechanism-cognitive restructuring-represents a distinct, cognition-focused pathway that, if effective, would offer a relatively specific and scalable approach. However, this same mechanistic specificity may also constitute its key limitation: although CRI often incorporates behavioral components such as stress management and relaxation, its emphasis on cognitive restructuring may still underaddress the multidimensional nature of death anxiety, particularly symptom burden and social isolation [31, 36]. Differing patient characteristics and contexts likely moderated response as well. Large, adequately powered, and pre-registered confirmatory studies are urgently needed to identify treatment moderators and strengthen the evidence base for integrating CRI into routine psycho-oncology care.

The evidence for secondary outcomes remains mixed and largely inconclusive. In our review, both ELMI and CRI showed some benefits, but the findings were limited and inconsistent across domains. For ELMI, the intervention yielded inconsistent effects on general psychological distress. Regarding quality of life, two ELMI studies observed improvements specifically in end-of-life quality of life, yet no significant changes in overall quality of life were detected. These findings resemble those of a recent meta-analysis of 12 RCTs, which reported only small effects on depressive symptoms and anxiety, with any quality-of-life gains not sustained beyond one month [37]. For CRI, significant improvements were consistently observed across general psychological distress outcomes, while quality-of-life evidence was sparse, resting on a single study that showed a significant but isolated improvement. This pattern mirrors the broader CBT literature, where mental health benefits are moderate but vary by age and delivery format, with quality-of-life data remaining limited [38]. Collectively, secondary outcomes are directionally positive but limited in magnitude, frequently inconsistent across domains. Future studies should prioritize consensus-based core outcome sets, longer follow-up periods, and standardized protocols, and should include moderator analyses before integrating these interventions into routine practice.

The active components hypothesized to reduce death anxiety were seldom implemented as distinct, replicable modules with prescribed dose in either ELMI or CRI studies, with core death-related and spiritual elements frequently embedded within general psychosocial support. This pattern reflects a broader field-wide gap between theoretical models of death anxiety and their operationalization in clinical studies. Future intervention development should therefore systematically report and compare intervention doses to enable dose-response analyses, ensure fidelity through manualized training [39], and isolate the unique contribution of death-related and meaning-based work from nonspecific support [40]. Only by designing interventions that directly activate the intended change processes, and by measuring those processes, can the clinical potential of both approaches be realized [41].

The methodological quality of eligible studies merits rigorous assessment. Of the nine included studies, four were judged at high risk of bias [16–18, 22], and the other five at moderate risk of bias [21, 23, 27–29]. Three studies failed to document full intervention protocols, markedly compromising research replicability. Furthermore, a lack of uniform intention-to-treat analysis in most studies diminishes the credibility of synthesized evidence. Future clinical studies should adopt intention-to-treat analysis as the primary statistical approach, with clear disclosure of missing data management strategies. Since full blinding tends to place undue burden on the target population, adopting partial blinding and streamlined study procedures can effectively enhance participant retention and methodological robustness.

Limitations

This review has several key limitations. First, substantial heterogeneity was observed across included studies, which prevented meta-analysis for most outcomes and restricted the ability to draw firm conclusions about specific intervention effects. Second, the interpretability of findings was constrained by inconsistent and often suboptimal methodological quality among the included studies. Moreover, even for outcomes where meta-analysis was feasible, the pooled effect sizes were small, further limiting the clinical significance of the observed effects. Third, although the inclusion criteria required a specific measure of death anxiety, some included studies also reported only non-specific outcomes alongside the required measure; the operational application of this criterion was therefore somewhat lenient, which may have introduced heterogeneity. Fourth, the geographic scope of included studies was relatively narrow, with most conducted in East Asia, North America, and Africa. Because cultural context can strongly influence the acceptability and effectiveness of psychological interventions [42], this restricted geographic distribution may reduce the generalizability of the results. Finally, study samples were predominantly elderly patients with cancer, who often had limited physical tolerance and rapid disease progression. These characteristics frequently led to shorter observation periods and greater challenges in maintaining rigorous experimental design. As a result, the reliability and robustness of the review’s conclusions may be somewhat weakened, limiting evaluation of the long-term effects of interventions.

Conclusion

This review identified two core psychological interventions for cancer patients: ELMI and CRI. The effect of ELMI on death anxiety was modest, while findings for CRI were inconsistent. These interventions may also yield limited improvements in end-of-life quality of life and general psychological distress. Given the very low certainty of evidence, neither can be recommended for routine practice. That said, the data do suggest several directions worth testing: delivering death-related and meaning-centered components as structured, active work rather than general support; using flexible formats and earlier initiation to counter the high attrition from disease progression; and offering booster sessions when prognosis allows. Death anxiety is multidimensional; cognitive restructuring alone may therefore be insufficient unless symptom burden and social isolation are also addressed. All these are provisional insights that should remain within research settings. Future trials must be large-scale and rigorously designed to establish long-term efficacy and evaluate personalized stepped-care models.

Abbreviations

ELMI

Explorations of Life Meaning Interventions

CRI

Cognitive Restructuring Interventions

CALM

Managing Cancer and Living Meaningfully

CBT

Cognitive Behavioral Therapy

CI

Confidence interval

RCTs

Randomized Controlledstudies

SMD

Standardized Mean Difference

BDI

Beck Depression Inventory

CAR

Controlled Anger Scale

DADDS

Death and Dying Distress Scale

DAQ

Death Anxiety Questionnaire

DASS

Depression Anxiety Stress Scales

DAS

Templer’s Death Anxiety Scale

DT

Distress Thermometer

GAD-7

Generalized Anxiety Disorder-7

GIPPEC

Global Institute of Psychosocial Palliative & End-of-Life Care

ITT

Intention-to-Treat

K10

Kessler Psychological Distress Scale

MaP

Meaning and Purpose

MQOL

McGill Quality of Life Questionnaire

PHQ-9

Patient Health Questionnaire-9

PP

Per-Protocol

QUAL-EC

Quality of Life at the End of Life-Cancer

Authors’ contributions

Author Contributions StatementGuiyun Wang: Conceptualization, Methodology, Software.Jiaojiao Li: Data curation, Writing-Original draft preparation.Jinnan Xiao: Writing-Reviewing and Editing.Yanxia Qi: Data curation.Siyuan Tang: Software, Validation.Jiaoyang Liu: Supervision.Yuehua Han: Supervision.Jinke Kuang: Supervision.Wenjing Xu: Supervision.

Funding

The study was funded by Youth Innovation Technology Project of Higher School in Shandong Province (No.: 2023KJL031), and Research Project of Hunan Nursing Association (Grant No.: HNKYQ202302).

Data availability

The datasets used and/or analysed during the current study are available from the corresponding author on reasonable request.

Declarations

Ethics approval and consent to participate

This study was not applicable because it was a meta-analysis.

Consent for publication

This study was not applicable because it was a meta-analysis.

Competing interests

The authors declare no competing interests.