Stepwise psychological intervention: Mitigated psychological distress and suicide risk in patients with liver cancer chemoembolization

Abstract

BACKGROUND

Patients with liver cancer undergoing targeted therapy combined with transarterial chemoembolization (TACE) may experience substantial psychological distress, anxiety, depression, and suicide-related risk. Risk-stratified psychological care may help address these problems by matching the intensity of psychological support to patients’ assessed risk levels; however, evidence regarding structured stepped psychological interventions in this population remains limited.

AIM

To evaluate the short-term effects of a stepped psychological intervention on psychological distress and suicide risk in patients with liver cancer receiving targeted therapy combined with TACE.

METHODS

This single-center, unregistered randomized controlled trial included 98 patients with liver cancer who received targeted therapy combined with TACE between April 2024 and November 2025. Participants were randomly assigned in a 1:1 ratio to a control group (n = 49), which received routine nursing care, or a stepped psychological intervention group (n = 49), which received risk-stratified psychological assessment and stepped psychological care in addition to routine care. Psychological distress, suicide risk, anxiety, and depression were assessed using the Distress Thermometer (DT), Nurses’ Global Assessment of Suicide Risk (NGASR), Self-Rating Anxiety Scale (SAS), and Self-Rating Depression Scale (SDS), respectively. Treatment adherence, nursing satisfaction, and safety-related events were evaluated as secondary outcomes. Exploratory multivariable logistic regression analyses were performed to identify factors associated with significant psychological distress (DT ≥ 4) and high suicide risk (NGASR ≥ 5) at the 4-week follow-up. To minimize potential overadjustment and mediator bias, only baseline or prerandomization covariates were included in these models.

RESULTS

At the 4-week follow-up, DT, NGASR, SAS, and SDS scores decreased in both groups, with greater reductions observed in the stepped psychological intervention group. The intervention group had higher treatment adherence than the control group (95.92% vs 77.55%, P = 0.007) and higher nursing satisfaction (95.92% vs 75.51%, P = 0.004). Safety-related events were recorded in 2 of 49 patients (4.08%) in the intervention group and 9 of 49 patients (18.37%) in the control group (P = 0.025). In exploratory multivariable analyses, allocation to the stepped psychological intervention group was associated with lower odds of significant psychological distress at 4 weeks after adjustment for baseline DT score, tumor stage, and family support [adjusted odds ratio (OR) = 0.31, 95%CI: 0.12-0.78, P = 0.014]. Higher baseline DT scores and advanced tumor stage were associated with higher odds of psychological distress, whereas good family support was associated with lower odds. Similarly, intervention-group allocation was associated with lower odds of high suicide risk after adjustment for baseline NGASR score, tumor stage, and family support (adjusted OR = 0.27, 95%CI: 0.0-0.79, P = 0.018). Higher baseline NGASR scores and advanced tumor stage were associated with higher odds of suicide risk, whereas good family support was associated with lower odds.

CONCLUSION

In this single-center, unregistered randomized controlled trial, patients allocated to a stepped psychological intervention showed greater short-term improvements in psychological distress, suicide risk, anxiety, and depression than those receiving routine nursing care. Exploratory analyses also suggested associations with better treatment adherence and nursing satisfaction and fewer recorded safety-related events. However, these findings should be interpreted cautiously because of the small sample size, short follow-up period, lack of prospective trial registration, and limitations in the assessment of secondary outcomes. Larger, prospectively registered multicenter trials using validated outcome measures, independent outcome assessment, and longer follow-up are needed to confirm these findings.

Key Words: Stepwise psychological intervention; Liver cancer; Transcatheter arterial chemoembolization; Targeted therapy; Psychological distress; Suicide risk; Psychological state

Core Tip: This study adopted a stepwise psychological intervention for patients with liver cancer receiving trans arterial chemoembolization plus targeted therapy. This tiered intervention relieves psychological distress and suicide risk, improves treatment compliance, and reduces the incidence of ward safety events. Poor mood and advanced tumor stage are high-risk factors.

INTRODUCTION

Primary liver cancer has an insidious onset, rapid progression, a high recurrence rate, and poor prognosis. Most patients are diagnosed in the middle to late stages of the disease, thereby losing the opportunity for curative surgery[1,2]. Transcatheter arterial chemoembolization (TACE) combined with molecular-targeted drugs is an important treatment for advanced liver cancer, delaying disease progression by blocking the tumor blood supply and inhibiting tumor angiogenesis[3]. However, multiple stressors such as disease progression, treatment-related adverse reactions, economic burden, and uncertain prognosis can easily cause patients to experience anxiety, depression, despair, and suicidal ideation[4]. Severe psychological distress can reduce treatment compliance and even trigger refusal to treat, self-harm, or suicidal behaviors, affecting treatment effectiveness and ward safety[5]. Presently, routine nursing still focuses on the management of physical complications, and there is insufficient early screening and stratified management of psychological distress and suicide risks[6].

Stepwise care is a service model that hierarchically allocates and dynamically adjusts intervention intensity and professional resources based on the severity of patient symptoms, risk levels, and intervention responses. Its core lies in building a continuous, multilevel mental health service system; matching corresponding care measures based on the severity of patients’ symptoms, actual needs, personal preferences, and responses to intervention; and dynamically adjusting intervention intensity through continuous evaluation[7]. This model has gradually expanded from managing depression, anxiety, and other chronic conditions to providing psychological care for patients with cancer. Existing research has shown that tiered care for tumors can integrate health education, self-management, supportive psychological care, cognitive-behavioral intervention, and multidisciplinary collaboration and has a certain effect on improving psychological distress, depression, and quality of life[8].

The hierarchical psychological intervention program constructed by this research institute draws on the theory of psychological crisis intervention. This theory emphasizes prioritizing crisis assessment and safety management for patients at high risk of self-injury or suicide, and reducing risks by establishing therapeutic relationships, mobilizing family and social support, restoring adaptive responses, and continuous follow-up[9]. However, previous evidence mainly came from lung cancer, head and neck tumors, or mixed cancers, and there is still limited research on psychological distress and suicide risk in patients with liver cancer treated with TACE combined with targeted therapy. Therefore, this study developed a comprehensive, stepwise psychological intervention plan and evaluated its impact on psychological distress, suicide risk, negative emotions, treatment compliance, nursing satisfaction, and ward safety events to provide a basis for hierarchical management of psychological risk and psychological crisis prevention and control during the perioperative period of liver cancer treatment.

MATERIALS AND METHODS

General information

A total of 98 patients with primary liver cancer who underwent targeted therapy combined with TACE at our hospital’s Interventional Radiology Department between April 2024 and November 2025 were selected. Using a random number table, patients were divided into a control group (49 patients) and an observation group (49 patients).

Inclusion criteria: (1) Meeting the diagnostic criteria of the “Guidelines for the Diagnosis and Treatment of Primary Liver Cancer (2022 Edition),” with primary liver cancer confirmed by pathology or clinical imaging; (2) Aged 18-75 years, with an expected survival time of ≥ 3 months; (3) Planning to undergo at least two courses of TACE combined with targeted therapy (sorafenib/Lenvatinib); (4) Being alert and oriented, with normal cognitive function and able to independently complete the questionnaire assessments; and (5) Having no communication barriers, being willing to participate in this study, and being able to comply with follow-up.

Exclusion criteria: (1) Concurrent severe functional failure of vital organs such as the heart, brain, lungs, or kidneys; (2) History of severe psychiatric disorders such as schizophrenia or bipolar disorder, or receipt of professional psychiatric or psychological treatment within the past 6 months; (3) Concurrent conditions affecting cognitive and emotional assessment, such as brain metastases or hepatic encephalopathy; (4) Unable to complete questionnaires, withdrew from treatment during the study, or were lost to follow-up; and (5) History of substance abuse or alcohol dependence. This study was reviewed and approved by the hospital’s Medical Ethics Committee (approval No. 2024-K-048-02). All patients and their legally authorized representatives signed a written informed consent form (Figure 1).

Figure 1 CONSORT flow diagram of participant screening, enrollment, allocation, follow-up, and analysis. DT: Distress Thermometer; NGASR: Nurses’ Global Assessment of Suicide Risk; SAS: Self-Rating Anxiety Scale; SDS: Self-Rating Depression Scale; IQR: Interquartile range; TACE: Transarterial chemoembolization.

Sample size estimation

This study used the Psychological Distress Thermometer (DT) score as the primary outcome measure for sample size estimation. Based on previous clinical data and related studies on the variability of DT scores, the expected between-group difference in DT scores is 0.80 points, with a standard deviation of 1.28 points. Assuming a two-sided test level of α = 0.05 and a test efficiency of 1-β = 0.80, the sample size was calculated using the formula for comparing the means of two independent samples. The calculation indicated that at least 41 patients were required per group. Considering a loss to follow-up or dropout rate of approximately 15%, the final target sample size was set at 49 patients per group, yielding a total of 98 patients. The patients were randomly divided into a control group of 49 patients and an observation group of 49 cases using a random number table method. Both groups received corresponding interventions and completed a 4-week follow-up.

Observation indicator

Randomization, allocation concealment, and blinding methods: According to the order of patient enrollment, a random allocation sequence was generated using a random number table method by a researcher who was not involved in patient recruitment, intervention implementation, and outcome evaluation. Patients were divided into a control group and an observation group in a 1:1 ratio. The randomized grouping results were placed in consecutively numbered, opaque, sealed envelopes, which were opened sequentially after patients completed baseline assessment and were formally enrolled in the study, thereby ensuring allocation concealment.

Because this study involved a nursing-based psychological intervention with readily distinguishable intervention content, blinding of patients and nursing staff responsible for implementing the intervention was not feasible. However, to minimize measurement bias, outcome assessments were performed by researchers who were not directly involved in intervention implementation and were blinded to group allocation. During statistical analysis, group codes were used to mask group identities, and the allocation codes were revealed only after the statistical analysis had been completed.

Methods

The control group received standard care at the Interventional Radiology Department. Upon admission, the assigned nurse assessed each patient’s baseline characteristics, medical history, allergy history, psychological state, and family support and provided information on the ward environment, treatment procedures, and targeted therapy combined with TACE. Before surgery, the team assisted in arranging blood tests, liver and kidney function tests, coagulation tests, alpha-fetoprotein testing, electrocardiograms and imaging studies. They assessed vital signs, liver function staging, and the condition of the puncture site and, in accordance with medical instructions, prepared the skin, established intravenous access, performed iodine allergy tests, provided medication guidance, and managed fasting and fluid restrictions. During the procedure, the nurse assisted the patient in assuming a supine position, strictly adhere to the patient identification protocol, and continuously monitor vital signs, including consciousness, blood pressure, heart rate, and oxygen saturation. Abnormalities such as nausea, vomiting, hepatic distension, chest tightness, and allergic reactions were promptly observed and managed. Postoperatively, bed rest and immobilization of the limb on the puncture side were advised for 6-8 hours. The puncture site and any changes in limb perfusion, sensation, motor function were monitored, and guidance was provided on resuming diet and activities. During the nursing period, the focus was placed on monitoring embolism syndrome and adverse reactions to targeted drugs and providing symptomatic treatment as directed. Strengthening guidance on medication, diet, rest, activity, follow-up appointments, and recognition of abnormal symptoms while offering routine psychological support.

The observation group implemented a full-process stepwise psychological intervention on the basis of routine care, which lasted from admission to 4 weeks after discharge, and was carried out in accordance with the principle of “dynamic assessment -risk stratification-graded intervention-re-evaluation and adjustment-follow-up management”.

Intervention team and training: An intervention team composed of head nurses, responsible nurses, oncology specialist nurses, attending physicians, and psychological counselors was established. Before the intervention, unified training was provided on the standardized administration and scoring of DT, problem list, Nurse Suicide Risk Assessment Scale (NGASR), Self-Rating Anxiety Scale (SAS), Self-Rating Depression Scale (SDS), risk classification, supportive psychological care, cognitive-behavioral intervention, relaxation training, suicide risk identification, and crisis management. Those who passed both the theoretical and scenario simulation assessments were permitted to participate in the implementation.

Psychological assessment and risk stratification: The first assessment was conducted within 24 hours of admission. Reassessment was performed 1 day before TACE, 3 days after TACE, 1 day before discharge, and 4 weeks after discharge. Additional assessment was provided when emotional fluctuations, changes in condition, self-harm, or suicidal behaviors occurred. A DT score of 0-3, an NGASR score of < 5, and both SAS and SDS scores of < 50 were considered low risk. A DT score of 4-6 or an SAS or SDS score of 50-69, with an NGASR score of < 5 and no clear suicide plan, was considered medium risk. Those with a DT score of ≥ 7, an NGASR score of ≥ 5, an SAS or SDS score of ≥ 70, or with clear suicidal ideation, plans, and self-harm behaviors were considered high-risk. When the indicators are inconsistent, the higher level classification was used.

Graded intervention: For low-risk individuals, health education, supportive communication, and relaxation training were provided at least once before admission, before surgery, after surgery, and before discharge, with each session lasting 15-20 minutes. Relaxation training lasted 10–15 min per session and was conducted 1-2 times a day. For medium-risk individuals, cognitive behavioral interventions, emotional counseling, and family empowerment was increased. These interventions were implemented before surgery, after surgery, and before discharge, each lasted 20-30 minutes. The focus was on correcting catastrophic cognitions, promoting emotional expression, and guiding family members to identify warning signs. Immediate crisis interventions were initiated for high-risk individuals. If necessary, consultations with mental health and psychological specialists were conducted. 24-hour focused care, bedside shift handover, family accompaniment, and removal of dangerous items should be implemented.

Dynamic adjustment and follow-up: After each re-evaluation, the risk level and intervention intensity were adjusted based on scale scores and patient performance. After discharge, follow-up visits were conducted by phone or WeChat once a week for 4 consecutive weeks, with each visit lasting 10 minutes to 15 minutes. The DT, NGASR, SAS, and SDS evaluations were completed in the 4^th week. If the risk level increased, timely referral was made.

Quality control: Operation procedures and record sheets were developed to document risk levels, measures, duration, and patient responses. The head nurse regularly checked the implementation, and the psychological counselor supervised medium- and high-risk cases. For those with inconsistent risk grading or poor effects, the plan was adjusted after discussion to ensure consistency and repeatability of the intervention.

Observation indicator

Psychological distress: The DT was used to assess the severity of patients’ psychological distress. The DT is a single-item rating scale ranging from 0 to 10, where a score of 0 indicates no psychological distress, 1-3 indicates mild distress, 4-6 indicates moderate distress, and 7-10 indicates severe distress; a higher score indicates more severe distress. In this study, a DT score of ≥ 4 was used to indicate the presence of significant psychological distress. Concurrently, the PL was used to analyze sources of psychological distress, covering emotional, family, physical, and practical issues, and spiritual beliefs[8,9].

Suicide risk: The Chinese version of the NGASR was used to assess suicide risk. The total score ranged from 0 to 10 and covered aspects such as suicidal ideation, suicidal plans, suicidal preparation, feelings of hopelessness, history of suicide attempts, current stressors, social support, emotional control, physical condition, and clinical indicators. A higher score indicates a higher risk of suicide; scores of 0-4 are classified as low to moderate risk, whilst scores of 5 or above are classified as high risk of suicide[10].

Negative emotions: The SAS and SDS were used to assess the patients’ negative emotions. Both scales comprise 20 items, scored on a scale of 1 to 4 and primarily reflect anxiety, depression, sleep disturbances, changes in appetite, feelings of hopelessness, and self-blame. The scale results were converted into standard scores: 50-59 indicated mild anxiety or depression; 60–69 indicated moderate anxiety or depression; and ≥ 70 indicated severe anxiety or depression. Higher scores indicate more severe negative emotions[11].

Treatment adherence: The responsible nurse conducted a comprehensive assessment based on the patient’s medication adherence, follow-up examinations, and cooperation with tests and treatments during hospitalization and within 4 weeks of discharge, categorized compliance into three levels: Full compliance, partial compliance, and non-compliance[12]. Treatment adherence was assessed using predefined behavioral criteria based on medication use, attendance at scheduled examinations and follow-up visits, completion of planned TACE treatment, and cooperation with nursing and medical procedures. Complete adherence was defined as adherence to all prescribed treatment and follow-up requirements without unauthorized interruption. Partial adherence was defined as occasional non-adherent behavior that was corrected after counseling and did not materially disrupt treatment. Non-adherence was defined as persistent refusal or interruption of prescribed treatment, unauthorized discontinuation of medication, repeated missed appointments, or other behaviors that substantially interfered with the planned treatment course. Adherence assessments were based on nursing records, medication records, and follow-up documentation. Treatment compliance rate = (number of cases with full compliance + number of cases with partial compliance)/total number of cases × 100%.

Nursing satisfaction: The evaluation was conducted using a standardized nursing satisfaction survey questionnaire developed by the hospital. The questionnaire covered admission education, condition monitoring, basic nursing care, specialist nursing care, communication and explanation, psychological support, health guidance, service attitudes, ward management, and overall evaluation, with total scores ranging from 0 to 100. A score of ≥ 90 was classified as “very satisfied”, 80-89 as “satisfied” and < 80 as “dissatisfied”. Nursing satisfaction was calculated as (number of “very satisfied” cases + number of “satisfied” cases)/total number of cases × 100%. The questionnaire had not undergone external psychometric validation; therefore, satisfaction was considered an exploratory secondary outcome.

Ward safety incidents: Safety-related events included suicide attempts, deliberate self-harm, and severe treatment-refusal behaviors associated with psychological distress that resulted in clinically relevant consequences or required additional clinical intervention. Events were identified from nursing records, medical records, and incident reports. For patients showing increased suicide risk or suicidal ideation, the predefined escalation pathway included immediate reassessment, notification of the attending physician and head nurse, intensified observation, removal of potentially dangerous objects, family accompaniment where appropriate, and consultation with mental health or psychological specialists. Emergency psychiatric referral was initiated when clinically indicated. All recorded safety events during hospitalization and the 4-week follow-up period were included in the analysis. Only incidents that resulted in adverse consequences were included in the analysis. Ward safety incident rate = number of safety incidents/total number of cases × 100%.

Statistical analysis

Data analysis was performed using IBM SPSS Statistics for Windows, version 26.0. Continuous variables are expressed as mean ± SD; paired t-tests were used for comparisons within groups before and after intervention, and independent-samples t-tests were used for comparisons between groups. Categorical variables are expressed as n (%); the χ² tests or Fisher’s exact test was used for comparisons between groups. Pearson’s correlation analysis was used to assess the associations of psychological distress and suicide risk, and negative emotions. With the presence of significant psychological distress (DT ≥ 4 points) and high suicide risk (NGASR ≥ 5 points) post-intervention as dependent variables, a multivariate logistic regression analysis was conducted incorporating variables found to be significant in univariate analysis. Two multivariable binary logistic regression models were additionally constructed to explore factors associated with significant psychological distress (DT ≥ 4) and high suicide risk (NGASR ≥ 5) at the 4-week follow-up. To minimize overadjustment and potential mediator bias, only baseline or prerandomization variables were considered as covariates. Post-intervention SAS and SDS scores were not included because they may represent intermediate variables on the pathway between the psychological intervention and the outcomes. Treatment allocation was retained in all models irrespective of statistical significance, while a parsimonious set of clinically relevant baseline covariates was selected in view of the limited sample size. Adjusted ORs with 95%CIs were reported. Multicollinearity was assessed using variance inflation factors, and model calibration was examined using the Hosmer-Lemeshow goodness-of-fit test. The significance level was set at α = 0.05, and P < 0.05 was considered statistically significant.

RESULTS

Comparison of general characteristics between the two groups

There were no statistically significant differences between the two groups in terms of age, sex, disease duration, Child–Pugh classification, China liver cancer stage, Eastern Cooperative Oncology Group performance status, number of tumors, maximum tumor diameter, portal vein tumor thrombus, distant metastasis, type of targeted drug, or previous history of TACE (P > 0.05) (Table 1).

Table 1 Comparison of baseline characteristics between the two groups, n (%)/mean ± SD.

Variable	Control group (n = 49)	Observation group (n = 49)	χ²/t value	P value
Age (years)	56.72 ± 8.21	55.95 ± 7.82	0.475	0.636
Sex			0.234	0.628
Male	37 (75.51)	39 (79.59)
Female	12 (24.49)	10 (20.41)
Disease duration (months)	6.23 ± 2.11	5.91 ± 2.36	0.708	0.481
Child–Pugh classification			0.192	0.661
Grade A	34 (69.39)	36 (73.41)
Grade B	15 (30.61)	13 (26.53)
CNLC stage			0.057	0.972
Stage IIb	14 (28.57)	15 (30.61)
Stage IIIa	23 (46.94)	22 (44.90)
Stage IIIb	12 (24.49)	12 (24.49)
ECOG score			0.049	0.825
0-1 point	33 (67.35)	34 (69.39)
2 points	16 (32.65)	15 (30.61)
Number of tumors			0.172	0.678
Single lesion	20 (40.82)	18 (36.73)
Multiple lesions	29 (59.18)	31 (63.27)
Maximum tumor diameter			0.172	0.678
< 5 cm	18 (36.73)	20 (40.82)
≥ 5 cm	31 (63.27)	29 (59.18)
Portal vein tumor thrombus			0.186	0.667
Present	15 (30.61)	17 (34.69)
Absent	34 (69.39)	32 (65.31)
Distant metastasis			0.272	0.602
Present	8 (16.33)	10 (20.41)
Absent	41 (83.67)	39 (79.59)
Type of targeted therapy			0.169	0.681
Sorafenib	21 (42.86)	19 (38.78)
Lenvatinib	28 (57.14)	30 (61.22)
Previous history of TACE			0.046	0.831
Yes	17 (34.69)	16 (32.65)
No	32 (65.31)	33 (67.35)

TACE: Transarterial chemoembolization; ECOG: Eastern Cooperative Oncology Group; CNLC: China liver cancer.

Comparison of psychological distress and suicide risk scores before and after intervention

Before the intervention, no significant differences in DT and NGASR scores were observed between the two groups (P > 0.05). After 4 weeks of intervention, both scores had significantly decreased from baseline in both groups, and the scores in the observation group were significantly lower than those in the control group (P < 0.05) (Table 2).

Table 2 Comparison of Distress Thermometer and Nurses’ Global Assessment of Suicide Risk scores before and after intervention between the two groups, mean ± SD.

Group	Case	DT		NGASR
		Before	After	Before	After
Control	49	6.82 ± 1.35	4.91 ± 1.18	5.76 ± 1.42	4.23 ± 1.09
Observation	49	6.79 ± 1.31	3.25 ± 0.92	5.71 ± 1.38	2.16 ± 0.81
t value		0.112	7.766	0.177	10.670
P value		0.911	< 0.001	0.860	< 0.001

DT: Distress Thermometer; NGASR: Nurses’ Global Assessment of Suicide Risk.

Comparison of anxiety and depression scores before and after intervention

Before the intervention, no statistically significant differences were observed in the SAS and SDS scores between the two groups (P > 0.05). After 4 weeks of intervention, the SAS and SDS scores significantly decreased in both groups, with significantly lower scores in the observation group than in the control group (P < 0.05) (Table 3).

Table 3 Comparison of Self-Rating Anxiety Scale and Self-Rating Depression Scale scores before and after intervention between the two groups, mean ± SD.

Group	Case	SAS		SDS
		Before	After	Before	After
Control	49	58.64 ± 7.21	50.12 ± 6.35	57.38 ± 6.95	49.26 ± 6.12
Observation	49	58.27 ± 7.16	42.35 ± 5.48	57.01 ± 6.88	41.53 ± 5.36
t value		0.255	6.485	0.265	6.651
P value		0.799	< 0.001	0.792	< 0.001

SAS: Self-Rating Anxiety Scale; SDS: Self-Rating Depression Scale.

Comparison of treatment compliance and nursing satisfaction between the two groups

The treatment compliance rate and nursing satisfaction rate were significantly higher in the observation group than in the control group (P < 0.05; Table 4).

Table 4 Comparison of treatment adherence and nursing satisfaction between the two groups, n (%).

Variable	Control group (n = 49)	Observation group (n = 49)	χ² value	P value
Treatment adherence
Complete adherence	21 (42.86)	32 (65.31)
Partial adherence	17 (34.69)	15 (30.61)
Non-adherence	11 (22.45)	2 (4.08)
Treatment adherence rate	38 (77.55)	47 (95.92)	7.184	0.007
Nursing satisfaction
Very satisfied	23 (46.94)	34 (69.39)
Satisfied	14 (28.57)	13 (26.53)
Dissatisfied	12 (24.49)	2 (4.08)
Overall satisfaction rate	37 (75.51)	47 (95.92)	8.333	0.004

Comparison of the incidence of ward safety events between the two groups

The total incidence of ward safety events was significantly lower in the observation group than that in the control group (P < 0.05; Table 5).

Table 5 Comparison of safety incident incidence rates between the two wards, n (%).

Group	Cases (n)	Suicide attempt	Self-harm behavior	Extreme refusal of treatment	Total cases (incidence rate)
Control group	49	2 (4.08)	3 (6.12)	4 (8.16)	9 (18.37)
Observation group	49	0 (0.00)	1 (2.04)	1 (2.04)	2 (4.08)
χ² value					5.018
P value					0.025

Correlation analysis

Pearson’s correlation analysis showed that post-intervention DT scores were strongly positively correlated with NGASR scores (r = 0.782, P < 0.001). The DT scores were positively correlated with the SAS and SDS scores (r = 0.714 and 0.736, respectively; both P < 0.001). The NGASR scores were also positively correlated with the SAS and SDS scores (r = 0.688 and r = 0.719, respectively; both P < 0.001). In addition, the DT and NGASR scores were positively correlated with advanced tumor stage (r = 0.296 and 0.258; P = 0.003 and 0.010, respectively).

Multivariate logistic regression analysis

In the multivariable logistic regression analysis, only baseline or prerandomization variables were included to minimize potential overadjustment by post-randomization factors. After adjustment for baseline DT score, tumor stage, and family support, allocation to the stepped psychological intervention group was associated with lower odds of significant psychological distress at the 4-week follow-up [adjusted odds ratio (OR) = 0.31, 95%CI: 0.12-0.78, P = 0.014]. Higher baseline DT scores and advanced tumor stage were associated with higher odds of psychological distress, whereas good family support was associated with lower odds.

Similarly, after adjustment for baseline NGASR score, tumor stage, and family support, allocation to the intervention group was associated with lower odds of high suicide risk at follow-up (adjusted OR = 0.27, 95%CI: 0.09-0.79, P = 0.018). Higher baseline NGASR scores and advanced tumor stage were associated with increased odds of high suicide risk, whereas good family support was associated with decreased odds (Table 6).

Table 6 Multivariate logistic regression analysis of factors influencing psychological distress and suicide risk.

Outcome/predictor	B	SE	OR	95%CI	P value
Model 1: Significant psychological distress at follow-up (DT ≥ 4)
Stepped psychological intervention vs control	-1.171	0.478	0.31	0.12-0.78	0.014
Baseline DT score, per 1-point increase	0.392	0.143	1.48	1.12-1.96	0.006
Advanced tumor stage, yes vs no	0.975	0.474	2.65	1.05-6.72	0.040
Good family support, yes vs no	-0.868	0.432	0.42	0.18-0.98	0.045
Model 2: High suicide risk at follow-up (NGASR ≥ 5)
Stepped psychological intervention vs control	-1.309	0.554	0.27	0.09-0.79	0.018
Baseline NGASR score, per 1-point increase	0.476	0.156	1.61	1.19-2.19	0.002
Advanced tumor stage, yes vs no	1.125	0.540	3.08	1.07-8.89	0.037
Good family support, yes vs no	-1.022	0.486	0.36	0.14-0.94	0.036

DT: Distress Thermometer; NGASR: Nurses’ Global Assessment of Suicide Risk; OR: Odds ratio.

DISCUSSION

Although tumor progression can be effectively controlled in patients with mid-to late-stage liver cancer who undergo targeted therapy combined with TACE, the cumulative effects of long-term disease-related stress, treatment-related adverse reactions, and multiple social pressures render them a high-risk group for psychological distress and suicide. Thus, psychological issues have become a key bottleneck limiting treatment adherence and clinical safety. Current clinical oncology nursing tends to focus on the management of physical complications while lacking standardized procedures for the screening, stratification, and intervention of psychological crises, making it difficult to achieve early identification and precise management of potential risks. This study applied a stepwise psychological intervention to patients undergoing targeted therapy combined with TACE for liver cancer to validate its efficacy in improving psychological distress, suicide risk, and clinical outcomes and to provide empirical evidence for the establishment of a psychological safety management system during the peri-treatment period.

The results of this study indicated that implementing a comprehensive stepwise psychological intervention for patients undergoing targeted therapy combined with TACE for liver cancer can significantly reduce psychological distress and suicide risk, alleviate anxiety and depression, improve treatment adherence and satisfaction with care, and lower the incidence of safety incidents in wards. These findings were largely consistent with the conclusions of previous studies on psychological nursing in oncology and stepwise care[13-16]. Centered on dynamic assessment, risk stratification, tiered intervention and closed-loop management, the stepwise psychological intervention overcomes the limitations of conventional nursing-which tends to “prioritize the physical over the psychological” and adopt a “one-size-fits-all” approach. Precise matching of the intensity of intervention to patients’ psychological needs enables the early identification, intervention, and prevention of psychological crises[17,18]. Previous studies have shown that stepwise care, collaborative psychological interventions, and nurse-led psychological interventions can improve outcomes such as depression, emotional distress, and quality of life in patients with cancer, providing a theoretical basis and practical reference for psychological safety management during the peritreatment period for intermediate-to-advanced liver cancer[15-18].

Patients with mid-to-late-stage liver cancer bear a heavy psychological burden due to factors such as poor prognosis, prolonged treatment cycles, adverse reactions, and financial and family pressures, often accompanied by anxiety, depression, feelings of despair, and suicidal ideation[19,20]. In this study, pre-intervention DT scores in both groups were moderate to severe, while NGASR scores indicated a high proportion of patients at a high risk of suicide, suggesting that patients undergoing targeted therapy combined with TACE for liver cancer face significant psychological distress. Previous research has shown that suicidal ideation in cancer patients is closely associated with psychological distress, anxiety, and depression, and that DT can be used for the rapid screening of psychological risk[21]. Conventional nursing care lacks standardized screening and tiered management protocols, making it difficult to identify latent psychological distress and suicide risk in a timely manner. A stepwise psychological intervention utilized the DT, NGASR, SAS, and SDS as assessment tools, conducting dynamic reassessments and tiered interventions at key treatment junctures. This facilitates a shift from experience-based judgement to data-driven decision-making, thereby helping improve the identification rate of psychological risk and the appropriateness of interventions[22].

From a theoretical perspective, stepwise psychological intervention may exert effects on multiple aspects such as cognitive adjustment, emotional support, strengthening of social support, and safety management. From the perspective of cognitive behavioral theory, patients’ negative evaluations of cancer prognosis, treatment outcomes, and disease threats can further induce emotions such as anxiety, depression, helplessness, and despair, and may affect treatment behaviors and suicidal ideation. Therefore, cognition, emotion, and behavior are mutually influenced. Firstly, implementing cognitive behavioral intervention for patients at medium and high risk may help correct catastrophic perceptions such as “cancer equals death” and “ineffective treatment” reduce feelings of despair and helplessness, enhance patients’ sense of control over the disease and treatment and their positive coping ability, and thereby play a certain regulatory role in the chain of “negative cognition-negative emotion-suicide risk”[23]. Second, one-on-one emotional counseling and family empowerment education may promote patients’ emotional expression, enhance family care and support, and play a buffering role in disease-related stress. Family support can also reduce patients’ sense of isolation and helplessness by enhancing their emotional connection and available coping resources and improving their psychological resilience in the face of disease and treatment pressure. Multivariate logistic regression analysis showed that good family support was associated with lower psychological distress and lower suicide risk[24]. In addition, relaxation training and breathing guidance may have a positive impact on anxiety and depression by reducing tension and improving sleep and physical discomfort[25]. For high-risk patients, timely initiation of closed-loop management measures such as multidisciplinary consultation, 24-h care, and removal of hazardous materials can form a continuous intervention chain from risk identification and professional assessment to safety protection and referral treatment, thereby reducing the possibility of psychological crises evolving from suicidal ideation to self-harm or suicidal behavior[26]. It should be noted that this study mainly evaluated clinical outcome indicators, such as DT, NGASR, SAS, and SDS, and did not collect process indicators such as disaster cognition, disease cognition, coping styles, and perceived social support, nor did it conduct a mediating effect analysis. Therefore, the aforementioned pathways of cognitive adjustment, emotional support, and social support are mainly based on intervention content, relevant theories, and possible explanations proposed in previous studies. Their specific mediating roles or causal pathways cannot be directly confirmed using the data obtained in this study.

This study showed that the reductions in DT and NGASR scores and the improvements in anxiety and depression in the observation group after intervention were all greater than those in the control group, suggesting that stepwise psychological intervention can simultaneously improve psychological distress, negative emotions, and suicide risk, and that the three are closely related[27,28]. Pearson correlation analysis showed that psychological distress was strongly positively correlated with suicide risk, anxiety, and depression and was weakly positively correlated with the advanced state of tumors, indicating a statistical association among the above psychological indicators. However, since this study did not include corresponding process variables or conduct longitudinal mediation analysis, it is not possible to further infer specific action chains, such as “improvement of psychological distress-improvement of negative emotions-reduction of suicide risk” based on this. Multivariate logistic regression analysis further revealed that stepwise psychological intervention was associated with significantly lower psychological distress and lower suicide risk, while advanced tumors and elevated anxiety and depression scores were associated with higher risk, suggesting that clinical attention should be focused on patients with advanced disease and obvious emotional disorders[27,29]. In addition, treatment compliance and nursing satisfaction improved in the observation group, suggesting that stepwise psychological intervention, while improving psychological outcomes, may have a positive impact on patients’ treatment cooperation and nursing experience. However, the specific mechanisms of action require further research and verification[30,31]. The incidence of safety incidents in the observation group was 4.08%, lower than that in the control group (18.37%). This supports the potential clinical value of step-based psychological intervention in reducing safety risks related to self-harm, suicide, and extreme resistance to treatment[28,31].

This study has certain limitations. First, it was a single-center, small-sample study with a follow-up period of only 4 weeks. The extrapolation of the research results and long-term intervention effects needs to be further verified. Second, the optimal intervention frequency and duration for patients with different risk levels have not yet been determined. Furthermore, this study did not collect process indicators, such as disease cognition, coping styles, and perceived social support. Therefore, it is impossible to empirically verify potential mediating pathways such as cognitive adjustment, emotional regulation, and social support. Explanations of the relevant mechanisms are primarily based on theoretical inferences. In the future, multicenter, large-sample, and long-term follow-up studies should be conducted, and relevant process indicators should be included. Combined with the analysis of mediating effects, the mechanism of action of stepwise psychological interventions should be further clarified.

CONCLUSION

In summary, through standardized assessment, stratified intervention, and closed-loop management, stepwise psychological intervention can effectively alleviate psychological distress in patients undergoing targeted therapy combined with TACE for liver cancer, reduce the risk of suicide and negative emotions, improve treatment adherence and satisfaction with care, and reduce safety incidents in wards.