Psychological intervention and health education effects on quality of life in patients after pancreatic cancer surgery: A meta-analysis

Abstract

BACKGROUND

Pancreatic cancer is one of the most aggressive malignancies, and surgery remains the only potentially curative treatment. However, postoperative complications, psychological distress, and treatment-related symptoms can severely impair quality of life (QoL) of patients. Although psychological interventions and health education have shown beneficial effects in other cancer populations, their effectiveness in postoperative patients with pancreatic cancer remains inconsistent and inadequately synthesized. We hypothesized that structured psychological and educational support would improve global health status, functional outcomes, and the symptom burden of postoperative patients with pancreatic cancer compared to standard care.

AIM

To determine how psychological intervention and health education affects the QoL of postoperative patients with pancreatic cancer.

METHODS

We searched PubMed, EMBASE, Cochrane Library, Web of Science, CINAHL, and MEDLINE from inception to October 2025. Two investigators independently screened studies, extracted data, and assessed bias. Quality was appraised using the Cochrane Risk of Bias tool (RoB 2.0), and the primary effect measure was the standardized mean difference (SMD) derived from a meta-analysis using RevMan 5.4.

RESULTS

Of the 1613 records initially identified, six studies involving 712 patients were included in the final analysis after applying exclusion criteria. The meta-analysis revealed that, compared with usual care, psychological interventions and health education significantly improved patients’ global health status [MD = 7.12, 95% confidence interval (CI): 4.91 to 9.32, P < 0.001], physical functioning (SMD = 0.61, 95%CI: 0.37 to 0.85, P < 0.001), role functioning (SMD = 0.64, 95%CI: 0.33 to 0.94, P < 0.001), emotional functioning (SMD = 1.03, 95%CI: 0.78 to 1.28, P < 0.001), dietary digestion (MD = -2.37, 95%CI: -3.12 to -1.63, P < 0.001), and nursing satisfaction (MD = 0.69, 95%CI: 0.50 to 0.88, P < 0.001).

CONCLUSION

Psychological interventions combined with health education can improve global health status, alleviate symptom burden, and enhance functional outcomes compared to standard care among postoperative patients with pancreatic cancer.

Key Words: Psychological intervention; Health education; Pancreatic cancer; Quality of life; Systematic review; Meta-analysis

Core Tip: This systematic review and meta-analysis is the first to synthesize evidence on the impact of psychological intervention and health education on the quality of life among postoperative patients with pancreatic cancer. Despite a low certainty of evidence, the results indicated significant improvements in global health, physical and emotional functioning, and digestion. These findings highlight the potential of integrated supportive care in a population with a high symptom burden and underscore the need for more rigorous trials.

INTRODUCTION

Pancreatic cancer is one of the most aggressive and lethal malignancies worldwide and surgery remains the only potentially curative treatment. Locally advanced pancreatic body or tail cancers can be treated by distal pancreatectomy with en bloc celiac axis resection. However, these lengthy procedures increase the chance of complete resection and are associated with long surgical durations, as well as increased complication and postoperative mortality rates[1]. Postoperative complications can further disrupt the ability to fulfil daily roles and participate in social functions, significantly impacting long-term quality of life (QoL)[2]. Patients with pancreatic cancer experience worse QoL compared to those with hepatobiliary or colorectal cancers, particularly in terms of psychological distress, cognitive function, and coping with the disease[3,4]. Additionally, adjuvant therapies commonly used postoperatively, such as cisplatin, may have toxic effects that affects both treatment efficacy and QoL.

In addition to surgical and chemotherapeutic interventions, adjuvant agents that might improve treatment efficacy and potentially ameliorate treatment-related toxicity are under investigation. For instance, fucoxanthin is a natural compound that enhances the sensitivity of pancreatic cancer cells to cisplatin. It does this by remodeling tumor cell energy metabolism and upregulating copper transporter SLC31A1 expression[5]. While such pharmacological advances aim to optimize oncological outcomes, they do not directly address the profound psychological and informational needs that arise during the postoperative and treatment phases. Indeed, surgery and adjuvant chemotherapy impose burdens that extend beyond physical suffering. The psychological impact of the diagnosis and surgery, concerns about prognosis, fear of treatment uncertainties, and postoperative lifestyle changes can trigger anxiety, depression, and post-traumatic stress disorder[6,7]. These negative emotions not only affect treatment compliance and motivation to undergo rehabilitation but might also indirectly influence disease progression and the QoL through neuroendocrine and immune regulatory mechanisms.

Psychological interventions and health education are essential components of supportive care and they have been widely applied to rehabilitate patients with various types of cancers[8,9]. Psychological interventions typically include cognitive-behavioral therapy, mindfulness-based stress reduction, and supportive group therapy, aiming to help patients adjust negative thoughts, reduce emotional distress, and enhance coping abilities[10]. Health education, through systematic information provision, self-management skill training, and lifestyle guidance, helps patients to understand their disease and treatment, which enhances self-efficacy and treatment compliance[11].

The 2024 United States national guidelines explicitly categorized Survivor/Family Education and Counseling and Psychological support as Level 2A recommendations for supportive care for all patients with any type of cancer[12]. Substantial evidence supports the notion that these interventions can significantly improve the QoL, emotional state, and functional recovery of patients with breast, colorectal, and cervical cancers[13,14]. For instance, a systematic review and meta-analysis of 26 studies and 11638 patients confirmed that psychological nursing and family education met the supportive care needs of patients with cervical cancer and significantly reduced anxiety and depression levels[14]. Psychological interventions combined with health education have been widely applied to patients with cancer undergoing treatment. A randomized controlled trial (RCT) found that this approach significantly improved the QoL, reduced anxiety and depression, alleviated pain, and increased nursing satisfaction, more effectively than standard care[15]. A structured model with integrated emotionadaptation psychological intervention with feedback-based health education has improved the psychological status and respiratory function, enhanced QoL scores, and reduced the incidence of treatmentrelated adverse effects among patients with lung cancer[16]. These results indicated that this combination can address the psychosocial and informational needs of patients with diverse types of cancer. However, the effectiveness of psychological interventions and health education remains inconclusive among patients with pancreatic cancer, particularly during the postoperative phase, because systematic evidence has not been consolidated. Furthermore, given the generally short overall survival and the rapid progression of pancreatic cancer, the psychological support required might differ from those of other types cancer. If so, then more targeted and timely intervention strategies would be needed. Therefore, we aimed to consolidate published clinical research through systematic review and meta-analysis of the literature. We evaluated the impact of psychological interventions and health education on the QoL of postoperative patients with pancreatic cancer. We also aimed to provide an evidence base for clinical practice and promote the development and optimization of comprehensive postoperative support models for pancreatic cancer.

MATERIALS AND METHODS

This systematic review and meta-analysis proceeded according to the preferred reporting items for systematic reviews and meta-analyses (PRISMA) guidelines[17].

Search strategy

We systematically searched the literature using the following electronic databases: PubMed, EMBASE, the Cochrane Library, Web of Science, CINAHL, and MEDLINE. The search timeframe spanned from the inception of each database to October 2025. The search strategy combined Medical Subject Headings (MeSH) terms and free-text keywords related to “pancreatic neoplasms”/”pancreatic cancer” “pancreaticoduodenectomy” AND “postoperative period”/“postoperative”/“after surgery”/“surgery” AND “psychological intervention”/“psychotherapy”/“supportive care”/“health education” AND “quality of life”/“QoL”. We also manually screened reference lists of selected studies to identify potentially relevant articles.

An example of the search strategy used in PubMed is as follows: [“Psychosocial intervention” (MeSH Terms) OR “palliative care” (MeSH Terms) OR “health education” (MeSH Terms) OR “intervention” (title/abstract) OR “consult” (title/abstract) OR “education” (title/abstract) OR “phone” (title/abstract) OR “nursing” (title/abstract) OR “care” (title/abstract)] AND [“pancreatic neoplasms” (MeSH Terms) OR “pancreatic cancer” (title/abstract) OR “pancreaticoduodenectomy” (MeSH Terms) OR “Pancreatobiliary” (title/abstract) OR “pancreatic head cancer” (title/abstract) OR “pancreatic tail cancer” (title/abstract)] AND [“quality of life” (title/abstract) OR “health related quality of life” (title/abstract) OR “HRQOL” (health-related quality of life) (title/abstract) OR “QLQ-30” (European Organization for Research and Treatment of Cancer Quality of Life Questionnaire-Core 30) (title/abstract) OR “PSQI” (Pittsburgh sleep quality index) (title/abstract) OR “SDS” (Self-rating Depression Scale) (title/abstract) OR “SAS” (Self-Rating Anxiety Scale) (title/abstract) OR “SF-36” (36-Item Short Form Health Survey) (title/abstract) OR “SF-12” (title/abstract) OR “WHOQOL-BREF” (World Health Organization quality of life) (title/abstract) OR “WHOQOL-100” (title/abstract) OR “QLQ-PAN26” (quality of life questionnaire-pancreatic cancer 26) (title/abstract)].

Inclusion criteria

Participants: Patients aged ≥ 18 years with a pathologically confirmed diagnosis of pancreatic cancer who underwent surgical pancreaticoduodenectomy, distal pancreatectomy, or total pancreatectomy, no distant metastasis, no severe cognitive impairment or mental illness, able to complete questionnaires independently.

Intervention: The experimental group had received any type of structured psychological intervention or health education and the control group received standard care.

Outcomes: Studies must include QoL scores quantified using validated European Organization for Research and Treatment of Cancer quality of life questionnaire-core 30 (EORTC QLQ-C30), functional assessment of cancer therapy, SF-36 scales.

Exclusion criteria

Non-English publications, text unavailable, insufficient data for extraction, duplicate publications, interventions consisted of physical therapy or pharmacological treatment.

Study selection and data extraction

Two reviewers independently screened the literature and extracted data. Disagreements were resolved through discussion with a third reviewer. The selection process began with the removal of duplicates using EndNote X9 software, followed by title and abstract screens to exclude obviously irrelevant records such as reviews and case reports. The remaining full-text articles were then assessed for eligibility. We extracted the following: Basic study information, namely first author, publication year, and country, the study characteristics of sample size, patient age, sex distribution, tumor stage, and surgical procedure; Intervention details of specific psychological content or health education, frequency, duration, and providers; The outcome measures of QoL scale scores, complication rates, and nursing satisfaction, and methodological information: Randomization, allocation concealment, blinding, and loss to follow up.

Statistical analysis

Data were statistically analyzed using ReviewManager 5 (The Cochrane Collaboration, Oxford, United Kingdom). The mean difference (MD) or standardized MD (SMD) with 95% confidence intervals (CI) were calculated for continuous outcomes. The MD and SMD were respectively applied when studies used the same or different measurement scales. Heterogeneity among studies was visually assessed using forest plots and quantitatively using the I² statistic. We interpreted I² values as: 0%-40% (might not be important), 30%-60% (moderately heterogeneous), 50%-90% (substantial heterogeneity), and 75%-100% (considerably heterogeneous)[18,19]. The quality of evidence for key outcomes was evaluated using the grading of recommendations assessment, development, and evaluation (GRADE) approach[20]. The methodological quality of RCTs was assessed using the revised Cochrane risk-of-bias tool for randomized trials (RoB 2)[21], and non-randomized studies were determined using the Newcastle-Ottawa Scale (NOS)[22]. The two reviewers independently assessed the quality of the methods and statistically analyzed the data and discrepancies were resolved by consensus.

RESULTS

Study selection

The initial database search identified 1613 records. After removing 727 duplicates, 886 unique records were screened based on titles and abstracts. Of these, 818 records were excluded, leaving 68 full-text articles for detailed assessment. Ultimately, four RCTs and two non-RCTs involving 712 patients met the inclusion criteria and were included in the systematic review and meta-analysis. Figure 1 shows a PRISMA flow diagram of the selection process.

Figure 1 Preferred reporting items for systematic reviews and meta-analyses study selection flow chart.

Characteristics of the included studies

The four RCTs and two non-RCTs, involving 712 patients proceeded in China and Sweden. The surgical procedures consisted of three pancreatoduodenectomies[23-25], one pancreatectomy[26], and two radical resections[24,27]. Psychological interventions primarily consisted of mindfulness meditation[24], the BrainLink intelligent biofeedback instrument [monitoring electroencephalography (EEG) signals, attention/relaxation training[24], targeted psychological nursing to monitor psychological changes and alleviate anxiety or fear][26], and psychological counseling was combined with cognitive-behavioral interventions[25]. Health education measures included dietary guidance[26], standardized discharge instructions[28], social comfort care[26], provision of written self-care advice via the Interaktor application[23], real-time symptom reporting and communication support with healthcare professionals via the application[23], self-management manuals[28], dissemination of health information and one-on-one question and answer (QA) via the WeChat platform[28], pain management, nutritional support, and systematic follow-up[25], nutritional education, WeChat group support, and online QA for postoperative care[27]. The QoL was assessed mainly using the EORTC QLQ-C30, while others used the QLQ-PAN26 or SF-36 scales.

Risk of bias assessment

Among the four RCTs, the RoB 2.0 tool identified one study as having a high risk of bias[28] and the other three were rated as having “some concerns”[24-26] (Figure 2). Table 1 shows that the methodological quality of the two non-RCTs assessed using the NOS was judged as good[23,27].

Figure 2 Risk of bias assessment for randomized controlled trials using Risk of Bias tool (RoB 2.0). A: Risk of bias summary; B: Risk of bias.

Table 1 Methodological quality assessment of non-randomized studies using Newcastle-Ottawa Scale.

Ref.	Selection	Comparability	Outcome
Zhang et al[27]	4	2	3
Gustavell et al[23]	4	2	3

Number indicates the score awarded to the study for corresponding item on Newcastle-Ottawa Scale.

Intervention outcomes and comparison

Table 2 provides detailed summary of the psychological intervention and health education measures applied in the studies, including their descriptions, duration, frequency, and outcome measures. Table 3 shows a GRADE assessment of evidence quality for key outcomes, in which the certainty of most outcomes is rated as “very low”.

Table 2 Summary of psychological interventions and health education measures.

Ref.	Intervention type	Added to standard nursing (yes/no)	Description of intervention	Intervention (n)	Control (n)	Duration	Frequency	Outcome measure (quality of life)
Zhang et al[28]	WeChat support + education + team-based continuing nursing	Yes	Self-management manual for patients, setting up specialized outpatient chemotherapy ward, standardized electronic follow up archives, discharge guidance with recording instructions WeChat platform for science popularization, interaction, QA, and scheduling	50	50	Psychological	Follow-up and support were provided continuously from discharge onward via WeChat and scheduled admissions	QLQ-PAN26
Miao[26]	Comfort care	Yes	Monitored patients’ psychological changes; made targeted protocols to managed disorders. Communicated with families to boost confidence in treatment	68	68	Postoperative hospital stay period	Closely monitored psychology; timely family communication	SF-36
Zhang et al[27]	Triangle theory-based continuity of care	Yes	Nutritional education, dietary counseling, WeChat group support, online QA on postoperative care	70	114	3 months	Weekly telephone follow-ups, monthly home visits post-discharge; regular online QA	Pancreatic cancer 26 (QLQ-PAN26)
Gustavell et al[23]	Interactive application support, self-care advice, healthcare professional connection	Yes	Interaktor application: Reported symptoms daily at home; Continuous access to written evidence-based self-care advice; Links for more information, could connect with healthcare professionals	26	33	Up to six Months after surgery	Daily symptom reporting (with daily application reminders); nurse contact as needed for alert-generating symptoms. Median 3 alerts /patient in first 4 weeks, median 7 alerts/patient in rest period. Self-care advice median 13, 5 views first 4 weeks. 11 views/patient during rest period	EORTC QLQ-C30; QLQ-PAN26
Mi et al[24]	Mindfulness meditation (MM) + BrainLink intelligent biofeedback	Yes	Routine nursing plus MM combined with BrainLink intelligent biofeedback instrument	73	72	12 weeks	MM: Daily 20 minutes each time; BrainLink training: Conducted at time of first chemotherapy infusion, then on 4, 8, and 12 weeks thereafter	EORTC QLQ-C30
Wang and Wu[25]	Multidisciplinary specialized nursing care	Yes	Perioperative health education (written/oral/multimedia), psychological counseling, cognitive-behavioral interventions	44	44	1 year	Biweekly telephone calls 1 month after discharge. Monthly thereafter; 24-hour consultation	EORTC QLQ-C30

QA: Question and answer; QLQ-PAN26: Quality of life questionnaire-pancreatic cancer 26; SF-36: 36-Item Short Form Health Survey; EORTC QLQ-C30: European Organization for Research and Treatment of Cancer quality of life questionnaire-core 30; MM: Mindfulness meditation.

Table 3 Grading of recommendations assessment, development, and evaluation analysis.

Outcome	Certainty assessment							Number of patients		Effect absolute (95%CI)	Certainty
	Number of studies	Study design	Risk of bias	Inconsistency	Indirect	Imprecise	Other considerations	IG	CG
Global health status	3	NR	Serious1,2,3	NS	NS	NS	None	142	149	MD = 7.12 (4.91 to 9.32)	Very low
Physical functioning	3	NR	Serious1,2,3	NS	NS	NS	None	137	145	SMD = 0.61 (0.37 to 0.85)	Very low
Emotional functioning	3	NR	Very serious1,2,3	NS	NS	NS	None	137	145	SMD = 1.03 (0.78 to 1.28)	Very low
Dietary digestion	2	NR	Serious1,2,3	NS	NS	NS	None	120	164	MD = -2.37 (-3.12 to -1.63)	Very low

¹Deviation from expected intervention measures.

²Outcome bias measurement.

³Missing data.

IG: Intervention; CG: Control; CI: Confidence interval; MD: Mean difference; NR: Nonrandomized; NS: Nonserious; SMD: Standardized mean difference.

Effects of psychological intervention and health education on QoL

Global health status: Global health status scores are described in three reports[23-25]. The meta-analysis results revealed significantly higher global health status scores (MD = 7.12; 95%CI: 4.91 to 9.32; P < 0.001; Figure 3), without significant heterogeneity (P = 0.97; I² = 0%) among the patients in the psychological intervention or health education group compared to those given standard care. Based on the GRADE assessment (Table 3), the certainty of evidence for global health status was rated as “very low” due to serious risk of bias and imprecision.

Figure 3 Forest plot comparison of psychological intervention and health education with standard care (European Organization for Research and Treatment of Cancer quality of life questionnaire-core 30). A: Global health status; B: Symptom scales; C: Functional scales. IG: Intervention; CG: Control; CI: Confidence interval.

Symptom scales: Symptom scales scores are described in two reports[23,24]. The meta-analysis results indicated significantly lower symptom burden scores in the intervention, than in the control group (MD = -0.62; 95%CI: -1.05 to -0.20;P = 0.004; Figure 3), indicating that the intervention reduced the overall symptom burden. However, heterogeneity was high (P = 0.05, I² = 74%).

Functional scales: Functional scale scores are described in two reports[23,24]. The meta-analysis revealed significantly higher functional scale scores in the intervention, than in the control group (MD = 1.91; 95%CI: 1.23 to 2.60; P < 0.001; Figure 3), although heterogeneity was considered (P = 0.001, I² = 90%). We analyzed functional dimensions in four subgroups as follows.

Physical functioning: Physical functioning scores were significantly higher in the intervention, than in the control group (SMD = 0.61; 95%CI: 0.37 to 0.85; P < 0.001; Figure 4), with no significant heterogeneity (P = 0.57; I² = 0%)[23-25]. The GRADE certainty of evidence was rated as ‘very low’ due to a serious risk of bias.

Figure 4 Forest plot comparison of psychological intervention and health education with standard care. A: Physical; B: Role; C: Emotional; D: Social functioning. IG: Intervention; CG: Control; CI: Confidence interval.

Role functioning: Role functioning scores described in two studies[23,26] were significantly higher in the intervention, than the control group (SMD = 0.64; 95%CI: 0.33 to 0.94; P < 0.001; Figure 4), without significant heterogeneity (P = 0.30; I² = 7%).

Emotional functioning: Three studies included emotional functioning[23-25]. Emotional functioning scores were significantly higher in the intervention, than in the control group (SMD = 1.03; 95%CI: 0.78 to 1.28; P < 0.001), without significant heterogeneity (P = 0.52; I² = 0%; Figure 4). The GRADE certainty of evidence was rated as “very low” due to a serious risk of bias.

Social functioning: Three studies[23-25] found significantly higher social functioning scores in the intervention, than in the control group (SMD = 0.84; 95%CI: 0.44 to 1.24; P < 0.001), but heterogeneity was high (P = 0.08, I² = 60%; Figure 4).

QLQ-PAN26 scale outcomes

Pancreatic cancer-specific symptoms in two studies[27,28] were assessed using the QLQ-PAN26 scale. Subgroup analysis results showed the following.

Pancreatic pain: Pain scores differed significantly between the intervention and control groups (SMD = -0.96; 95%CI: -1.05 to -0.87; P < 0.001), with high heterogeneity (P = 0.006; I² = 87%; Figure 5).

Figure 5 Forest plot comparison of psychological intervention and health education with standard care (European Organization for Research and Treatment of Cancer quality of life questionnaire-pancreatic cancer 26). A: Pancreatic pain; B: Dietary digestion; C: Nursing satisfaction. IG: Intervention; CG: Control; CI: Confidence interval.

Digestion: Digestion outcomes of two studies were reported[27,28]. Scores were significantly better in the intervention, than in the control group (MD = -2.37; 95%CI: -3.12 to -1.63; P < 0.001), with no significant heterogeneity (P = 0.92; I² = 0%). The GRADE certainty of evidence was rated as “very low” (Figure 5).

Nursing satisfaction: Nursing satisfaction in two studies[27,28] was significantly better in the intervention, than in the control group (MD = 0.69; 95%CI: 0.50 to 0.88; P < 0.001), without significant heterogeneity (P = 0.92; I² = 0%; Figure 5).

Publication bias

Figures 6, 7 and 8 show outcomes in funnel plots. Visual assessments of these plots did not reveal obvious asymmetry, which suggests a low likelihood of substantial publication bias. However, the statistical power to detect such bias was limited due to the paucity of studies in each meta-analysis (n < 10).

Figure 6 Funnel plot (European Organization for Research and Treatment of Cancer quality of life questionnaire-core 30). A: Global health status; B: Symptom; C: Functional scales. MD: Mean difference.

Figure 7 Funnel plot of detailed functional scales. A: Physical; B: Role; C: Emotional; D: Social functioning. SMD: Standardized mean difference.

Figure 8 Funnel plot (European Organization for Research and Treatment of Cancer quality of life questionnaire-pancreatic cancer 26). A: Pancreatic pain; B: Digestion; C: Nursing satisfaction. MD: Mean difference.

DISCUSSION

Very low-quality evidence indicates that an integrated supportive care model incorporating psychological interventions or health education significantly improves global health status, functional scores, and symptom scores in patients after pancreatic cancer surgery compared with standard care.

Global health status

Very low-quality evidence suggests that psychological interventions and health education are more effective than standard care in improving global health status scores (MD = 7.12; 95%CI: 4.91 to 9.32; P < 0.001)[23-25]. The global health status item of the EORTC QLQ-C30 scale assesses the macrolevel perception of overall QoL and general health of patients. This improved in macro-level perception of health might be because such interventions specifically address key gaps in conventional postoperative care. While routine management focuses on physiological monitoring and complication control, it tends to underestimate the profound informational and psychological needs that emerge after major cancer surgery[29]. In contrast, the interventions evaluated herein placed greater emphasis on the psychological wellbeing and informational needs of patients ranging from appbased symptom management and self-care guidance using Interaktor[23] to structured telehealth education and WeChat[28] together with multidisciplinary supportive nursing[25]. Collectively, these approaches appear to foster a deeper understanding of the illness, enhance self-efficacy, and strengthen support perceived by patients. Consequently, the improved global health status scores might reflect not only better symptom control but also an elevated sense of coherence and control over one’s health.

Patients recovering from pancreatic cancer surgery often experience feelings of isolation and insufficient support. Therefore, healthcare teams should prioritize strengthening support systems for patients[30]. The interventions examined herein consistently emphasized patient-provider communication and interaction. For instance, the comfort nursing model created a therapeutic atmosphere conducive to psychological and social wellbeing by closely monitoring psychological changes in patients, actively communicating with families, and optimizing the ward environment[26]. Similarly, a continuing nursing team provided accessible and efficient remote post-discharge support via a WeChat platform for knowledge dissemination, one-on-one QA, and clear chemotherapy scheduling[28]. Mindfulness meditation combined with biofeedback intervention[24] offers structured meditation training and objective EEG data feedback, which promoted deep psychosomatic regulation and emotional support. Furthermore, a smartphone-based interactive application[23] has enabled dynamic symptom management and a professional feedback loop via daily symptom reporting and alert generation, allowing proactive nurse intervention and timely self-care guidance. Continuity of care based on the Triangle theory[27] further reinforced self-management skills and the nutritional status of patients via stratified nutritional interventions and systematic follow ups, which indirectly contribute to an overall improvement in QoL. Collectively, this continuous, accessible professional and emotional support can enhance a sense of security among patients, positively influencing their perception of overall QoL.

Functional scales

Our meta-analysis results showed that psychological interventions and health education significantly improved physical, role and emotional functioning compared with standard care[21-23].

The improved physical and role functioning might stem from the multifaceted approaches of the interventions. A direct approach includes structured protocols for early mobilization and graded activity, which directly counteract postoperative physical deconditioning and promote functional recovery[25]. Indirectly, these interventions also alleviate common symptoms such as pain and digestive discomfort, which often impede daily activities. Furthermore, digital tools such as interactive applications help patients enhance their self-management skills, support active participation in rehabilitation and thus facilitate the transition back to normal roles and responsibilities[23]. Moreover, we found that the intervention models focusing on psychological support and health education showed the most pronounced effect on emotional functioning, which is of notable clinical significance. This is particularly relevant given the exceptionally heavy psychological burden borne by patients with pancreatic cancer. The incidence of depression and anxiety is remarkably high, and this vulnerability is further exacerbated by the poor prognosis of pancreatic cancer, rapid symptom progression, and challenging treatment course[31,32]. Standard postoperative care is primarily focused on physical recovery, and often overlooks disordered emotional states. In contrast, psychological interventions and health education specifically address such issues. For instance, mindfulness meditation and biofeedback training are practical tools that allows patients to regulate emotions and enhance a sense of control over anxiety and depressive symptoms[24]. Additionally, the consistent, supportive communication inherent in models such as tailored psychological nursing[26] and interactive e-health support[23] mitigates feelings of isolation and fosters a therapeutic alliance, addressing core relational needs during a vulnerable period.

In summary, a comprehensive supportive care model integrating psychological intervention and health education, by simultaneously addressing physical symptoms, psychological distress, and self-worth, can promote holistic functional recovery. Therefore, psychological and health education interventions are vital for patients with pancreatic cancer undergoing major cancer surgery.

Symptom management

Very low-quality evidence suggests that intervention significantly reduces overall symptom scores. This indicates that psychological interventions and health education comprehensively alleviated distressing physical symptoms during recovery after pancreatic cancer surgery. This integrated effect primarily stems from significant improvements in two key areas: Digestive symptoms and pancreatic pain. The meta-analysis revealed a clear and consistent beneficial effect on symptoms associated with the digestive system, as outcomes were significantly better in the intervention groups. This finding holds considerable clinical importance, as patients frequently face challenges after pancreatectomy such as exocrine insufficiency, early satiety, and fat malabsorption, leading to weight loss, deteriorating nutritional status, and a severely negative impact on QoL. The studies included in our meta-analysis addressed this pathophysiological challenge through structured nutritional education, personalized dietary counseling, and the proactive management of gastrointestinal symptoms. For example, app-based daily symptom tracking enabled early identification and guidance to prevent postoperative nausea and constipation and avoid exacerbating these symptoms[23]. This proactive, educational approach empowers patients to make informed dietary choices, effectively manage enzyme replacement therapy, and adapt to new digestive physiology, thereby mitigating one of the most common postoperative sequelae. Concurrently, pancreatic pain was significantly reduced, but was accompanied by considerable statistical heterogeneity, which is both a limitation and an informative signal. This likely mirrors the diverse and multimodal nature of pain management strategies applied among the studies. Some interventions utilized direct, standardized approaches such as regular pain assessment, structured analgesic guidance, and non-pharmacological methods such as relaxation training and comfortable positioning[23,26]. Others might have achieved pain reduction more indirectly, as a secondary benefit of improved psychological wellbeing and reduced anxiety, which decrease pain perception and catastrophizing[27,28]. This suggests that pain management within supportive care is not a singular approach but rather encompasses physical, psychological, and educational dimensions. The concurrent improvement in nursing satisfaction further reinforced this model, suggesting that structured, attentive, and supportive care fosters a more positive and collaborative patient-caregiver relationship, which in itself can be therapeutic and enhance care delivery. Overall, integrated interventions based on psychological support and health education address the primary sources of postoperative suffering through proactive symptom management. This laying a foundation for recovery and improving overall health status and functioning.

Limitations

Heterogeneity of interventions: Intervention content, intensity, duration, and providers substantially varied among the six studies analyzed herein. This spectrum ranged from comfort nursing, which emphasizes environmental and psychological support[26], to smartphone application-based symptom management[23] and mindfulness meditation combined with biofeedback[24]. These interventions likely targeted different mechanisms, consequently leading to varied effects among different QoL domains, and would explain the elevated statistical heterogeneity in some outcomes. Furthermore, such clinical and methodological diversity introduces significant implementation bias and comparability bias. The effects of the different interventions varied. Therefore, which components genuinely improved QoL was difficult to determine, and also complicated the formulation of specific clinical practice recommendations. Therefore, the pooled effect estimates reflect the average effect of multiple supportive strategies based on psychological intervention and health education.

Variability in intervention intensity and frequency: The intensity of interventions significantly varied among the included studies. For example, some studies required daily symptom reporting[23] or daily meditation sessions[24], while others described the frequency of intervention delivery only in general or unspecified terms. Such differences in intensity directly influenced the dose-effect relationship. These variations introduced bias into the evaluation of effects and prevented the determination of an optimal level of psychological and educational support. Specifically, it remains unclear whether the improvements stem from the core content of the interventions or simply from the frequency of patient-provider contact. This variability in intervention intensity and frequency is a common issue in many supportive care models. Future studies should rigorously document and standardize the intensity of interventions to clarify this relationship.

Differences in control conditions: The definition and specific components of standard care probably varied among the studies and clinical settings. Standard care might already include basic health education and psychological support in some centers[23], or it might be minimal or limited in others[23,26,28]. This variability in the control conditions affects the comparability of study outcomes. This represents a form of selection bias at the study design level, as the baseline standard of care is not equivalent. Consequently, the estimated effect size (SMD/MD) might be underestimated in settings where standard care is already robust, or overestimated where it is minimal.

Inconsistent follow up duration: The analyzed studies assessed the QoL at widely varying time points, ranging from an immediate postoperative hospital stay until one year after surgery. Given that QoL is a dynamic measure that evolves as patients recover, these inconsistent follow up schedules can hinder valid comparisons and obscure the time-sensitive effects of interventions. This introduces measurement bias and attrition bias of a temporal nature. Measured effects early might capture a transient postoperative adjustment, whereas later assessments might reflect adaptation or the emergence of long-term complications.

Additional potential biases: Beyond the inherent limitations of the analyzed studies, our review process might also be biased. Although the funnel plot did not show obvious asymmetry, the dearth of studies (< 10) caused difficulties with judging publication bias. The subjective nature of QoL outcome measures renders them particularly susceptible to detection bias in unblinded studies. Furthermore, the inability to blind most interventions might lead to over-optimistic reporting by patients or assessors.

CONCLUSION

The present study provides a cautiously optimistic view of the conclusion that psychological interventions and health education can improve the QoL in patients after undergoing pancreatic cancer surgery. Whereas the impact of this surgery on QoL is established, evidence supporting specific strategies to improve postoperative QoL, particularly through psychological and educational support, remain limited. Therefore, future investigations should focus on larger, multicenter, methodologically rigorous RCTs. Such trials should apply standardized intervention protocols and incorporate long-term follow up and health economic evaluations to further validate the effects of these interventions, ultimately improving the wellbeing of patients.