Root Cause Analysis–based Risk Management Nursing Protocol For Patients With Gastric Ulcer

Summary

This protocol describes a root cause analysis-based risk management nursing workflow for hospitalized patients with gastric ulcers, including risk identification, causal analysis, individualized intervention, dynamic reassessment, discharge education, and follow-up.

Abstract

Gastric ulcer is a chronic digestive disorder characterized by recurrence, complications, and reduced quality of life. Although pharmacological treatment remains essential, variability in nursing processes, incomplete risk identification, and insufficient follow-up may hinder recovery and long-term disease control. Root cause analysis (RCA) is a structured quality-improvement method used to identify underlying causes of care-related risks and to guide targeted corrective action. This article describes a reproducible RCA-based risk management nursing workflow for hospitalized patients with gastric ulcers. In this retrospective comparative implementation, 240 patients admitted between January 2021 and December 2024 were identified from hospital records and classified, according to the nursing model documented during hospitalization, into a routine-care group and an RCA-based nursing group, with 120 patients in each group. The workflow includes multidisciplinary team formation, identification of representative nursing risk events, timeline reconstruction, root-cause classification, individualized intervention planning, dynamic inpatient reassessment, discharge education, and structured post-discharge follow-up. Outcome assessment procedures include evaluation of nursing effectiveness, complications, recurrence, patient satisfaction, and quality of life using gastroscopy, hospital records, a structured satisfaction assessment, and the Short Form-36 (SF-36). In this retrospective comparative study, patients managed under the RCA-based workflow showed more favorable outcomes than those receiving routine nursing care. This protocol provides an operational framework for integrating RCA into gastric ulcer nursing practice and may support reproducible risk management in other chronic disease care settings.

Introduction

Gastric ulcer is a common digestive disorder with a multifactorial etiology involving Helicobacter pylori infection, abnormal gastric acid secretion, drug-induced mucosal injury, dietary habits, and psychosocial factors¹. With changes in lifestyle, eating patterns, and stress exposure, peptic ulcer disease continues to impose a substantial clinical burden worldwide². Many patients with gastric ulcer experience persistent abdominal pain, poor appetite, and sleep disturbance, and those who develop complications such as bleeding or perforation often show marked impairment in health-related quality of life³. Although pharmacological treatment, including acid suppression and H. pylori eradication, remains central to disease management, the quality, consistency, and continuity of nursing care still influence recovery, recurrence control, and long-term well-being⁴. Conventional nursing models often focus on symptom observation and support for medical treatment, while giving less attention to recurrent care-process risks, patient-specific barriers to adherence, and structured continuity management⁵. As a result, nursing quality may vary across patients, and improvements in quality of life may remain limited⁶. Strengthening nursing quality and promoting sustained recovery in patients with gastric ulcer, therefore, remain important goals in clinical nursing practice.

Root cause analysis (RCA) has been introduced into healthcare as a structured quality-improvement method for examining adverse events, process failures, and recurrent care-related risks⁷. Rather than addressing only the visible outcome of a problem, RCA seeks to identify underlying causal factors across personnel, workflow, communication, education, and management processes⁸. In nursing settings, RCA-based approaches have been used to connect risk analysis with targeted corrective action and practice improvement⁹. Compared with conventional nursing management, an RCA-based approach offers several methodological advantages. It improves risk identification by shifting assessment from surface-level symptom management to cause-oriented analysis of repeated nursing problems. It strengthens intervention design by linking identified causes to specific and operational corrective measures, such as medication guidance, dietary management, psychological support, monitoring, and follow-up. It also improves outcome monitoring because the same risk framework used during causal analysis can be applied during reassessment and continuity management, thereby increasing procedural consistency and reproducibility across stages of care¹⁰.

From a practical perspective, this protocol is most suitable for hospitalized patients with gastric ulcers who require coordinated nursing management across admission, inpatient care, discharge preparation, and early post-discharge follow-up. It is particularly useful in settings where recurrent problems such as poor medication adherence, inadequate diet control, insufficient recognition of warning symptoms, psychological stress, or loss to follow-up are commonly observed. The protocol is also better suited to wards that can support multidisciplinary review, standardized documentation, and follow-up contact after discharge. Its applicability may be more limited in settings with highly incomplete nursing records, insufficient staffing for RCA review, lack of structured follow-up resources, or emergency clinical situations in which immediate stabilization takes priority over process-based nursing reassessment.

Despite the growing use of RCA in healthcare quality management, reproducible descriptions of RCA-based nursing workflows for gastric ulcers remain limited. In addition, when quality of life is evaluated in patients with chronic digestive disease, the Chinese version of the Short Form-36 (SF-36) has demonstrated acceptable validity and reliability in populations including patients with chronic gastritis and peptic ulcer¹¹. Disease-specific patient-reported outcome instruments for peptic ulcer have also shown good psychometric properties and may provide useful support for structured outcome assessment¹². On this basis, the present study describes and evaluates an RCA-based risk management nursing protocol for hospitalized patients with gastric ulcers. By constructing a gastric ulcer-specific nursing risk framework, identifying key risk factors through RCA, and developing individualized management plans, this study aims to provide a more standardized and operational approach to nursing care within a retrospective comparative study framework, with observed outcomes including nursing effectiveness, complications, recurrence, patient satisfaction, and quality of life.

Protocol

All procedures involving human participants were performed in accordance with the ethical standards of the institutional and national research committee and with the Declaration of Helsinki. This retrospective comparative study was approved by the Ethics Committee of Luzhou People’s Hospital (approval no. LLW202601008). De-identified clinical data extracted from hospital records were used for the retrospective review. Written informed consent from patients or their legal representatives was obtained before follow-up contact and questionnaire completion when required by the ethics committee.

1. Identification of eligible patients and definition of study groups

NOTE: The overall study workflow is shown in Figure 1.

1. Define the medical-record review period as January 2021 to December 2024.
   1. Search the hospital electronic medical record system for patients admitted during this period with a documented diagnosis of gastric ulcer.
2. Screening of eligible cases
   1. Confirm gastric ulcer by reviewing the gastroscopy report for each candidate case.
   2. Include cases with a gastroscopy-confirmed diagnosis, complete inpatient nursing records, available baseline clinical information, and follow-up data sufficient for endpoint assessment.
   3. Exclude cases with incomplete key diagnostic records, severe hepatic dysfunction, severe renal dysfunction, severe mental disorders, malignant tumors, or other conditions that prevent reliable comparison of nursing outcomes.
      CAUTION: Exclude cases with missing key source documents rather than inferring eligibility from partial records.
3. Confirmation of analyzable records
   1. Review baseline demographic data, admission nursing assessments, gastroscopy findings, inpatient nursing records, discharge documentation, and follow-up records for each eligible case.
   2. Retain only cases with sufficiently complete records for group classification and endpoint extraction.
4. Classification of the study groups
   1. Review the documented nursing model for each retained case.
   2. Classify cases as the control group when the record documents routine nursing care only.

2. Definition of routine nursing care in the control group

1. Definition of baseline assessment
   1. Extract admission records of vital signs, abdominal symptoms, appetite, sleep, bowel habits, medication history, and prior ulcer-related treatment history.
   2. Extract baseline gastroscopy findings from the endoscopy report or medical record.
2. Definition of inpatient routine care
   1. Extract records of symptom observation, vital-sign monitoring, and physician notification for acute deterioration.
   2. Extract medication administration records, including missed doses, delayed administration, and documented adverse reactions when available.
   3. Extract records of routine medication education, dietary advice, and basic emotional support.
3. Definition of discharge instruction
   1. Extract discharge records covering medication adherence, diet control, warning symptoms, outpatient review timing, smoking cessation advice, and alcohol restriction.
   2. Mark undocumented routine nursing items as undocumented.
      CAUTION: Do not assume completion of a routine nursing step when the corresponding record is absent.

3. Definition of RCA-based risk management nursing in the observation group

1. Definition of the RCA team
   1. Confirm whether the documented RCA process involved a quality-control nurse, senior clinical nurses, a gastroenterologist, and a clinical pharmacist.
   2. Define the team coordinator as the quality-control nurse and define the remaining members as contributors to event review, clinical-risk clarification, and medication-related analysis.
2. Definition of representative nursing risk events
   1. Review nursing notes, shift handover records, medication administration records, physician notification records, discharge education records, and follow-up records for recurrent or clinically meaningful nursing problems.
   2. Select representative events when they show repeated occurrence, direct relevance to complications or recurrence, delayed recognition of warning symptoms, poor adherence affecting treatment continuity, insufficient discharge understanding, or loss to follow-up.
   3. Group target events into the following categories: poor medication adherence, high psychological stress, improper diet, delayed recognition of gastrointestinal bleeding, insufficient discharge understanding, and loss to follow-up.
      NOTE: Select representative events based on recurrence frequency, clinical relevance, and modifiability, rather than solely on severity.
3. Definition of event-timeline reconstruction
   1. Reconstruct the care timeline for each representative event from admission to discharge and follow-up.
   2. Record the sequence of assessment, education, medication administration, symptom reporting, physician notification, discharge instruction, and follow-up contact.
   3. Mark the first point at which omission, delay, misunderstanding, communication failure, incomplete documentation, or interruption of follow-up became evident.
4. Definition of root-cause analysis
   1. Review whether each representative event was discussed in a structured RCA meeting.
   2. Confirm whether cause-oriented questioning was used to identify why the event occurred, why it was not prevented earlier, and which care-process failure allowed it to continue.
   3. Classify documented root causes into one or more of the following domains: knowledge deficit, patient behavior, nurse-patient communication, workflow failure, follow-up failure, or insufficient psychosocial support.
      NOTE: Focus on root-cause classification of modifiable care-process factors rather than on individual blame.
5. Definition of individualized risk assessment
   1. Review whether the patient’s knowledge level, medication adherence risk, dietary management risk, psychological stress, discharge understanding, literacy level, family support, and follow-up reliability were assessed.
   2. Record each domain as low risk, moderate risk, or high risk when such classification is documented or can be directly mapped from the nursing assessment form.
6. Definition of targeted intervention application
   1. Record repeated medication education and adherence reminders when poor medication understanding, missed doses, or delayed administration were identified.
   2. Record individualized diet instruction when improper food intake, uncertainty about trigger foods, or poor dietary compliance was identified.
   3. Record psychological support and repeated communication when anxiety, stress, poor cooperation, or avoidance behavior was identified.
   4. Record teach-back education and family participation when discharge understanding was incomplete, or communication barriers were present.
   5. Record strengthened warning-sign instruction when there was a risk of delayed recognition of bleeding, severe pain, or other acute deterioration.
   6. Record structured follow-up scheduling before discharge when recurrence risk or loss-to-follow-up risk was judged to be moderate or high.
      NOTE: Define intervention adjustment triggers as persistent nonadherence, new warning symptoms, poor discharge understanding, or unreliable follow-up contact.
7. Definition of dynamic reassessment
   1. Review whether medication adherence, dietary compliance, psychological status, discharge understanding, and warning-sign recognition were reassessed during hospitalization and before discharge.
   2. Record whether the nursing plan was adjusted when moderate-risk or high-risk problems persisted after the previous intervention.
8. Definition of discharge management
   1. Review whether reliable contact information was confirmed before discharge.
   2. Review whether discharge instruction covered medication use, diet control, warning symptoms, review timing, and conditions requiring immediate medical attention.
   3. Review whether patient understanding was checked by repeat explanation or teach-back.
9. Definition of post-discharge follow-up
   1. Confirm whether follow-up was performed once every 2 weeks during the first month after discharge and once monthly thereafter until the end of the 6 month follow-up period.
   2. Define telephone contact as the primary follow-up mode and archived mobile-message follow-up as an adjunct mode when written reminders or repeated contact were needed.
   3. Extract follow-up records covering medication adherence, dietary compliance, abdominal symptoms, recurrence-related warning signs, outpatient review completion, and lifestyle management.
   4. Record nursing advice provided after identification of nonadherence, symptom worsening, or misunderstanding of home management.
   5. Record urgent referral advice when melena, hematemesis, syncope, sudden worsening abdominal pain, or other signs of acute deterioration are reported.
      CAUTION: Treat reported bleeding symptoms, syncope, or sudden severe abdominal pain as warning events requiring urgent medical evaluation rather than routine follow-up advice.
10. Definition of follow-up completion
    1. Extract the follow-up date, contact mode, patient response, identified risk problems, nursing advice, and next scheduled contact from each follow-up record.
    2. Record at least two re-contact attempts before classifying a patient as temporarily unreachable.

4. Extraction of study variables

1. Extraction of baseline variables
   1. Extract age, sex, medical history, and other baseline demographic and clinical characteristics from the medical record.
   2. Extract baseline gastroscopy findings and admission nursing assessments from the source documents.
2. Extraction of inpatient nursing variables
   1. Extract records related to symptoms, medication use, diet management, psychological status, education delivery, and discharge instruction during hospitalization.
   2. Extract records of RCA team review, representative event identification, root-cause classification, risk reassessment, and intervention adjustment for the observation group.
3. Extraction of follow-up variables
   1. Extract follow-up records from the hospital archive, telephone follow-up log, and archived mobile-message documentation when available.
   2. Extract post-discharge data on complications, recurrence, quality of life, and nursing satisfaction.

5. Assessment of outcomes

1. Definition of follow-up timing
   1. Define the follow-up period as 6 months after discharge.
   2. Use the latest available follow-up review and gastroscopy record within the predefined window for endpoint assessment.
2. Assessment of ulcer healing
   1. Review baseline and follow-up gastroscopy reports.
   2. Estimate ulcer area as length × width using the maximum length and width recorded in the endoscopy report.
      NOTE: Use the original endoscopy report at each time point, rather than a retrospective visual estimate from stored images, whenever possible.
3. Classification of nursing effectiveness
   1. Classify a case as markedly effective when no ulcer lesion is detected on follow-up gastroscopy.
   2. Classify a case as effective when the estimated ulcer area is reduced by more than 50% relative to baseline.
   3. Classify a case as ineffective when no significant improvement is observed.
4. Recording of complications and recurrence
   1. Record upper gastrointestinal bleeding, perforation, and other predefined complications during follow-up.
   2. Define recurrence as reappearance of a gastric ulcer on follow-up gastroscopy after prior healing, or clear worsening of the original lesion during follow-up, including rehospitalization due to gastric ulcer when supported by medical records.
5. Assessment of quality of life
   1. Administer the SF-36 questionnaire at the end of follow-up under nurse guidance, or extract the completed questionnaire from the archived follow-up record when already available.
   2. Score the eight domains according to the official scoring method and convert each domain to a 0–100 scale.
6. Assessment of nursing satisfaction
   1. Administer the hospital nursing satisfaction questionnaire at the end of follow-up, or extract the completed questionnaire from the archived record when available.
   2. Score the questionnaire across the five domains of service attitude, communication clarity, response timeliness, health education, and continuity support.
   3. Classify responses as satisfied, basically satisfied, or dissatisfied according to the predefined hospital scoring rule.
      NOTE: Interpret nursing satisfaction results as supportive outcome data rather than as a stand-alone externally validated patient-reported measure.

6. Performance of statistical analysis

1. Preparation of the dataset
   1. Enter all extracted data into an electronic data file.
   2. Check all entries twice for completeness, internal consistency, and transcription accuracy.
   3. Code categorical variables numerically and define missing values explicitly before analysis.
2. Definition of the analysis population
   1. Include all eligible cases with complete baseline grouping information in the descriptive analysis.
   2. Define the final analysis sample for each endpoint according to the availability of the corresponding follow-up data.
      NOTE: Do not describe the analysis as intention-to-treat because this was a retrospective comparative study.
3. Analysis of the data
   1. Import the cleaned dataset into statistical analysis software.
   2. Use two-sided tests for all comparisons.
   3. Review continuous variables for approximate distributional suitability before applying parametric tests.
4. Comparison of continuous variables
   1. Express continuous variables as mean ± standard deviation.
   2. Compare groups using independent-samples t tests when the data approximately follow a normal distribution.
5. Comparison of categorical variables
   1. Express categorical variables as counts and percentages.
   2. Compare groups using chi-square tests, or use Fisher’s exact test when expected cell counts are <5.
6. Definition of significance
   1. Consider differences statistically significant when P < 0.05.
   2. Report very small P values as P < 0.001.

Results

Study-group classification and analysis of the population

The study flow is summarized in Figure 1. A total of 286 candidate cases were initially identified from hospital records during the predefined review period. After screening for gastroscopy-confirmed diagnosis, recording completeness, and applying predefined exclusion criteria, 240 eligible patients were included in the retrospective comparative analysis, with 120 patients in the control group and 120 in the observation group. Follow-up completion and endpoint-specific analysis samples are summarized in Table 1. The detailed eligibility criteria and record-review checklist used for case screening and study-group classification are provided in Supplementary Table 1.

Implementation of the RCA-based nursing workflow

Implementation indicators for the RCA-based nursing workflow are summarized in Table 2. In the observation group, 186 representative nursing risk events were reviewed. The most frequently identified dominant root-cause domains were knowledge deficit, patient behavior, and nurse–patient communication. The most common triggered nursing actions were repeated medication education, individualized diet instruction, teach-back plus family involvement, psychological support or repeated communication, strengthened warning-sign education, and reinforced follow-up scheduling. Dynamic reassessment classified 52 patients as low risk, 44 as moderate risk, and 24 as high risk at the key inpatient reassessment stage. The worksheet used for representative event review and root-cause recording is shown in Supplementary Table 2. The dynamic risk reassessment and intervention trigger framework used during hospitalization and before discharge is summarized in Supplementary Table 3.

Follow-up process indicators

Follow-up process indicators are summarized in Table 3. At least one successful post-discharge contact was achieved in all patients in both groups. Completion of the full planned 6 month follow-up schedule was higher in the observation group than in the control group. Temporarily unreachable cases were more frequent in the control group, though all were successfully recontacted within the allowed recontact window. The post-discharge follow-up checklist used for telephone- or messaging-based follow-up is provided in Supplementary Table 4.

Example of RCA-guided targeted nursing action

An example of protocol translation from RCA finding to targeted nursing action is shown in Figure 2. In a representative event category involving poor medication adherence after discharge, RCA review identified incomplete understanding of medication purpose, insufficient reinforcement during discharge teaching, and weak follow-up reliability as the main modifiable contributors. These findings were translated into repeated medication education, teach-back confirmation, family-assisted reminders, and reinforced early follow-up scheduling.

Comparison of nursing effectiveness between groups

Nursing effectiveness outcomes are summarized in Table 4. At the end of the 6 month follow-up period, the observation group showed a higher total effective rate than the control group (91.67% vs. 64.17%), and the between-group difference was statistically significant (χ² = 26.37, P < 0.001).

Comparison of complication and recurrence rates between groups

Complication and recurrence outcomes are summarized in Table 5. During the 6 month follow-up period, the observation group had a lower complication rate than the control group (10.83% vs. 26.67%, P < 0.001). The recurrence rate was also lower in the observation group than in the control group (3.33% vs. 12.50%, P = 0.009).

Comparison of quality of life between groups

Quality-of-life outcomes are summarized in Table 6. At baseline, no statistically significant between-group differences were observed across the eight SF-36 domains (all P > 0.05). At the end of the 6 month follow-up period, both groups showed higher SF-36 scores than at baseline. Follow-up scores in the observation group were higher than those in the control group across physical functioning, role-physical, bodily pain, general health, vitality, social functioning, role-emotional, and mental health (all P < 0.001).

Comparison of nursing satisfaction between group

Nursing satisfaction outcomes are summarized in Table 7. At the end of follow-up, the observation group showed a higher total satisfaction rate than the control group (90.00% vs. 66.67%), and the between-group difference was statistically significant (χ² = 19.25, P < 0.001). The framework and scoring rule of the nursing satisfaction questionnaire used at follow-up assessment are presented in Supplementary Table 5.

Linking protocol stages to measured outputs

The correspondence between major protocol stages and measured outputs is summarized in Table 8. Patient screening and study-group classification determined the analyzable cohort. RCA event review, root-cause classification, reassessment, and targeted intervention generated implementation indicators. Follow-up completion reflected continuity-management fidelity. Gastroscopy-based records of healing, complications, and recurrences, SF-36 scores, and nursing satisfaction scores served as the main comparative outcome measures.

Figure 1: Workflow of the retrospective comparative study of the root cause analysis (RCA)-based nursing protocol for patients with gastric ulcer. Eligible patients with gastric ulcers were identified from hospital records during the predefined study period and were classified into the control or observation group according to the nursing model documented during hospitalization. Baseline data, inpatient nursing records, follow-up records, implementation indicators, and comparative outcomes were extracted for retrospective analysis. Please click here to view a larger version of this figure.

Figure 2: Example of translation from RCA finding to targeted nursing action. The figure illustrates how a representative nursing risk event was reviewed through timeline reconstruction and root-cause classification, and how the identified modifiable contributors were translated into specific corrective nursing actions, dynamic reassessment, and follow-up management. Please click here to view a larger version of this figure.

Item	Observation group	Control group	Total
Candidate cases screened from hospital records	138	148	286
Excluded after eligibility review	18	28	46
Included in retrospective comparative analysis	120	120	240
Completed primary clinical endpoint assessment at 6 months	120	120	240
Completed SF-36 follow-up assessment	117	115	232
Completed nursing satisfaction assessment	118	116	234
Temporarily unreachable during at least one scheduled contact	8	19	27
Successfully reached after re-contact attempts	8	19	27
Persistently unreachable after two re-contact attempts	0	0	0
Completed full planned follow-up schedule	112	98	210

Table 1: Study-group classification, follow-up completion, and analysis population. The table summarizes the number of candidate cases identified from hospital records, excluded cases after eligibility review, included cases in the retrospective comparative analysis, completed primary endpoint assessments, completed questionnaire-based follow-up assessments, temporarily unreachable cases, successfully re-contacted cases, and final analysis samples.

Indicator	Count
Patients in observation group	120
Representative nursing risk events reviewed	186
RCA meetings completed	24
Mean representative events reviewed per meeting	7.8
Root-cause categories identified: knowledge deficit	58
Root-cause categories identified: patient behavior	41
Root-cause categories identified: nurse–patient communication	29
Root-cause categories identified: workflow failure	24
Root-cause categories identified: follow-up failure	19
Root-cause categories identified: insufficient psychosocial support	15
Triggered intervention: repeated medication education	63
Triggered intervention: individualized diet instruction	49
Triggered intervention: teach-back plus family involvement	37
Triggered intervention: psychological support / repeated communication	32
Triggered intervention: strengthened warning-sign education	28
Triggered intervention: reinforced follow-up scheduling	26
Low-risk classification at key inpatient reassessment	52
Moderate-risk classification at key inpatient reassessment	44
High-risk classification at key inpatient reassessment	24

Table 2: Implementation indicators of the RCA-based nursing workflow in the observation group. The table presents the number of representative nursing risk events reviewed, the number of RCA meetings completed, the dominant root-cause categories identified, the frequencies of triggered nursing interventions, and the distribution of low-, moderate-, and high-risk classifications at key reassessment points.

Indicator	Observation group, n (%)	Control group, n (%)
At least one successful post-discharge contact	120 (100.00)	120 (100.00)
Completed follow-up during first month	116 (96.67)	108 (90.00)
Completed full planned 6-month follow-up schedule	112 (93.33)	98 (81.67)
Temporarily unreachable at least once	8 (6.67)	19 (15.83)
Re-contact successful after temporary loss	8 (100.00 of unreachable)	19 (100.00 of unreachable)
Required reinforced follow-up scheduling	26 (21.67)	11 (9.17)
Reported nonadherence during follow-up and received targeted nursing advice	31 (25.83)	44 (36.67)
Reported warning symptoms during follow-up and were advised urgent clinical review	7 (5.83)	14 (11.67)

Table 3: Follow-up process indicators in the control and observation groups. The table summarizes the proportions of patients who completed at least one successful post-discharge contact, completed early follow-up, adhered to the full planned 6-month follow-up schedule, experienced a temporary loss of contact, received successful re-contact, required reinforced follow-up scheduling, reported nonadherence, and reported warning symptoms requiring urgent clinical review.

Group	Markedly effective (n, %)	Effective (n, %)	Ineffective (n, %)	Total effective rate (n, %)
Observation group	80 (66.67)	30 (25.00)	10 (8.33)	110 (91.67)
Control group	58 (48.33)	19 (15.83)	43 (35.83)	77 (64.17)

Table 4: Comparison of nursing effectiveness between the observation and control groups. The table shows the distribution of cases in the observation and control groups as markedly effective, effective, or ineffective, along with a comparison of the total effective rate.

Group	Complications, n (%)	Recurrence, n (%)
Observation group	13 (10.83)	4 (3.33)
Control group	32 (26.67)	15 (12.50)

Table 5: Comparison of complication and recurrence outcomes between the observation and control groups. The table presents the numbers and percentages of patients with complications and recurrence during the 6 month follow-up period in the observation group and control group.

Domain	Control Group at Baseline	Observation Group at Baseline	P Value (Baseline)	Control Group at 6 Months	Observation Group at 6 Months	P Value (at 6 Months)
Physical functioning	61.24 ± 8.31	61.78 ± 8.05	>0.05	72.15 ± 7.44	81.26 ± 6.83	<0.001
Role-physical	58.37 ± 9.12	58.96 ± 8.87	>0.05	69.83 ± 8.25	80.42 ± 7.14	<0.001
Bodily pain	56.48 ± 8.76	56.92 ± 8.41	>0.05	68.57 ± 7.62	78.39 ± 6.95	<0.001
General health	57.62 ± 7.95	58.11 ± 8.02	>0.05	69.41 ± 7.26	79.84 ± 6.57	<0.001
Vitality	55.83 ± 8.14	56.29 ± 8.07	>0.05	67.92 ± 7.38	78.76 ± 6.64	<0.001
Social functioning	57.15 ± 8.63	57.68 ± 8.28	>0.05	69.77 ± 7.71	80.33 ± 6.88	<0.001
Role-emotional	58.04 ± 8.58	58.49 ± 8.34	>0.05	70.26 ± 7.92	81.11 ± 6.97	<0.001
Mental health	56.91 ± 8.02	57.33 ± 7.88	>0.05	68.84 ± 7.41	79.52 ± 6.72	<0.001

Table 6: Comparison of SF-36 scores between groups at baseline and at 6 month follow-up. The table presents SF-36 domain scores for physical functioning, role-physical, bodily pain, general health, vitality, social functioning, role-emotional, and mental health for the observation and control groups at baseline and at the end of the 6 month follow-up period. Data are presented as mean ± standard deviation.

Group	Satisfied (n, %)	Basically Satisfied (n, %)	Dissatisfied (n, %)	Total Satisfaction Rate (n, %)
Observation group	78 (65.00)	30 (25.00)	12 (10.00)	108 (90.00)
Control group	52 (43.33)	28 (23.33)	40 (33.33)	80 (66.67)

Table 7: Comparison of nursing satisfaction between the observation and control groups. The table shows the numbers and percentages of patients who were satisfied, basically satisfied, or dissatisfied with nursing care in the observation group and control group, together with a comparison of the total satisfaction rate.

Protocol stage	Main documented output	Table / Figure location
Eligibility screening and study-group classification	Candidate cases, excluded cases, included cases, analyzable population	Table 1; Figure 1
RCA representative event review	Number of risk events reviewed	Table 2
Root-cause classification	Distribution of dominant root-cause domains	Table 2
Dynamic risk reassessment	Low-, moderate-, and high-risk distribution	Table 2
Targeted intervention triggering	Frequencies of corrective nursing actions	Table 2
Post-discharge follow-up implementation	Contact completion, schedule adherence, temporarily unreachable cases	Table 3
RCA-guided protocol translation in practice	Example of root cause → corrective action	Figure 2
Gastroscopy-based healing evaluation	Nursing effectiveness categories	Table 4
Complication and recurrence monitoring	Follow-up complication and recurrence rates	Table 5
Patient-reported quality of life	SF-36 domain scores	Table 6
Nursing experience evaluation	Satisfaction category distribution	Table 7

Table 8: Mapping of major protocol stages to implementation indicators and outcome measures. The table links the main stages of the protocol, including eligibility screening, RCA event review, root-cause classification, dynamic reassessment, targeted intervention, follow-up implementation, and final endpoint assessment, to their corresponding documented outputs and measurable outcomes.

Supplementary Table 1: Eligibility criteria and record-review checklist used for case screening and study-group classification. The table includes the main screening items for study-period eligibility, gastroscopy-confirmed diagnosis, completeness of key clinical records, major exclusion criteria, and final eligibility for inclusion in the retrospective comparative analysis.Please click here to download this file.

Supplementary Table 2: RCA event review worksheet used for recording representative nursing risk events and root-cause findings. The worksheet includes event category, timeline reconstruction, deviation points, immediate and underlying causes, root-cause category, and proposed corrective actions.Please click here to download this file.

Supplementary Table 3: Dynamic risk reassessment and intervention trigger sheet used during hospitalization and before discharge. The table defines the main reassessment domains, risk levels, and corresponding nursing actions for medication adherence, dietary compliance, psychological status, discharge understanding, and follow-up reliability.Please click here to download this file.

Supplementary Table 4: Post-discharge follow-up checklist used for telephone or messaging-based follow-up. The checklist was used to record medication adherence, dietary compliance, abdominal symptoms, warning signs, lifestyle status, outpatient nt review completion, nursing advice, and the next scheduled follow-up contact. Please click here to download this file.

Supplementary Table 5: Nursing satisfaction questionnaire framework and scoring rule used at follow-up assessment. The table presents the five satisfaction domains, domain score ranges, total score rule, and the category definitions for satisfied, basically satisfied, and dissatisfied responses.Please click here to download this file.

Discussion

This study evaluated a root cause analysis (RCA)-based risk management nursing protocol for hospitalized patients with gastric ulcers and compared its documented clinical performance with that of routine nursing care. Under the conditions of this retrospective comparative study, the RCA-based workflow was associated with higher nursing effectiveness, lower complication and recurrence rates, better SF-36 scores, and higher nursing satisfaction. These findings support the practical value of organizing gastric ulcer nursing care as a structured process rather than as a series of loosely connected routine tasks. In this protocol, RCA was used not simply as an audit tool, but as an operational framework linking representative risk-event identification, cause-oriented review, targeted correction, reassessment, and continuity management. More importantly, the apparent strength of the protocol lies not simply in adding more nursing tasks, but in reorganizing nursing care into a structured sequence of risk identification, causal analysis, targeted intervention, dynamic reassessment, and follow-up management¹³.

Several steps appear to be particularly important for the successful implementation of this protocol. Representative risk events should be collected systematically rather than selected casually, because the quality of RCA depends on whether recurrent and clinically meaningful nursing problems are captured at the start of the process¹⁴. Timeline reconstruction and repeated “why” questioning are also essential for distinguishing immediate causes from modifiable underlying causes. Without this step, nursing teams may focus only on visible patient behaviors, such as missed medication doses or poor dietary adherence, while overlooking contributing factors such as incomplete education, unclear communication, weak discharge preparation, or fragile follow-up systems¹⁵. Individualized intervention planning and dynamic reassessment are equally important because gastric ulcer management is influenced by multiple changing factors, including symptom burden, adherence behavior, diet, and psychological stress¹⁶. In practice, the protocol worked best when these steps were treated as interrelated decisions rather than isolated documentation tasks. Once a risk pattern was identified, the nursing response had to move quickly from description to action, and then from action to reassessment.

The protocol also differs from standard nursing management in several concrete ways. Routine nursing care typically emphasizes symptom monitoring, medication administration, general health education, and discharge instructions. By contrast, the RCA-based protocol adds a structured review layer that asks why recurrent problems occurred, where the care pathway first deviated, and which modifiable factors should be prioritized for correction. This difference is methodologically important. In standard practice, nonadherence or poor diet control may be recorded as patient problems. In the RCA-based workflow, the same issues are further examined as process-linked events that may involve communication gaps, incomplete teach-back, insufficient family engagement, or unreliable follow-up. This shift makes intervention design more specific and makes continuity management more consistent across hospitalization and follow-up. The observed patterns in SF-36 and nursing satisfaction are consistent with this interpretation. The RCA-based workflow may have improved not only clinical management but also the overall care experience, possibly because more structured medication guidance, diet education, emotional support, and continuity management addressed factors that directly shape quality of life and patient perception of care^17,18.

Several common implementation problems should nevertheless be recognized. One practical difficulty is that representative risk events may initially be described too broadly, which can weaken causal analysis and lead to generic corrective actions. This problem can be reduced by clearly defining event categories and by reconstructing the timeline around a specific deviation point rather than a vague overall outcome. A second challenge is that RCA meetings can drift toward blame attribution or repetitive discussion when team roles are unclear. In practice, this can be addressed by using a designated coordinator, keeping the discussion focused on modifiable process factors, and requiring each reviewed event to end with at least one actionable nursing response. A third difficulty is implementation burden. Compared with routine nursing, the RCA-based workflow requires additional time for event review, reassessment, and follow-up documentation. However, this burden is partly offset when the workflow is standardized, as repeated event categories, predefined root-cause domains, and structured follow-up checklists reduce duplication and improve the protocol's usability in day-to-day nursing work. In this sense, the protocol may increase front-end organizational effort while improving downstream workflow clarity and corrective efficiency.

Several limitations should be considered when interpreting these findings. First, this was a single-center retrospective comparative study with a moderate sample size, which may limit the generalizability of the results. Second, the follow-up period was limited to 6 months, and longer observation is needed to determine whether the observed reduction in recurrence is sustained over time. Third, although the protocol was structured, some observed benefit may reflect increased attention and more frequent contact in the observation group in addition to the RCA framework itself¹⁹. Fourth, nursing satisfaction was assessed using a structured hospital-based questionnaire developed for internal clinical evaluation. Although the questionnaire content was reviewed and refined for routine use, formal external psychometric validation was not performed, which may limit comparability with other studies. Finally, the study was conducted within one institutional workflow, and local staffing patterns, communication practices, follow-up resources, and documentation quality may influence implementation in other settings.

Despite these limitations, the protocol remains broadly relevant. The main transferable value of this approach lies in its structured sequence of risk-event identification, root-cause clarification, targeted response, and continuity management. Its usability is likely to be greatest in clinical settings where recurrence prevention depends heavily on patient education, adherence support, symptom monitoring, and post-discharge contact. In such settings, the protocol offers a workable balance between standardization and individualization: the RCA structure provides a stable framework, while the triggered interventions remain adaptable to patient-specific risks. For this reason, the workflow may be useful not only in gastric ulcer care but also in other chronic digestive disorders and clinical nursing settings in which outcomes depend on adherence, symptom monitoring, health education, and post-discharge follow-up²⁰. Future work may further refine the protocol by integrating RCA with evidence-based nursing pathways, digital follow-up systems, and disease-specific risk-stratification tools. Overall, the present study indicates that RCA-based risk management in nursing may serve as a practical, potentially scalable framework for improving the quality and consistency of care for patients with gastric ulcers.

Materials

List of materials used in this article

Name	Company	Catalog Number	Comments
Data abstraction form	Self-developed in this study	N/A	Used for retrospective extraction of baseline, inpatient, and follow-up data
Dynamic risk reassessment sheet	Self-developed in this study	Supplementary Table S3	Used to define reassessment domains and intervention triggers in the RCA-based workflow
Eligibility screening checklist	Self-developed in this study	Supplementary Table S1	Used for retrospective case screening and eligibility confirmation
Endoscopy video processor	Olympus	CV-190	Compatible video processor for endoscopic examination
Gastric ulcer education manual	Self-developed in this study	N/A	Used for discharge education and teach-back support in the RCA-based nursing workflow
Health-related quality-of-life instrument	QualityMetric	SF-36v2 Health Survey	Used for 8-domain quality-of-life assessment
Messaging platform for follow-up communication	Tencent	Weixin / WeChat	Used as an adjunct communication tool for post-discharge follow-up when applicable
Nursing satisfaction questionnaire	Self-developed in this study	Supplementary Table S5	Structured 5-domain, 100-point instrument used for hospital-based satisfaction assessment
Post-discharge follow-up checklist	Self-developed in this study	Supplementary Table S4	Used to standardize telephone or messaging-based follow-up review
RCA event review worksheet	Self-developed in this study	Supplementary Table S2	Used for event logging and root-cause review
Statistical analysis software	IBM	SPSS Statistics 25.0	Used for statistical analysis
Video gastroscope	Olympus	GIF-HQ190	Used for diagnostic and follow-up upper gastrointestinal endoscopy