Association of Optimized Nursing Care with Clinical and Patient-Reported Outcomes in Maintenance Hemodialysis Patients: A Retrospective Before-After Study Authors: Hongmei Zhao, Changgeng Liu, Lin Song, Weihui Wang Published: 2026 ABSTRACT AbstractObjectiveTo evaluate the association of optimized nursing care with clinical and patient-reported outcomes in maintenance hemodialysis patients.MethodsThis single-center retrospective before-after study included 112 maintenance hemodialysis patients treated between January 2022 and December 2024. Patients in the conventional care period (January 2022-June 2023; n=53) received standard hemodialysis nursing care, whereas patients in the optimized care period (July 2023-December 2024; n=59) received structured optimized nursing care incorporating risk stratification, standardized complication prevention, health education, self-management support, and psychosocial screening. Outcomes included dialysis adequacy, complications, hospitalization, laboratory indicators, quality of life, psychological status, patient satisfaction, and nursing care quality scores.ResultsAfter adjustment for age, sex, BMI, dialysis vintage, comorbidities, and vascular access type, the optimized care period was associated with higher Kt/V (β=0.102, 95% CI: 0.042–0.161, P=0.001), higher KDQOL-SF scores (β=8.980, 95% CI: 5.271–12.689, P<0.001), lower hospitalization frequency (IRR=0.654, 95% CI: 0.465–0.921, P=0.015), lower HADS-Total scores (β=−4.036, 95% CI: −5.188 to −2.884, P<0.001), and higher PSQ-18 scores (β=8.006, 95% CI: 4.795–11.217, P<0.001). The adjusted association with overall complication occurrence was not statistically significant (OR=0.443, 95% CI: 0.106–1.857, P=0.265). Nursing care quality scores were moderately correlated with KDQOL-SF (r=0.448, P<0.001) and weakly correlated with Kt/V (r=0.271, P=0.004).ConclusionIn this single-center retrospective before-after study, optimized nursing care was associated with better dialysis adequacy, quality of life, psychological status, patient satisfaction, and hospitalization outcomes. Given the observational design and possible temporal confounding, prospective multicenter studies are needed to confirm these findings.Keywords: hemodialysis, nursing care quality, patient outcomes, dialysis adequacy, quality of life, retrospective study Introduction Chronic kidney disease represents a growing global health challenge, emerging as the ninth leading cause of death globally.1 As the disease progresses to end-stage renal disease (ESRD), hemodialysis becomes the predominant modality of renal replacement therapy, accounting for approximately 74.0% of ESRD patients in many countries.2 Hemodialysis care is resource-intensive and requires specialized nursing support not only for safe treatment delivery, but also for symptom monitoring, self-management, treatment adherence, and psychosocial support. Recent evidence indicates that registered nurse staffing levels and nursing processes are associated with patient safety, treatment adequacy, complication prevention, and patient-reported outcomes.3,4 These findings are consistent with the growing emphasis on patient-centered dialysis care, in which life participation, symptom burden, satisfaction, and individual preferences are considered important dimensions of care quality.5,6 Current nursing care models in hemodialysis settings exhibit considerable variability in practice standards and implementation protocols. Conventional care often focuses on routine vital sign monitoring, standard dialysis procedures, and problem-based education. In contrast, optimized nursing care models usually integrate risk stratification, standardized prevention bundles, structured education, adherence support, and psychological assessment. Recent studies suggest that such nursing interventions may be associated with better self-management, fluid and dietary adherence, quality of life, and satisfaction among hemodialysis patients.7,8 Despite growing recognition of nursing care quality as a determinant of hemodialysis outcomes, several limitations persist in current evidence. Nursing interventions differ substantially across institutions, making it difficult to compare or reproduce care models. Although haemodialysis nursing-sensitive quality indicators have been developed in Delphi-based studies, their external validation and routine use across different dialysis settings remain limited.9,10 In addition, many studies focus mainly on clinical indicators, whereas fewer studies jointly evaluate patient-reported outcomes such as quality of life, satisfaction, psychological distress, and adherence-related support.6,11,12 Real-world studies that quantify nursing care quality and relate it to both clinical and patient-centered outcomes may therefore provide useful evidence for quality improvement in dialysis units. This study aimed to evaluate the association of optimized nursing care with clinical and patient-reported outcomes in maintenance hemodialysis patients. The before-after design was based on a real-world change in nursing practice at our center, where optimized nursing care was implemented after the conventional care period. We compared dialysis adequacy, complication occurrence, quality of life, hospitalization, psychological status, satisfaction, and nursing care quality scores between the two periods, while acknowledging that this design supports association rather than causal inference. Materials and Methods Study Design and Population This single-center retrospective before-after study was conducted at the Hospital from January 2022 to December 2024. The study protocol adhered to the principles of the Declaration of Helsinki and was approved by the hospital’s institutional ethics committee. Inclusion Criteria. Patients were eligible for inclusion if they: (1) were aged 18 years or older; (2) had been receiving regular hemodialysis for at least 3 months; (3) underwent hemodialysis 2–3 times per week; and (4) had complete clinical records available for analysis. Exclusion Criteria. Patients were excluded if they: (1) had concurrent malignancy; (2) had severe infection or immunological disorders; (3) had an estimated life expectancy of less than 6 months; or (4) had incomplete follow-up data. A total of 112 patients who met the eligibility criteria were included and divided into two groups based on the period of nursing care received: the conventional care group (n=53, January 2022 to June 2023) and the optimized care group (n=59, July 2023 to December 2024). The two groups represented sequential care periods rather than concurrent cohorts. Nursing Intervention Protocols Conventional Care Group Patients in the conventional group received standard hemodialysis nursing care according to institutional protocols, including: (1) Routine vital sign measurement before each dialysis session; (2) Standard dialysis performed according to institutional protocols, with vital signs monitored every 30–60 minutes during the session; (3) Routine vascular access management, including assessment of pulsation and murmur prior to arteriovenous fistula (AVF) or arteriovenous graft (AVG) cannulation, standard cannulation techniques, and post-dialysis hemostasis; (4) Providing verbal education on fundamental dialysis treatment principles, dietary recommendations (low-sodium, low-potassium, low-phosphorus diet), and medication adherence reminders; (5) Routinely monitoring for common dialysis-related complications, including observing symptoms of hypotension (dizziness, nausea, muscle cramps), and implementing standard interventions as needed; (6) Ensuring patients attend regular follow-up appointments according to standard clinical practice and completing post-treatment care. Optimized Care Group Patients in this group received enhanced nursing care incorporating the following components, and the key distinguishing elements from conventional care are summarized in Supplementary Table S1. Before implementation, nurses received standardized training on risk stratification, intradialytic hypotension (IDH) prevention, vascular access care, infection control, health education, and psychosocial screening. Protocol implementation was monitored through routine nursing documentation, monthly nursing quality reviews, and feedback from nursing supervisors. (1) Systematic Assessment and Risk Stratification. Beyond routine vital signs, assessment included evaluation of interdialytic weight gain (IDWG), volume status indicators, and psychosocial status. Patients were stratified according to risk for IDH, malnutrition, and cardiovascular complications based on established risk factors. Nutritional risk was assessed using serum albumin levels and clinical evaluation quarterly. (2) Standardized Operational Protocols. An evidence-based IDH prevention bundle was implemented based on European Best Practice Guidelines (EBPG) recommendations, including the use of cool dialysate temperature (35.5–36.0°C) for high-risk patients, ultrafiltration rate limitation (≤10 mL/kg/hour), sodium profiling when indicated, and optimization of antihypertensive medication timing. Enhanced vascular access management followed Kidney Disease Outcomes Quality Initiative (K/DOQI) guidelines, incorporating systematic pre-cannulation assessment, rope-ladder or buttonhole cannulation technique, patient education on upper limb exercises to promote AVF patency, and standardized central venous catheter care bundle with chlorhexidine-based antiseptic protocols.13 (3) Enhanced Health Education and Self-Management Support. A structured health education program was delivered through scheduled sessions (minimum twice monthly) using standardized educational materials. Education content encompassed disease knowledge, individualized dietary counseling (protein intake 1.0–1.2 g/kg/day, sodium restriction <2000 mg/day, potassium and phosphorus management), fluid management strategies with daily self-monitoring (IDWG target <4% of dry weight), medication education and adherence-related problem solving.14 Psychological assessment was incorporated into routine care, with supportive counseling and referral provided for patients with identified distress. (4) Complication Prevention. Enhanced infection control measures were implemented, including strict hand hygiene monitoring, standardized aseptic technique for vascular access procedures, and surveillance for catheter-related bloodstream infections. Cardiovascular risk factors were systematically addressed through blood pressure management, anemia control, and mineral bone disorder monitoring. Outcome Measures Primary Outcomes Dialysis Adequacy Kt/V (urea clearance index) was calculated using the second-generation Daugirdas formula based on pre- and post-dialysis blood urea nitrogen levels, treatment time, and ultrafiltration volume.15 A Kt/V ≥1.2 was considered adequate dialysis according to KDOQI guidelines.16 Complication Rates Complications were extracted from dialysis and medical records. Intradialytic hypotension was defined as a symptomatic blood pressure decrease requiring nursing or medical intervention during dialysis, as documented in dialysis records. Infectious complications included catheter-related bloodstream infection and arteriovenous fistula infection documented in the medical record. Vascular access complications included access dysfunction, thrombosis, or bleeding requiring additional intervention. Cardiovascular events included acute coronary syndrome, heart failure exacerbation, clinically significant arrhythmia, or stroke. For the main regression analysis, overall complication occurrence was analyzed at the patient level, defined as at least one of the prespecified complications during the corresponding observation period. Quality of Life Health-related quality of life was assessed using KDQOL-SF, a disease-specific instrument validated for patients with chronic kidney disease.17 The KDQOL-SF comprises 36 generic items from the Short Form-36 (SF-36) and 43 kidney disease-specific items, covering physical and mental health domains, symptom burden, and effects of kidney disease on daily life. The instrument has demonstrated excellent internal consistency reliability (Cronbach’s α = 0.73–0.95 for subscales) and construct validity in diverse hemodialysis populations.18 Secondary Outcomes Laboratory Parameters such as hemoglobin, serum albumin, serum calcium, phosphorus, intact parathyroid hormone (iPTH), and C-reactive protein (CRP) levels were measured using standard laboratory methods. For analysis, values closest to the end of each patient’s observation period and obtained under clinically stable conditions were used. Nutritional status was evaluated using serum albumin and hemoglobin levels as surrogate markers, which have been validated as reliable indicators of nutritional status in hemodialysis patients.19 Unplanned hospitalization rates and length of hospital stay were recorded during the study period. A composite nursing care quality score (0–100 points) was developed by the research team based on expert consensus and previous literature on nursing quality indicators in hemodialysis settings.3,20 The scoring system comprised five domains: operational compliance (30 points, including aseptic technique, vascular access care, and dialysis parameter settings), health education completion (20 points, including frequency, content coverage, and patient comprehension), complication prevention measures (25 points, including monitoring frequency and intervention implementation rate), nursing documentation completeness (15 points), and patient adherence management (10 points). Quality assessments were conducted monthly by trained nursing supervisors. Content validity was established through review by a panel of five nephrology nursing experts (Content Validity Index = 0.92). Internal consistency reliability was assessed in a pilot sample of 30 patients, yielding a Cronbach’s α of 0.81, indicating acceptable reliability.21 Psychological Status Anxiety and depression were assessed using the Hospital Anxiety and Depression Scale (HADS), a 14-item self-report questionnaire specifically designed for patients with physical illness.22 The HADS comprises two subscales: anxiety (HADS-A, 7 items) and depression (HADS-D, 7 items), each scored from 0 to 21. Scores of 0–7 indicate normal status, 8–10 indicate borderline abnormal, and 11–21 indicate abnormal levels of anxiety or depression. HADS has been validated in hemodialysis populations with good reliability (Cronbach’s α = 0.82–0.90).23 Patient Satisfaction: Nursing care satisfaction was evaluated using the Patient Satisfaction Questionnaire-18 (PSQ-18).24 The PSQ-18 assesses seven dimensions of care: general satisfaction, technical quality, interpersonal manner, communication, financial aspects, time spent with provider, and accessibility. Scores were transformed to a 0–100 scale, with higher scores indicating greater satisfaction. KDQOL-SF, HADS, and PSQ-18 assessments were obtained from routine follow-up records at the end of the corresponding care period. Data Collection Data were systematically extracted from the hospital’s electronic medical record system. Clinical information, including demographic characteristics, dialysis parameters, laboratory results, and complication records, was obtained from medical charts. Nursing care documentation provided information on intervention implementation and quality assessment scores. Follow-up data, including quality of life assessments and patient satisfaction surveys, were collected during routine clinical visits. Patients with incomplete key outcome or follow-up data were excluded from the final analysis, and no statistical imputation was performed. Statistical Analysis Statistical analyses were performed using SPSS version 26.0 (IBM Corp., Armonk, NY, USA) and GraphPad Prism version 9.0 (GraphPad Software, San Diego, CA, USA). Continuous variables were tested for normality using the Shapiro–Wilk test. Normally distributed continuous variables were presented as mean ± standard deviation and compared between groups using independent-samples t-tests. Non-normally distributed continuous variables were expressed as median (interquartile range) and analyzed using Mann–Whitney U-tests. Categorical variables were presented as frequencies (percentages) and compared using chi-square tests or Fisher’s exact tests, as appropriate. Because this was a retrospective study, all eligible patients during the prespecified study period were included. No formal a priori sample size calculation was performed, and the sample size may have limited power for low-frequency complication outcomes. Correlation between nursing care quality scores and patient outcomes was assessed using Pearson correlation coefficients. The strength of correlations was interpreted according to Cohen’s guidelines: |r| < 0.3 as weak, 0.3 ≤ |r| < 0.5 as moderate, and |r| ≥ 0.5 as strong.25 Multivariable regression analyses were performed to evaluate the association between nursing care model and patient outcomes while adjusting for potential confounders selected according to clinical relevance and previous literature. Multiple linear regression was used for continuous outcomes (Kt/V, KDQOL-SF, HADS-Total, and PSQ-18 scores). Binary logistic regression was employed to analyze patient-level overall complication occurrence, with results expressed as odds ratios (OR) and 95% confidence intervals (CI). Poisson regression was used to model hospitalization frequency, with results expressed as incidence rate ratios (IRR) and 95% CI. Overdispersion in the hospitalization count model was assessed using the Pearson chi-square statistic divided by the residual degrees of freedom. All multivariable models were adjusted for age, sex, body mass index, dialysis vintage, diabetes mellitus, hypertension, cardiovascular disease, and vascular access type. The quality of life model additionally included serum albumin as a covariate given its established association with health-related quality of life in hemodialysis patients.26 All statistical tests were two-sided, and P-values < 0.05 were considered statistically significant. Results Baseline Demographic and Clinical Characteristics A total of 137 maintenance hemodialysis patients were initially screened during the study period. Twenty-five patients were excluded because of dialysis duration less than 3 months (n=8), incomplete key clinical or follow-up records (n=7), concurrent malignancy, severe infection, or immunological disorders (n=5), or insufficient follow-up for outcome assessment (n=5). Finally, 112 eligible patients were included and divided according to the period of nursing care received: the conventional care group (n=53) and the optimized care group (n=59). As shown in Table 1, no significant differences were observed between the two groups in baseline demographic or clinical characteristics. The mean age was 58.52 ± 12.35 years in the conventional group and 57.89 ± 11.78 years in the optimized group (P > 0.05). Both groups had similar distributions of sex, body mass index, dialysis vintage, primary etiology of end-stage renal disease, vascular access type, and comorbidities (all P > 0.05), indicating comparability at baseline. Table 1Baseline CharacteristicsCharacteristicConventionalGroup (n=53)OptimizedGroup (n=59)t/χ2/ZPDemographic CharacteristicsAge (years)58.49 ± 12.4757.86 ± 11.790.273a0.785Male, n (%)31 (58.49)31 (52.54)0.195b0.659BMI (kg/m2)22.95 ± 3.2423.15 ± 3.340.309a0.758Dialysis vintage (months)38.00 (19.00, 59.00)36.00 (18.50, 71.50)0.213c0.834Primary etiology, n (%)0.873b0.832Diabetic nephropathy18 (33.96)17 (28.81)Hypertensive nephrosclerosis13 (24.53)18 (30.51)Chronic glomerulonephritis14 (26.42)17 (28.81)Others8 (15.09)7 (11.86)Vascular access type, n (%)0.288b0.592Arteriovenous fistula39 (73.58)47 (79.66)Central venous catheter14 (26.42)12 (20.34)Comorbidities, n (%)Hypertension44 (83.02)47 (79.66)0.045b0.832Diabetes mellitus22 (41.51)18 (30.51)1.032b0.310Cardiovascular disease17 (32.08)14 (23.73)0.599b0.439Notes: Data are presented as mean ± standard deviation, median (interquartile range), or n (%).a, Independent-samples t-test; b, Chi-square test; c, Mann–Whitney U-test (Z value reported).Abbreviation: BMI, body mass index. Baseline Characteristics Notes: Data are presented as mean ± standard deviation, median (interquartile range), or n (%).a, Independent-samples t-test; b, Chi-square test; c, Mann–Whitney U-test (Z value reported). Abbreviation: BMI, body mass index. Comparison of Dialysis Adequacy and Laboratory Parameters Patients in the optimized care period had higher Kt/V values than those in the conventional care period (1.27 ± 0.18 vs 1.18 ± 0.17, P = 0.008), suggesting better dialysis adequacy (Table 2). With respect to mineral metabolism, serum phosphorus levels were lower in the optimized care group (1.68 ± 0.35 mmol/L vs 1.82 ± 0.38 mmol/L, P = 0.045), whereas serum calcium and iPTH showed no significant between-group differences (P > 0.05). CRP levels were also lower in the optimized care group (median 5.84 mg/L vs 8.56 mg/L, P = 0.029). Table 2Comparison of Dialysis Adequacy and Laboratory Parameters Between Two GroupsParameterConventionalGroup (n=53)OptimizedGroup (n=59)t/ZPDialysis adequacyKt/V1.18 ± 0.171.27 ± 0.182.712a0.008Mineral metabolismSerum calcium (mmol/L)2.16 ± 0.232.19 ± 0.210.722a0.472Serum phosphorus (mmol/L)1.82 ± 0.381.68 ± 0.352.030a0.045iPTH (pg/mL)265.80 (182.45, 408.50)248.25 (175.60, 382.75)0.856b0.392Inflammatory markerCRP (mg/L)8.56 (5.12, 14.68)5.84 (3.28, 10.42)2.185b0.029Notes: Data are presented as mean ± standard deviation or median (interquartile range) a, Independent-samples t-test; b, Mann–Whitney U-test (Z value reported).Abbreviations: Kt/V, urea clearance index; iPTH, intact parathyroid hormone; CRP, C-reactive protein. Comparison of Dialysis Adequacy and Laboratory Parameters Between Two Groups Notes: Data are presented as mean ± standard deviation or median (interquartile range) a, Independent-samples t-test; b, Mann–Whitney U-test (Z value reported). Abbreviations: Kt/V, urea clearance index; iPTH, intact parathyroid hormone; CRP, C-reactive protein. Changes in Clinical Outcomes and Psychological Status At the patient level, intradialytic hypotension occurred in 9 of 53 patients in the conventional care group and 7 of 59 patients in the optimized care group; infectious complications in 4 of 53 and 3 of 59; cardiovascular events in 3 of 53 and 3 of 59; and vascular access complications in 6 of 53 and 5 of 59, respectively (Figure 1). The patient-level adjusted analysis of overall complication occurrence is reported in the multivariable regression section. Comparison of Major Complication Rates Between Two Groups. Data are presented as n/N (%) of patients who experienced each prespecified complication during the corresponding observation period. (A) Intradialytic hypotension, (B) infectious complications, (C) cardiovascular events, and (D) vascular access complications. Patients receiving optimized care experienced significantly better clinical outcomes across multiple domains (Table 3). Unplanned hospitalization frequency was lower than in the conventional care period (0.98 ± 1.09 vs 1.53 ± 1.20 times per year, P = 0.012). Quality of life assessed using the KDQOL-SF was higher in the optimized care group (69.78 ± 9.14 vs 61.88 ± 9.11, P < 0.001). Serum albumin concentrations were also higher (37.36 ± 4.15 g/L vs 35.42 ± 4.28 g/L, P = 0.017), whereas hemoglobin levels did not reach statistical significance (102.68 ± 13.95 g/L vs 97.85 ± 14.62 g/L, P = 0.076). Table 3Hospitalization Rates and Quality of Life Between GroupsOutcome VariableConventionalGroup (n=53)OptimizedGroup (n=59)tPHospitalization times (n/year)1.53 ± 1.200.98 ± 1.092.5420.012Quality of life scoreKDQOL-SF61.88 ± 9.1169.78 ± 9.144.574<0.001Nutritional indicatorsHemoglobin (g/L)97.85 ± 14.62102.68 ± 13.951.7880.076Albumin (g/L)35.42 ± 4.2837.36 ± 4.152.4340.017Notes: Data are presented as mean ± standard deviation.Abbreviation:KDQOL-SF, Kidney Disease Quality of Life Short Form. Hospitalization Rates and Quality of Life Between Groups Notes: Data are presented as mean ± standard deviation. Abbreviation:KDQOL-SF, Kidney Disease Quality of Life Short Form. Patient-reported psychological and satisfaction outcomes favored the optimized care period (Table 4). HADS-Anxiety scores were lower in the optimized care group compared with the conventional care group (7.18 ± 2.87 vs 9.01 ± 3.15, P = 0.002), as were HADS-Depression scores (6.69 ± 2.91 vs 8.76 ± 3.28, P < 0.001). Patient satisfaction measured by the PSQ-18 was higher in the optimized care group (77.83 ± 8.36 vs 69.81 ± 8.39, P < 0.001). Table 4Comparison of Psychological Status and Patient Satisfaction Between GroupsVariableConventionalGroup (n=53)OptimizedGroup (n=59)tPHADS-Anxiety score9.01 ± 3.157.18 ± 2.873.2170.002HADS-Depression score8.76 ± 3.286.69 ± 2.913.539<0.001HADS-Total score17.77 ± 5.8213.87 ± 5.243.732<0.001PSQ-18 score69.81 ± 8.3977.83 ± 8.365.065<0.001Notes: Data are presented as mean ± standard deviation.Abbreviations:HADS, Hospital Anxiety and Depression Scale; PSQ-18, Patient Satisfaction Questionnaire-18. Comparison of Psychological Status and Patient Satisfaction Between Groups Notes: Data are presented as mean ± standard deviation. Abbreviations:HADS, Hospital Anxiety and Depression Scale; PSQ-18, Patient Satisfaction Questionnaire-18. Multivariable Regression Analysis To account for potential confounding factors, multivariable regression analyses were performed to evaluate the independent association between nursing care model and patient outcomes (Table 5). Table 5Multivariable Regression Analysis of the Association Between Nursing Care Model and Patient OutcomesOutcomeModelEffect Size95% CIPKt/VUnadjustedβ = 0.1010.041, 0.1620.001Adjustedβ = 0.1020.042, 0.1610.001KDQOL-SFUnadjustedβ = 7.8994.477, 11.321<0.001Adjustedβ = 8.9805.271, 12.689<0.001Complication occurrenceUnadjustedOR = 0.4460.109, 1.8200.260AdjustedOR = 0.4430.106, 1.8570.265Hospitalization timesUnadjustedIRR = 0.6430.459, 0.9010.010AdjustedIRR = 0.6540.465, 0.9210.015HADS-Total scoreUnadjustedβ = −4.129−5.885, −2.373<0.001Adjustedβ = −4.036−5.188, −2.884<0.001PSQ-18 scoreUnadjustedβ = 8.0234.884, 11.163<0.001Adjustedβ = 8.0064.795, 11.217<0.001Notes: Effect sizes are presented for optimized care group compared to conventional care group. β, regression coefficient; OR, odds ratio; IRR, incidence rate ratio; CI, confidence interval. Adjusted models included age, sex, BMI, dialysis vintage, diabetes mellitus, hypertension, cardiovascular disease, and vascular access type as covariates. Quality of life model additionally adjusted for serum albumin. Multiple linear regression was used for continuous outcomes; binary logistic regression for complication occurrence; Poisson regression for hospitalization times. Multivariable Regression Analysis of the Association Between Nursing Care Model and Patient Outcomes Notes: Effect sizes are presented for optimized care group compared to conventional care group. β, regression coefficient; OR, odds ratio; IRR, incidence rate ratio; CI, confidence interval. Adjusted models included age, sex, BMI, dialysis vintage, diabetes mellitus, hypertension, cardiovascular disease, and vascular access type as covariates. Quality of life model additionally adjusted for serum albumin. Multiple linear regression was used for continuous outcomes; binary logistic regression for complication occurrence; Poisson regression for hospitalization times. After adjusting for age, sex, BMI, dialysis vintage, comorbidities (diabetes mellitus, hypertension, and cardiovascular disease), and vascular access type, the optimized care group remained significantly associated with higher Kt/V values (β = 0.102, 95% CI: 0.042–0.161, P = 0.001) and higher quality of life scores (β = 8.980, 95% CI: 5.271–12.689, P < 0.001). The detailed regression coefficients for all covariates are presented in Tables 6 and 7. Table 6Multiple Linear Regression Analysis of Factors Associated with Kt/VVariableβSE95% CIPIntercept1.0150.1500.717, 1.313<0.001Group (Optimized vs Conventional)0.1020.0300.042, 0.1610.001Age (years)0.0010.001−0.001, 0.0040.250Sex (Male vs Female)0.0530.031−0.009, 0.1150.093BMI (kg/m2)0.0020.005−0.008, 0.0110.750Dialysis vintage (months)0.0010.0000.000, 0.0010.045Diabetes mellitus (Yes vs No)0.0440.033−0.020, 0.1090.178Hypertension (Yes vs No)−0.0420.040−0.121, 0.0370.291Cardiovascular disease (Yes vs No)−0.0430.034−0.109, 0.0240.207Vascular access (AVF vs CVC)0.0190.037−0.054, 0.0920.610Notes: Model R2 = 0.193, Adjusted R2 = 0.122.Abbreviations: β, regression coefficient; SE, standard error; CI, confidence interval; AVF, arteriovenous fistula; CVC, central venous catheter. Multiple Linear Regression Analysis of Factors Associated with Kt/V Notes: Model R2 = 0.193, Adjusted R2 = 0.122. Abbreviations: β, regression coefficient; SE, standard error; CI, confidence interval; AVF, arteriovenous fistula; CVC, central venous catheter. Table 7Multiple Linear Regression Analysis of Factors Associated with KDQOL-SF ScoreVariableβSE95% CIPIntercept78.39711.29355.995, 100.799<0.001Group (Optimized vs Conventional)8.9801.8705.271, 12.689<0.001Age (years)−0.0090.075−0.158, 0.1410.909Sex (Male vs Female)0.1051.857−3.578, 3.7880.955BMI (kg/m2)−0.2140.283−0.775, 0.3480.452Dialysis vintage (months)−0.0120.015−0.043, 0.0190.436Diabetes mellitus (Yes vs No)1.4491.929−2.377, 5.2750.454Hypertension (Yes vs No)−0.2832.371−4.986, 4.4200.905Cardiovascular disease (Yes vs No)0.3771.990−3.571, 4.3250.850Vascular access (AVF vs CVC)−2.5852.197−6.943, 1.7720.242Serum albumin (g/L)−0.2580.223−0.700, 0.1840.250Notes: Model R2 = 0.202, Adjusted R2 = 0.123.Abbreviations: β, regression coefficient; SE, standard error; CI, confidence interval; AVF, arteriovenous fistula; CVC, central venous catheter. Multiple Linear Regression Analysis of Factors Associated with KDQOL-SF Score Notes: Model R2 = 0.202, Adjusted R2 = 0.123. Abbreviations: β, regression coefficient; SE, standard error; CI, confidence interval; AVF, arteriovenous fistula; CVC, central venous catheter. For hospitalization outcomes, Poisson regression analysis showed that the optimized care period was associated with lower hospitalization frequency (IRR = 0.654, 95% CI: 0.465–0.921, P = 0.015) after adjustment for measured covariates. No substantial overdispersion was detected in this model (Pearson chi-square/df = 1.13), so the Poisson regression model was retained. Regarding psychological outcomes, the adjusted analysis demonstrated that optimized care was significantly associated with lower HADS-Total scores (β = −4.036, 95% CI: −5.188 to −2.884, P < 0.001) and higher patient satisfaction scores (β = 8.006, 95% CI: 4.795–11.217, P < 0.001). However, the association between nursing care model and overall complication occurrence did not reach statistical significance in the adjusted model (OR = 0.443, 95% CI: 0.106–1.857, P = 0.265), although the point estimate suggested a protective trend. Relationship Between Nursing Care Quality Score and Clinical Outcomes To further explore the relationship between nursing care quality and key patient-reported and clinical outcomes, correlation analyses were performed between nursing care quality scores and KDQOL-SF and Kt/V (Figure 2). The mean nursing care quality score was higher in the optimized care group than in the conventional care group (84.75 ± 5.11 vs 65.98 ± 4.11, P < 0.001). In exploratory analyses, nursing care quality scores were moderately correlated with KDQOL-SF (r = 0.448, R2 = 0.200, P < 0.001) and weakly correlated with Kt/V values (r = 0.271, R2 = 0.073, P = 0.004). These correlations were consistent with, but do not prove, the adjusted associations observed in the multivariable models. Correlations Between Nursing Care Quality Score and Patient Outcomes. Scatter plots demonstrate the relationship between nursing care quality score and (A) Kt/V values and (B) quality of life scores (KDQOL-SF) in 112 hemodialysis patients. Each point represents an individual patient. Solid lines indicate linear regression fits. (A) A modest but statistically significant positive correlation was observed between nursing care quality score and Kt/V (r = 0.271, R2 = 0.073, P = 0.0039). (B) A moderate positive correlation was observed between nursing care quality score and quality of life (r = 0.448, R2 = 0.200, P < 0.001). Discussion The present study investigated the association between nursing care quality and patient outcomes in maintenance hemodialysis patients. In this single-center retrospective before-after study, the optimized nursing care period was associated with higher Kt/V values, higher KDQOL-SF scores, fewer unplanned hospitalizations, lower psychological distress, and higher patient satisfaction after adjustment for measured confounders. These findings suggest that structured nursing care may be relevant not only to dialysis delivery but also to patient-centered outcomes. However, because the two groups were separated by time period, the findings should be interpreted as associations rather than evidence of causality. The higher Kt/V values observed in the optimized care period are consistent with previous reports indicating that standardized nursing protocols and nursing-led prevention of intradialytic instability may support treatment completion and dialysis efficiency.27,28 In our study, the optimized care group had a higher mean Kt/V than the conventional care group, and this association remained significant in multivariable analysis. The complication findings require cautious interpretation. Intradialytic hypotension, infectious complications, cardiovascular events, and vascular access complications were numerically less frequent in the optimized care period at the patient level, but the adjusted association between care model and overall complication occurrence did not reach statistical significance. Therefore, these results should not be interpreted as showing an independent reduction in complications. The low number of complication events and the modest sample size likely limited statistical power for this outcome. The direction of the observed differences is clinically plausible because intradialytic hypotension is common in hemodialysis patients,29 and the optimized protocol incorporated risk stratification, cool dialysate use for selected high-risk patients, ultrafiltration rate limitation, aseptic technique reinforcement, and standardized vascular access care based on established dialysis nursing and hemodynamic management principles.30,31 Nevertheless, the present data support only a hypothesis-generating trend for complication outcomes. A key patient-centered finding was the association between optimized nursing care and better health-related quality of life. The adjusted difference in KDQOL-SF scores exceeded the commonly used minimally important difference for this instrument,18 suggesting potential clinical relevance. This finding is consistent with the concept that quality of life in dialysis patients is shaped not only by biochemical and dialysis adequacy parameters, but also by symptom burden, communication, education, and the perceived quality of interpersonal care.32 In the current study, patients in the optimized care period also had significantly lower HADS-Total scores and higher PSQ-18 satisfaction scores after multivariable adjustment, further suggesting that optimized nursing care may be associated with benefits extending beyond dialysis adequacy to psychological well-being and care experience. These results are particularly relevant to Patient Preference and Adherence because patient satisfaction, psychological well-being, and perceived support may influence engagement with dialysis treatment, dietary and fluid restrictions, medication use, and self-management behaviors.6,8 The correlation analyses provide additional context for interpreting these findings. Nursing care quality showed a moderate correlation with quality of life but only a weak correlation with Kt/V. The multivariable regression analysis confirmed that the group-level difference in Kt/V remained statistically significant after adjustment, but the modest correlation coefficient and low explained variance indicate that dialysis adequacy remains multifactorial. Kt/V is influenced by dialyzer characteristics, blood flow rates, treatment duration, ultrafiltration, body size, and residual kidney function.16 Nursing care quality is therefore likely to represent one contributor within a broader clinical system, whereas patient-reported outcomes may be more sensitive to education, communication, symptom monitoring, and psychosocial support. Several interconnected mechanisms may explain why optimized nursing care was associated with better outcomes. For dialysis adequacy, real-time monitoring, earlier recognition of intradialytic symptoms, and prompt nursing intervention may have reduced shortened or interrupted dialysis sessions, thereby supporting delivery of the prescribed dialysis dose. For patient-reported outcomes, structured education and adherence-related problem solving may have helped patients better understand fluid control, dietary management, and medication routines.33 Regular psychosocial screening and supportive counseling may also have contributed to lower anxiety and depression scores, consistent with evidence that psychosocial interventions can improve depression, quality of life, and fluid adherence in hemodialysis patients.34 These pathways align with previous observations that symptom burden and psychological distress are major determinants of health-related quality of life in chronic kidney disease and dialysis populations.35 The lower adjusted hospitalization frequency in the optimized care period may reflect the combined influence of closer monitoring, improved self-management support, and earlier response to clinical deterioration. Given the healthcare burden associated with hospitalization in dialysis patients,26,36 this association may have practical implications for nursing quality improvement, although economic value should be assessed directly in future cost-effectiveness studies. Several limitations of this study should be acknowledged. Although we performed multivariable regression analyses to adjust for measured confounders and examined potential secular trends in institutional factors, unmeasured time-varying confounders cannot be completely excluded. The retrospective before-after design means that the two groups were drawn from sequential time periods rather than concurrent cohorts. The single-center setting limits generalizability to other dialysis units, regions, and international practice environments. The sample size was modest and no formal a priori sample size calculation was performed; consequently, the study may have been underpowered for low-frequency outcomes such as complication analyses. Moreover, the locally developed nursing care quality score showed acceptable internal consistency and inter-rater reliability, but it requires external validation before broader use. Patients with incomplete key outcome or follow-up data were excluded and no imputation was performed, which may have introduced selection bias. Future studies should use prospective multicenter designs, ideally with concurrent comparison groups or randomized implementation where feasible, larger samples, repeated outcome assessment, validated nursing quality tools, and formal evaluation of adherence, patient preferences, and cost-effectiveness. Conclusion This single-center retrospective before-after study suggests that optimized nursing care was associated with better dialysis adequacy, quality of life, psychological status, patient satisfaction, and hospitalization outcomes in maintenance hemodialysis patients after adjustment for measured confounders. Although some complication proportions were numerically lower during the optimized care period, the adjusted patient-level association with overall complication occurrence was not statistically significant. These findings support the potential value of structured nursing care for patient-centered hemodialysis management, but prospective multicenter studies are needed to confirm these associations. Funding Statement This research received no specific grant from any funding agency in the public, commercial, or not-for-profit sectors. Data Sharing Statement The datasets used and/or analyzed during the current study are available from the corresponding author on reasonable request. Ethics Approval and Consent to Participate This study was approved by the Ethics Committee of People’s Hospital of Aba Tibetan and Qiang Autonomous Prefecture (Approval No. 2024-037). Informed consent was waived for this retrospective study due to the exclusive use of de-identified patient data, which posed no potential harm or impact on patient care. The procedures were conducted in accordance with the ethical standards set forth by the Committee on Human Experimentation and the Helsinki Declaration of 1975, as revised in 2000. Author Contributions All authors made a significant contribution to the work reported, whether that is in the conception, study design, execution, acquisition of data, analysis and interpretation, or in all these areas. They took part in drafting, revising, or critically reviewing the article. All authors have given final approval of the version to be published, have agreed on the journal to which the article has been submitted, and agree to be accountable for all aspects of the work. Disclosure The authors declare that they have no competing interests in this work. References Associated Data Data Availability Statement The datasets used and/or analyzed during the current study are available from the corresponding author on reasonable request.