Comparing nurse-assisted outpatient follow-up with standard care among patients with decompensated liver cirrhosis: a systematic review and meta-analysis

Abstract

Nurse-assisted care has been shown to improve outcomes in these patients when compared with standard care. This study aimed to compare nurse-assisted outpatient follow-up with standard care in patients with decompensated liver failure. MEDLINE, Embase, and Cochrane Central databases were searched for randomized controlled trials (RCTs) comparing specialized or nurse-assisted care with standard care in patients with decompensated liver cirrhosis. Outcomes of interest were mortality, 30-day readmission, model for end-stage liver disease (MELD) score, Child–Pugh score, and mean number of hospital stays. Evaluations were reported as risk ratios and mean differences, with 95% confidence intervals (CIs) using weighted random-effects models. The analysis included 668 patients from seven studies (three RCTs and four observational studies). Compared with standard care, nursing and multidisciplinary care showed no significant reduction in mortality (risk ratio: 0.78, 95% CI: 0.53–1.16, P = 0.12, I² = 69%); however, upon subgroup analysis according to type of study significant reduction was noted among RCTs [risk ratio: 0.53, 95% CI: 0.30–0.94, P = 0.03, I² = 0%). Meta-analysis also showed a significant reduction in 30-day readmission rates (risk ratio: 0.39, 95% CI: 0.25–0.59, P < 0.0001, I²=0%), which were consistent upon subgroup analysis. There was no significant difference in Child–Pugh score (mean difference: 0.07, 95% CI: ‐0.79 to 0.93, P = 0.83, I² = 0%), MELD score (mean difference: 0.17, 95% CI: ‐1.33 to 1.67, P = 0.82, I² = 0%) and mean difference in number of hospital stay (mean difference: ‐1.59, 95% CI: ‐5.68 to 2.51, P = 0.45, I² = 89%). Results were consistent upon subgroup analysis except for the mean number of hospital stays, which showed a significant reduction among observational studies (mean difference: ‐4.20, 95% CI: ‐8.18 to ‐0.22, P = 0.04).

Introduction

Liver cirrhosis is a chronic progressive condition in which the normal liver tissue undergoes fibrosis and scarring. This can destroy normal liver function and lead to liver failure [1]. Liver cirrhosis causes over 1 million deaths every year, and the majority of the cases are preventable. The most common risk factors leading to cirrhosis include infection by the hepatitis B or C virus, alcohol consumption, and metabolic syndrome [2]. It has been reported that around 4–12% of patients with liver cirrhosis develop at least one decompensating event [3]. Decompensated cirrhosis is the acute deterioration of a cirrhotic patient, which manifests as jaundice, ascites, hepatic encephalopathy, hepatorenal syndrome, or variceal bleeding [3,4].

Nurse-assisted care has shown remarkable results in terms of safety and efficacy for the management of chronic diseases. Interventions categorized as nurse-assisted care included a wide range of approaches, such as telephone follow-up, outpatient education, structured discharge programs, home visits, and multidisciplinary nursing support [5]. Equal or better outcomes related to quality of life, symptom control, and disease-specific clinical outcomes have been reported with nurse-assisted care than with standard physician-led care programs [5]. Nursing care involves assessment, diagnosis, planning, implementation, and evaluation [6]. These steps ensure patient-centered quality care, which improves outcomes among those with chronic diseases, such as liver cirrhosis.

Despite several advantages of nurse-assisted care in the management of patients with decompensated liver cirrhosis, a comprehensive review of the current literature is lacking in this field. This systematic review and meta-analysis aimed to review the current literature and compare the efficacy of nurse-assisted care to standard care in outpatient follow-up among patients with decompensated liver cirrhosis.

Methods

The Preferred Reporting Items for Systematic Review and Meta-Analyses guidelines and the risk of bias in systematic reviews and assessment of multiple systematic reviews (AMSTAR) 2 were both followed when doing this meta-analysis [7,8]. The International Prospective Register of Systematic Reviews (PROSPERO), maintained by the National Institute for Health Research, contains information about this study (identification no. CRD42023431545). Because the information was accessible to the general public, institutional review board approval was not necessary.

Data sources and search strategy

MEDLINE, EMBASE, and Cochrane CENTRAL were comprehensively searched from inception through May 2023 by two independent reviewers (J.G. and X.W.). We extracted studies based on abstracts and titles. A full-text appraisal was sought when required. Medical subject headings, phrases, and keywords were used to find keywords for ‘nursing care’, ‘multidisciplinary care’, and ‘decompensated liver cirrhosis’.

Study selection

We included studies if they were: (a) randomized controlled trials (RCTs) or observational studies that determined the impact of care in different arms, (b) reported either of mortality, 30-day readmission, model for end-stage liver disease (MELD) score, Child–Pugh score, or mean number of hospital stay as one of their outcomes. A third investigator (Y.W.) was consulted in case of any disagreement regarding study selection. All articles were then uploaded to EndNote Reference Library (Version X7.5; Clarivate Analytics, Philadelphia, Pennsylvania, USA) software to remove any duplicates.

Interventions categorized as nurse-assisted care included a wide range of approaches, such as telephone follow-up, outpatient education, structured discharge programs, home visits, and multidisciplinary nursing support.

Data extraction and assessment of study quality

Two reviewers (J.G. and X.W.) independently extracted from the selected studies the characteristics of the studies, patient demographics, summary events, number of events, sample sizes, and treatment type. Summary events were also extracted for outcomes of interest, and the mean difference with SD from baseline. The quality of studies across six categories (selection bias, performance bias, detection bias, attrition bias, reporting bias, and other bias) was evaluated using the Cochrane Risk of Bias Tool (CRBT).

Statistical analysis

Outcomes of interest consisted of mortality, 30-day readmission, MELD score, Child–Pugh score, and the mean number of hospital stays. RevMan (version 5.4.1; The Nordic Cochrane Centre, The Cochrane Collaboration; Copenhagen, Denmark) was used to conduct the meta-analysis. The outcomes of interest were provided as risk ratios with 95% confidence intervals (CIs) and were aggregated using an inverse variance weighted random-effects model. Forest plots were used to graphically display the pooled analyses. Continuous outcomes of interest were presented as mean differences with 95% CIs and were pooled using an inverse variance weighted random-effects model. When the mean was not available, we used the median for analysis. When the change from the baseline was not reported, we calculated the difference in means between the baseline and the posttreatment measurements. Its SD was derived from the baseline and the follow-up by assuming their correlations were 0.5. The Higgins I² was utilized to assess heterogeneity between trials. A 25–50% number was regarded as low, 50–75% moderate, and more than 75% serious. In all cases, a P value less than 0.05 was considered significant.

Results

Search results

Our initial search yielded 3091 potentially relevant articles of which 961 were selected for full-text review (Fig. 1). Upon further exclusions, seven studies (three RCTs and four observational studies), with a total of 668 patients, were shortlisted for data extraction [9–14, 15]. Egger’s regression was NS for publication bias (t = 1.10, P = 0.514).

Fig. 1.

PRISMA flow diagram of study identification for meta-analysis. PRISMA, Preferred Reporting Items for Systematic Review and Meta-Analyses.

Study characteristics and quality assessment

Study characteristics and baseline demographics have been summarized in Table 1. RCTs were deemed to be of generally low risk of bias according to the CRBT.

Table 1.

Baseline characteristics of included studies

Title	Authors	Patient source	Patient description	Study design	No. of patients	Intervention	Control	Follow-up
How to improve care in outpatients with cirrhosis and ascites: a new model of care coordination by consultant hepatologists	Morando et al. [9]	General Hospital Padova	Admitted with acute complications to cirrhosis	Consecutive allocation after discharge to intervention or control by a 2 : 3 ratio	100	Care management check-up, including structured diagnostics, treatment, and follow-up after 1–12 weeks	Standard outpatient care by primary physician and ‘on-demand’ hepatologist	9–11 Months
Efficacy of a chronic disease management model for patients with chronic liver failure	Wigg et al. [10]	Hepatology Unit of Flinders Medical Center, Adelaide	Admitted with chronic liver failure-related complications	Randomized, controlled, parallel-group study design	60	Home visit by nurse 1 week after discharge, weekly nurse telephone calls, and telephone reminders of appointments. Nurse visits involved decision-making and self-management support, diet, and medication education	Standard inpatient, hospital outpatient, and primary care management	12 Months
An educational tool for the prophylaxis of hepatic encephalopathy	Garrido et al. [11]	Center for Liver Diseases, Padova University Hospital	Stable outpatients	RCT	39	Education on HE, including basic information, bowel emptying, medicine, and monthly telephone calls	Standard care	12 Months
Rehabilitation for cirrhotic patients discharged after hepatic encephalopathy improves survival	Andersen et al. [12]	Abdominal Center, University Hospital Bispebjerg,	Alcoholic liver cirrhosis	A prospective study with historic control from the same center	33	A session with a nurse after discharge, alcohol school for some patients, and help with social services were offered	Standard outpatient control, as defined by a responsible physician	20 Months
The impact of outpatient clinical care on the survival and hospitalization rate in patients with alcoholic liver cirrhosis	Majc and Tepes [13]	Department of Gastroenterology, Murska Sobota General Hospital	Liver cirrhosis and active drinkers of alcohol	Prospective study, historic control from other departments of the same hospital	199	Education on alcohol abstinence, diet, and adjusting diuretic therapy	No regular outpatient control	5 Years
A program that reduces early readmissions, mortality at 60 days, and healthcare costs in decompensated cirrhosis	Morales et al. [14]	Hepatology Unit, Germans Trias y Pujol Hospital, Badalona	Liver cirrhosis and discharged after admission with complications	Prospective study, historic control group from the same center	192	HEPACONTROL program. 7-day follow-up visit after discharge with a hepatologist, medicine adjustment, lab tests, or diagnostics as needed; leaflet on warning signs of decompensation	Outpatient visit within 2 months after discharge, no lab tests of diagnostics available	7 Months minimum
A simple educational tool for reducing 30-day hospital readmissions in patients with decompensated cirrhosis	Kumral et al. [15]	The University of Virginia (hospital name not reported)	Liver cirrhosis and acute decompensating event	Prospective study, historic control from the same center	45	Nurse teaching session with the patient and family members, booklet, a digital scale, and pill organizer, 72-h postdischarge phone call	Standard care is not defined	1 Month

RCT, randomized controlled trial.

Outcomes

Mortality

Two RCTs and three observational studies reported mortality as an outcome. A significant difference in mortality was reported by the RCTs (risk ratio: 0.53, 95% CI: 0.30–0.94, P = 0.03, I² = 0%), but the observational studies did not show a significant difference (risk ratio: 0.92, 95% CI: 0.62–1.36, P = 0.67, I² = 71%). Cumulative analysis of both RCTs and observational studies showed no significant difference in mortality (risk ratio: 0.78, 95% CI: 0.53–1.16, P = 0.12, I² = 69%) (Fig. 2).

Fig. 2.

Forest plot comparing nursing care with standard care for mortality. CI, confidence interval; RCT, randomized controlled trial.

30-Day readmission

One RCT and three observational studies reported 30-day readmission as an outcome. Cumulative analysis also showed a significant difference in 30-day readmission (risk ratio: 0.39, 95% CI: 0.25–0.59, P < 0.0001, I² = 0%). Results were consistent upon subgroup analysis [RCT (risk ratio: 0.36, 95% CI: 0.16–0.80, P = 0.01)]; observational studies (risk ratio: 0.39, 95% CI: 0.24–0.66, P = 0.0004, I² = 0%) (Fig. 3).

Fig. 3.

Forest plot comparing nursing care with standard care for 30-day readmission. CI, confidence interval; RCT, randomized controlled trial.

Model for end-stage liver disease score

Three RCTs reported the MELD score of their patients. There was no significant difference found in the MELD score of patients provided with nurse-assisted care versus standard care (mean difference: 0.17, 95% CI: ‐1.33 to 1.67, P = 0.82, I² = 0%) (Fig. 4).

Fig. 4.

Forest plot comparing nursing care with standard care for MELD score. CI, confidence interval; MELD, model for end-stage liver disease.

Child–Pugh score

Two RCTs reported the Child–Pugh score. There was no significant difference found in the Child–Pugh score of patients provided with nurse-assisted care versus standard care (mean difference: 0.07, 95% CI: ‐0.79 to 0.93, P = 0.83, I² = 0%) (Fig. 5).

Fig. 5.

Forest plot comparing nursing care with standard care for Child–Pugh score. CI, confidence interval; RCT, randomized controlled trial.

Mean number of hospital stays

Two RCTs and one observational study reported the mean number of hospital stays. No significant difference was shown by the subgroup analysis of RCTs (mean difference: ‐0.54, 95% CI: ‐5.23 to 4.14, P = 0.82, I² = 91%) and cumulative analysis of RCTs and observational studies (mean difference: ‐1.59, 95% CI: ‐5.68 to 2.51, P = 0.45, I² = 89%); however, the observational study showed a significant difference in the mean number of hospitals stays (mean difference: ‐4.20, 95% CI: ‐8.18 to ‐0.22, P = 0.04) (Fig. 6).

Fig. 6.

Forest plot comparing nursing care with standard care for the mean number of hospital stays. CI, confidence interval; RCT, randomized controlled trial.

Discussion

In this meta-analysis evaluating the impact of nursing and multidisciplinary care on outcomes in patients with liver cirrhosis, we report several key findings. Nursing care was associated with a decreased risk of mortality. Moreover, it is also associated with a decreased risk of 30-day readmission rates in these patients.

Patients with decompensated liver cirrhosis must receive nursing care and multidisciplinary care. They offer comprehensive and coordinated support to address these patients’ complex requirements [16]. Nurses evaluate and monitor patients, administrate and educate on medications, manage symptoms, assist with fluid and dietary management, and educate patients and their families [17]. The heterogeneity in nurse-assisted interventions (telephone follow-ups, home visits, and outpatient education) could partially explain the variability in observed outcomes. Multidisciplinary care is the collaborative effort of a group of healthcare professionals to optimize patient outcomes. It includes care coordination, psychosocial support, and education for the empowerment of patients and their families [18]. By combining these methods, healthcare teams can provide comprehensive care and enhance the health of patients with decompensated liver cirrhosis [19].

From a health economics perspective, the reduction in 30-day readmissions and the mean number of hospital stays associated with nurse-assisted follow-up is highly relevant. Prior research has demonstrated that structured nurse-led interventions, although requiring upfront investment in training and program implementation, can generate substantial cost savings by reducing unnecessary hospitalizations and optimizing resource utilization. These findings support the cost-effectiveness of nurse-assisted care models, especially in healthcare systems with limited resources [10,14]

This meta-analysis synthesizes the best available evidence on mortality and readmissions in patients with decompensated cirrhosis who received nurse-assisted follow-up, including nurse-assisted multidisciplinary interventions. Several parameters, including objective outcomes such as mortality and readmissions, improved significantly across all categories of studies. In each of the three standardized hospital follow-up investigations, mortality rates were significantly reduced. In terms of readmission rates, however, we observed inconsistent results across the various types of interventions, indicating that there are currently no data to support a specific type of nurse-assisted or multidisciplinary postdischarge intervention.

In a recent review of nursing care for patients with liver cirrhosis, the involvement of carers or family members was identified as a crucial component of both treatment and follow-up [16]. Several studies demonstrate the positive impact and significance of utilizing family nursing and involving families in acute and chronically ailing patients healthcare [20–22]. As described in previous studies, there is a need for randomized clinical studies of standardized nursing care and chronic care models with multidisciplinary involvement in the treatment of liver cirrhosis, as the efficacy of such programs has been extensively demonstrated in other major chronic diseases, such as heart failure and chronic obstructive pulmonary disease [16]. Future research should concentrate on evaluating and validating nurse- and physician-led clinics and rehabilitation programs to provide individualized healthcare services to patients and their families and optimize the use of healthcare resources. Involving a patient’s family in his or her healthcare has proved beneficial in the treatment of other chronic diseases [23].

The absence of statistically significant effects on longer-term or disease-severity outcomes such as mortality, Child–Pugh, and MELD scores in our pooled analyses may reflect limited statistical power and relatively short follow-up reported by many included studies rather than a true lack of effect. Several systematic reviews of postdischarge and nurse-led interventions highlight substantial heterogeneity in study design, small sample sizes, and short time horizons – factors that constrain the ability to detect differences in hard endpoints such as survival or biochemical scores. Larger randomized trials with longer follow-up and prespecified survival or disease-progression endpoints are therefore needed to determine whether the observed reductions in early readmissions translate into durable improvements in mortality or liver-specific severity measures [24,25].

Nurse-assisted outpatient follow-up tends to target transitional-care processes (early postdischarge contact, medication reconciliation, patient education, symptom surveillance, and rapid outpatient response), which act directly to prevent common, short-term causes of early readmission (treatment errors, dehydration, medication nonadherence, and failure to detect early decompensation). These mechanisms can rapidly reduce short-term readmissions without immediately changing underlying liver disease severity (MELD/Child–Pugh) or long-term survival, which are driven by biological progression and may require sustained, disease-modifying interventions and longer observation to detect benefit. This pathway – improved process of care and self-management leading to fewer early readmissions but only gradual effects on disease-severity metrics – is described in both cirrhosis and other chronic-disease settings [24,26].

Reduced 30-day readmissions and a lower number of hospital stays are clinically important and potentially cost-saving. Several controlled programs in cirrhosis and a broader literature on quality-improvement and nurse-led models report reductions in early readmissions and, in many cases, favorable economic outcomes. For example, the HEPACONTROL program reported lower early readmission rates, lower 60-day mortality, and reduced early readmission costs compared with historical controls. More broadly, systematic economic reviews of quality-improvement interventions show that multicomponent programs can reduce readmissions but that net cost impact varies by intervention design, patient population, and the time horizon used for the economic evaluation; interventions that engage patients and caregivers often provide greater value. We therefore emphasize that while clinical benefits for readmissions are consistent, formal cost-effectiveness analyses – ideally conducted alongside randomized trials with clear time horizons and transparent accounting of program costs (training, nurse time, and infrastructure) versus savings (avoided admissions, shorter length of stay) – are required to guide implementation decisions across different health systems [14,27].

Limitation

There are several limitations that should be considered while interpreting the findings of this review. All randomized studies were classified as having a high risk of bias, whereas all nonrandomized studies were rated as having low evidence certainty. Given the nature of the interventions in question, concealing participants and personnel was neither practical nor logical, which helps to explain the high risk of performance bias in the randomized studies. Some studies’ performance bias may have been mitigated by objective outcomes such as hospital admission and mortality. Because of numerous subjective outcomes, such as self-efficacy, lifestyle enhancements, and quality of life, detection bias could be a potential source of bias in the studies included in this review. Studies were conducted in countries with diverse cultures, religions, and etiologies of cirrhosis; therefore, comparing results from various healthcare settings should be considered [9–15]. Lastly, with regard to the multifaceted educational interventions, it should be determined which component of the educational program was responsible for the positive outcomes, or if it was the combination of the educational program and closer contact with healthcare professionals that was effective. Future studies with stratified outcomes could strengthen the evidence supporting the use of educational aids in postdischarge follow-up. Another important limitation of this meta-analysis is the lack of economic data reported in the included studies. Although nurse-assisted care showed benefits in reducing readmission rates and hospital stays, we could not directly assess its cost-effectiveness because of insufficient data. Future studies should incorporate economic evaluations to better determine the financial impact and feasibility of nurse-assisted interventions across different healthcare systems.

Conclusion

This meta-analysis compared nurse-assisted care with standard care in patients with decompensated liver cirrhosis and demonstrated several important findings. Nurse-assisted care was consistently associated with a significant reduction in 30-day readmission rates and showed potential benefits in reducing the number of hospital stays; however, no statistically significant improvements were observed in mortality, Child–Pugh scores, or MELD scores, likely reflecting the limited sample sizes, heterogeneous interventions, and relatively short follow-up durations across included studies.

From a clinical perspective, these results highlight the value of nurse-assisted interventions – such as structured follow-up, patient education, and multidisciplinary support – in improving short-term outcomes and transitional care after hospital discharge. Given the substantial burden of readmissions in cirrhosis, the implementation of nurse-assisted care models may help optimize healthcare resource utilization and improve quality of care.

Nevertheless, the long-term impact of these interventions on survival and disease severity remains uncertain. Future well-designed, large-scale RCTs with longer follow-up periods are required to clarify whether reductions in early readmissions translate into durable benefits in mortality and liver disease progression. In addition, future research should incorporate formal cost-effectiveness analyses and explore the role of family involvement and multidisciplinary integration to further strengthen the evidence base.

Acknowledgements

None.

Conflicts of interest

There are no conflicts of interest.