Liver cirrhosis (LC) is a progressive chronic disease with high mortality from complications such as infections, variceal bleeding and hepatic encephalopathy. Evolving evidence since 2012 supports earlier secondary prevention and lifestyle interventions to slow progression, creating a role for nursing interventions in both stable and advanced stages of LC. Patients have reported gaps in continuity and holistic support in usual outpatient care, and nursing-led follow-up may address self-management, learning and complication prevention.
Outpatient nursing interventions in LC have been infrequently studied and findings uncertain. The Quality Liver Nursing Care Model (QLiNCaM) was developed to offer structured, individualized RN follow-up inspired by successful RN models in chronic heart failure, person-centred care and nursing theory. The QLiNCaM was evaluated within a randomized controlled trial across six Swedish hospitals from 2016 to 2022, with previously reported effects on some clinical and patient-reported outcomes but no effect on the primary outcome of health-related quality of life.
The present study aimed to explore aspects that influenced implementation of the nurse-based outpatient follow-up intervention (QLiNCaM) at the six participating clinics, based on experiences reported by RNs, physicians and managers. Key questions were:
This was an explorative, deductive qualitative study. Directed content analysis guided by the PARiHS conceptual framework (Promoting Action on Research Implementation in Health Services) structured the analysis. PARiHS frames successful implementation as the interaction of three elements: evidence (research, clinical experience, patient experience, local information), context (culture, leadership, evaluation) and facilitation (roles, skills and attributes). Each element and sub-element lies on a high–low continuum predicting implementation success.
Six publicly funded Swedish hospitals (two county hospitals and four university hospitals) in mid and southern Sweden participated. While the hospitals endorsed the intervention, none had an existing nurse-led LC follow-up program. The intervention study sought 500 patients but recruited 167, reflecting recruitment challenges in routine clinical settings. Informants for this implementation study included seven of eight RNs who delivered the intervention, a purposive sample of ten physicians providing standard follow-up, and twelve managers responsible for outpatient physicians and nurses. In total, 29 informants contributed data via 22 interviews.
The QLiNCaM intervention consisted of structured, individualized RN follow-up routines and recommendations tailored to each patient’s needs. RNs received training prior to and during the study (single-day training, a three-day course in motivational interviewing and communication, and tutorial meetings). The randomized controlled design meant both control and intervention patients were treated at the same clinics, creating risks of contamination. To mitigate this, only one physician per clinic had full insight into intervention content, and RNs’ interactions with control patients (for example, during telephone calls or procedures) were potential sources of contamination.
Semi-structured interviews were conducted at two time points per clinic to capture changes over the extended participation period (2018–2022). The first interviews (November 2018–January 2019) focused on RNs’ new role (n = 7). Subsequent interviews (December 2019–May 2022) addressed implementation and study performance and included individual and group interviews with RNs, physicians and managers (total interviews = 22; combined duration = 881 minutes; mean ~40 minutes). Interviews were performed via Zoom, face-to-face or telephone and were audio recorded. Interview guides were informed by the Consolidated Framework for Implementation Research and piloted before use.
Interviews were transcribed verbatim and analysed as a single dataset using directed content analysis. A coding scheme based on PARiHS elements and sub-elements was applied. Statements relevant to the aim were coded and sorted into indicators of high or low probability for implementation success. Codes were grouped into categories within the PARiHS domains. Analysis was iterative, conducted primarily by the first author with review and consensus discussions with the second author. NVivo software supported analysis.
The study adhered to the Helsinki Declaration and was approved by the Regional Ethics Board in Uppsala (ID 2016/146). Informants provided informed consent and received no compensation. Data contain potentially identifying information and are restricted; access may be requested from Region Dalarna per ethics requirements. Rigour addressed credibility, dependability, confirmability and transferability: broad data coverage across informant roles, iterative coding with consensus, use of interviewers not involved in the intervention for most interviews, and adherence to COREQ reporting guidelines.
Analysis identified 23 distinct factors spanning the PARiHS elements that influenced implementation probability. Eleven factors reduced the likelihood of successful implementation; examples included staffing shortages, lack of clarity about roles and procedures, insufficient resources, limited supportive leadership, a fragile mandate for RNs to implement the intervention, and a moral dilemma experienced by staff when patients in need could not all be randomized to the intervention.
Twelve factors increased the likelihood of successful implementation. The intervention was perceived as clinically relevant and safe, aligned with clinic goals, and compatible with RNs’ scope of practice. Managers who actively facilitated nurses’ roles within multidisciplinary teams helped strengthen the nursing mandate and professional growth. Practical features—simple procedures that RNs could perform independently—were also facilitating factors. The randomized design's contamination risks required role restrictions and careful management.
Recruitment challenges were notable: the trial aimed for 500 participants but achieved 167, and recruitment difficulties varied across clinics and over time. The extended participation periods at individual clinics (three to seven years) and differing clinical research experience influenced implementation dynamics.
Implementation of the QLiNCaM nurse-based follow-up was shaped by interacting elements of evidence, context and facilitation as described by the PARiHS framework. Successful implementation was supported when the intervention was viewed as valuable and feasible for clinic practice, when managers actively supported nurse roles, and when procedures were simple and within RNs’ independent practice. Barriers included workforce shortages, unclear mandates and ethical tensions introduced by randomization in clinical populations with unmet needs.
The authors conclude that the PARiHS framework captures relevant perspectives needed to translate evidence into everyday nursing practice and that adequate perceived benefits are required to drive implementation success. They note that staff shortages are a barrier to conducting clinical research and to implementation in practice, and that moral dilemmas may arise when trial design restricts access to interventions deemed beneficial.
The findings suggest that scaling nurse-based LC follow-up programs will require explicit role definitions, managerial support, allocation of resources, and attention to ethical concerns when using randomized designs in routine care. Simple, independently executable RN procedures that align with clinical goals and professional competence may facilitate wider adoption. Future implementation efforts should address recruitment barriers and contextual readiness at participating sites.