Protocol for the process evaluation of a randomised clinical trial of incremental-start versus conventional haemodialysis: the TwoPlus study

Abstract

Introduction

Process evaluation provides insight into how interventions are delivered across varying contexts and why interventions work in some contexts and not in others. This manuscript outlines the protocol for a process evaluation embedded in a hybrid type 1 effectiveness-implementation randomised clinical trial of incremental-start haemodialysis (HD) versus conventional HD delivered to patients starting chronic dialysis (the TwoPlus Study). The trial will simultaneously assess the effectiveness of incremental-start HD in real-world settings and the implementation strategies needed to successfully integrate this intervention into routine practice. This manuscript describes the rationale and methods used to capture how incremental-start HD is implemented across settings and the factors influencing its implementation success or failure within this trial.

Methods and analysis

We will use the Consolidated Framework for Implementation Research (CFIR) and the Reach, Effectiveness, Adoption, Implementation and Maintenance (RE-AIM) frameworks to inform process evaluation. Mixed methods include surveys conducted with treating providers (physicians) and dialysis personnel (nurses and dialysis administrators); semi-structured interviews with patient participants, caregivers of patient participants, treating providers (physicians and advanced practice practitioners), dialysis personnel (nurses, dieticians and social workers); and focus group meetings with study investigators and stakeholder partners. Data will be collected on the following implementation determinants: (a) organisational readiness to change, intervention acceptability and appropriateness; (b) inner setting characteristics underlying barriers and facilitators to the adoption of HD intervention at the enrollment centres; (c) external factors that mediate implementation; (d) adoption; (e) reach; (f) fidelity, to assess adherence to serial timed urine collection and HD treatment schedule; and (g) sustainability, to assess barriers and facilitators to maintaining intervention. Qualitative and quantitative data will be analysed iteratively and triangulated following a convergent parallel and pragmatic approach. Mixed methods analysis will use qualitative data to lend insight to quantitative findings. Process evaluation is important to understand factors influencing trial outcomes and identify potential contextual barriers and facilitators for the potential implementation of incremental-start HD into usual workflows in varied outpatient dialysis clinics and clinical practices. The process evaluation will help interpret and contextualise the trial clinical outcomes’ findings.

Ethics and dissemination

The study protocol was approved by the Wake Forest University School of Medicine Institutional Review Board (IRB). Findings from this study will be disseminated through peer-reviewed journals and scientific conferences.

Trial registration number

NCT05828823.

STRENGTHS AND LIMITATIONS OF THIS STUDY.

• Using a mixed-methods approach that collects both quantitative and qualitative data will provide a more comprehensive and nuanced understanding of the determinants that influence the adoption and implementation of innovative dialysis approaches in the various clinical sites.

• The study design will facilitate mapping the determinants of the adoption and implementation of innovative dialysis approaches in clinical settings.

• The evaluation process will provide insights to inform the real-world scalability of theintervention.

• The process evaluation and the intervention implementation take a collaborative approach by engaging various key stakeholders at various stages of the intervention implementation.

• One potential limitation of this study is that the process evaluation, encompassing surveys and interviews, necessitates a dedicated time commitment from the staff, thereby influencing the extent of their participation.

Introduction

Most individuals diagnosed with kidney dysfunction requiring dialysis (KDRD), a condition commonly known as end-stage kidney disease (ESKD), who begin chronic haemodialysis (HD) are prescribed a conventional HD treatment at an intensity of thrice-weekly treatments, each delivering dialysis urea clearance index of single-pool Kt/V (spKt/Vurea) ≥1.2, irrespective of an individual’s underlying clinical characteristics.¹ This conventional HD treatment was validated in randomised clinical trials that involved solely prevalent HD patients with mean dialysis vintage >2 years who had virtually no residual kidney function (RKF).^{2 3} Conventional HD was then extrapolated as an ‘optimal’ dialysis dose for all patients, including those with incident KDRD and ongoing levels of appreciable RKF. This abrupt transition from non-dialysis chronic kidney disease to thrice-weekly HD underappreciates the progressive nature of kidney disease, whereby the decline in kidney function has been gradual and ongoing, including at the time of HD initiation. On average, one in four patients with new-onset KDRD has ongoing kidney function that can complement an initial schedule of twice-weekly HD.¹⁴,⁶ Patients are expected to adhere to thrice-weekly HD care, which may not align with their individual priorities and needs. Currently, there is little emphasis on person-centred HD, with virtually all care delivered in a standardised, one-size-HD-fits-all approach.⁷

Observational data suggest that incremental HD—initially delivered twice weekly and later increased thrice weekly in parallel with a decline in RKF or changes in clinical status—yields good symptom control,⁴,⁶⁸ confers similar or better patient survival^{1 14 15} and quality of life,^{16 17} making this a promising patient-centred HD.^{613 14 18},²³ However, limited comprehensive and rigorous prospective controlled trials are focused on testing the hypothesis that incremental-start HD is either non-inferior or more effective than conventional HD.

Epidemiologic studies showed independent regional variation in KDRD incidence,²⁴ timing of HD initiation and patient outcomes independent of patient characteristics.²⁵,²⁷ Furthermore, HD treatment effects can differ across regions.²⁸ Differences in state- and local-level healthcare (Medicaid eligibility, capacity of municipal hospitals and federally-funded clinics), which influence access to general medical and subspecialty care, are believed to broadly explain regional variations in dialysis care.^{29 30} However, local practice style and environment (eg, level of competition and incentive structure) could also vary across regions and may underlie differences in KDRD-related healthcare and outcomes.^{29 31} Therefore, incorporating mixed-methods analysis is critical for evaluating incremental-start HD implementation across various organisations.

The TwoPlus randomised controlled trial

The TwoPlus study is a randomised controlled trial (RCT) that compares incremental-start HD with conventional HD in patients with newly diagnosed KDRD and RKF. The primary objective of the study is to test the hypothesis of non-inferiority of incremental-start HD versus conventional HD for safety at an average 2-year follow-up. The primary outcome is a composite of all-cause mortality, emergency department visits and hospitalisations. Secondary outcomes will compare the effects of incremental-start HD and conventional HD on biophysical and biopsychosocial domains. The engagement of diverse stakeholders will be incorporated throughout all phases of the trial. This will include dialysis administration, patients, caregivers, physicians, advanced practice providers, dialysis nurses, dialysis dietitians and dialysis social workers. Such engagement is essential to delineate factors that facilitate or hinder intervention implementation.³² A separate, detailed description of the TwoPlus RCT will be published.

The TwoPlus process evaluation aims and objectives

The aims of the process evaluation are to understand the context in which the intervention is delivered and identify factors that may influence implementation fidelity and sustainability. Specific objectives include:

1. Assess organisational readiness to adopt incremental-start HD across the participating organisations.

2. Evaluate the adoption, reach and fidelity of incremental-start HD utilising Proctor’s Implementation Outcomes Framework.³³

3. Assess barriers and facilitators to the adoption of incremental-start HD utilising the Consolidated Framework for Implementation Research (CFIR) framework.³⁴ This includes the characterisation of inner setting characteristics, intervention characteristics, inner setting factors that mediate implementation and external factors that influence the adoption of HD intervention at each enrollment centre.

4. Understand participants’ (patients, caregivers, treating providers and dialysis personnel) experiences of the delivery and receipt of the intervention.

5. Assess if the intervention of incremental-start HD was delivered as intended (fidelity).

Methods and analysis

This protocol was written in accordance with the Good Reporting of a Mixed Methods Study (GRAMMS) checklist.³⁵ (See attached Research Checklist.)

The TwoPlus trial design

The study uses a type 1 hybrid effectiveness-implementation approach with a primary focus on evaluating whether incremental-start HD achieves its intended outcomes (effectiveness). It also examines how the intervention is implemented and the factors influencing its implementation success or failure.³⁶ The study will consist of implementation preparation (Stage 1, months 1–12, from the date of project start) followed by study conduct and dissemination (Stage 2, months 13–60). Study participants encompass patients, caregivers, treating physicians and advanced practice providers, dialysis nurses, dialysis dietitians and dialysis social workers. Over 2 years, we will enroll 350 adults started on chronic HD who will be individually randomised to incremental-start HD (intervention group, n=175) versus conventional HD (control group, n=175) and 140 caregivers of patient participants, proportionally derived from the two treatment groups (figure 1). Eligibility criteria for patient and caregiver participation are summarised in online supplemental table 1. Patients randomised to incremental-start HD will receive treatment with twice-weekly HD and adjuvant pharmacotherapy (loop diuretics, sodium bicarbonate buffer and potassium binders as needed); the HD treatment will subsequently step to ‘full-dose’ thrice-weekly HD based on RKF levels and clinical needs. Criteria for progression, summarised in online supplemental table 2, will guide the transition from twice-weekly to thrice-weekly HD. Patients randomised to conventional HD will be treated to thrice-weekly HD according to usual care. To monitor RKF, timed urine collection will be obtained at baseline and every 3 months (or more often when clinically indicated) during the study.

Figure 1. The TwoPlus study design and timeline. HD, haemodialysis; KDRD, kidney dysfunction requiring dialysis; M, month of study participation as time point for questionnaire data collection; spKt/V, single pool Kt/V urea dialysis clearance per HD treatment; stdKt/V, standard Kt/V urea clearance per week.

Study setting

The TwoPlus RCT is conducted at national medical institutions and their affiliated outpatient and inpatient dialysis units. Patient and caregiver enrollment will take place at national healthcare systems and their affiliated outpatient and inpatient dialysis units. A Clinical Coordinating Centre and Data Coordinating Centre, separate from enrollment healthcare systems, will oversee participant enrollment, safety monitoring and data collection. Selection of healthcare systems in the TwoPlus RCT is determined by considering factors such as the patient population, ensuring an adequate volume and sociodemographic diversity for effective study recruitment. Selection criteria also considered the healthcare systems’ expertise in clinical trials, availability of research infrastructure and the study team’s engagement in preparing study implementation.

Process evaluation design

The process evaluation will utilise an explanatory sequential mixed-methods approach. This two-phase design will begin collecting and analysing quantitative data obtained from local providers, that is, Dialysis Treating Providers Advisory Panel, and dialysis nurses, that is, Dialysis Nurse Advisory Panel, to assess readiness for study implementation and factors that influence the implementation of the TwoPlus intervention, followed by subsequent collection and analysis of qualitative data. Qualitative data will include semi-structured interviews among local providers, that is, Dialysis Treating Providers Advisory Panel, and dialysis nurses, that is, Dialysis Nurse Advisory Panel, dialysis dieticians, that is, Dialysis Dietician Advisory Panel, and dialysis social workers, that is, Dialysis Social Worker Advisory Panel, and focus group interviews with study investigators and stakeholder partners.

Process evaluation guiding frameworks

The process evaluation is guided by the CFIR and Proctor’s Implementation Outcomes Framework.³³ CFIR is a typology of constructs likely to influence the implementation of evidence-based interventions: (a) intervention characteristics, (b) outer setting, (c) inner setting, (d) characteristics of individuals and (e) the implementation process. These constructs are widely used to evaluate intervention implementation processes.^{37 38} CFIR provides a conceptual structure for evaluating implementation efforts. The interview guide will be informed by the five constructs of CFIR. In addition, we will use Proctor’s Implementation Outcomes Framework for translating research into ‘real-world’ settings. It will guide the implementation evaluation process via a collection of distinct outcomes (intervention adoption, intervention fidelity and reach) that indicate the successful implementation of evidence-based strategies.

Process evaluation

Figure 2 shows how RE-AIM and CFIR guide the process evaluation. Process evaluation constructs related to implementation preparation and study conduct will be evaluated at each stage of the study (table 1). During Stage 1 of the study, the study team will train representatives at the participating organisations assigned to the intervention arm on study-related processes of care to prepare for intervention implementation. A survey will be deployed to treating providers and dialysis personnel after training completion and before enrollment. During Stage 2, semi-structured interviews will be conducted with Stakeholder Advisory Panels (ie, Patient and Caregiver Advisory Panel, Treating Providers Advisory Panel, Dialysis Nurse Advisory Panel, Dialysis Dietitian Advisory Panel and Dialysis Social Worker Advisory Panel) at varied months during the study. Focus group meetings with study investigators and Stakeholder Partners will take place every 6 months. A Fidelity Monitoring Tool will capture protocol adherence at each site. Information obtained during surveys and semi-structured interviews with Stakeholder Advisory Panels will be used to evaluate implementation outcomes.

Figure 2. Framework for mixed methods process evaluation. CFIR, Consolidated Framework for Implementation Research; HD, haemodialysis; RE-AIM, Reach, Effectiveness, Adoption, Implementation and Maintenance.

Table 1. Process evaluation constructs.

Outcome/construct	Measures (quantitative/qualitative)	Level of analysis	Source	Timepoints
Implementation preparation (Stage 1)
Intervention characteristics	Organisational readiness to change34	Provider institution	Dialysis treating provider advisory panel Dialysis nurse advisory panel	−3 to −1 months before recruitment starts up
	Intervention feasibility and acceptability35	Provider
Inner setting characteristics	Implementation climate36 Organisational social context scale36 Implementation leadership37	Provider institution
External factors	External change support38 39	Provider institution
Study conduct and dissemination (Stage 2)
Reach	Number (%) of patients with incident KDRD eligible for study participation Number (%) of eligible patients who consent to study participation	Patient Healthcare system Dialysis organisation	Study administrative record	Beginning to the end of the recruitment period
Adoption	Number (proportion) of participants and participating dialysis units with >90% intervention fidelity			Beginning to end-of-study event
Intervention characteristics, inner settings characteristics, adaptation and sustainability	Semi-structured interviews	Patient and caregiver	Patient and caregiver advisory panel	+9 months, +18 months, +27 months, +36/37 months
		Provider	Dialysis treating provider advisory panel Dialysis nurse advisory panel	+10 months, +17 months, +28 months, +38/39 months
			Dialysis dietitian advisory panel Dialysis social worker advisory panel	+11 months, +23 months, +29/30 months
	Focus group meeting	Study teams	Investigators Stakeholder partners	Every 6 months
Intervention fidelity	Fidelity monitoring tool	Patient	Study administrative record	Every 6 months

Surveys during Stage 1 will use validated instruments.³⁴,³⁹ Semi-structured questionnaires during Stage 2 will be informed by CFIR and Proctor implementation outcomes instruments.^{33 66} Anticipated enrollment start and end are M11 and M36, respectively. Anticipated end of follow-up is M54. Note: ‘−‘ and ‘+’ months are calculated from the month of recruitment start-up at each enrollment centre.

CFIR, Consolidated Framework for Implementation Research; EMR, electronic medical record; KDRD, kidney dysfunction requiring dialysis; M, month, from the time of project start.

The research team consists of study investigators at each enrollment centre, an Implementation Science Team with expertise in qualitative and mixed-methods research, and Stakeholder Partners comprised of patient representatives, treating physicians, dialysis nurses, dialysis dietitians and dialysis social workers. The Stakeholder Partners will lead their respective Advisory Panel meetings and will (i) summarise their Panel’s recommendations, (ii) share the feedback from their Advisory Panel with other Stakeholder Advisory Panels and (iii) be members of and participate in meetings with the study investigators.

Participants in process evaluation

Participants in the process evaluation encompass treating providers and dialysis nurses from across all enrollment centres in Study Stage 1 and stakeholder advisory panels during Study Stage 2. A subgroup of patients participating in the RCT and their caregiver participants will form the Patient and Caregiver Advisory Panel. Treating providers (physicians and advanced practice practitioners, dialysis nurses, dialysis dieticians and dialysis social workers) will form the Dialysis Treating Providers Advisory Panel, Dialysis Nurse Advisory Panel, Dialysis Dietician Advisory Panel and Dialysis Social Worker Advisory Panel, respectively (online supplemental figure 1 and table 3).

Quantitative data collection

Surveys

During Stage 1, prior to commencing study recruitment, local nephrology providers and dialysis personnel will receive orientation on intervention- and study-related procedures. Following training, and prior to enrollment start-up, surveys will be administered to local providers, that is, Dialysis Treating Providers Advisory Panel, and dialysis nurses, that is, Dialysis Nurse Advisory Panel, to assess readiness for study implementation. We will use validated instruments, summarised in online supplemental table 4, adapted to the TwoPlus study intervention.³⁹,⁴⁴

Fidelity monitoring tool

Adherence to intervention will be monitored using a Fidelity Monitoring Tool. This tool will capture participants’ adherence to study protocol and HD schedules (online supplemental table 5). This will be achieved by analysing the data collected during the RCT through the electronic transfer of patient participants’ dialysis treatments and related medications from the dialysis electronic medical records. Data transfer will be de-identified, take place quarterly and be stored in a centralised electronic study database. Fidelity data will be assessed at 6 monthly intervals and shared with the study teams of each enrollment centre. This tool will allow the study teams to critically analyse performance and identify areas for improvement.

Qualitative data collection

During study conduct, stakeholder engagement consists of implementing study-related processes of care and providing feedback about barriers, facilitators and adaptation to intervention implementation.

Semi-structured interviews

These will take place every 8–12 months between the Implementation Science Team and local providers, that is, Dialysis Treating Providers Advisory Panel, and dialysis nurses, that is, Dialysis Nurse Advisory Panel, dialysis dieticians, that is, Dialysis Dietician Advisory Panel, and dialysis social workers, that is, Dialysis Social Worker Advisory Panel. Interviewers will use a semi-structured interview guide designed for this study (online supplemental table 6). The semi-structured interview guide consists of open-ended, qualitative questions that allow and encourage participants to elaborate as they like, which we expect to yield rich personal narratives and provide insight into processes of care and experiences related to the study intervention. The interviews will gather insights from patients and caregivers enrolled in the study, treating physicians and advanced practice providers, dialysis nurses, dialysis dietitians and dialysis social workers regarding their experiences with different HD schedules, timed urine collection, patient well-being, satisfaction and methods to seamlessly integrate varied HD schedules and assessment of residual kidney function into routine workflows. During semi-structured interviews with providers, we will probe perspectives on the feasibility of patient-provider encounters four times per month when patients receive twice-weekly or thrice-weekly HD. In addition, through the engagement of dialysis administration, we will collect the number of patient and physician or advanced practice provider encounters that took place every month during the study period.

Focus group meetings

The Implementation Science Team will also hold semi-annual focus group meetings with study investigators and Stakeholder Partners to gather operational adaptations and other perspectives related to the implementation of incremental-start HD. Focus group meetings will use a semi-structured focus group guide designed specifically for the purposes of this study (online supplemental table 6). Participants will be recruited using purposive sampling, and the focus group guide will consist of open-ended, qualitative questions that allow and encourage participants to elaborate as they like. We expect this approach to yield rich personal narratives and provide insight into the intervention implementation, potential barriers and facilitators to implementation, and recommendations for adaptations.

Adaptation to intervention

Adaptations are often made to successfully translate new interventions into other settings and ensure their fit with the local context, for example, the organisation and the client population.³⁴ Identifying processes and types of adaptations to study-related processes of care will be important for understanding how to improve the fit between incremental-start HD and the context in which practitioners use the intervention.⁴⁵ Thus, knowledge about adaptation is essential in evaluating the effectiveness of an intervention.⁴⁶ To seamlessly integrate the intervention into the usual workflows at outpatient dialysis centres, all stakeholders for whom incremental-start HD is pertinent (ie, patient representatives, treating providers, dialysis personnel and dialysis administration) will be engaged in pre-implementation planning that will take place during Stage 1 of the study.³²

We will collect data concerning adaptations to the processes of care related to core components of incremental-start HD at the outpatient dialysis units and the relationship of adaptations to subsequent study fidelity. The study team has identified the following core components of incremental-start HD are listed in: (i) collection of timed urine collection no less often than every 3 months from the date of randomisation, (ii) blood tests in relation to timed urine collection, (iii) communication of urine collection results to patient participants and (iv) review and adjustment of pharmacoadjuvant therapy with patient participants (online supplemental table 7). Performance on core components will be monitored with a) the use of the Fidelity Monitoring Tool, subscore on adherence to process elements, and b) semi-structured interviews with the advisory panels.

Sampling and recruitment for surveys

Using a HIPAA-compliant REDCap data collection instrument, we will distribute an online survey to all individual dialysis nurses and treating providers at each enrollment centre. The de-identified survey takes <15 min to complete (online supplemental table 2).

Sampling and recruitment for semi-structured interviews

Members for Advisory Panels will be selected from each enrollment centre at the beginning of each semi-structured interview period (online supplemental table 3). A list of advisory panel members who match our sampling criteria will be generated for each interview period. The interviewee will be contacted by a member of the Implementation Sciences Team, who will be blinded to the intervention allocated to the patient selection. To help ensure the generalisability of experiences, we aim to engage stakeholders of different gender, age groups, race and socioeconomic status. We will use purposive sampling to capture the equal engagement of patient participants from each treatment group, and caregivers linked to patients assigned to each treatment group. A written informed consent for RCT participation will be obtained from the patients, which will indicate whether they want to participate in semi-structured interviews. At the time of informed consent, a dedicated discussion regarding the criteria for progression from twice- to thrice-weekly HD, as outlined in online supplemental table 2, is held with the prospective participant. These criteria and expectations are periodically reviewed and reinforced by the study team and treating providers throughout the study. A telephone consent, with a waiver of in-person signed informed consent, will be obtained from caregivers; and members of stakeholder advisory panels (patients and caregivers participating in the study; treating physicians and advanced practice providers, dialysis nurses, dialysis dietitians and dialysis social workers from medical institutions where the study is conducted) before each interview. Where relevant, reasons for not participating in a semi-structured interview will be elicited. The final sample for semi-structured interviews will depend on the heterogeneity of the recruited participants and the point at which data saturation is reached.

Tracking advisory panel activities

We will use REDCap to capture how stakeholders engaged and report how they influenced the study. Details, by study stage, will include which stakeholder groups were involved, when, where, why and how they were engaged; level of engagement; and how challenges to engagement were overcome.⁴⁷

Confidentiality, members of the advisory panels

Audio recordings of semi-structured interviews will be triggered after sociodemographic information is collected from the participants. The digital audio recording will be transcribed and labelled with the participant’s Study ID. No identifying information will be linked to the audio recording or transcription.

Participant honourarium

Members of the Advisory Panels will be provided a modest honourarium to recognise their time and effort in taking the survey and participating in semi-structured interviews. There will be a $50.00 compensation per dialysis nurse and per dialysis administrator for their time spent participating in the survey and $50.00 compensation per patient, caregiver, dialysis nurse, dialysis dietician, dialysis social worker, physician and advanced practice provider for their time spent participating per semi-structured interview.

Data analysis

Quantitative data analysis

We will use descriptive statistics and comparative analyses to characterise intervention reach, adoption and fidelity within each group and over time.

Survey analyses

Survey scores (means and SD) will be compared between enrollment centres using a paired t-test and a significance level of p≤0.05. We will use generalised estimating equations to model survey scores, adjusting for dialysis centre size and staffing ratios (dialysis personnel/dialysis patient population ratio; and treating providers/dialysis patient population ratio) and accounting for potential clustering effects within dialysis centres.⁴⁸

Reach

Reach will be assessed by the participation rate and the representativeness of study participants. It will be measured as the number/proportion of patients with KDRD eligible for study participation and the number/proportion of eligible patients who consent to study participation.

Adoption

This represents the intention to use the intervention and involves the absolute number, proportion and representativeness of patients willing to adopt the interventions. Adoption will be assessed as the proportion of participants and participating dialysis units with >90% intervention fidelity, in both treatment groups, by the end of the study.

Fidelity

Fidelity will be defined as the consistency of the prescribed and delivered HD prescription according to the randomised intervention and recommendations made by the investigators or the treating providers for the patient’s follow-up in the study. We will assess intervention fidelity using quantitative data and summarise it as a fidelity score based on the Fidelity Monitoring Tool.

Qualitative analyses

All interviews and focus group discussions will be transcribed verbatim, checked for accuracy with corrections made as appropriate, anonymised and imported into NVivo for data management.⁴⁹

Following the principle of constant comparison, transcripts will be analysed as interviews are conducted. First-cycle codes will be derived directly from the data.⁵⁰ Codes will consist of a short phrase generated by the researcher that captures the essence or attributes of a data segment.⁵⁰ We will use open coding and will apply codes to data sections that the analysing researcher deems appropriate. We will use constant comparison to examine the data, both within a given interview and across interviews. We will use NVIVO qualitative data analysis software to analyse the data and record codes. When a substantial number of interviews (~8) have been completed, we will begin searching for themes by analysing the initial codes to determine how the codes can be grouped into themes.

Thematic interpretation will include the development of a coding framework by 2–3 study team members with experience in qualitative research using the Framework Approach.^{51 52} This 5-step process combines (a) familiarisation, (b) developing a theoretical framework, (c) indexing, (d) summarising data in an analytical framework (in this case, using CFIR constructs) and (e) data synthesis and interpretation.⁵² Thematic interpretation will focus on individual and provider factors associated with successful or unsuccessful adoption and/or maintenance of the intervention components at different sites.

Approach to monitoring adaptations in implementing study-related processes of care

The adaptation process will be guided by FRAME (the Framework for Reporting Adaptions and Modifications–Expanded)⁵³ and Framework for Reporting Adaptations and Modifications to Evidence-based Implementation Strategies (FRAME-IS),⁵⁴ which are comprehensive tools that enable documentation of adaptations made to an intervention and/or implementation strategy. The structured measurement and evaluation of the intervention process will provide context for interpreting study effectiveness results and inform replication and scale-up.⁵⁵ These frameworks ask questions related to when during the implementation process the modifications were made, whether the modifications were planned or unplanned and proactive or reactive, the individuals involved in the decision to modify programme elements, and the nature of context (eg, format, setting) and content modifications (eg, tailored, substituted, removed elements).

The transcribed data material will be read twice and then condensed, coded and categorised in a coding scheme constructed to map intervention components, implementation strategies and modifications components from the Adaptation-Impact Framework (AIF).⁵³⁵⁶,⁵⁸ The points of interest in the frameworks are presented in online supplemental table 7.^{33 53 54 58} The coded material will be compared and discussed until a consensus is reached among the reviewers. Afterwards, discussions will be held with the rest of the research team, and results will be developed in the form of general patterns related to adaptations and modifications of the intervention components and implementation strategies.

Ensuring rigor

To enhance rigor and trustworthiness, we will conduct the study and report our findings in accordance with the Consolidated Criteria for Reporting Qualitative Research (COREQ).⁵⁹ We will ensure credibility by attaining prolonged engagement in each interview and continuing interviews until reaching data saturation. We will repeat interviews if/as needed to clarify or expand on any issues not completely explored in initial interviews. We will analyse interviews as they are completed and use constant comparison to test for data saturation.

Internal validity

To increase internal validity, interviews will be audio recorded, professionally transcribed verbatim and checked for accuracy against the corresponding recording. Each interview will be analysed by at least two investigators. To maximise credibility, we will use analytic triangulation, whereby researchers analyse independently and discuss any discrepancies in their codes to arrive at an agreed-upon conclusion. To minimise bias, each interviewer will record field notes regarding details of the interview, the participant, the setting and any thoughts or impressions before and after the interviews, emphasising self-reflexivity (the process by which investigators become aware of their own biases). To optimise confirmability, we will record memos to create an audit trail of all coding decisions.

We will also use peer debriefing with a collaborator to increase credibility. We will engage two specialists with expertise in qualitative research who are not involved in data collection and initial analysis to review the work and evaluate for potential biases, methodological improvements, and/or a varying approach to the data analysis/interpretation.

Triangulation and data synthesis

Our process evaluation is not designed to make immediate changes to the study unless ethical issues require them; early semi-structured interviews and intervention fidelity evaluations will be reported to the site study teams and may be considered in intervention implementation. The aim of the triangulation protocol is to produce meta-themes that cut across individual methods.³⁵

Integration of findings

Aggregate survey responses will be utilised prior to enrollment start-up to evaluate what sites need additional training and coaching on intervention implementation.

During the study conduct, the results of the fidelity monitoring tool will be discussed semi-annually with the study teams at each enrollment centre to continuously optimise intervention implementation. The study teams will then communicate the information to their local treating providers and dialysis personnel.

The process evaluation data stemming from semi-structured interviews and focus group meetings will be analysed before the trial results are known, and the main trial will be analysed independently of the process evaluation findings. Once both analyses are complete, a combined analysis of qualitative and quantitative data will be undertaken to develop explanatory hypotheses about why the intervention appeared to be implemented more successfully in one site than another, and some components were implemented whereas others were not. This may lead us to identify factors that are plausibly and/or consistently related to the successful or unsuccessful delivery of the intervention components and changes to the patient’s experience of care.⁶⁰ Following statistical analysis of trial outcomes, the qualitative data may be re-examined in the light of the trial results to help explain them. If appropriate, the trial statistician may conduct additional analysis to test hypotheses generated from integrating the process evaluation data with the trial outcomes. The final stage of analysis will be to draw together the findings from the broader quantitative analysis to understand why the intervention did (or did not) work in all or some contexts and identify implications for longer-term implementation if appropriate.

Patient and public involvement

No patient and public involvement was sought during the development of this protocol.

Discussion

In recent years, incremental-start HD emerged as a feasible and safe model of patient-centred HD.¹ Clinical trials testing alternative HD models are critical to build the evidence base for safe and effective clinical practice.⁶¹ The TwoPlus study aims to determine the clinical effectiveness of incremental-start HD. In addition, elements of process evaluation, which are part of incremental-start HD implementation, will determine the potential wide-scale operationalisation, adoption and sustainability of the intervention.

Incremental-start HD involves the introduction of distinctive dialysis practices at outpatient dialysis networks. These practices include twice-weekly HD schedules, longitudinal monitoring of residual kidney function and modification of HD prescription based on changes in residual kidney function. In addition, the clinical management at outpatient dialysis units is dynamic (ie, intermittent dialysis sessions per week; laboratory tests done monthly or more often; patient evaluation by nurse personnel at each dialysis treatment; patient evaluation by treating providers four times per month; and patient evaluation by a dietician and social worker at least once per month) and inherently relies on patient-provider dyads and multidisciplinary collaboration. Thus, multistakeholder engagement and bidirectional partnership are imperative to integrate, sustain and evaluate these distinct interventions during implementation preparation, study conduct and dissemination. Our multilevel, partnered intervention will unify multi-stakeholder feedback to inform and efficiently implement incremental-start HD across varied healthcare systems and dialysis organisations.

The hybrid design allows for testing the implementation outcomes and assessing patient clinical outcomes. With the combined use of two implementation frameworks, that is, RE-AIM and CFIR, we aim to comprehensively incorporate a process evaluation to identify salient factors that facilitate or impede successful implementation in standard clinical practice. We will gauge diverse perspectives from the clinic implementers, patient and caregiver participants, and clinic leaders regarding incremental-start HD in comparison to conventional HD. The process evaluation process will allow us to better understand the contextual factors that may influence the future implementation and maintenance of incremental-start HD in other similar clinical settings.⁶² Insights obtained through semi-structured interviews and focus group meetings will be used to optimise recruitment strategies, patient monitoring processes and intervention incorporation into usual workflows at outpatient dialysis centres, while ensuring that the core components of the study protocol, listed in online supplemental table 7, remain unchanged. Furthermore, longitudinal monitoring of intervention adherence throughout the study is important, particularly given the use of an intent-to-treat analysis. The Fidelity Monitoring Tool was developed during the study design phase with input from investigators and stakeholder partners, focusing on key elements essential for intervention delivery and patient safety, and aligning with the core components of the study protocol. This tool was created specifically for this study and has not been used previously. The investigators and stakeholder partners considered that a 90% fidelity rate would represent satisfactory intervention adherence. High adherence would ensure that the study population closely mirrors those receiving the intervention as intended, thereby strengthening the evaluation of the intervention’s effectiveness.

We point out that the TwoPlus research proposal used the terminology of KDRD in lieu of ESKD for several reasons. The decision was influenced by the 2020 KDIGO consensus conference, which highlighted concerns about the current nomenclature for kidney disease based on patient-reported views on terminology usage.⁶³ A key takeaway from this conference was the widespread dissatisfaction with terms like ‘end-stage’ or ‘dialysis dependence’” as descriptors of kidney disease. These terms are considered demoralising and stigmatising, carrying negative connotations for patients. By transitioning from the ESKD framework to KDRD, which incorporates stages of disease evolution akin to chronic kidney disease stages, the hope is to encourage a shift away from the binary mindset of disease present/absent toward a perspective that recognises the ongoing evolution of kidney function even after dialysis initiation.⁶⁴ Using terminology without the ‘end-stage’ undertone may facilitate greater acceptance of dialysis prescriptions tailored to individual patient needs.⁶⁵

Our study has several strengths. First, we are using a combination of qualitative and quantitative methods, which will be triangulated to provide a coherent and comprehensive picture of factors that influence the adoption of TwoPlus intervention. The process evaluation process includes individual, interpersonal and organisation/health facility-level determinants. This will provide insights to inform the real-world scalability of the clinic intervention. Second, the intervention implementation takes a collaborative approach by engaging various key stakeholders at various stages of the intervention implementation. Third, the research team working on the process evaluation is different from the main trial team, which increases the chance that the interviewer will be blinded to the intervention during interviews with patient participants.

Limitations and complexities of this study are worth noting. While we will provide objective fidelity data, incremental-start HD requires resources beyond this study’s scope, so the results might not be representative of how the intervention will be sustained after the trial. Large-scale adoption and sustainability of incremental-start HD will depend on changes in dialysis reimbursement model. The study will be implemented across varied national healthcare systems and dialysis organisations, which may bring substantial variations in the implementation process, including organisational resources, culture, staffing and workflow. As part of the process evaluation, we will record the variations in the implementing sites and dialysis clinics and any adaptations that will be implemented in the study sites. The study will operate through the workflow of the participating sites, which are subject to the clinic guidelines and may influence the fidelity of the intervention. Changes in the US bundled healthcare payment for dialysis, planned to be released in 2025, may impact the provision of phosphate binders and other oral medications by the dialysis providers. Although the study does not specifically track the actual amounts of medications the patient will take at home, prescribed medications will be documented, with any differences between treatment groups noted in the analysis. Finally, the process evaluation, encompassing surveys and interviews, necessitates a dedicated time commitment from the staff, thereby influencing the extent of their participation. This study will provide important information about implementing an innovative incremental-start HD and potential factors that will influence the scalability of the intervention. This study will also contribute knowledge on the barriers and facilitators to stakeholder engagement in a wide-scale clinic study. These insights can inform the development of future implementation strategies in this field.

Ethical considerations and dissemination

The study has received approval from the Wake Forest University School of Medicine Institutional Review Board (IRB Number:00092986; IRB Director Email: jbmoore@wakehealth.edu). All protocol modifications will be communicated to relevant parties through appropriate systems/channels. The process evaluation results will be published in peer-reviewed journals and presented at various meetings and conferences.

Ethical approval

The study protocol has received ethical approval for all enrollment centres participating in the Two Plus study and encompasses the components of process evaluation. Through stakeholder engagement (ie, patients, caregivers, physicians, advanced practice providers, dialysis nurses, dialysis dieticians and dialysis social workers) and dissemination of findings globally, we hope to build on a patient-centred and sustainable model of incremental-start HD.