Dietary Counselling Interventions in Malnutrition Research: Achieving an International Consensus on Best Practices Using an Amended Delphi Process

ABSTRACT

Introduction

The evidence for the use of dietary counselling interventions in the management of malnutrition is inconsistent. Lack of consistency limits the ability to compare studies, impacting research, clinical practice and policy development.

Aims

To establish an international consensus on minimum requirements for dietary counselling interventions used for the prevention and management of malnutrition in adults in a clinical research context.

Methods

An international expert panel with seven members was formed. An amended Delphi study was conducted with 37 initial consensus statements generated from a targeted search of the literature. The agreement was assessed across two Delphi rounds and two online meetings. The full consensus was defined as the assignment of a score of 5 (strongly agree). A preliminary external validation was conducted with clinical and/or research dietetic professionals when delegates at the International Congress of Dietetics (ICD) 2021 voted during an online meeting on agreement with statements relating to patient assessment and delivery of a dietary counselling intervention.

Results

Consensus was achieved for 27 statements across four themes, with justifications and commentary. Minimum requirements included eight statements for ‘nutritional assessment’ (theme 1), seven for ‘nutritional intervention content’ (theme 2), five for ‘nutritional intervention delivery’ (theme 3) and seven for ‘monitoring and follow‐up’ (theme 4). The ICD delegates voted on elements of assessment (77 of 140 delegates in agreement) and delivery of the intervention (37 of 101 delegates in agreement), demonstrating that consensus on these aspects of nutritional support interventions was low to moderate in this delegate sample.

Conclusions

Consensus was achieved by an international group of experts to define the minimum essential factors for delivering dietary counselling interventions for the management of malnutrition in a research context; however, further external validation is required. The minimum requirements have the potential to influence research, clinical practice and policy development.

Summary

• Lack of consistency in the content and description of dietary counselling interventions for the management of malnutrition limits the ability to compare studies, impacting research, clinical practice and policy development.

• An international group of seven experts achieved consensus on 27 statements across four themes essential for delivering dietary counselling interventions in the research context. Minimum requirements included eight statements for nutritional assessment, seven for nutritional intervention content, five for nutritional intervention delivery and seven for monitoring and follow‐up.

• Preliminary validation of the consensus statements undertaken at an online international dietetic conference demonstrated that consensus on patient assessment and delivery of the intervention was low to moderate in this sample of clinical and research dietitians. Further validation is required through the circulation of the final statements to clinical researchers, external experts and representatives of professional bodies for peer review.

1. Introduction

Evidence‐based practice to address protein‐energy malnutrition (herein referred to as malnutrition) is currently undermined by a lack of consensus on the terms used to define and describe dietary counselling interventions in research studies. During the conduct of a recent Cochrane review [1], it became evident that despite the inclusion of 94 randomised controlled trials (RCTs) including more than 10,000 participants, there was no convincing evidence for dietary counselling provided with or without oral nutritional supplements (ONS). For the purposes of the Cochrane review, the terms dietary advice or dietary counselling (herein referred to as ‘dietary counselling’) were defined as ‘instruction in the modification of food intake given with the aim of improving nutritional intake by a dietitian or other healthcare professional’. Scrutiny of the studies included in the review revealed that the term dietary counselling was defined and used inconsistently within the RCTs. Supporting Information S1: 1 provides examples of the inconsistencies found in one section of the Cochrane review that evaluated the effect of dietary advice with or without ONS [1]. This section of the review included 24 studies with 3523 participants where the comparison was made between dietary advice and no advice. The examples in Table S1 have clearly demonstrated inconsistencies in the terminology used to describe the intervention, the content and delivery of dietary counselling interventions including the materials and procedures used, modes of delivery, who provided the intervention and the length and intensity of the intervention. A subsequent review evaluating 58 dietary counselling RCTs found that the reporting quality of oral nutrition interventions remained suboptimal, with the expertise of the intervention provider, intensity and adherence to the intervention being the most frequently inadequately described [2].

Malnutrition has been defined as a state resulting from a lack of intake or uptake of protein and energy that leads to altered body composition and body cell mass leading to diminished physical and mental function and impaired clinical outcome from disease [3, 4, 5]. It can occur at any stage in life but, for the purposes of this study, the focus was on adults (≥ 18 years) in clinical settings. Although the exact prevalence of malnutrition is unknown, its reported prevalence in adults ranges from up to 10% in the community and up to 70% or more in hospitals, residential aged care and rehabilitation [6, 7, 8]. Malnutrition results in severe downstream consequences, including increased morbidity, complications, institutionalisation, hospitalisation and mortality; accompanied by overburdening of health and care systems and high healthcare costs [9, 10, 11, 12].

One of the most used strategies to address malnutrition is dietary counselling with or without the use of ONS; however, despite its widespread use in clinical practice, research supporting the efficacy of dietary counselling is conflicting as demonstrated in the recent Cochrane review [1]. The observed differences in outcomes may result from patient‐ and healthcare setting‐related factors such as differences due to stage and extent of disease, differences in the funding and the resourcing of nutrition and dietetic services, but may also be due to how the dietary counselling intervention was delivered, and the components of the care process that were considered part of the intervention being evaluated in clinical research (e.g., assessment, diagnosis, care planning and delivery, monitoring and evaluation) [1, 2, 4, 13, 14]. Other factors that may contribute to the differences include who delivered the intervention (e.g., dietitian, nurse, physician); their qualifications and experience for effective delivery; and how the intervention was delivered (e.g., mode of delivery, frequency and number of follow‐up contacts) [1, 15]. These inconsistencies in the understanding of dietary counselling interventions may have contributed to the variations in its use in clinical practice.

While the Cochrane review showed potential benefits of dietary counselling with or without ONS in terms of weight gain [1], the inconsistencies summarised in Table S1 mean it has been impossible to fully or accurately evaluate interventions that include dietary counselling to determine its effectiveness in managing malnutrition in adults in clinical settings. This has the potential to negatively impact on patient care and in turn may negatively impact policy, funding and perceptions of the value of dietetic care. The lack of agreed definition, in addition to undermining the evidence base, may also be obscuring variation in the effectiveness of dietary counselling interventions resulting from different methods and/or modes of delivery. When conducting clinical research there is a need to balance scientific rigour with the potential for translation of the research into clinical practice. This is especially relevant when conducting research on interventions, such as dietary counselling, which involves individualisation to specific patient needs [16], a fundamental component of providing dietary care according to the Nutrition Care Process [17].

To enable accurate translation into clinical practice as well as comparability of research studies, it is essential that consensus is gained on the minimum requirements for an intervention targeting malnutrition to be termed dietary counselling and that the interventions themselves are described as fully as possible when conducting malnutrition intervention research. The benefits of consensus in research may be fourfold: (1) to assist dietetic researchers to design RCTs to evaluate the benefits of different dietary counselling interventions; (2) to enable systematic reviewers to fully evaluate the impact of dietary counselling on patient‐centred, functional and clinical outcomes, as well as cost‐effectiveness; (3) to control the type of nutrition intervention to allow examination of other predictive variables such as clinical condition, composition of nutrition support teams, care setting, family care networks, intensity of intervention and training and experience of staff; and (4) to allow policymakers and governing bodies to create funding models which avoid healthcare waste and properly fund effective dietary interventions. A consensus may also guide dietitians in clinical practice to provide optimal care by clearly translating the nutrition care process to the dietary management of malnutrition.

Therefore, the goal of this research was to establish an international consensus on minimum requirements for dietary counselling interventions used in research for the prevention and management of malnutrition in adults in clinical settings.

2. Methods

An amended Delphi process was used to achieve the research goal of gaining consensus between experts. Defined as ‘a method for achieving consensual agreement among expert panellists through repeated iterations of anonymised opinions and of proposed compromise statements from the group moderator’ [18], the Delphi process was adapted for online circulation of questionnaires [19] and is summarised in Figure 1. Developed by the RAND Corporation in 1948, the basic methodology of the Delphi process has remained essentially the same since its inception [20]. The respondents are to be experts in the relevant field, their responses are to remain anonymous, there must be two or more rounds before a consensus can be reached, and finally responses from each round should be analysed by researchers and reported back to respondents after each round.

Figure 1.

Summary of the amended Delphi process utilised to gain international consensus on dietary counselling interventions for malnutrition in a research context.

2.1. Convening the Expert Panel

From September to December 2019, dietitian experts were invited to take part in the amended Delphi process. Experts for this project were dietitians with a strong history of peer‐reviewed publications in the topic area (dietary counselling interventions for the management of malnutrition), who were invited to become members of the expert panel by the lead authors (C.E.W. and C.B.). Eligible experts were identified based on international reputation and/or recommendation of existing panel members. Five experts were approached by the lead authors, both face‐to‐face during the ESPEN Congress 2019 and by email, none of whom declined. The aim was to achieve global representation to take account of variations in clinical practice that might arise in different types of healthcare economies. The expert panel comprised seven members, which is in accordance with the recommended panel size for the Delphi process (6–11 panel members) [21].

2.1.1. Expert Panel

The expert panel comprised dietitian researchers from six countries across four continents: Dr Elizabeth Weekes & Dr Christine Baldwin (UK; Europe); Prof Marian A. E. de van der Schueren (The Netherlands; Europe); Prof Heather Keller (Canada; North America); Dr Alison Steiber (USA; North America); Dr Skye Marshall (Australia; Oceania); and Dr Su Lin Lim (Singapore; Asia). Although experts were sought to represent the continents of Africa and South America, none were identified which fulfilled the eligibility criteria. Email communications between members of the expert panel during February and March 2020 were used to confirm membership of the expert panel, understanding of the brief and to arrange the preliminary meeting.

2.1.2. Ethics

Ethical approval was deemed unnecessary for this initial phase of the project since only members of the authorial team were involved in the Delphi Rounds. Each expert confirmed their willingness to contribute to the Delphi Rounds on initial invitation to take part in the project. The authors acknowledge that validation of the consensus document will, in the future, require ethical approval.

2.2. Preliminary Meeting

An online meeting was convened with the expert panel on 7 May 2020 to discuss the scope, aims and objectives of the project, to discuss the design and conduct of the literature review and to confirm the process used to derive terms, definitions and concepts to be submitted for consensus‐building. At the same time, the panel members established the criteria to be used to determine the strength and stability of consensus and confirmed the validation strategy.

Inclusion criteria were agreed as: (1) dietary counselling interventions aimed at achieving behaviour change, (2) the intervention goal was the management of malnutrition, (3) intervention recipients were adults (aged ≥ 18 years), (4) the context was clinical research. There was no restriction on intervention setting (community, hospital, geographical region, etc). Exclusion criteria were enteral and parenteral nutrition interventions; dietary counselling given in the context of clinical practice (i.e., not as part of clinical research) and ONS as a standard prescription (i.e., ONS without any dietary counselling).

2.3. Initial Consensus Statements: Literature Review

Under the guidance of the lead authors, a targeted literature search [22] was undertaken from June to August 2020 by an external independent researcher who was not part of the expert panel (Figure 1). The aim of the review was to scope definitions of dietary counselling (and related terminology such as dietary advice) and to identify the essential or desirable components of dietary counselling interventions required to achieve behaviour change. PubMed was searched using the Clinical Queries function for systematic reviews of dietary counselling and malnutrition, for local and international clinical guidelines and for pathways on the management of malnutrition. Reference lists of relevant publications were scrutinised for additional reports of interest and the ‘related citations’ function in PubMed was used to identify additional but related reports of interest for key papers. A detailed examination of 94 studies of dietary counselling interventions included in the Cochrane review [1], that used the Template for Intervention Description and Replication (TIDieR) checklist to extract detailed information, was consulted [15]. The aspects identified as relevant to dietary counselling were compared, collated and discussed by the independent researcher and the lead authors (C.E.W. and C.B.) and organised into five themes (Nutritional assessment, Nutritional intervention, Delivery of intervention, Coordination of care and Monitoring) and potential consensus statements based on the Nutrition Care Process were mapped to each theme [17] (Figure 1). The literature‐informed draft themes and statements were presented to the full expert panel during an online meeting on 2 September 2020, and the final content of the first Delphi questionnaire, including the initial consensus statements, was agreed by all expert panel members.

2.4. Delphi Rounds

2.4.1. Round 1

After the online meeting, the first Delphi questionnaire (Supporting Information S2) was circulated to the full expert panel on 2 September 2020 and returned to the project lead by all panel members by 18 September 2020. Experts were asked to rate each statement using a 5‐point Likert scale (ranging from 1 ‘strongly disagree’ to 5 ‘strongly agree’) regarding their importance as aspects to be included in research studies of dietary counselling interventions for malnourished patients. All statements had space for comments to qualify responses made and each of the five themes had space for additional comments. Responses to all questions were collated, analysed and summarised by the project lead.

2.4.2. Round 2

The Round 2 questionnaire (Supporting Information S3) comprised statements from Round 1 that met the agreement criteria of full consensus (for comment only) and statements where full consensus was not achieved during Round 1. For the statements for which consensus had not been reached, participants were invited to reconsider their Round 1 scores in light of anonymised written comments made by other group members. The 2nd questionnaire was circulated to the full expert panel on 16 April 2021 and returned to the project lead by all panel members by 30 April 2021. Responses to all questions were collated and summarised by the project lead according to the methods of Holey's [23].

2.4.3. Consensus Meetings

Following Round 2, the full expert group met online on 2 and 21 June 2021 to discuss all statements for which 100% consensus had not been reached. Discussion occurred until agreement was reached on the inclusion, modification or exclusion of each statement. This was an iterative process of organising responses, which involved the elimination of some aspects judged to be unnecessary and the merging of other responses. At the same time, supporting comments were discussed and agreed for each statement. Both meetings were recorded and transcribed by one researcher (C.E.W.) and then the transcription was validated by a second researcher (C.B.) listening to the recordings and any differences in understanding were resolved by discussion (Figure 1).

2.5. Data Analysis

On receipt of each questionnaire (Delphi Rounds 1 and 2), the project lead compiled an Excel database of the scores assigned by panel members for each statement, together with all comments made about each statement, and any additional comments made at the end of each section theme.

Statements were defined as meeting 100% consensus when all panel members assigned a score of 5 (strongly agree). For all statements where 100% was not achieved, the median score (range) was calculated. Disagreement was defined as a median score ≤ 3.

2.6. Preliminary External Validation

The first step towards external validation of the proposed consensus statements took place during an online symposium at the International Congress of Dietetics (ICD) on 2 September 2021. The ICD attracts delegates from across the world with clinical or research experience, or both, and a wide range of years in the profession and clinical specialties; the symposium was free to attend for all delegates (i.e., no additional fee to attend); however, attendance was optional for all delegates and not part of the main plenary programme. The aim was to determine if consensus could be achieved on the clinical components of nutritional support interventions for malnutrition in adults, assuming there might be differences in responses on the other themes due to different healthcare economies, clinical expertise and dietetic autonomy. Delegates were presented with a summary of the project and asked to poll online on the inclusion of relevant concepts. The results from polling were presented as the number of positive responses for inclusion of specific components.

3. Results

The final list of 27 statements across four themes, together with a rationale for each statement and commentary is presented in Table 1. The process by which these consensus statements were derived and agreed is described below.

Table 1.

Consensus statements, rationales and commentary.

Statement	Rationale	Commentary (including examples)
1. Nutritional assessment for a dietary counselling intervention for the management of malnutrition should include:
1.1 Anthropometric measurements and use of composite screening and assessment tools, for example, weight, height, mid‐arm circumference, triceps skinfold thickness, bio‐electrical impedance analysis (BIA), dual X‐ray absorptiometry (DEXA) or CT scans (ADA, BDA, BAPEN, ESPEN)	1. To provide a full description of participants nutritional status and body composition at study baseline. 2. To measure the impact of the intervention on nutritional status and body composition.	The consensus group recommends the use of validated nutrition screening and assessment tools or processes. Clinical judgement will be required to determine the appropriate method to use based on clinical condition, care setting, resources and time available in the context of the research question. Body weight alone is insufficient as the sole determinant for establishing nutritional status and researchers should consider the interpretation of data on history of weight change, including the potential impact of changes in fluid status. Serial body composition measurements will provide important data on the impact of the intervention on changes in fat mass and lean body mass.
1.2 Physical examination for features of malnutrition as indicated by the research question (BDA, ADA etc.)	To provide data on participants 1. Fluid status. 2. Body composition. 3. Clinical manifestations of micro‐nutrient deficiencies as indicated by the research question.	Consider: The potential impact of shifts in fluid status leading to misinterpretation of weight change. Loss of muscle mass associated with poor outcomes. Loss of fat mass associated with increased risk of pressure sores. Micronutrient deficiencies might impact some clinical outcomes, for example, wound healing, infections and so forth.
1.3 Review of relevant biochemical and laboratory data as indicated by the research question (BDA, ADA etc.)	To provide data on participants 1. Clinical condition. 2. Fluid and/or electrolyte status. 3. Vitamin and mineral status as indicated by the research question.	Clinical judgement will be required to determine whether laboratory data are required to answer the research question. For example, laboratory data may be necessary; To describe the participant's clinical condition, for example, disease severity, inflammatory status and fluid status. To assess the risk of the re‐feeding syndrome (electrolytes & micronutrients) in the included population. To explore dietary quality (vitamin and mineral status) in the context of malnutrition or if participants are likely to be at risk of specific deficiencies, for example, vitamin D in housebound. Access to relevant lab data may be limited in some care settings, for example, GP clinic or community services and may not be required to answer some research questions. The inclusion of laboratory data may incur additional costs for the study and additional blood tests for the participant (over and above usual care).
1.4 Review of medical and clinical history and assessment of clinical and therapeutic factors as indicated by the research question (BDA, ADA etc.)	To provide data on participants 1. Clinical history. 2. Factors that might impact future dietary intake as indicated by the research question.	Clinical judgement will be required to determine which clinical data are needed to answer the research question, for example, previous, current or future medications or treatments, presence of inflammation. Particularly relevant if the population has conditions that affect food intake, for example, dysphagia, or are likely to receive treatments or medications that might impact on dietary intake or nutritional status, for example, major GI surgery, chemotherapy or bone marrow transplant.
1.5 Assessment of current dietary intake with reference to previous intake as indicated by the research question (BDA, NICE etc.)	1. To provide baseline data on current dietary intake (in comparison with previous intake) as indicated by the research question. 2. To establish likely causes of reduced dietary intake. 3. To measure the impact of the intervention on dietary intake.	Dietary counselling needs to provide individualised advice, and this cannot be done without first assessing the current diet, which facilitates understanding of someone's previous food habits and how much they might have changed or will be impacted by any subsequent intervention. Diet history is not always possible due to time constraints or the capacity of the participant. Some assessment of previous dietary intake may be necessary to inform how quality and quantity of diet has changed. The method chosen for assessment of intake will be determined by the research question and resources available but the importance of using validated techniques, for example, review of food record charts or multiple pass dietary recall, diet history or dietary diary is emphasised.
1.6 Assessment of psychological, social and environmental factors likely to impact dietary intake (ADA, ASPEN, BAPEN, BDA, DAA, ESPEN)	1. To provide baseline data on psychological, social and/or environmental factors that might impact on dietary intake and/or adherence as indicated by the research question. 2. To identify factors that might be impacting on dietary intake.	Psychological and social barriers to eating and drinking will need to be documented if they are likely to impact dietary intake and/or affect adherence to the intervention. Local protocols should be followed to mitigate any issues identified. When assessing environmental factors that impact on dietary intake it is important to distinguish between factors that are specific to individual (e.g., lack of cooker) which is amenable to dietary intervention or factors that are more specific to the care setting (e.g., drug round interruptions during meals) which are not amenable to a dietary counselling intervention. Assessment of these factors are particularly important in situations of food insecurity and when safety of preparation is a consideration. The use of validated tools (e.g., food insecurity tools) if available is recommended.
1.7 An estimate of energy, macro and micronutrient requirements and fluid, as indicated by the research question (BDA, NICE etc)	1. To provide baseline data on participants estimated nutritional requirements for comparison with dietary intake data and/or dietary reference values.	Understanding nutritional requirements is important: To determine the gap between the current intake and requirements. To enable development of the diet prescription. For monitoring of changes in nutrition intake during the study. Estimation of energy and protein requirements is the minimum necessary for an intervention in the management of malnutrition. An estimate of micronutrient requirements is likely to be necessary if there is suspicion of micronutrient deficiency or if the intervention involves supplementation with vitamins or trace elements. Comparing dietary intake data with estimated requirements will provide data that might inform how well participants adhered to the intervention.
1.8 Specify the proportions of participants identified as at‐risk or malnourished and the criteria used for categorisation of nutritional (risk) status (ADA, ASPEN, BDA, ESPEN)	1. To provide data on the nutritional (risk) status of the study population at baseline. 2. To measure the impact of the intervention on nutritional (risk) status.	This represents a synthesis of the information gathered in the previous sections and a statement of the proportions of the population in each nutritional (risk) category. Such a statement lets the reader and systematic reviewers know the proportion of participants in each nutritional (risk) status category. Research papers should include a clear statement on how malnutrition has been defined.
2. A dietary counselling intervention for the management of malnutrition is not a ‘one‐off’ event but is a process that takes place over multiple sessions. The intervention should include:
2.1 More than one counselling session based on the outcomes of the nutritional assessment and as indicated by the research question	1. At least two sessions are necessary to assess if participants adhered to the intervention provided at baseline. 2. Research is needed to determine the optimal intensity and/or frequency of contacts to achieve behaviour change.	To facilitate and promote behaviour change, dietary counselling should not be seen as a one‐off event but as a process that ideally takes place over several sessions. Currently, it is not possible to determine the optimal intensity and/or frequency of contacts needed to achieve behaviour change in dietary counselling interventions.
2.2 Engagement with the participant and/or carers in a conversation to identify, set and monitor nutritional goals and to identify supportive actions to promote facilitators and reduce barriers to intervention	1. To provide data on patient‐relevant nutritional goals. 2. To identify (and support) potential facilitators of behaviour change. 3. To identify (and address) potential barriers to behaviour change as indicated by the research question.	Work with participants and/or carers to set goals will help to facilitate achievement of participant‐focused expected outcomes. Goals should be: Tailored to what is reasonable based on a participant's circumstances. Appropriate to expectations for treatments and outcomes. Goals need to be SMART and designed to resolve or improve the nutrition and dietetic diagnosis, its aetiology and/or its signs and symptoms.
2.3 Education regarding the need for nutritional support and the potential consequences of malnutrition	To provide participants with a clinical rationale for the nutritional intervention as indicated by the research question.	The type and amount of education provided will need to be tailored to the participant's likely understanding of the concepts involved and needs to be fully described. May need to include family, friends or carers if the participant lacks the capacity to understand the educational components of the intervention and/or has communication difficulties. Input of family, friends and carers needs to be fully described.
2.4 Should be individualised to the participant's specific needs (ADA, ASPEN, BAPEN, BDA, DAA, ESPEN, NICE)	To ensure the intervention meets the specific needs of the participant as indicated by the research question.	Dietary counselling involves negotiation with the participant about what is ideal versus what might be possible for them. The participant‐centred care approach is essential and the types of adjustments made for individuals need to be fully described. Needs to recognise if the participant is at a relevant stage of change and assess motivation. The use of validated tools to assess readiness to change is recommended, and their use should be fully described.
2.5 Any combination of the following strategies as indicated by the research question (and the participant's needs) (a) Advice on fortification of food and drink to enhance nutritional content, use of supplements or medical nutrition as appropriate (BAPEN, ESPEN) (b) Advice on shopping and meal preparation, menu planning and/or other social and environmental strategies where appropriate (BAPEN, BDA, ESPEN) (c) Advice on the appropriate use of ONS (d) Advice to staff and/or carers on the provision of practical support, for example, feeding assistance and verbal encouragement (e) Counselling using validated psychological techniques to support behaviour change	To provide a ‘toolbox’ of intervention strategies with the choice of intervention being informed by an assessment of the likely causes of inadequate dietary intake.	The intervention should: Comprise of any combination of relevant strategies as indicated by the research question. Be individualised to each participant's nutritional requirements, dietary habits, nutritional status, clinical condition, support networks and access to resources. Promote the use of validated tools and techniques in its delivery, for example, motivational interviewing. Should be fully articulated in the research protocol and described in any subsequent papers. Some interventions are setting specific and dependent on the presence of relevant support structures. a. Food fortification might take place in hospitals, care homes or other residential units and this will require close collaboration with the catering department. Participants and/or their carers might also be advised and supported with food fortification at home. When used in a research context, the development and testing of fortified recipes should be fully described in the protocol and any subsequent research papers. b. Modifications that are specific to the individual, for example, mobility issues limiting shopping opportunities or lack of cooking equipment, may be amenable to a dietary counselling intervention. c. Clear description is required of how much, when and how ONS were provided including details of any resources provided, for example, recipe leaflets, mixing containers or straws. d. Family, friends or carers may require advice and tips on feeding assistance and/or how to provide verbal encouragement. e. Environmental modifications likely to maximise dietary intake may be relevant in any care setting, for example, minimisation of drug round interruptions on hospital wards or noisy conversations in care home during meals or changing the dining environment to be more conducive to eating. f. Clear statement needed on any validated psychological techniques used, for example, motivational interviewing, cognitive behavioural therapy and a rationale for their use.
2.6 Appropriate inclusion of family and carers in communications on participant nutritional care (ADA, ASPEN, BAPEN, BDA, DAA, ESPEN, NICE)	People with malnutrition often have complex support networks and working with existing networks can have a crucial role in making the intervention successful.	Often necessary when participants have cognitive, sensory, communication or learning difficulties. While this may not always be achievable in the hospital setting, the inclusion of family and carers during the intervention may be crucial to facilitating participant compliance on discharge or if they are seen in the community.
2.7 Resources are provided in appropriate language or format and at relevant levels to summarise and support counselling, advice and education where appropriate (ADA, ASPEN, BDA, DAA, ESPEN)	To support the participant to remember and act on any advice provided.	Written resources are often necessary to support the participant after leaving a meeting that might be lengthy and includes complex information (or of short duration in an acute setting before discharge). Provision of materials to participants may be a weblink, summary email, booklet, a sheet of paper with 3‐−4 instructions written out or something pictorial. Researchers should consider participant's cognitive ability, access to and ability to use relevant IT. Consider any specific learning, sensory or communication difficulties. Consider participant and stakeholder input in the design, content, validity testing, feasibility and piloting of new resources.
3. Delivery of a dietary counselling intervention for the management of malnutrition should:
3.1 Have a defined location, duration, frequency and mode of intervention that is clearly articulated to the participant and/or carer at the outset and is fully described in the research protocol (BDA etc.)	To promote replicability of the study by providing a full and accurate description of each component of the intervention.	Location, duration, frequency and mode of delivery (e.g., face‐to‐face, telephone calls, group sessions, online) may all individually or in any combination impact on the effectiveness of the intervention. When conducting systematic reviews, it is often difficult to locate and extract data on how, where and when the intervention was delivered, making comparisons between different interventions almost impossible. It is important to be able to understand how context impacts choice of intervention delivery, for example, changes in delivery of interventions due to the COVID pandemic. Therefore, researchers should fully describe the location, duration, frequency and mode of delivery of the intervention.
3.2 Includes documentation of the nutritional care provided as per professional and institutional requirements that includes location, duration, frequency and format of treatment, onward referral on discontinuation (if appropriate) and dropouts (date and reason if known) according to the participant's needs and/or research protocol (ADA, ASPEN, BAPEN, BDA, DAA, ESPEN, NICE)	1. To adhere to professional, ethical and legal requirements to document care. 2. To facilitate communication with other HCPs involved in the participant's care.	Documentation of the details of a dietary intervention in medical, dietetic and care records should be provided: To facilitate communication with other HCPs involved in the participant's care. Should include details of nutritional risk screening, diagnosis, assessment of risk factors, nutritional requirements, nutrition therapy, goals and outcomes for nutrition therapy, including estimated time to reach goals, as well a note of who is responsible for the follow‐up. Documentation may be on paper, in electronic record systems or web‐based (with appropriate information governance protections). Local and national (professional) standards may apply.
3.3 Liaison (written, verbal or electronic) with relevant health and/or social care professionals where necessary to promote the implementation of the nutrition care plan (ADA, ASPEN, BAPEN, BDA, DAA, ESPEN, NICE)	To facilitate any support required from other HCPs to implement the care plan.	Communication of the treatment plan to the participant/carer and relevant health and social care professionals is necessary to facilitate multi‐disciplinary care planning, treatment and monitoring.
3.4 Onward referral to other health and social care professionals where appropriate in accordance with local and/or national policies and access at the end of the research intervention	To inform management plans for subsequent nutritional care.	Clinical judgement of the research dietitian in consultation with the multi‐disciplinary team will inform decisions about referral to other services and should be carried out in accordance with local policies and consideration of availability of appropriate services.
3.5 An agreed plan for the end of the intervention or early discontinuation	To mitigate any risks identified during the course of the study.	Participants and staff need to know the circumstances under which the research intervention will be expected to end and any circumstances where the intervention might need to end early. A clear and comprehensive exit strategy should be included in the research protocol and fully articulated to participants and/or their carers.
4. Monitoring of a dietary counselling intervention should follow the nutritional assessment and treatment plan and should include:
4.1 Documentation of monitoring arrangements—who, when, where, frequency, what and how—as indicated by the research question	1. To promote replicability of the study by providing a full and accurate description of each component of the intervention. 2. To monitor the impact of the intervention.	Judged to be important: To comply with R&D governance requirements. To ensure that successful interventions can be replicated in a research context or translated into clinical practice.
4.2 Documentation on how agreed goals have been achieved, modified or adhered to	To monitor the impact of the intervention.	Necessary to determine if a participant reaches their goals. Failure to achieve goals and reasons for poor(er) adherence should be explored. This should be a part of the behaviour change/counselling model/framework chosen. In some models it would be a core part; but in others it may not be so relevant.
4.3 Assessment of changes in dietary intake and behaviour	To measure the impact of the intervention on dietary intake (macro‐ and/or micronutrients) as indicated by the research question.	Important to promote the use of validated methods to assess changes in dietary intake and food‐related behaviour. Adherence to the various components of the intervention should be assessed and described separately, for example, adherence with ONS provision described separately from adherence to advice provided on snack intake.
4.4 Assessment of anthropometric, body composition and/or biochemical measurements where appropriate to assess effectiveness of nutritional intervention (ADA, ASPEN, BAPEN, BDA, DAA, ESPEN, NICE)	To measure the impact of the intervention on body composition and biochemical status as indicated by the research question	Important to promote use of standardised equipment, for example, weighing scales, stadiometers and standard operating procedures for measurement, for example, measurement of mid‐arm circumference or skinfold thickness, lab‐defined ‘acceptable ranges’ for biochemical and other laboratory data.
4.5 Assessment of changes in clinical, environmental, functional, psychological or social factors likely to affect nutritional status and/or dietary intake	To measure the impact of the intervention on relevant outcomes: Clinical Functional Participant‐centred Economic	Important to promote the use of validated tools to measure nutritional, clinical, functional, QoL and composite outcomes, for example, QoL, ADL, nutritional assessment tools, symptom severity tools, short physical performance battery
4.6 Feedback to the participant/carer on behaviour changes and outcomes	To support the participant in maintaining/modifying/developing relevant onward goals. To support subsequent participant management.	Positive feedback to reinforce beneficial changes in behaviour and any positive outcomes, for example, meeting participant‐identified goals. Feedback and discussion around barriers and facilitators should be a 2‐way process.
4.7 Documentation of all assessments and any changes to the nutrition care plan, subsequent monitoring arrangements or onward referral	1. To adhere to professional, ethical and legal requirements to document care. 2. To support future audits of research process.	To comply with R&D governance requirements.

3.1. Literature Review

To identify the essential or desirable components of dietary counselling interventions required to achieve behaviour change in a research context, recommendations on the detection and management of malnutrition in clinical settings were reviewed from the following clinical guideline development groups: American Dietetic Association, American Society for Parenteral and Enteral Nutrition, British Association for Parenteral and Enteral Nutrition, British Dietetic Association, Dietitians Australia, European Society for Clinical Nutrition and Metabolism and National Institute for Clinical Excellence. Recommendations were mapped to relevant components of the Nutrition Care Process [17], and the five themes that appeared most consistently across the literature from diverse countries and healthcare systems. Systematic reviews of dietary counselling interventions were reviewed to identify any further components that might only be relevant in a research context, for example, outcome assessment. Within the five themes, a total of 37 statements were drafted for inclusion in the first Delphi questionnaire (Supporting Information S2; Figure 2).

Figure 2.

Summary of themes and statements from Delphi rounds 1 and 2.

3.2. Delphi Round 1

A summary of the consensus statistics and resulting actions from Delphi Rounds 1 and 2 is presented in Table 2. Following Delphi Round 1, eight of 37 (22%) statements achieved a complete score of 5 (100% consensus, i.e., all panel members assigned a rating ‘strongly agree’) (Table 2). At this stage, there was 100% consensus on the use of anthropometric measurements and review of relevant medical history, clinical, therapeutic and other factors in theme 1 (nutritional assessment). In both theme 2 (components of a dietary counselling intervention) and theme 3 (delivery of intervention), 100% consensus was not achieved for any of the statements. In theme 4 (coordination of care), there was 100% consensus on the need for full documentation and appropriate inclusion of family and carers. In theme 5 (monitoring), there was 100% consensus on the need to provide feedback to patients and/or carers on behaviour change and outcomes. Since 100% consensus was not achieved in the remaining 29 statements, these were included (with comments from all panel members) in the Round 2 questionnaire (Supporting Information S2). The statements where there was greatest disagreement (i.e., median score ≤ 3) related to which components of the intervention should be included is theme 2 (Table 2).

Table 2.

Summary of scores and actions resulting from Delphi Rounds 1 and 2.

	Statement	DELPHI 1 consensus statistics and actions	DELPHI 2 consensus statistics and actions	Discussion points for consensus meetings
Nutritional assessment for a dietary counselling intervention for the management of malnutrition should include:
1.1	Anthropometric measurements	100% strongly agree Median score = 5 Include in consensus document	N/A	Use of validated tools? Assessment of fluid status, body composition and physical examination.
1.2	Review of relevant biochemical and laboratory data	71% agree/strongly agree Median score = 4 (3−5) Round 2 Delphi	57% agree/strongly agree Median score = 4 (3–5) Discuss at consensus meeting	Access in different care settings, for example, inpatient, outpatient, community.
1.3	Review of relevant medical/clinical history	100% strongly agree Median score = 5 Include in consensus document	N/A	Acknowledge role of clinical judgement. Combine with 1.6?
1.4	Assessment of previous dietary intake	71% agree/strongly agree Median score = 4 (3−5) Round 2 Delphi	57% agree/strongly agree Median score = 4 (2–5) Discuss at consensus meeting	Use of validated methods for assessing dietary intake? Combine 1.4 and 1.5?
1.5	Assessment of current dietary intake	86% agree/strongly agree Median score = 4 (3−5) Round 2 Delphi	86% agree/strongly agree Median score = 5 (3–5) Discuss at consensus meeting
1.6	Assessment of clinical, therapeutic and other factors	100% strongly agree Median score = 5 Include in consensus document	N/A	Combine with 1.3?
1.7	Assessment of psychological and social factors	86% agree/strongly agree Median score = 5 (3–5) Round 2 Delphi	86% agree/strongly agree Median score = 5 (3−5) Discuss at consensus meeting	Combine 1.7 and 1.8? Use of validated assessment tools?
1.8	Assessment of environmental factors	67% agree/strongly agree Median score = 5 (3–5) Round 2 Delphi	86% agree/strongly agree Median score = 5 (4–5) Discuss at consensus meeting
1.9	An estimate of nutritional requirements	86% agree/strongly agree Median score = 5 (3−5) Round 2 Delphi	86% agree/strongly agree Median score = 5 (3−5) Discuss at consensus meeting	Estimate energy and protein requirements as a minimum. Fluid requirements?
1.10	A statement of the nutritional or dietetic diagnosis	86% agree/strongly agree Median score = 4.5 (1−5) Round 2 Delphi	100% strongly agree Median score = 5 Discuss comments at consensus meeting	Need for clear definition of malnutrition and criteria used to define status.
A dietary counselling intervention for the management of malnutrition should include:
2.1	Advice and support on the fortification of food and drink	57% agree/strongly agree Median score = 4 (3–5) Round 2 Delphi	71% agree/strongly agree Median score = 4 (2−5) Discuss at consensus meeting	Combine and suggest as potential strategies for a dietary counselling intervention (alone or in any combination).
2.2	Advice and support on shopping and meal preparation	57% agree/strongly agree Median score = 4 (2−5) Round 2 Delphi	57% agree/strongly agree Median score = 4 (2−5) Discuss at consensus meeting
2.3	Advice and support on the appropriate use of ONS	57% agree/strongly agree Median score = 4 (3−5) Round 2 Delphi	71% agree/strongly agree Median score = 4 (2−5) Discuss at consensus meeting
2.4	Advice on the provision of practical support, for example, feeding assistance	43% agree/strongly agree Median score = 3 (3−5) Round 2 Delphi	57% agree/strongly agree Median score = 4 (2−5) Discuss at consensus meeting
2.5	Advice on environmental modifications	14% agree Median score = 3 (3−4) Round 2 Delphi	57% agree/strongly agree Median score = 4 (2−5) Discuss at consensus meeting
2.6	Education regarding need for nutritional support and consequences of malnutrition	86% agree/strongly agree Median score = 5 (3−5) Round 2 Delphi	100% agree/strongly agree Median score = 5 (4−5) Discuss at consensus meeting
2.7	Counselling using validated psychological techniques to support behaviour change	86% agree/strongly agree Median score = 4 (3−5) Round 2 Delphi	57% agree/strongly agree Median score = 4 (1−5) Discuss at consensus meeting	Emphasise use of validated techniques and consider training needs of dietitians.
2.8	Goal‐setting activities individualised to the patient's needs	100% strongly agree Median score = 5 Include in consensus document	N/A	SMART goals.
2.9	All advice and support should be individualised to the patient's specific needs	100% agree/strongly agree Median score = 5 (4−5) Round 2 Delphi	100% strongly agree Median score = 5 Discuss comments at consensus meeting	Individual versus group settings?
2.10	Provision of written materials (or other formats if required, for example, online resources)	71% agree/strongly agree Median score = 4 (3−5) Round 2 Delphi	71% agree/strongly agree Median score = 4 (3−5) Discuss at consensus meeting	Is there ever a situation where written resources might not be provided?
Delivery of a dietary counselling intervention for the management of malnutrition should:
3.1	Include a documented nutrition care plan that has been agreed with the patient and/or their carer	100% strongly agree Median score = 5 Include in consensus document	N/A	Clear statement of location, duration, frequency and mode of delivery. Combine?
3.2	Be provided in a location that is convenient and acceptable to the patient/carer	50% agree/strongly agree Median score = 4 (1−5) Round 2 Delphi	14% agree Median score = 3 (1−5) Discuss at consensus meeting
3.3	Have a defined duration and frequency that is clearly articulated to the patient/carer at the outset	71% agree/strongly agree Median score = 4 (3−5) Round 2 Delphi	86% agree/strongly agree Median score = 4 (3−5) Discuss at consensus meeting
3.4	Include advice on shopping, menu planning and eating out	29% agree/strongly agree Median score = 3 (3−5) Round 2 Delphi	57% agree/strongly agree Median score = 4 (2−5) Discuss at consensus meeting	Include in 2.2?
Co‐ordination of care in dietary counselling interventions to manage malnutrition should include:
4.1	Documentation of the assessment, nutritional diagnosis, nutrition care plan and plans for monitoring	100% strongly agree Median score = 5 Include in consensus document	N/A	Remove this section and combine relevant statements with previous section?
4.2	Verbal communication to the patient/carer and relevant health and social care professionals	100% agree/strongly agree Median score = 5 (4−5) Round 2 Delphi	86% agree/strongly agree Median score = 5 (3−5) Discuss at consensus meeting
4.3	Liaison with relevant health and social care professionals	86% agree/strongly agree Median score = 5 (3−5) Round 2 Delphi	100% agree/strongly agree Median score = 5 (4−5) Discuss at consensus meeting
4.4	Appropriate inclusion of family and carers	100% strongly agree Median score = 5 Include in consensus document	N/A
4.5	Onward referral to other health and social care professionals where appropriate	86% agree/strongly agree Median score = 5 (4−5) Round 2 Delphi	100% agree/strongly agree Median score = 5 (4−5) Discuss at consensus meeting
4.6	Agreed plan for the end of intervention	86% agree/strongly agree Median score = 5 (3−5) Round 2 Delphi	100% strongly agree Median score = 5 Discuss comments at consensus meeting
Monitoring of dietary counselling interventions to manage malnutrition should include:
5.1	Documentation of monitoring arrangements—who, when, what and how	86% agree/strongly agree Median score = 5 (4−5) Round 2 Delphi	100% strongly agree Median score = 5 Discuss comments at consensus meeting	Combine and re‐structure to reflect monitoring of research rather than clinical outcomes.
5.2	Assessment of adherence to the intervention including exploration of any factors affecting nonadherence	57% agree/strongly agree Median score = 5 (3−5) Round 2 Delphi	86% agree/strongly agree Median score = 5 (3−5) Discuss at consensus meeting
5.3	Assessment of changes in dietary intake and behaviour	86% agree/strongly agree Median score = 5 (4−5) Round 2 Delphi	100% agree/strongly agree Median score = 5 (4−5) Discuss at consensus meeting
5.4	Assessment of anthropometric measurements	71% agree/strongly agree Median score = 5 (3−5) Round 2 Delphi	100% strongly agree Median score = 5 Discuss comments at consensus meeting
5.5	Assessment of changes in clinical, environmental, functional, psychological or social factors likely to affect nutritional status and/or dietary intake	86% agree/strongly agree Median score = 5 (1−5) Round 2 Delphi	100% agree/strongly agree Median score = 5 (4−5) Discuss at consensus meeting
5.6	Feedback to the patient/carer on behaviour changes and outcomes	100% strongly agree Median score = 5 Include in consensus document	N/A
5.7	Documentation of all assessments and any changes to the nutrition care plan, subsequent monitoring arrangements or onward referral	86% agree/strongly agree Median score = 5 (3−5) Round 2 Delphi	100% agree/strongly agree Median score = 5 (4−5) Discuss at consensus meeting

Note: Shaded boxes are statements where 100% consensus was achieved (i.e., all expert panel members ‘trongly agree’).

3.3. Delphi Round 2

Following Delphi Round 2, 100% consensus was achieved for a further five statements relating to nutritional or dietetic diagnosis in theme 1, individualisation of the intervention to patients' specific needs in theme 2, the need for an agreed plan for the end of the intervention in theme 4 and documentation of monitoring arrangements and assessment of anthropometric measurements in theme 5 (Table 2). Disagreement (median score ≤ 3) occurred for only one statement relating to the location of the intervention in theme 3. Since Delphi Rounds 1 and 2 had resulted in full consensus for only 13 (35%) of statements, two online consensus meetings were convened.

3.4. Online Meetings (June 2021)

Across the two online meetings, all statements were reviewed by the expert panel, then refined and merged to promote consensus on the essential components of a dietary counselling intervention for the management of malnutrition in adults in a clinical research context. Supporting Information S4 provides a summary of the discussions and actions resulting from these two meetings. At the first meeting, the expert panel agreed that the theme ‘Coordination of care’ was more relevant to clinical practice without a research context and therefore the statements were modified and moved to relevant sections where appropriate (100% consensus). During the second meeting, the final consensus document was drafted with four themes: nutritional assessment (eight statements), components of the intervention (seven statements), delivery of dietary counselling interventions (five statements) and monitoring (seen statements) (Figure 2), and a rationale and commentary for each statement was agreed with 100% consensus (Table 1).

3.5. Preliminary External Validation

3.5.1. ICD Polling

The ICD symposium provided an opportunity to engage with an international group of interested clinical and research‐active dietetic professionals in a preliminary way. The symposium (entitled ‘What is dietary counselling? Consensus around definitions of dietary counselling in the management of disease‐related malnutrition’) took place online on 2 September 2021. Having originally been timetabled for a face‐to‐face symposium at the ICD in South Africa in 2020, it was postponed and moved online due to the COVID‐19 pandemic. Following a presentation on the need for consistency in dietary counselling interventions in clinical research and the aims and objectives of the consensus project, delegates who attended the session were asked to vote on two themes broken down into 13 items related to the consensus statements (see Table 3 for polling questions and responses). Due to a limited time slot for the symposium, the attending delegates were asked to vote only on elements of assessment and delivery of the intervention, since it was hypothesised that differences in healthcare economies and clinical specialties might affect consensus on these aspects. The voting results showed that consensus on these aspects of nutritional support interventions was low to moderate in this international sample of clinical and research dietitian delegates. More delegates voted on assessment than on questions related to intervention. The online polling system allowed delegates to indicate agreement but not disagreement; therefore, the percentage of agreement for each question could not be determined; and the number of delegates in the session for each question was also unavailable. The online polling process was managed by the ICD 2021, and although the authors formally requested relevant information, the ICD secretariat was unable to provide data. Furthermore, it was not possible to determine the characteristics of voters such as nationality, healthcare economy or professional background. Therefore, responses were considered as a very preliminary indication of the views of a relatively small, possibly unrepresentative, group of dietetic professionals. Due to these limitations, no amendments were made to the consensus statements based on polling results. Further validation is required through the circulation of the final statements to clinical researchers, external experts and/or representatives of professional bodies for peer review. Delegates were also asked to contact the project lead if they wished to engage further in the review of the guidance statements. Seven delegates contacted the main author directly after the conference of which two were from the UK and one each were from Ireland, the Netherlands, Sweden, Canada and India.

Table 3.

Polling questions and online responses of ‘What is dietary counselling?’ Symposium at the online International Congress of Dietetics, 2021.

Question	Options	Responses (N = unknown)
What are the key intervention components that must be included in any research study to be able to say that ‘dietetic counselling’ was used to address malnutrition?	Baseline nutrition assessment Advice/support on fortification of food and drink Advice/support on shopping, meal preparation and other social/environmental strategies Advice/support on appropriate use of ONS Advice/support on provision of practice support, for example, eating assistance Education on need for nutrition and consequences of malnutrition Counselling using validated psychological techniques to support behaviour change Goal‐setting	101 59 69 46 37 73 53 48
What should be included in a baseline nutrition assessment for a research study that includes a ‘dietetic counselling’ intervention for malnutrition?	Anthropometrics Biochemical/laboratory data relevant to the research question Review of medical and clinical history; factors relevant to research Assessment of current food/dietary intake Assessment of psychological, social and environmental factors that impact dietary intake Estimate of energy, fluid, macro and micronutrient requirements relevant to the research question	140 99 112 130 100 77

4. Discussion

This Delphi process allowed the first‐ever international consensus to be achieved via an expert panel on the minimum requirements for delivering dietary counselling interventions for the management of malnutrition in adults in a clinical research context. Consensus was achieved for 27 statements across four themes, with justifications and commentary. Minimum requirements included eight statements for nutritional assessment, seven for nutritional intervention content, five for nutritional intervention delivery and seven for monitoring and follow‐up. The large number of consensus statements which are required to meet the criteria for dietary counselling for malnutrition creates a stark contrast with basic interventions such as leaflet distribution or standardised ONS prescription accompanied by basic advice, which also claim to be, but are not, ‘dietary counselling’. Many statements emphasise the requirement for individualised care and the requirement for clinical judgement, which aligns with the foundations and recommendations upon which all dietetic intervention is built [17]. This recognition and context provide further enlightenment on the poor outcomes associated with standardised care [1, 17].

4.1. Nutritional Screening and Assessment

When selecting patients for study inclusion, researchers might use either a nutrition screening tool or a formal nutritional assessment. The expert panel recommended the use of validated nutrition screening tools or a formal screening process to identify potential participants as this will allow not only replication of findings but also will facilitate comparison of study populations. The tool used should be fully described including the cut‐off criteria for establishing nutrition risk status, again to facilitate comparison of populations. In line with best practice recommendations, the screening tool selected should be validated for the target population [17]. In line with the Nutrition Care Process, the team did not recommend the use of a specific screening tool or process, being mindful of variations within and between different healthcare jurisdictions and settings as well as varying reliability and validity depending on the setting and patient type [17].

When nutritional assessment is undertaken to make a diagnosis of malnutrition, the expert panel recommended that a full description of the tool and/or process used should be provided, including any criteria for categorising nutritional status. As per nutrition screening, no specific tool or process was recommended, allowing for the need for the assessment process to be selected based on the validity and reliability of the tool for the clinical population being studied [17]. For several elements of the nutritional assessment, clinical judgement will play a role, and it will be necessary to make a judgement about the exact nature of the data included, which will be driven by the research question. These judgements need to be made explicit and fully described in the research literature and emphasise the need that dietary counselling be performed by an appropriately trained and qualified healthcare professional.

4.2. Components of the Nutritional Intervention

This theme captured the essence of a dietary counselling intervention and encompassed more than just the food‐based component since it includes how the intervention is delivered, the resources used to deliver it, and the appropriate inclusion of family and carers and other healthcare staff. The inclusion of family, carers and other healthcare professionals was noted to be required for people with supportive care needs, such as those with cognitive or other impairments (e.g., hearing) that might affect their ability to understand or assimilate information. The expert panel agreed that more than one counselling session is necessary for an intervention to be considered dietary counselling, with a minimum of two sessions required to permit monitoring of the impact of the intervention. Due to the observed inconsistencies in length, frequency and intensity of interventions reported in the Cochrane review (1), it was impossible to determine the optimal frequency and/or intensity of contacts to achieve meaningful behaviour change. This is an area requiring further research.

Dietary counselling interventions have been poorly described, and the consensus statements demonstrate the need to capture and report all components of the intervention [24]. To allow replication and comparison between studies, it is important that the rationale for the specific intervention is described, the goals of the intervention are clearly articulated, and that there is a description of the development of any resources used in the study as well as any patient or stakeholder input into the design, content, validity testing, feasibility evaluation and piloting of any intervention [15]. Other elements of the study should also be described as per reporting guidelines, such as patient demographics and medical status [15]. If developing new resources for a dietary intervention study, including the target audience in the creation and review of these materials is ideal to ensure that they are meeting patient and stakeholder needs [25].

The consensus statements reiterate that dietary counselling is individualised to specific patients and that researchers should include a full description, where intellectual property permits, of how this was achieved. Whilst documentation for individual participants isn't required, the general approach to the provision of dietary counselling should be reported, including how decisions were made and which resources and techniques were used to provide care. If any techniques designed to support behaviour change were used to deliver the intervention these should be described together with a description of any training provided to the person delivering the intervention. The authors acknowledge that word limits for journals may impact how much description of a dietary counselling intervention might be permitted within a publication; however, journals are increasingly encouraging the inclusion of online supplementary material that could be used to describe an intervention in more detail.

4.3. Delivery of the Intervention

When conducting systematic reviews, it is often difficult to locate and extract data on the delivery of the intervention. A Cochrane review examining dietary counselling found that the aspects most frequently missing in studies were a description of the location, duration, frequency and mode of intervention [1, 2]. These factors individually or in any combination may impact on the effectiveness of the intervention; however, due to poor reporting, it is unknown to what extent they impact efficacy and therefore translation to practice and comparisons of interventions is hampered. Frameworks and tools to assist researchers in measuring and reporting how individualised, non‐standardised, interventions were delivered would assist in meeting the consensus statements.

Clear articulation of the end of the intervention or early discontinuation is needed so that research participants and researchers are clear about the circumstances under which an intervention needs to end. It is equally important to clearly define the distinction between the end of active intervention and arrangements for any follow‐up period (and handover from research to clinical services) as this will allow a clear understanding of how the effects of an intervention might persist beyond the active research period.

4.4. Monitoring of the Intervention

This theme related directly to the recommendation that dietary counselling is not a ‘one‐off’ intervention and emphasises the need to know whether the intervention was effective as well as providing an opportunity for modification of the intervention according to individual patient needs. Monitoring in a research context is different from clinical monitoring and needs to focus not only on outcome assessments but also on measurements that indicate adherence to the continued intervention. Findings from the Cochrane review noted how few studies reported explicitly on monitoring, how adherence was measured (if at all), and how many participants adhered to the intervention [1].

5. Limitations

This consensus represented the views and experiences of an expert panel and the literature and guidelines sourced and may not represent the views of the dietetic research community nor the guidance of all national dietetic associations globally. Global representation in the expert panel was sought to mitigate this limitation; however, the lack of representation from Africa and South America may have implications for the global application of these consensus statements. Furthermore, it must be noted that the invited expert panel members were not selected as representatives of their respective national dietetic associations.

From the anonymised Delphi rounds, consensus was achieved for only 35% of the statements. Most of the consensus statements result from subsequent online discussions, and it is possible that group dynamics influenced the outcome. Although consensus discussions were found to be required due to the nuanced context of the draft statements and allowed for not only consensus but meaningful commentary, this possible limitation underpins the importance of external validation.

The preliminary external validation took place during the COVID‐19 pandemic during an online symposium at the ICD in September 2021. At the time, sophisticated online hosting was not available which meant it was not possible to fully present all consensus statements nor to quantify and describe delegates in the session. The responses of delegates indicate that a minimum of 140 delegates attended the session; however, it is likely a greater number attended as the votes for disagreement with the statements were not captured. Furthermore, as there was no denominator for the responses, and percentages of agreement could not be calculated, limiting the potential for external validation. Consensus by delegates was low to moderate for the statements presented (assessment and delivery of the intervention), and the potential reasons for these discrepancies, such as style and method of presentation, as well as delegate suitability, merit further investigation through formalised external validation.

Of the original five themes for dietary counselling interventions for malnutrition, the consensus statements were reduced to four, representing a merging of themes. Multiple statements did not reach consensus for this study, and while not necessarily a limitation for the purposes of an intervention in the clinical research context, it should be noted that the consensus statements may not reflect requirements for dietary counselling in practice.

6. Implications and Recommendations

The consensus statements as they currently stand require formal and extensive validation through further Delphi Rounds. The statements will be circulated to experts in the field (i.e., clinical academics with a track record in conducting research on the management of malnutrition), national professional bodies such as the British Dietetic Association and multidisciplinary organisations that focus on the detection and management of malnutrition. The authors will pay particular attention to identifying representatives of organisations in South America, Africa and other parts of the world not covered during the initial phase of this project, fo example, Russia and the Middle East. During the anonymised Delphi Rounds, respondents will be asked to rate each statement for inclusion in the final consensus document using a 5‐point Likert scale (ranging from 1 ‘strongly disagree to 5 ‘strongly agree’). The Delphi Rounds will be repeated until international consensus can be achieved.

Once these consensus statements undergo external validation, they have the potential to improve the comparability and replicability of dietary counselling interventions in a clinical research context and to guide translation into clinical practice. Validated consensus statements may also be used in systematic reviews of dietary counselling interventions for malnutrition in adults as a checklist to evaluate how well the interventions were conducted and reported, which may assist in the interpretation and confidence of findings.

7. Conclusion

Preliminary international consensus via a panel of experts was gained for 27 statements across four themes of the nutrition care process, which were considered a minimum requirement for dietary counselling interventions for the management of malnutrition in adults in a clinical research context. Further external validation by dietetic researchers and professional organisations will support this consensus and if adopted may influence research, clinical practice and policy development.

Author Contributions

C.E.W. and C.B. convened the expert group and led the targeted review. All authors were involved in the design and conduct of the research. C.E.W. produced the first draft of the manuscript. All authors reviewed, amended and provided final approval of the manuscript.

Ethics Statement

Ethical approval was deemed unnecessary for this initial phase of the project since only members of the authorial team were involved in the Delphi Rounds. Each expert confirmed their willingness to contribute to the Delphi Rounds on initial invitation to take part in the project. The authors acknowledge that validation of the consensus document will, in the future, require ethical approval.

Conflicts of Interest

The authors declare no conflicts of interest.

Peer Review

The peer review history for this article is available at https://www.webofscience.com/api/gateway/wos/peer-review/10.1111/jhn.70028.

Supporting information

Supporting information.

Supporting information.

Supporting information.

Supporting information.

Data Availability Statement

The data that support the findings of this study are available from the corresponding author upon reasonable request.