Advertisement

Severe Neurological Impairment: A delphi consensus-based definition

  • John Allen
    Correspondence
    Corresponding author. Trinity Centre for Health Sciences, Department of Paediatrics, Tallaght University Hospital, Belgard Road, Tallaght, Dublin, 24, Ireland.
    Affiliations
    Department of Paediatrics, Tallaght University Hospital, Tallaght, Dublin, Ireland

    Discipline of Paediatrics and Child Health, Trinity College, The University of Dublin, Ireland

    Trinity Research in Childhood Centre, Trinity College, The University of Dublin, Ireland
    Search for articles by this author
  • Maria Brenner
    Affiliations
    Trinity Research in Childhood Centre, Trinity College, The University of Dublin, Ireland

    School of Nursing and Midwifery, Trinity College, The University of Dublin, Ireland
    Search for articles by this author
  • Julie Hauer
    Affiliations
    Division of General Pediatrics, Boston Children's Hospital, Boston, MA, USA
    Search for articles by this author
  • Eleanor Molloy
    Affiliations
    Department of Paediatrics, Tallaght University Hospital, Tallaght, Dublin, Ireland

    Discipline of Paediatrics and Child Health, Trinity College, The University of Dublin, Ireland

    Trinity Research in Childhood Centre, Trinity College, The University of Dublin, Ireland

    Trinity Translational Medicine Institute, St. James' Hospital, Dublin, Ireland

    Neonatology, Coombe Women and Infants' University Hospital, Dublin, Ireland

    Neonatology, Our Lady's Children's Hospital, Crumlin, Dublin, Ireland
    Search for articles by this author
  • Denise McDonald
    Affiliations
    Department of Paediatrics, Tallaght University Hospital, Tallaght, Dublin, Ireland

    Discipline of Paediatrics and Child Health, Trinity College, The University of Dublin, Ireland

    Trinity Research in Childhood Centre, Trinity College, The University of Dublin, Ireland
    Search for articles by this author
Open AccessPublished:September 11, 2020DOI:https://doi.org/10.1016/j.ejpn.2020.09.001

      Highlights

      • Severe neurological impairment (SNI) is a term which has not been consistently defined.
      • Lack of an agreed definition can result in variance in reporting of research, including incidence and outcome measures.
      • The Delphi method has been successfully used as a method of reaching consensus, including the creation of definitions.
      • A consensus-based, international, multi-disciplinary definition of the term SNI has been formulated.
      • This definition can be used to perform consistent, reproducible research and for adequate service-planning.

      Abstract

      Background

      Severe Neurological Impairment (SNI) is a term for which there is no consistently used definition. This may hamper consistency in the reporting of research in the area and communication between professionals involved in the care of those with SNI.

      Objective

      We aimed to create an international, multidisciplinary, consensus-based definition of SNI.

      Design

      The Delphi method was employed to reach consensus on the definition of SNI.

      Method

      An international, multi-disciplinary expert panel was recruited. The process proceeded over three rounds with feedback provided to panellists between each of them. Consensus was defined as 70% agreement. A working definition was created and, following presentation at an international meeting and consultation with parent representatives, further refined, to create a finalised definition.

      Results

      Thirty-four expert panellists commenced the process. Six items reached the threshold of consensus. The finalised definition is as follows: “Severe Neurological Impairment describes a group of disorders of the central nervous system which arise in childhood, resulting in motor impairment, cognitive impairment and medical complexity, where much assistance is required with activities of daily living. The impairment is permanent but can be progressive or static.”

      Conclusion

      A consensus-based definition of SNI which includes multi-disciplinary, international and parental input has been created. This should prove useful for clinical, research and resource-planning purposes.

      Keywords

      1. Introduction

      Severe Neurological Impairment (SNI) is a term that is commonly used in medical literature. However, there is no consistently used definition and, indeed, there is wide variation in the application of the term [
      • Allen J.
      • Molloy E.
      • McDonald D.
      Severe neurological impairment: a review of the definition.
      ]. Fundamental details about the term SNI are not clear e.g. whether the impairment is progressive, static or both; or, whether it involves the central or peripheral nervous system or both. The lack of a clear definition of SNI hampers consistency in research, communication, care-delivery and service-planning.
      SNI as a term is already in use and, therefore, must be defined properly. SNI describes a cohort of children with unique needs who are not adequately described by any other terms that are currently employed. Children with Medical Complexity (CMC) and Children with Complex Chronic Conditions (CCC) are terms that have been defined previously [
      • Cohen E.
      • Kuo D.Z.
      • Agrawal R.
      • Berry J.G.
      • Bhagat S.K.
      • Simon T.D.
      • et al.
      Children with medical complexity: an emerging population for clinical and research initiatives.
      ,
      • Feudtner C.
      • Feinstein J.A.
      • Zhong W.
      • Hall M.
      • Dai D.
      Pediatric complex chronic conditions classification system version 2: updated for ICD-10 and complex medical technology dependence and transplantation.
      ], while, more recently, children's complex care needs have been defined [
      • Brenner M.
      • Kidston C.
      • Hilliard C.
      • Coyne I.
      • Eustace-Cook J.
      • Doyle C.
      • et al.
      Children's complex care needs: a systematic concept analysis of multidisciplinary language.
      ]. Children with SNI fit both definitions, as do children with complex congenital heart disease, diabetes, arthritis and severe asthma. Terms with a more neurological focus such as neurodisability ([
      • Morris C.
      • Janssens A.
      • Tomlinson R.
      • Williams J.
      • Logan S.
      Towards a definition of neurodisability: a Delphi survey.
      ]) capture the involvement of “the brain and/or neuromuscular system” and “functional limitations” but do not speak to the severity or medical complexity of those with SNI. Therefore, SNI is an important term that requires a clear, consistent definition. This would assist with reliability and reproducibility of research, enumeration of prevalence, quantification of medical complexity within the group and service-planning.
      Any new definition should be meaningful to all who care for children with SNI. Consequently, any effort to create a definition should include every stakeholder and be as internationally applicable as possible. In this study, we aimed to create an international, multidisciplinary, consensus-based definition of SNI.

      2. Methods

      2.1 The Delphi method and rationale for its use

      We chose the Delphi method as the process to reach a consensus on a definition of SNI. The method has several strengths which informed this choice: It is an accepted method for the creation of definitions in the medical literature [
      • Morris C.
      • Janssens A.
      • Tomlinson R.
      • Williams J.
      • Logan S.
      Towards a definition of neurodisability: a Delphi survey.
      ,
      • Sudore R.L.
      • Lum H.D.
      • You J.J.
      • Hanson L.C.
      • Meier D.E.
      • Pantilat S.Z.
      • et al.
      Defining advance care planning for adults: a consensus definition from a multidisciplinary Delphi panel.
      ,
      • Singer M.
      • Deutschman C.S.
      • Seymour C.W.
      • Shankar-Hari M.
      • Annane D.
      • Bauer M.
      • et al.
      The third international consensus definitions for sepsis and septic shock (Sepsis-3).
      ]; the ability to use an online questionnaire format was ideal for a multidisciplinary and international panel of experts; the anonymous nature of the process minimises power-dynamics which may occur for reasons such as reputation or personality and may also promote participation [
      • Hasson F.
      • Keeney S.
      • McKenna H.
      Research guidelines for the Delphi survey technique.
      ]; it is cost-effective and does not require special expertise to deliver.

      2.2 Preparatory phase

      In preparation for the Delphi process, we carried out a review of the literature, the details of which have been previously published [
      • Allen J.
      • Molloy E.
      • McDonald D.
      Severe neurological impairment: a review of the definition.
      ]. We searched the databases of PubMed and Cochrane using the term “Severe Neurological Impairment”. Forty-two articles were included for full-text analysis, with 23 including an explicit definition of SNI. In the other articles, the authors definition of the term was implied by the participants included in the study. Seven themes were extracted from the various definitions of SNI in the literature which are shown in Table 1. Motor impairment, intellectual disability, communication difficulties and increased care needs were included in the definition in 80%, 70%, 30% and 13% of papers respectively. Dependence on others for decision-making, chronicity and distinction between disorders of the central nervous system and peripheral nervous system were less frequently included [
      • Allen J.
      • Molloy E.
      • McDonald D.
      Severe neurological impairment: a review of the definition.
      ]. This established, with certainty, that there was no universally accepted definition of SNI. Themes from the literature review were used to inform the design of the survey instruments in the later stages of the Delphi process. A flow chart illustrating the stages of the Delphi process is shown in Fig. 1.
      Table 1Themes extracted from definitions of Severe Neurological Impairment during preparatory phase literature review.
      Intellectual disability
      Limitations in mobility
      Communication difficulties
      Increased care needs
      CNS vs PNS
      Dependence on others for decision-making
      Chronicity

      2.3 Expert panel

      International experts in paediatric complex care and neurodisability were asked to assist in facilitating the Delphi process. These facilitators were well-published and prominent figures in the area of paediatric complex care, neurodisability and palliative care. This process resulted in expert facilitators in 5 countries, namely, Ireland, the United Kingdom, the United States of America, Canada and Australia.
      These expert facilitators were asked to disseminate an invitation to participate in their region. A multidisciplinary panel was stipulated by the authors. The definition of “expert” was not provided but we expected that all facilitators would have a tacit knowledge of the meaning of this term. With each invitation there was a request to further disseminate the call to appropriate colleagues, employing a snowball effect in recruitment. Those who wished to participate were required to email the research team independently to express their interest; therefore, participants were anonymous.

      2.4 Consensus

      Consensus was sought on items to be included in the final definition. A priori, consensus was defined as approval (either agree or strongly agree) of 70% of participants or greater. Any item which reached 70% was automatically brought forward to the definition [
      • Diamond I.R.
      • Grant R.C.
      • Feldman B.M.
      • Pencharz P.B.
      • Ling S.C.
      • Moore A.M.
      • et al.
      Defining consensus: a systematic review recommends methodologic criteria for reporting of Delphi studies.
      ]. Conversely, any item which received less than 30% approval was automatically excluded from the definition.

      2.5 Survey instruments and rounds

      From the outset, a limit of 3 rounds was set for the process. Each survey round was conducted online. The first round was an open round, during which, panellists were asked to list any words, phrases, themes or concepts that they associated with SNI. The first round was open for a two-week period, with a reminder email after 1 week. Free-text responses were copied verbatim, key phrases were extracted and similar responses were grouped into themes by one of the authors (JA) and reviewed by another author (DMc). These themes, as well as the data that had been previously gathered from the literature review [
      • Allen J.
      • Molloy E.
      • McDonald D.
      Severe neurological impairment: a review of the definition.
      ], were then used to inform the following rounds.
      The second round sought to assess agreement with items to be included in the final definition using a 5-point Likert scale. The panellists were also invited to leave comments on each of the statements to be used in the following round as an element of feedback provided to participants. Participants were also asked to indicate wording preferences for use in the final definition. This round remained open for 3 weeks. Medians and modes of the responses were calculated using Microsoft Excel®. These statistics were then used, along with the written responses, to provide feedback to participants in round 3 of the process.
      In the third round, any items which had not clearly reached consensus or had not been excluded due to non-consensus (as defined above) were brought forward. These questions were then posed again, along with feedback from the previous round. This feedback was anonymous and included selected written responses, medians and modes. Panellists were asked to vote again having reviewed this new information. The third round was open for 3 weeks. All items which had reached consensus were included in a working definition which was formulated by the research team.

      2.6 Parental views

      Parents of children with significant disability and medical complexity were approached to give their input into the definition. Two parents volunteered to participate which we felt was acceptable, although less than ideal. A meeting was arranged between the authors and parent representatives. The process and the working definition were discussed with particular reference to wording, relevance and any areas not addressed. They felt it accurately described a cohort of children with significant disability of neurological origin, while at the same time being inclusive of those with particular underlying diagnoses and those with no specific diagnosis. Some minor wording adjustments were made.

      2.7 Presentation at an international meeting

      The working definition was presented at the joint meeting of the American Academy of Cerebral Palsy and Developmental Medicine and the International Alliance of Academies of Childhood Disability in September 2019. We used TurningPoint© audience participation software to anonymously assess agreement of attendees with the working definition. An open-floor discussion was held to inform the final definition. Wording of the definition was finalised by the research team following this meeting.

      3. Results

      3.1 Round 1

      Thirty-eight healthcare professionals responded to the initial invitation to participate in the process. Thirty-four panellists completed the first round within the specified time period. Twenty-two panellists were working in Ireland, six in the USA, four in Australia, one in Canada and one in the UK. Fourteen physicians, fourteen allied health professionals (Occupational Therapy (OT); Physiotherapy; Speech and Language Therapy (SLT); Dietetics) and six nurses took part.
      Nineteen themes were identified from the responses in round 1 of the Delphi process and the literature review, namely: motor impairment; aetiology; cognitive impairment; quality of life; central nervous system (CNS) versus peripheral nervous system (PNS); communication; medical complexity; sensory impairments; increased care needs; dependence on medical technology; multi-organ dysfunction; progressive or static; emotional impairment; chronicity; a quantification of severity; family-reported issues; life limitation; social functioning; and dependence for decision-making.

      3.2 Round 2

      In the second round, 30 panellists participated. However, one of these panellists, having commenced the process, did not provide responses to the questions; therefore, 29 participants completed round 2. This included twelve physicians, five nurses and thirteen allied health professionals (two SLT, four physiotherapists, five OT, one dietician and one clinical psychologist).
      More than 70% of participants agreed that the following themes were essential to the definition of SNI: motor impairment (82.8%); cognitive impairment (79.3%); medical complexity (75.9%); increased care needs (89.7%); and that it may include disorders that are progressive or static (89.7%). The following themes were automatically excluded as they reached less than 30% agreement: Emotional impairment (13.8%); dependence on medical technology (24.1%); multiorgan dysfunction (20.7%); family-reported issues (27.6%). It is important to note that dependence on medical technology, multiorgan dysfunction and family-reported issues often form part of the term “medical complexity”, as does increased care needs, but participants were asked whether these terms should be included, independently of medical complexity, to emphasise their importance. Nine items were brought forward to round 3 for participants to vote upon again, following feedback, as previously described. The items brought forward were: aetiology (48.3%); CNS versus PNS (44.8% voted CNS alone); quality of life (44.8%); life-limitation (34.5%); communication (51.7%); social functioning (55.2%); sensory impairments (41.4%); quantification of severity (62.1%); and dependence for decision-making (34.5%). One item, chronicity, was excluded after round 2 as the clear agreement that the disorder could be progressive or static rendered this theme irrelevant.

      3.3 Round 3

      Thirty-one panellists participated in round 3 of the Delphi process. One participant answered the first question only; therefore, for the remaining questions, the denominator was 30. The panel included thirteen physicians, six nurses and twelve allied health professionals (two SLT, four physiotherapists, four OT, one dietician and one clinical psychologist).
      Following round 3, one further item was included in the definition, namely that SNI referred to disorders of the CNS (73.3%). All other items (listed below) saw a decline in agreement; thus, did not reach the 70% cut-off for consensus and were excluded from the final definition. The items excluded after this round were: aetiology (33.3%); quality of life (10%); life-limitation (26.7%); communication (40%); social functioning (36.7%); sensory impairments (26.7%); quantification of severity (46.7%); and dependence for decision-making (26.7%). A summary of the results from rounds 1 to 3 is shown in Table 2.
      Table 2Summary of results from round 1 to 3.
      Number of participantsRound 1 nRound 2 n (%)Round 3 n (%)
      Number of participants343031
      Profession
      Physician141213
      Nurse656
      Allied health professional (total)141312
      Physiotherapist644
      Occupational Therapist554
      Speech and Language Therapist/Speech Pathologist222
      Dietician111
      Clinical Psychologist011
      Country
      Ireland221919
      USA656
      Australia454
      UK111
      Canada101
      Themes
      Motor impairment24 (82.8)Included after R2
      Cognitive impairment23 (79.3)Included after R2
      Aetiology14 (48.3)10 (32.3)
      Quality of life13 (44.8)3 [
      • Ananth P.
      • Melvin P.
      • Feudtner C.
      • Wolfe J.
      • Berry J.G.
      Hospital use in the last year of life for children with life-threatening complex chronic conditions.
      ]
      Life limitation10 (34.5)8 (26.7)
      Social functioning16 (55.2)11 (36.7)
      CNS vs PNS vs both
      CNS alone13 (44.8)22 (73.3)
      PNS alone0 (0)0 (0)
      Both11 (37.9)3 [
      • Ananth P.
      • Melvin P.
      • Feudtner C.
      • Wolfe J.
      • Berry J.G.
      Hospital use in the last year of life for children with life-threatening complex chronic conditions.
      ]
      Do not specify5 (17.2)5 (16.7)
      Communication15 (51.7)12 (41.4)
      Medical complexity22 (75.9)Included after R2
      Multi-organ dysfunction6 (20.7)Excluded after R2
      Dependence on medical technology7 (24.1)Excluded after R2
      Family-reported issues8 (27.6)Excluded after R2
      Increased care needs26 (89.7)Included after R2
      Sensory Impairments12 (41.4)8 (26.7)
      Dependence for decision making10 (34.5)8 (26.7)
      Emotional Impairment4 (13.8)Excluded after R2
      Quantification of severity18 (62)14 (46.7)
      Progressive vs static vs bothIncluded after R2
      Progressive alone1 (3.4)
      Static alone2 (6.9)
      Both26 (89.7)
      ChronicityTheme removed after R2 and incorporated into progressivity
      Chronic19 (67.9)
      Acute1 (3.6)
      Both8 (28.6)
      Do not specify1 (3.6)
      CNS = Central Nervous System; PNS = Peripheral Nervous System; R2 = Round 2; Items in bold = reached consensus threshold and included in final definition.

      3.4 Finalising the definition

      After the third round was complete, a working definition of SNI was created. The working definition was as follows (Fig. 2):
      Fig. 2
      Fig. 2Working definition of Severe Neurological Impairment (SNI).
      The definition was presented at an international, multi-disciplinary conference and audience response software was used to gauge agreement with our working definition. Fourteen attendees chose to download and interact via the software. All participants agreed, in general, with the working definition. An open floor discussion followed and comments from the audience were invited on the working definition. After the conference, and using the feedback received, 2 items from the working definition were slightly altered for the final version. “A group of children” was changed to “which arise in childhood” to reflect that children with SNI can progress to become adults with SNI; “complex medical needs” became “medical complexity”. This latter change was made to better align the term medical complexity with the definition of CMC as proposed by Cohen et al. (i.e. “substantial family-identified health care service needs”, “1 or more chronic clinical condition(s)”, functional limitations and high health care use) [
      • Cohen E.
      • Kuo D.Z.
      • Agrawal R.
      • Berry J.G.
      • Bhagat S.K.
      • Simon T.D.
      • et al.
      Children with medical complexity: an emerging population for clinical and research initiatives.
      ], in an attempt to avoid ambiguity. Thus, the final definition is as follows (Fig. 3):
      Fig. 3
      Fig. 3Finalised definition of Severe Neurological Impairment (SNI).

      4. Discussion

      A consensus-based definition of SNI has been created through this Delphi process. SNI is a term which is commonly used in the medical literature but is not used consistently [
      • Allen J.
      • Molloy E.
      • McDonald D.
      Severe neurological impairment: a review of the definition.
      ]. Lack of an agreed definition may result in variance in reporting of prevalence and outcome measures. This is an issue which has previously been demonstrated in the literature with other terminology such as “disability”. For example, in 2004, Bajekal et al. published a review of disability estimates and definitions in the UK[
      • Bajekal M.
      • Harris T.
      • Breman R.
      • Woodfield K.
      Review of Disability Estimates and Definitions: A Study Carried Out on Behalf of the Department for Work and Pensions.
      ]. They found that estimates of adults with disabilities varied from 8.6 million to 11 million, with much of this variation attributed to how disability is defined.
      The development of a consensus-based definition of SNI should improve the precision of reporting of research and, consequently, may improve outcomes in health, quality of life, as well as decision making at the end of life (EOL) for those with SNI. As an example, care at EOL for children with SNI often occurs in the hospital, with many receiving intensive life-sustaining therapies in the last 2 days of life [
      • Ananth P.
      • Melvin P.
      • Feudtner C.
      • Wolfe J.
      • Berry J.G.
      Hospital use in the last year of life for children with life-threatening complex chronic conditions.
      ,
      • DeCourcey D.D.
      • Silverman M.
      • Oladunjoye A.
      • Balkin E.M.
      • Wolfe J.
      Patterns of care at the end of life for children and young adults with life-threatening complex chronic conditions.
      ]. Reproducible research, aided by a uniform definition of SNI, could help to optimise care at an incredibly difficult time for families.
      Strengths of this definition of SNI include that it was developed in a systematic manner, including literature review, expert opinion and lived-experience. An open round was used to reduce bias. Criteria for consensus were defined a priori in line with best practice; consensus was set at 70% agreement, a generally accepted benchmark [
      • Diamond I.R.
      • Grant R.C.
      • Feldman B.M.
      • Pencharz P.B.
      • Ling S.C.
      • Moore A.M.
      • et al.
      Defining consensus: a systematic review recommends methodologic criteria for reporting of Delphi studies.
      ]. The feedback between the second and third round was both numerical and narrative. Narrative responses were anonymised and represented different points of view to avoid the introduction of bias. The provision of both forms of feedback provided the opportunity to consider all points of view, including some more subtle differences in opinion. Hallowell [
      • Hallowell M.R.
      Techniques to Minimize Bias When Using the Delphi Method to Quantify Construction Safety and Health Risks.
      ], citing Best [
      • Best R.J.
      An experiment in Delphi estimation in marketing decision making.
      ], describes how this technique improves accuracy in the Delphi technique.
      Another significant strength of this definition is the focus on diseases and injury of the central nervous system, a group of disorders with risk for multiple comorbid problems not seen with peripheral disorders. Other terminology combines these 2 distinct groups [
      • Cohen E.
      • Kuo D.Z.
      • Agrawal R.
      • Berry J.G.
      • Bhagat S.K.
      • Simon T.D.
      • et al.
      Children with medical complexity: an emerging population for clinical and research initiatives.
      ,
      • Feudtner C.
      • Feinstein J.A.
      • Zhong W.
      • Hall M.
      • Dai D.
      Pediatric complex chronic conditions classification system version 2: updated for ICD-10 and complex medical technology dependence and transplantation.
      ,
      • Morris C.
      • Janssens A.
      • Tomlinson R.
      • Williams J.
      • Logan S.
      Towards a definition of neurodisability: a Delphi survey.
      ]. Such a definition can also benefit children with impairment of the CNS yet without a specific diagnosis. Going forward, including subcategories of intellectual and motor disability will be an essential part of health outcomes research.
      A multi-disciplinary panel of experts was included to incorporate a diverse range of expertise and to ensure that the definition was useful across the disciplines. The physician and allied health professional groups had similar numbers, while it would initially appear that nurses had fewer representatives. However, both the physician and allied health professional groups are relatively heterogeneous i.e. physicians include experts in paediatric palliative care, neurodisability, community paediatrics etc. On the other hand, the nursing group is quite homogenous and was comprised of nurses with an interest in caring for children with neurodisability and/or medical complexity.
      The views of international experts were sought and participants from 5 countries signed up to participate. However, we acknowledge that there are limitations to the panel that was recruited. All participants were from English-speaking countries and there were no participants from middle or low-income countries. Therefore, the resulting definition may not be considered truly international. It was also noted that panellist participation was skewed towards one country which may also be considered a limitation of the study. Some differences in terminology were noted in this process, including the use of cognitive impairment rather than intellectual disability in the definition. Although efforts were made to ensure that the panel was multi-disciplinary, we also acknowledge that there are professionals involved in the care of children with SNI who were not included in our panel. For example, professionals in the educational sector may have provided useful and alternative insights into the definition of SNI. These issues would be important to address in any future work on updating the definition of SNI.
      A further limitation of this study was that we recruited only 2 parent representatives. Unfortunately, there were no parent representative organisations which we could approach locally to recruit more parents. Our attempts to recruit parents through community disability services were unsuccessful as these organisations had no formal parent forums. We were also cognisant that we did not want to impose any additional significant time burdens on the 2 parents that had volunteered. For these reasons, parental involvement was less than ideal. Going forward, we envisage the creation of a parental forum from within our cohort, as we recognise this is a deficit and is important to rectify in any future work.
      Although there were a relatively small number of participants, this is considered appropriate in Delphi methodology and produces reliable results [
      • Akins R.B.
      • Tolson H.
      • Cole B.R.
      Stability of response characteristics of a Delphi panel: application of bootstrap data expansion.
      ]. Thus, a large panel size is not considered vital to render the conclusions of the Delphi process valid.
      Perhaps, a more important consideration in Delphi reliability is participant attrition. Panellist dropout is acknowledged to be a major problem in the Delphi process and may cause response bias [
      • Toma C.
      • Picioreanu I.
      The Delphi technique: methodological considerations and the need for reporting guidelines in medical journals.
      ]. The dropout rate from round 1 to round 3 was only 9%, so, retention of our expert panellists throughout the process was excellent and could be considered a further strength of the study.
      Although many items did not reach the level of consensus for inclusion in the definition of SNI, they are still important considerations in the clinical care of these children. It is important to emphasise that this definition is a descriptive term for a group of children with similar issues and is not intended as a diagnosis. The endorsement of the definition of SNI by professional organisations will be important to ensure that the definition is adopted widely. Definitions should not be static and it is probable that this definition will need to be adapted in the future. In particular, future directions should involve input from developing countries.

      5. Conclusion

      A multi-disciplinary, consensus-based definition of SNI has been devised, with indications that it could be utilised internationally. This definition may be useful for researchers, healthcare providers, young people with SNI and their families. It provides a means of improving precision in reporting of prevalence and research data, with the ultimate aim of improving outcomes for this unique and vulnerable population.

      Declaration of competing interest

      Maria Brenner, Julie Hauer, Eleanor Molloy and Denise McDonald have no conflicts of interest to declare.

      Acknowledgements

      John Allen has received funding from the National Children’s Research Centre, Dublin, Ireland (Grant Number D/19/2 ).

      Appendix A. Supplementary data

      The following is the Supplementary data to this article:

      Ethical approval

      Ethical approval was received from the Research Ethics Committee of Tallaght University Hospital, Dublin 24, Ireland on September 14, 2018 (Ref:2018/09 Chairman's Action [
      • Singer M.
      • Deutschman C.S.
      • Seymour C.W.
      • Shankar-Hari M.
      • Annane D.
      • Bauer M.
      • et al.
      The third international consensus definitions for sepsis and septic shock (Sepsis-3).
      ])

      References

        • Allen J.
        • Molloy E.
        • McDonald D.
        Severe neurological impairment: a review of the definition.
        Dev. Med. Child Neurol. 2020; 62: 277-282
        • Cohen E.
        • Kuo D.Z.
        • Agrawal R.
        • Berry J.G.
        • Bhagat S.K.
        • Simon T.D.
        • et al.
        Children with medical complexity: an emerging population for clinical and research initiatives.
        Pediatrics. 2011; 127: 529-538
        • Feudtner C.
        • Feinstein J.A.
        • Zhong W.
        • Hall M.
        • Dai D.
        Pediatric complex chronic conditions classification system version 2: updated for ICD-10 and complex medical technology dependence and transplantation.
        BMC Pediatr. 2014; 14: 199
        • Brenner M.
        • Kidston C.
        • Hilliard C.
        • Coyne I.
        • Eustace-Cook J.
        • Doyle C.
        • et al.
        Children's complex care needs: a systematic concept analysis of multidisciplinary language.
        Eur. J. Pediatr. 2018; 177: 1641-1652
        • Morris C.
        • Janssens A.
        • Tomlinson R.
        • Williams J.
        • Logan S.
        Towards a definition of neurodisability: a Delphi survey.
        Dev. Med. Child Neurol. 2013; 55: 1103-1108
        • Sudore R.L.
        • Lum H.D.
        • You J.J.
        • Hanson L.C.
        • Meier D.E.
        • Pantilat S.Z.
        • et al.
        Defining advance care planning for adults: a consensus definition from a multidisciplinary Delphi panel.
        J. Pain Symptom Manag. 2017; 53: 821-832 e1
        • Singer M.
        • Deutschman C.S.
        • Seymour C.W.
        • Shankar-Hari M.
        • Annane D.
        • Bauer M.
        • et al.
        The third international consensus definitions for sepsis and septic shock (Sepsis-3).
        J. Am. Med. Assoc. 2016; 315: 801-810
        • Hasson F.
        • Keeney S.
        • McKenna H.
        Research guidelines for the Delphi survey technique.
        J. Adv. Nurs. 2000; 32: 1008-1015
        • Bajekal M.
        • Harris T.
        • Breman R.
        • Woodfield K.
        Review of Disability Estimates and Definitions: A Study Carried Out on Behalf of the Department for Work and Pensions.
        2004
        • Ananth P.
        • Melvin P.
        • Feudtner C.
        • Wolfe J.
        • Berry J.G.
        Hospital use in the last year of life for children with life-threatening complex chronic conditions.
        Pediatrics. 2015; 136: 938-946
        • DeCourcey D.D.
        • Silverman M.
        • Oladunjoye A.
        • Balkin E.M.
        • Wolfe J.
        Patterns of care at the end of life for children and young adults with life-threatening complex chronic conditions.
        J. Pediatr. 2018; 193: 196-203.e2
        • Diamond I.R.
        • Grant R.C.
        • Feldman B.M.
        • Pencharz P.B.
        • Ling S.C.
        • Moore A.M.
        • et al.
        Defining consensus: a systematic review recommends methodologic criteria for reporting of Delphi studies.
        J. Clin. Epidemiol. 2014; 67: 401-409
        • Hallowell M.R.
        Techniques to Minimize Bias When Using the Delphi Method to Quantify Construction Safety and Health Risks.
        in: Construction Research Congress 2009. 2009: 1489
        • Best R.J.
        An experiment in Delphi estimation in marketing decision making.
        J. Market. Res. 1974; 11: 447-452
        • Akins R.B.
        • Tolson H.
        • Cole B.R.
        Stability of response characteristics of a Delphi panel: application of bootstrap data expansion.
        BMC Med. Res. Methodol. 2005; 5: 37
        • Toma C.
        • Picioreanu I.
        The Delphi technique: methodological considerations and the need for reporting guidelines in medical journals.
        J. Int. J. Publ. Health Res. 2016; 4: 47-59