Article Text

Download PDFPDF

Severe illness getting noticed sooner: SIGNS-for-Kids—initial validity assessment of a paediatric illness recognition tool for caregivers
  1. Seth Bernard Gray1,2,3,
  2. Karen Dryden-Palmer1,3,
  3. Calvin He1,3,
  4. Ciara Tremblay1,3,
  5. Leah Marsot-Schiffman1,3,
  6. Dirk Huyer4,
  7. Christopher S Parshuram1,2,3,5
  1. 1Critical Care Medicine, The Hospital for Sick Children, Toronto, Ontario, Canada
  2. 2Paediatrics and Critical Care Medicine, University of Toronto, Toronto, Ontario, Canada
  3. 3Center for Safety Research, SickKids, Toronto, Ontario, Canada
  4. 4Office of the Chief Coroner of Ontario, Toronto, Ontario, Canada
  5. 5Child Health Evaluative Sciences Program, SickKids Research Institute, Toronto, Ontario, Canada
  1. Correspondence to Dr Christopher S Parshuram; chris{at}

Statistics from

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.


Timely identification of severe illness provides opportunity to prevent clinical deterioration and reduce morbidity and mortality.1–4 An expert panel consisting of parents and multidisciplinary providers created the Severe Illness Getting Noticed Sooner (SIGNS)-for-Kids as a public health tool to help caregivers identify and articulate the manifestations of severe illness in children.5 The tool was created during a consensus development workshop to incorporate expert opinion and previously published symptom checklists of severe illness.6 7 The SIGNS tool consists of 5 domains (behaviour, breathing, skin, fluids and response to usually effective treatment) and 13 subdomains. The objective of this study was to review the sensitivity of the SIGNS-for-Kids items in a cohort of children with fatal illnesses as an initial validation step.


A retrospective review of paediatric deaths was performed using records from the local coroner’s office. Information available included summative coroner reports and comprehensive medical records. Eligible patients had an index deterioration event described between January 2013 and December 2018, were ≤18 years and ≥36 weeks gestational age at the time of event, had observations or reports for >6 hours prior to the index event and were not in an intensive care unit (ICU) or under anaesthetist supervision at the time of index event. Index deterioration events were defined as cardiac arrest, respiratory arrest, intubation, transfer to ICU, interfacility transfer or urgent surgical procedure. We excluded patients with trauma or sudden unexplained death. The main outcome was the presence of SIGNS items. Secondary measures were the time before index event that SIGNS was present and the time to death from the index event. Analyses were descriptive and were performed using SAS V.9.4 (SAS Institute).


We screened 200 records and excluded 150 children with either traumatic death (n=77), less than 6 hours of documentation before the index event (n=61), ineligible age (n=7) or a diagnosis of sudden unexplained death (n=5). The 50 studied children had mean (SD) age of 32.0 (50.8) months. The index events were cardiac arrest in 25 (50%), endotracheal intubation in 12 (24%), interfacility transfer in 8 (16%) and respiratory arrest in 5 (10%). The diagnoses associated with the index event included respiratory in 33 (66%), circulatory in 24 (48%), infectious in 18 (36%) and endocrine/metabolic in 8 (16%).

One or more SIGNS criteria were present prior to index event in 48 (96%) children. Items from two SIGNS categories were present in 14 (28%), of which the most common combinations were behaviour/breathing (n=5) and behaviour/fluids (n=5). Items from three SIGNS categories were present in 6 (12%), with the most common combinations being behaviour/skin/fluids (n=2) and behaviour/breathing/skin (n=2).

SIGNS was present ≥24 hours before initial escalation in 28 (56%) children, and death occurred within 48 hours after index event in 42 (84%) (figure 1). Proportions of children with each SIGNS item present and median duration prior to index deterioration are presented in table 1.

Table 1

Proportion of children with SIGNS items present prior to index event

Figure 1

Data were abstracted from Ontario Coroners records of 50 paediatric (>8 hours of age and <18 years) non-traumatic deaths with escalation events (intubation, transfer to ICU, CPR) that occurred outside an ICU or Anaesthetist supervised area; 48 (96%) had one or more SIGNS criteria documented in the available records. More than half of cases had SIGNS criteria documented for 24 hours before index escalation event and 42 (84%) died within 48 hours after escalation.

Breathing abnormality was observed in 26 (52%) with median duration of 36.0 (8.3–49.5) hours before index event. The most common breathing abnormality was ‘noticeable breathing’ in 20 (40%). Behaviour abnormality was observed in 22 (44%) with median duration of 24.0 (8.3–43.8) hours prior to event. The most common behaviour abnormality, reduced independent actions, occurred in 16 (32%). Fluid abnormality was present in 21 (42%) with median duration of 24.0 (7.0–55.0) hours prior to event. The most common fluid abnormality, minimal intake, occurred in 13 (26%). No children had jaundice in the first month of life or a purple rash. Death occurred within 48 hours after index event in 42 (84%).


This retrospective evaluation suggests the potential utility of the SIGNS criteria to identify children with severe illness. Almost all children with fatal illness had one or more SIGNS, and SIGNS items were present long enough to permit timely intervention and care escalation. The four main limitations of this study are: its retrospective design, the selection of children with fatal illnesses who may have progressed to death even if intervention had occurred earlier, the exclusion of children with sudden unexplained death (where SIGNS should be absent), and parental recall and clinician ascertainment biases that may follow recognition of severe illness. Future prospective studies in children with and without critical illness are needed to further evaluate the validity and caregiver usability prior to clinical implementation.

Ethics statements

Patient consent for publication

Ethics approval

This study involves human participants and was approved by The Hospital for Sick Children Research Ethics Board. Reference number 1000064139. This was a retrospective, secondary use study reviewing coroner office records of deceased patients. It would not have been possible to obtain patient consent.


The authors would like to acknowledge the SIGNS-for-Kids Investigators Steering Committee for development of the tool and Kristen Middaugh RN for study coordination. Thank you to Andrew Stephen and Kathy Kerr of the Office of the Chief Coroner for study site coordination.



  • Contributors KD-P and DH conceptualised and designed the study, interpreted the data and critically reviewed the manuscript for important intellectual content. CH, CT and LM-S conceptualised and designed the study, acquired the data and reviewed the manuscript for important intellectual content. SBG and CSP conceptualised and designed the study, analysed and interpreted the data, drafted the initial manuscript and critically revised the manuscript for important intellectual content. All authors approved the final manuscript as submitted and agree to be accountable for all aspects of the work.

  • Funding The Study was in part supported by unrestricted grant from the Canadian Patient Safety Institute.

  • Competing interests None declared.

  • Provenance and peer review Not commissioned; externally peer reviewed.