Article Text

Download PDFPDF

Neurocognitive outcomes in children following immersion: a long-term study
  1. Maria Patricia Manglick1,
  2. Frank I Ross1,
  3. Mary-Clare Waugh2,
  4. Andrew J A Holland1,3,
  5. Daniel T Cass1,
  6. Soundappan S V Soundappan1,3
  1. 1 Department of Paediatric Surgery, The Children’s Hospital at Westmead, Sydney, New South Wales, Australia
  2. 2 Kids Rehabilitation Department, The Children’s Hospital at Westmead, Sydney, New South Wales, Australia
  3. 3 Sydney Medical School, The University of Sydney, Sydney, New South Wales, Australia
  1. Correspondence to Mrs Maria Patricia Manglick, Department of Paediatric Surgery, The Children’s Hospital at Westmead, Sydney, NSW 2145, Australia; patricia.manglick{at}


Objective To investigate long-term neurocognitive outcomes after a near-drowning incident in children who were deemed neurologically intact on discharge from hospital.

Design A prospective cohort study of near-drowning children.

Setting 95 drowning and near-drowning admissions, 0–16 years of age, from January 2009 to December 2013, to The Children’s Hospital at Westmead, Sydney, NSW, Australia.

Participants 23 children both met the criteria and had parental consent for the study.

Main outcome measures Identification of the long-term deficits in behaviour, executive function, motor skills, communicative skills and well-being over a 5-year period. Assessment was undertaken at 3–6 months, 1 year, 3 years and 5 years after near-drowning at clinic visits. Physical developmental screening and executive function screening were done using Behavior Rating Inventory of Executive Function-Preschool version (BRIEF-P) and BRIEF.

Result 95 drowning and near-drowning episodes occurred during the study period. 10 (11%) children died, 28 were admitted to the paediatric intensive care unit and 64 directly to a ward. 3 children died in emergency department, 7 children had severe neurological deficit on discharge from the hospital. 23 were subsequently recruited into the study; 5 (22%) of these children had abnormalities in behaviour and/or executive function at some during their follow-up.

Conclusion Children admitted to hospital following a near-drowning event warrant long-term follow-up to identify any subtle sequelae which might be amenable to intervention to ensure optimal patient outcome.

  • Near-drowning
  • Long-term Effects
  • Neurocognitive Outcomes
  • Immersion

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.

What is already known on this topic?

  • The effect of duration of submersion in near-drowning on airway status, cardiovascular circulation and the resulting hypoxic brain injury in children has been well documented.

  • Follow-up of the children over a year or two after discharge and their outcome has been reported only in a few studies.

What this study adds?

  • This 5-year follow-up study showed 22% of the patients recruited with parent’s consent showed behaviour problems, poor communication, executive function and learning difficulties.

  • These children face higher tasks in life compared with typically developing children.

  • The study supports the establishment of a long-term follow-up of children after a near-drowning event to detect problems for early intervention.


Submersion is the act of being completely covered by a liquid, immersion is being partly covered by a liquid (includes face) and drowning is death by suffocation after submersion in a liquid within 24 hours of the incident.1 2

In Australia, aquatic submersion is one of the leading causes of morbidity and mortality in children.3 The Royal Life Saving Society: National Drowning Report 2014 found that drowning remains as the second most common cause of unintentional injury death in children aged 0–4 years (8%) and 5–14 years (4%). Most incidences occur in private pools, public swimming pools, baths, ponds and lakes.4 5 The fatal and non-fatal incidence rates in children among the indigenous population in Australia (Aboriginal and Torres Strait Islander) remain 44% higher than for non-indigenous children. Drowning episodes happened more often in remote and regional areas and were higher for indigenous women than men. The severity of events differed statistically with indigenous status and by remoteness.6 7 A study on drowning deaths between 1988 and 1989 in the UK in 149 children under 14 years of age found a majority of the incidences occurred in baths, domestic pools, ponds, rivers, canals and lakes. Subsequent prevention strategies reduced the number by 30% (n=104) a decade later.8–10 In the USA, drowning remains second to motor vehicle injuries as a cause of death in children aged 1–14 years and the leading cause of death in children aged 1–5 years.11–14 Morbidity from submersion occurs in 12%–27% of survivors.15–20 Preschool-aged male children are at greatest risk of submersion injury, and residential swimming pools are the most common location in this age group.20 21

Assessment of the long-term neurocognitive outcome in children after a near-drowning incident had not been undertaken at our institution prior to this study. Patients with minor respiratory problems and normal outcome with <3 days length of stay at this hospital were not routinely followed up on discharge but were referred to their local doctor. These children may be grossly neurologically intact on discharge from hospital but the long-term cognitive sequelae might not manifest until much later. Our study was designed to follow-up this group of children who are neurologically intact on discharge to objectively quantify the neurocognitive outcomes.


Study design and data source

This study was a national ethics approved prospective study of children with a near-drowning episode admitted to The Children’s Hospital at Westmead (CHW), Sydney, Australia, over a 5-year period, from 2009 to 2013. Parents of near-drowning children with an Injury Severity Score of 9 (online supplementary appendix 1) were approached prior to the discharge for recruitment into the study. The parents were contacted by phone 2 weeks later to enquire on the health of the child and to remind them to return a completed participation slip. Those enrolled were contacted at 3–6 months, 1 year, 3 years and 5 years after drowning episode for follow-up review. Children with near-drowning incidents and severe neurological deficits at the time of discharge were followed up in the multidisciplinary Brain Injury Service of Kids Rehabilitation Department, CHW (figure 1).

Supplementary file 1

Figure 1

Flow chart of drowning admissions to The Children’s Hospital at Westmead (CHW) and dropout process. BI, brain injury; ED, emergency department; ICU, intensive care unit.

Data collected included age, gender, hypothermia, asystole, presence of apnoea, respiratory infection from exposure to contaminated pools and waterways, length of stay and health status at discharge from the hospital. The socioeconomic status was determined based on the suburb and postcode of the Australian Bureau of Statistics–Socio-Economic Indexes for Areas (Data 2006)22 (table 1). The duration under water, often an estimate, was provided by the family, friends and/or paramedics. The types of resuscitative efforts by bystanders and paramedics at the scene and the duration of resuscitation were also collected.

Table 1

Drowning and near-drowning admissions to CHW, years 2009–2013

At each review, clinical assessment was performed by a senior clinical nurse consultant who interviewed the parents and observed the child to evaluate physical health, social interactions, communication and learning skills, and gross and fine motor skills (online supplementary appendix 2). The children were screened for the developmental milestones using published, age-compatible early skills testing exercises (online supplementary appendix 3). The neurocognitive outcome was screened using the Behavior Rating Inventory of Executive Function-Preschool version (BRIEF-P) for children 2 to <6 years of age, and a BRIEF Parent Form for children ≥6 years of age (online supplementary appendix 4). The BRIEF assessment was completed by the parent and analysed by the rehabilitation department clinical psychologist.

The BRIEF-P and BRIEF are screening tools used for clinically based assessment of everyday behaviour. These tools constitute Inhibitory Self-Control Index (ISCI), the Flexibility Index (FI) and the Emergent Metacognition Index (EMI).

T scores (M=50, SD=10) are used to interpret the child’s level of executive functioning on the BRIEF. These scores provide information about a child’s score relative to the score of children in the standardisation sample. Traditionally, a T score ≥65 represents clinically significant impairment.


Ninety-five children who encountered a drowning or near-drowning episode were admitted to the hospital in the 5-year period from 2009 to 2013 (table 1). Twenty-eight were admitted to the paediatric intensive care unit (PICU), and 10 patients died (three in emergency department on arrival, six in PICU and one patient with severe neurological deficit died 16 months later) (figure 2). Of the 95 children, 65% were male. Ninety-three children were <4 years of age, and two were 7–9 years of age. Seven children were followed up by Kids Rehabilitation Brain Injury Services on discharge and were not part of the study group (table 1).

Figure 2

Drowning and near-drowning admissions to The Children’s Hospital at Westmead (CHW), years 2009–2013. *Non-English speaking; #No obvious neurological deficit. ED, emergency department; ICU, intensive care unit; ISS, Injury Severity Score; MVA, motor vehicle accident.

Of the 79 children with no obvious deficits at discharge, nine had developmental problem prior to the near-drowning incident and so were excluded. Eleven (16%) of the remaining 70 children were of non-English-speaking background and 46 (78%) were from disadvantaged areas (table 1). Some mothers were non-drivers, had transport difficulties or had additional young children to look after at home, making attendance at the specified follow-up clinics difficult. Twenty-five were recruited for the prospective follow-up study, although two were subsequently excluded (one had a motor vehicle accident after the drowning incident and another had been identified with a chromosome deletion), leaving 23 in the prospective group (figure 2). Two of these patients had a brief stay in PICU with a mean length of stay of 6.5 days and 21 were ward admissions with a mean length of stay of 2.3 days. The estimated duration of submersion was ≤3 min (table 1).

Eight of the incidents occurred in home pools, six in public pools, two in baths and seven in waterways (dam, lake, river, lagoon, fishpond/tank) (table 1). All children had cardiopulmonary resuscitation except one who did not require any resuscitation. Seventeen of the 23 children were ≤4 years of age, five were 5–7 years old and one was 8 years of age. Eighteen children had good clinical outcomes and returned to their premorbid level of function. Five (5/23) (22%) of the children were identified with late complications: these included behavioural issues, learning difficulties, poor concentration and sleep difficulties (table 2). Not all families were able to attend the scheduled screening follow-up reviews (table 3).

Table 2

BRIEF assessment on five patients, a cohort with abnormal GEC scores

Table 3

Near-drowning follow-up in 23 children, years 2009–2013, normal and abnormal BRIEF scores

The BRIEF assessment showed that four children had poor inhibitive impulse (ISCI), emotional instability (FI) and poor working memory (EMI), hence an overall impairment in executive function with a Global Executive Composite (GEC) score ≥65. These children had behaviour problems, and speech and learning problems with poor coordination (gross and fine motor skills). One child with a GEC score <65 had behavioural issues, and speech and learning problems with poor coordination. Two of these children had difficulty sleeping at night since the incident (table 2). The mean values of the two groups of children, those with normal outcome and those with poor executive function, are tabulated (table 4).

Table 4

Near-drowning in 23 children, BRIEF assessment-GEC, mean values


In this study, 5/23 (22%) children, a small cohort of the total 59 near-drowning children, had late behaviour problems, as well as poor communication, executive dysfunction and/or learning difficulties. This is likely to impact their school tasks and functioning at home. The study shows, for a 95% CI, the probability these children will develop neurocognitive sequelae is 0.07<p<0.44 using Clopper-Pearson binomial.23 The National Injury Health and Welfare, Australia population survey reports that, in the general population, 10% of the children under 12 years of age have behaviour problems and learning difficulties.24 25

The neurological status of previously normal children with near-drowning is dependent on the duration of submersion, airway status and cardiovascular circulation. The duration of submersion, very often underestimated due to the stress of the situation, determines the level of hypoxic-ischaemic injury. This affects the brain, lungs and kidneys, and predisposes them to neurological morbidity, poor learning, memory, executive function, concentration and planning skills. Pulmonary complications such as bacterial infections and lung damage may also occur as a result of aspiration of water.26 27 The severity of the initial ischaemic brain insult depends on the effectiveness of immediate resuscitation, subsequent transfer to hospital and postresuscitation management.8 26–33

In Australia, drowning and near-drowning incidents generally occur between mid-spring (October) and early autumn (March). Hypothermia is less of a problem as outdoor swimming pools, public pools and other aquatic bodies are often 25–29°C in temperature. Children with a near-drowning episode in warm temperatures >15°C, typical in most Australian waters, backyard or public pools and other bodies of water, are not likely to have brain cell preservation similar to the incidences in icy waters.

The children identified with impairment should be referred to the rehabilitation team for investigation of the delayed development.27 The parents, carers and teachers will also be better informed of the child’s development, be more understanding and tolerant of the child’s behaviour problems. It is of great importance that these children are given the necessary guidance resources, therapy and remedial support to help them manage the broad spectrum of issues they encounter in daily life: adapting to social environment, acquiring life skills and better academic outcomes.

Strengths and limitations of this study

Limitations of this study include incomplete follow-up data at all designated time points, the small numbers recruited and poor response from non-English-speaking population. Eleven eligible children (16%) were of non-English-speaking background (table 1).22 Despite the offer of interpreter services and regular media announcement to the community, parents were not always convinced that immersion/submersion would cause later problems especially when the children were ‘normal’ at discharge from the hospital. The study showed five (22%) children recruited had significant behaviour problems, and learning and sleeping difficulty. Further long-term follow-up is needed to validate and to determine outcomes more accurately.


Children admitted to the hospital following a near-drowning event may have long-term sequelae with learning, developmental or behavioural challenges that may not be apparent at the time of discharge. Long-term follow-up in an appropriate developmental or paediatric rehabilitation unit is recommended, especially if the child requires a PICU admission, has had >3 min submersion or has obvious physical cognitive and/or behavioural deficits on discharge. Further prospective study of the more subtle sequelae is needed to assist in targeting interventions for these children with more subtle deficits.



  • Contributors MPM, DTC and SSVS were responsible for the concept, design and plan of the study. MPM and FIR acquired the data. MPM, SSVS, DTC and MCW interpreted the data. All authors reviewed, edited and critically revised several drafts and the final drafts for important content. All authors approved the final manuscript. SSVS and DTC supervised the study and MPM is the guarantor. Suzanne Benson, clinical psychologist and neuropsychologist, Kids Rehab Unit, provided advice, guidance and computerised analyses of the BRIEF-P and BRIEF (Behavior Rating Inventory of Executive Function) assessment.

  • Funding The study was conducted with financial support of the Department of Surgery, Centre for Trauma Care Prevention Education and Research, The Children’s Hospital at Westmead, Sydney.

  • Competing interests None declared.

  • Patient consent Signed consent of participation in the study was obtained from parents of patients as part of the application approval by the National Ethics Committee Australia.

  • Ethics approval The study protocol was approved by the National Human Research Ethics Committee and the Low Negligible Risk and Site Specific Application approval (LNR/SSA) by the Governance Research Services of the Children’s Hospital at Westmead, Sydney.

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