Article Text

Download PDFPDF

Unintentional pediatric submersion-injury-related hospitalizations in the United States, 2003
  1. R H Cohen1,
  2. K C Matter2,
  3. S A Sinclair2,
  4. G A Smith1,
  5. H Xiang1
  1. 1
    The Ohio State University College of Medicine, Center for Injury Research and Policy, The Research Institute at Nationwide Children’s Hospital, Columbus, Ohio, USA
  2. 2
    Center for Injury Research and Policy, The Research Institute at Nationwide Children’s Hospital, Columbus, Ohio, USA
  1. Dr H Xiang, The Ohio State University College of Medicine, Center for Injury Research and Policy, The Research Institute at Nationwide Children’s Hospital, Columbus, Ohio, 700 Children’s Drive, Columbus, OH 43205, USA; huiyun.xiang{at}


The objective of this study was to examine the demographic characteristics and hospital resource utilization of submersion-injury-related hospitalizations among persons ⩽20 years of age in the USA in 2003. All 1475 pediatric submersion-injury-related hospital discharges in the Kids’ Inpatient Database were identified by ICD-9-CM diagnosis code or external cause of injury code. These cases represent an estimated 2490 pediatric submersion-injury-related hospitalizations nationwide. Inpatient costs for these estimated hospitalizations were ∼$10 million. The overall pediatric submersion-injury-related rate of hospitalization was 3.0 per 100 000 persons. Children aged 0–4 years had the highest rate of hospitalization (7.7 per 100 000 persons). Children with permanent submersion-injury-related morbidity accounted for 5.8% of hospital admissions and 37.3% of hospital costs in our study, and children with submersion-injury-related in-hospital death accounted for 11.6% of hospital admissions and 20.0% of hospital costs in our study. Prevention of submersion injury using focused, proven strategies deserves increased attention.

Statistics from

Submersion injury is second only to motor vehicle crashes as a cause of unintentional injury death among children 1–20 years of age in the USA. Among children age 0–20 years, there were 1126 unintentional non-boat-related submersion deaths and 4453 unintentional non-boat-related submersion injuries treated in US emergency departments (EDs) in 2003.1 More than 50% of children treated in US EDs for submersion injuries in 2001–2002 required hospitalization.2 The World Health Organization (WHO) recommends that researchers describe the magnitude of the morbidity and economic impact of submersion injury.3 However, few studies have examined submersion injuries that require hospitalization.46 The objective of this study was to describe the characteristics of unintentional submersion-injury-related hospitalizations among persons ⩽20 years of age in the USA, and the costs associated with these hospitalizations.


The Kids’ Inpatient Database (KID), part of the Healthcare Cost and Utilization Project (HCUP), collected hospital discharge information on treatments and resource utilization from 3438 non-federal, short-term, general hospitals (including academic medical centers and pediatric hospitals) in 36 states. Few US hospitals are federal. KID’s large sample size allows a wide range of analyses, including description of the economic burden of rare pediatric conditions.

All data were reported at the discharge level; thus, individuals hospitalized multiple times in 2003 may have had multiple records in the 2003 KID. Multiple records for the same individual are not linked in KID. Only pediatric discharges, defined as all discharges where a patient was 20 years of age or younger at the time of admission, were included. The KID does not include patients who were treated in the ED and released, nor does it include patients who died before being admitted to hospital.7

Information on hospital charges provided to KID by participating states was edited by professional coders. Professional fees and non-covered charges were removed, and values were rounded to the nearest dollar. Charges of zero dollars were set to “missing”, and charges greater than US$1 million or less than US$25 were set to “inconsistent.”8 Hospital charges in KID represent the amount billed by hospitals for the services and do not reflect the actual costs or the specific reimbursements that hospitals received for services. Therefore, we converted charges into costs using the cost-to-charge ratio file created by HCUP. This file uses cost information from the Centers for Medicare and Medicaid Services and provides an estimate of all-payer inpatient cost-to-charge conversion for nearly every sampled hospital in 2003 KID.9

International Classification of Diseases, Ninth Revision, Clinical Modification (ICD-9-CM) diagnosis code 994.1 (drowning and non-fatal submersion, bathing cramp, or immersion) was used to identify 1397 submersion-injury-related discharges. Additional cases (n = 78) were identified by submersion-related external cause of injury codes (E-codes) E910.0–E910.9. Most cases (88.7%) had at least one E910 code. Discharges with E-codes indicating assault, intentional self-injury, or undetermined intent (n = 33) were excluded from our analysis.

The WHO recommends that outcomes of submersion injury be classified as death, morbidity, and no morbidity.10 All of our cases were patients who were alive at the time of hospital admission, with discharges retrospectively divided into three outcome groups. A discharge was considered a death if the KID record indicated that the patient had died in hospital. A discharge was classified as submersion with no permanent morbidity if the patient had been discharged from hospital alive without cerebral injury. Cases in which the patient was discharged from hospital alive with major cerebral sequelae were categorized as permanent morbidity. Cases of submersion with permanent morbidity were identified by diagnosis codes 384.1 (anoxic brain injury) and 780.03 (persistent vegetative state).

Data analyses were conducted using SAS V9.1.2 (SAS Institute, Inc, Cary, North Carolina, USA) and SUDAAN V9.0.1 (Research Triangle Institute, Research Triangle Park, North Carolina, USA) statistical software. The actual sample size is a statistically unweighted number and is noted when presented. All other results are national estimates calculated using statistical weights provided with KID. The estimated number of cases was rounded to the nearest 10 when reported in the text of this article. Weighted percentages and 95% CIs of demographic factors were calculated across patient outcome groups of death, morbidity, and no morbidity. Rates of hospitalization per 100 000 persons were calculated by age and gender using population data from the 2003 US Census.11 Hospital costs per discharge were reported by mean, median, and overall sum for each outcome category, and were rounded to the nearest 100 when reported in the text of this article. Weighted percentages and 95% CIs were calculated for length of stay (LOS) across outcome categories. This study was approved by the Institutional Review Board of The Research Institute at Nationwide Children’s Hospital.


The KID collected data on 2 984 129 hospital discharges between 1 January 2003 and 31 December 2003.12 Of these discharges, 1475 were related to unintentional submersion injury of a person ⩽20 years of age. On the basis of these data, an estimated 2490 (95% CI 2230 to 2750) pediatric submersion-injury-related hospitalizations occurred nationwide in 2003. All further results in the text refer to the nationwide estimated cases, and actual sample values are noted when presented in the tables. An estimated 82.5% (95% CI 80.2 to 84.6) of cases had no permanent morbidity, 5.8% (95% CI 4.7 to 7.2) of cases resulted in permanent cerebral morbidity, and 11.6% (95% CI 9.9 to 13.7) of cases had an outcome of in-hospital death.

Table 1 presents national estimates and weighted percentages of selected demographic characteristics of patients hospitalized for submersion injury in 2003. Children aged 1–4 years constituted 53.1% of cases in this study. Age was not a significant factor in outcome. Males had nearly twice the number of submersion injury-related hospitalizations as females (62.4% versus 33.1%, respectively), but the outcome of submersion injury did not differ significantly by gender.

Table 1 Actual sample, national estimates, and weighted percentages of selected characteristics of pediatric patients hospitalized for submersion injury, 2003

Figure 1 illustrates the rate of hospitalization by age and gender. The overall rate of submersion-injury-related hospitalization for persons aged 0–20 years was 3.0 per 100 000 population; the rate for males was 3.6 per 100 000, and the rate for females was 2.0 per 100 000.

Figure 1 Rate of submersion-injury-related hospitalization by age and gender.

Box 1: Recommended strategies for prevention of submersion injury

  • Pool safety strategies

    • Close supervision by adult care giver*

    • Well-maintained four-sided pool fencing to prevent direct entry from house or yard*

      • Fence should include self-closing, self-latching gate*

    • Pool alarms

    • Pool covers

    • Cardiopulmonary resuscitation training for pool owners and children’s care givers

    • Poolside telephone to quickly contact emergency medical services

  • Prevention of submersion injury in young children

    • Do not allow unsupervised access to toilet

    • Remove water from pails and buckets immediately after use

  • Prevention of submersion injury in older children and adolescents

    • Swimming lessons

    • Water safety education

    • Avoid alcohol use during water recreation

  • Role for healthcare professionals

    • Educate about prevention of submersion injury

    • Advocate water safety legislation

*indicates key prevention point

Sources: Wintemute15; Quan et al17; US Consumer Product Safety Commission21; Stevenson et al22; American Academy of Pediatrics Committee on Injury, Violence, and Poison Prevention.24

Table 2 presents information on LOS and inpatient costs of submersion-injury-related hospitalizations by outcome categories. The LOS in an estimated 59.0% of submersion-injury-related hospitalizations was 0–1 days. Overall, as LOS increased, the percentage of submersion-injury-related hospitalizations resulting in morbidity and death increased. Patients ⩽20 years of age hospitalized for submersion injury in 2003 in the USA incurred hospital costs of an estimated US$9 550 700. Cases of submersion with permanent morbidity had the highest mean (US$52 400) and median (US$42 000) hospital costs, and cases of submersion with no permanent morbidity had the lowest mean (US$4400) and median (US$1800) hospital costs. Discharges categorized as in-hospital deaths or as submersion with permanent morbidity accounted for a disproportionately greater percentage of hospital costs (20.0% and 37.3%, respectively) than their percentage of hospitalizations (11.6% and 5.8%, respectively). Discharges categorized as submersion with no permanent morbidity accounted for 82.5% of estimated discharges, but incurred only 42.4% of hospital costs in our study.

Table 2 Length of stay and hospital costs for pediatric submersion-injury-related hospitalizations, 2003


Males had nearly twice the number of hospitalizations as females, a finding that is consistent with previous research.451314 We found an overall rate of 3.0 per 100 000. The National Center for Injury Prevention and Control (NCIPC) found a rate of 2.8 submersion-injury-related hospitalizations per 100 000 children ⩽20 years old, an estimate that they noted was unstable because of small sample size.1 Ellis and Trent5 studied pediatric and adult age groups in California in 1995 and obtained a rate of 2.8 per 100 000. Our overall hospitalization rate was 3.6 per 100 000 for males and 2.0 per 100 000 for females, which are both lower than previously published estimates.515 NCIPC estimated a rate of 3.9 per 100 000 for males and 1.6 per 100 000 for females in 2003.1

Children aged 0–4 years had the highest rate of hospitalization for submersion injury (7.7 per 100 000). This is consistent with earlier research showing that this age group (1–4 year olds, 0–4 year olds, and 1–5 year olds) experiences the highest percentage of submersion-injury-related hospitalization,56 mortality,16 and overall injury.17 However, our hospitalization rate is lower than previously published rates calculated using California data.513

Earlier research has found that males aged 15–19 years have higher rates of submersion-related mortality than those aged 5–14 years.16 We did not see increased rates of submersion-injury-related hospitalization among males aged 15–20 years compared with those aged 5–14 years. The high case fatality rates observed in older children117 probably means that many adolescent male submersion victims died before being admitted to hospital and were therefore not included in our study.

To our knowledge, our study is the first to estimate hospital costs for pediatric submersion injuries using a national database. Our findings are consistent with previous regional research showing that most patients had a short LOS,513 and that patients with good outcomes had far lower mean hospital charges than patients with poor outcomes.46

Key points

  • In 2003 in the USA, there were an estimated 2490 submersion-injury-related hospitalizations.

  • The cost of these hospitalizations was approximately US$10 million.

  • Males and children 0–4 years of age had the highest rates of submersion-injury related hospitalization.

Several limitations of our study should be considered when interpreting the results. Although we excluded any case with a known E-code indicating intentional injury, ∼10% of the submersion cases in our study had no E-code listed. Some may have been intentional injuries. In addition, we examined KID data that represent only a single year of reported injuries, and there is yearly variation in the number of submersion-injury-related hospitalizations.1 Our determination of factors affecting hospitalization outcome was limited by the information available in the KID. We could not control for pre-hospital factors such as water temperature, duration of submersion, and immediate resuscitation by bystanders.18 We were unable to link multiple discharges for a single patient; however, Langley et al19 reported that 97% of admissions for “accidents caused by submersion, suffocation, and foreign bodies” (E-codes 910–915) were first-time admissions.

We have underestimated the true cost of submersion-injury-related hospitalization, as 371 cases (25.1%) lacked the information necessary to convert charge data into costs. Costs in our study do not include pre-hospital care, non-covered charges, physicians’ fees, or costs for patients who died before being admitted to hospital or who were treated in the ED and released without inpatient admission. Also, indirect costs, such as lost earnings, make up the majority of injury-related costs and are not collected in KID.1420

Despite its limitations, this study has implications for prevention strategies and public policy. Our findings that males and children aged 0–4 years have higher rates of submersion-injury-related hospitalization show that these groups should be the focus of targeted prevention efforts. The high mortality and high mean hospital costs seen among pediatric patients hospitalized for submersion injuries make submersion injury an important public health concern, despite its relatively low incidence.

Several primary preventive strategies are already known. Adult supervision of young children, although unable to prevent all submersion injury, is important.1721 Well-maintained four-sided pool fencing, preventing direct entry to the pool from the house or yard, decreases the number of pool-related submersion injuries.1522 Fences should include self-latching, self-closing gates, which should not be propped open.2122 Box 1 shows additional recommendations for decreasing pediatric submersion injuries. Also, many clinicians rarely, or never, discuss submersion injury.23 They are encouraged to provide age-appropriate risk awareness education and submersion injury prevention counseling, and support legislative measures to increase water safety.24



  • Competing interests: None.

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.