|
Abstract
|
The efficient study of drowning has been hampered by the use of non-uniform terminology. At the 2002 World Congress on Drowning, a consensus guideline for the definition of drowning was established: "the process of experiencing respiratory impairment as the result of submersion/immersion in a liquid medium." The guideline further states that the so called "Utstein style" reporting of drowning incidents should be limited to 3 outcomes, morbidity, no morbidity, and mortality. This precludes Non-Utstein Drowning Term Use such as "wet," "dry," "near," "active," "passive," and "secondary" drowning. In 2005, The World Health Organization (WHO) issued a bulletin incorporating the World Congress on Drowning Definition. This systematic review sets out to identify the prevalence of Non-Utstein Drowning Term Use.
Data Sources
We searched all English-language articles, published from January 1, 2003 to July 15, 2010 using Medline/Pubmed and Web of Science. The search terms used were: "drowning," "dry drowning," "drowned," "wet drowning," "active drowning," "silent drowning," and "secondary drowning."
Data Selection
Using a novel citation management system, (Endnote Web), the titles and abstracts of the initial search yield were independently screened for inclusion/exclusion criteria by 2 of the authors.
Inclusion criteria
(1) articles relevant to the public health, impact, surveillance, treatment, pathophysiology, or prevention of drowning, and (2) article titles that contained the terms "drowning," "immersion," "submersion," or "water-related injury/mortality/safety." Exclusion criteria: (1) purely forensic or microbiologic articles (2) opinion/editorials, (3) non-peer reviewed publications, 4) non-human studies, 5) inability to obtain reference. Data Extraction: 2 trained abstractors using a standardized data tool independently reviewed the final article cohort. Elements extracted were: presence of NDTU; presence of any drowning definition.
Results
The primary measure was the composite prevalence and 95% confidence interval (95%CI) of any NDTU. The initial search yielded 1,148 articles. 119 were excluded because the article was non-English, 789 did not meet inclusion criteria, and 13 were excluded due to an inability to obtain full text. The final cohort consisted of 227 articles from 131 journals. The IRR for the data extraction by the 2 reviewers was very high [Kappa (95%CI) = 0.929 (0.88,0.98)]. The prevalence (95%CI) of NDTU was 42.7 (35.6,48.4)%., with 97% (n=94/97) attributed to the use of "near drowning." The terms "active," "passive," and "silent" drowning were not found in any of the articles. 37/227 (16%) contained an explicit drowning definition, of those, 18/37 48% (95%CI=32,64) used the Utstein-style definition. For those NDTU-containing journals, the mean (SD) Impact factor was 2.24 (1.84) [range=0.132-13.66) and the median cited half-life was 6.1 years.
Conclusion
The term "near drowning" continues to persist in high and low impact medical journals despite the World Congress on Drowning, and WHO 2005 articles strongly suggesting against its use. The presence of conflicting terminology hampers the study of prevention, rescue, and treatment of drowning among the scientific community. The journals in which this and other non-Utstein drowning terms appear serve as targets for editorial intervention.
Language: en |