Analysis of existing literature

A literature review is a way to critique, summarise and compare the outcomes of previous studies investigating drowning. It can be used to investigate the burden of drowning in a specific setting, identify trends or patterns of drowning over time, compare trends and patterns of drowning between different settings, investigate and compare the effectiveness of previously implemented drowning reduction interventions, and highlight gaps and appraise strengths and weaknesses in previous drowning-related research. Before starting a literature review, it is important to develop a search strategy:

• Clearly define the goal of the review and specify a research question: what is the topic or issue of interest? What key words best describe it? These key words will make up your search terms. Consider including synonyms associated with drowning as key words, such as ‘near-drowning’, ‘submersion’ and ‘immersion’.
• Set search criteria: would you like to only review research from a certain country? From a certain time period? Are you only interested in studies that use specific approaches to collecting eir data? This will help you set search parameters.
• Choose which sources you would like to search: these should have some relevance to drowning. These can be academic databases (for example, PubMed and Medline which contain literature on a broad range of public health topics, including drowning) or sources of grey literature (government reports investigating injury burden, or websites of institutions involved in drowning prevention).

Once a search strategy has been developed, it is time to methodically search each data source selected, applying the search terms and search criteria defined. Keep a record of how many results each search produces. Note how many search results are duplicates between sources, how many are not directly relevant to the research question, and how many do not fit within the search criteria. This information can be documented through making a PRISMA flow diagram.

Obtain full text versions for all relevant studies and summarise their outcomes in a data extraction table. It is useful to critically appraise each study included in the review; making comments about its quality, reliability, and the generalisability of its outcomes. Many free quality appraisal tools are available online, each designed for assessing different types of studies. Quality appraisal is an important part of the literature review as it acknowledges which studies have more validity than others, showing which results are most likely to be accurate and trustworthy.  

It is useful to compare the similarities and differences in the outcomes of each study. The repetition of a study outcome between sources suggests that it is true. Outcomes of a literature review may reveal a high risk population who would most benefit from a drowning prevention intervention, or may guide the selection of a drowning prevention intervention to trial. If too few results are identified through the initial search, try a) adding additional search terms, b) broadening the search criteria, or c) searching additional databases.

Advantages

• No primary data collection required. 
• Prevents the repetition of research that has been previously done. 
• Can be performed by one researcher. 
• Limited equipment is required. 
• Data from numerous sites, populations and participants is able to be analysed and compared instantly. 
  
  

Disadvantages

• Not all academic publications are free to view and membership to academic databases tends to be very expensive. 
• Old publications, or studies analysing old data, may not report results that reflect the current context. 
• Often, studies selected to include in a review are not directly comparable due to different methodologies used and differences in study populations. 
• Can be time consuming to perform. 
• There may be limited previous research investigating drowning in specific contexts.
  
  

Context

It is beneficial to perform a literature review when previous studies on drowning have already been conducted in a relevant setting and time period. Although not all research studies identified will be directly relevant to your work, they may show how drowning affects similar populations or identify existing approaches to reducing drowning in comparable settings.