Abstract
Background:
Preventing drowning in adults is a complex, multifaceted injury prevention issue. The adult drowning rate in Australia is continuing to rise. In 2023, more than half of those who drowned were aged over 45 years. While there has been a call for robust, evidence-informed drowning prevention interventions across the life course, evidence of evaluated, drowning prevention programmes for adults is lacking. For more than a decade, drowning prevention researchers and practitioners in Western Australia (WA) have worked in partnership to design and evaluate evidence-informed programmes. This paper describes formative research to develop ‘Make the Right Call’ (MTRC), an adult water safety programme in WA.
Methods:
A staged, mixed-methods, formative evaluation was undertaken to inform the design of MTRC.
Stages:
The approach involved cluster analysis of coronial data 2008–2018 (n = 93), interviews with adults aged 45–64 years (n = 10) and 65 years and older (n = 15), theory mapping, survey design, content and face validity testing of a baseline questionnaire with content experts (n = 11) and a refined instrument to collect knowledge, and data on norms and water-based activity. The subsequent MTRC programme comprised a media campaign, community swimming and safety classes, lifejacket trade-in, subsidised first-aid training and community education.
Discussion:
The results of each stage informed the design and evaluation of a new evidence-informed and theory-driven drowning prevention programme for adults aged 45 years and older, delivered by the peak drowning prevention agency in WA, Royal Life Saving Western Australia. An investment in mixed-methods research by the partnership added rigour and credibility to the programme and evaluation design.
Conclusion:
To the best of our knowledge, this is the first adult water safety programme to apply a theory-driven, evidence-informed approach to its development, implementation and evaluation. The partnership was vital in maximising and accelerating the acceptable transfer of results to enhance knowledge creation and, ultimately, the design of the MTRC drowning prevention programme.
Keywords
Introduction
Preventing adults from drowning is a complex, multifaceted issue for researchers, practitioners and policymakers (1). In countries such as Australia, drowning among people aged 45 years and older is a growing concern. For example, in 2023, adults 45 years and older represented more than half of the drowning deaths (2), a figure above the 10-year average. Consistent with other high-income countries (HICs), life expectancy is increasing in Australia (3), and the normal ageing process is linked with loss of muscle strength, agility, balance, flexibility and bone density, further increasing the risk of drowning as adults age (3,4). Internationally, there is some understanding of the risks of unintentional fatal and non-fatal drowning among adults, including older adults (3,4). For example, some risk factors (age, male gender, ethnicity, low socioeconomic status, and the use of alcohol and medications) (3,5–7) are consistent globally. However, other factors are specific to cultural and social norms (e.g. gender roles), occupation (e.g. commercial fishers) and geopolitical environments (e.g. migration and armed conflict) (5,7–9). Accordingly, this presents a complex set of individual, environmental and socioecological factors that need careful consideration in the design and delivery of drowning prevention interventions segmented by age.
A life-course approach is helpful in addressing public health issues (10). It can be used to profile age segments, assess exposure to hazards and differentiate between risk factors at the individual, social and economic level (11,12). Importantly, application of a life-course approach can facilitate an understanding of the temporal aspects of risk practices and environments (12). Public health researchers, practitioners and policymakers can subsequently develop tailored pre-vention interventions (13). Historically, this approach has been common practice when addressing drowning prevention for other age groups (14,15). However, until recently, interventions targeted at adults 45 years and older have mostly been absent from the drowning prevention landscape (4–6). Of interest, publications describing the methods that underpin the design, implementation and evaluation of drowning prevention programmes are either inconsistent (1,16) or lacking for specific segments of the population, notably adults 45 years and above. Accordingly, as highlighted in a recent review of adult drowning interventions, there is a call for multi-level, multi-strategy, context-specific, adult-focused prevention interventions across policy and practice (5). Contemporary evidence that identifies effective interventions that contribute to prevention efforts is an essential first step in addressing the challenge; there is a need for evidence that extends beyond epidemiology and risk factors (1) to better capture factors related to programme design, implementation and evaluation to enhance the existing evidence base about what works, for whom and why.
The literature reinforces this broader gap between health research (e.g. evidence-informed programmes, guidelines and policies) and practice (i.e. what is routinely delivered in real-world community settings) (17). One mechanism to accelerate the ‘know’ ‘do’ pipeline is the movement away from a model whereby researchers focus on dissemination of research findings only, to one of reciprocity between researchers and practitioners (18) – coined ‘knowledge translation,’ a dynamic and iterative process that includes the synthesis, exchange, application and dissemination of the evidence to improve the health of populations (19).
In this context, for more than a decade, researchers and practitioners in WA have been working in partnership to translate knowledge into action through evidence-informed drowning prevention interventions. The partnership has been previously used to inform the design, delivery and evaluation of interventions targeted at toddlers (20,21), young adults (22,23) and culturally and linguistically diverse parents (24). Consequently, interventions for toddlers and young people are now relatively mature, based on the theory and application of comprehensive health promotion strategies (25). In particular, public education campaigns have been delivered for young people to reduce alcohol-related drowning risk (23,26) and toddlers in relation to parental supervision for children 0–4 years (20). These campaigns have been regularly evaluated, and demonstrated an increase in awareness, knowledge and safe behaviours (20,21,27–29).
However, as highlighted above, the evidence base and resulting interventions to prevent adults drowning, in particular older adults, is limited. Recognising the importance of timely evidence and reciprocity as vital components of the decision-making process (1,17), there has been a more recent and consistent call for methodological advancements and wider publication of evidence underpinning the design, implementation and evaluation of all drowning prevention interventions. Accordingly, the partnership sought to establish a theory-driven, health promotion intervention to address the stated gaps described earlier. This paper describes the formative research and development of the ‘Make the Right Call’ (MTRC) water safety programme for adults in WA.
Methods
A staged, mixed-methods approach was used to gain an in-depth understanding of behaviours, processes, practices, and relationships in context (30) for the development of the MTRC programme. This consisted of four stages: cluster analysis of coronial data; in-depth individual interviews; theory mapping; and survey instrument design, content and face validity testing. The study obtained ethical approval from the Curtin University Human Research Ethics Committee (HRE2019-0347) and Justice Human Research Ethics Committee (CF/16/17315).
Make the Right Call (MTRC) programme
The subsequently developed adult water safety programme, MTRC, is delivered by the peak drowning prevention agency in WA, Royal Life Saving Society Western Australia (RLSSWA). The programme aims to increase knowledge and skills amongst people aged 45 years and older regarding preventing drowning in and around water. The MTRC programme uses a multi-strategy approach combining educational, behavioural and environmental strategies and comprises a media campaign (radio, print and social media), community swimming and safety skills classes, lifejacket trade-in, subsidised first-aid training and community education (https://royallifesavingwa.com.au/programs/make-the-right-call).
Context: the partnership
Formative programme development was under-taken by the partnership; the partnership between Curtin University (researchers) and RLSSWA (practitioners) has existed for more than 10 years. It takes place within a system of interactions between researchers, practitioners, programme ambassadors and volunteers, and varies in intensity and level of engagement depending on the nature of the research and the findings to be disseminated (31). The researchers were early- and mid-career (EMC) public health researchers, a PhD scholar, an honours student and a research assistant. A practitioner staff member from the agency was co-located with the researchers 1 day per week. The EMC researchers had over 25 years of mixed-methods evaluation research and practice expertise across health promotion priority areas, including injury prevention, physical activity promotion, alcohol and other drugs, migrant health, and sexual health. The practitioners consisted of a senior health promotion and research manager, a health promotion officer, and a research and evaluation officer. The senior health promotion and research manager had over 18 years of experience in designing and delivering health promotion interventions in the drowning domain and has been part of the partnership since its inception.
Stages
The following section outlines the stages undertaken by the partnership to inform the design and evaluation of MTRC.
Stage 1: Cluster analysis
In this stage, we examined the heterogeneity of older adults who have drowned, and identified population subgroups in WA. The full methods have been reported elsewhere, however for this stage, older adults is used to describe people 65 years and older, as defined in the National Australian Water Safety Strategy 2016–2020 (32). Briefly, we used cluster analysis to segment the population by examining coronial data 2001–2018 (n = 93) and identified four groups: (1) men who boat and fish in company, (2) affluent men with poor health, (3) non-drinkers who boat and fish, and (4) older men who slipped or fell. Males aged 65–74 years were particularly at risk while participating in various aquatic activities such as boating, fishing (including rock-fishing) and swimming for recreation. This stage provided insights into an underserved area and was used to directly inform Stages 2, 3 and 4 of the MTRC programme development.
Stage 2: In-depth interviews
Semi-structured, in-depth interviews were used to explore factors associated with drowning among adults aged 65 years and over in WA. A series of semi-structured interviews examined knowledge, attitudes, beliefs and behaviours (n = 15). The interviews were analysed using a hermeneutic phenomenological interpretation (33,34) whereby three steps were taken to identify common patterns of meaning from an individual’s lifeworld. The full methods and findings are reported elsewhere (33). Nine constitutive patterns emerged: Lumps and Bumps, Sunset Mentality, the Jaws Effect, Unpredictability versus Ability, Moving Every Muscle, Swim for Survival, Drown-Proofing, Risk and Responsibility, and Bondi Bias (33). Findings suggested that individuals who perceived themselves as strong swimmers had a decreased perception of risk, while participants who self-identified as weak swimmers were more likely to avoid risks and modified their behaviour. Low water safety literacy and a lack of understanding of the risk factors of drowning were findings that had a direct impact on the survey development and the strategy mix for the MTRC programme.
In addition, the drowning prevention agency conducted a series of semi-structured, in-depth interviews (n = 10) to explore factors associated with drowning among adults aged 45–64 years in WA. The interview schedule and pre-interview demographic survey were adapted from the materials used for the 65+ age group. The interviews with 45–64-year-olds found overlap in seven of the nine patterns, with some overlap in Lumps and Bumps, particularly for women. However, no overlap was seen in the theme of Sunset Mentality. The point of difference between the younger target group and interviews with the older adult target group (over 65 years) was the continuum as people move through different life stages, from parents of young children to reaching retirement and finally old age, which was used to inform the strategy mix for the MTRC programme.
Stage 3: Theory mapping
Behavioural theory was used to structure the ‘water safety and aquatic’ questions in the survey and to inform campaign messaging. Findings from the cluster analysis and interviews informed the selection of constructs from social cognitive theory (SCT) (35), and the health belief model (HBM) (36), which were used in the survey question design and testing. Subsequently, the partnership employed SCT (35), HBM (36) and gain framing (37,38) to guide the development of MTRC campaign messages and strategies. During message and strategy development, specific behavioural theory constructs were used to identify appropriate cognitions, emotions and processes to target in campaign material execution (21). A sample of SCT and HBM constructs, our rationale and proposed execution used to inform messages and strategies are provided in Table 1.
Sample summary of theory mapping for MTRC campaign messaging.
Stage 4: Survey development
This stage included four steps: (1) pilot survey development, (2) content validity, (3) face validity, and (4) (formative) survey refinement.
Step 1: A draft pilot survey was developed that consisted of validated and reliable questions sourced from previous injury prevention surveys, with some adaptation to minimise measurement error and improve conciseness (39). The survey consisted of 73 questions in five key domains: (1) demographics, (2) general wellbeing and daily activities, (3) travel patterns, (4) water safety and aquatic participation, and (5) media consumption. The pilot survey was created using Qualtrics XM software.
Step 2: Content validity involves the evaluation of a new survey instrument to ensure it accurately captures the construct domains it is intended to measure (40,41). A content evaluation panel of 5–10 experts is considered sufficient (40). Accordingly, a content validity questionnaire (CVQ) was embedded into the pilot survey. The CVQ was used to assess each item using a three-point scale (i.e. not necessary, useful but not essential, and essential) (42). Accounting for attrition and non-response, a larger than necessary sample (n = 15) of researchers and practitioners, working in either drowning prevention and/or with extensive understanding of qualitative research and questionnaire development was invited to participate. The survey was distributed via the Qualtrics ‘email survey link’ to experts, with basic instructions on completing the content validity testing. Eleven experts (n = 11) undertook this, equating to a 73% completion rate.
Content validity ratios (CVRs) were calculated for each survey item using the method developed by Lawshe (43) and recalculated by Ayre and Scally (40). The CVR quantifies content validity (43). The proposed cut-offs described by Ayre and Scally were used for the analysis (40) whereby the minimum and maximum CVR were 0.65 and 1, respectively. All 73 survey items achieved a CVR between 0.75 and 0.99. The lowest CVR (0.75) aligned with two questions from Domain 1 (i.e. What is your age? and What is your postcode?). The highest CVR (0.99) aligned with nine questions in the pilot survey, seven related to Domain 5 – media consumption (e.g. Do you use any of the following social media?) and two from Domain 2 – general wellbeing (e.g. Are you restricted in everyday activities because of this/any of these conditions?).
Step 3: A related concept to content validity is the face validity of a survey, a subjective determination of whether a set of items appears (at face value) to assess the concept of interest (44). Table 2 provides a sample of the experts’ comments.
Example face validity comments from experts (n = 11).
Step 4: Based on the content- and face validity findings, 19 questions were removed from the pilot survey, and nine questions were added to form the formative survey. Eight questions were rewritten to increase readability and/or clarity. The final formative instrument was a population-level, cross-sectional online survey consisting of 63 questions, administered over a 4-week period from 7 January to 4 February, 2020. Snowball sampling was used by distributing a Qualtrics survey link to key community organisations across WA for promotion within their networks. These organisations included local councils, shires, community health centres, churches, multicultural centres and sporting clubs. Members were encouraged to forward the survey link to relevant family and friends. The sample size (n = 384) was calculated to ensure a margin of error of 5% and a confidence level of 95%. The final complete data sample (n = 414) was used to inform the design of the MTRC campaign.
Discussion
In this paper, we describe a staged, mixed-method approach to exploring factors associated with drowning among adults in WA. Our approach, facilitated by a partnership between researchers and practitioners, provided an evidence-informed and more complete picture of water safety perspectives and practices among Western Australian adults. In Stage 1, we used coronial drowning data to identify four segments of people who had drowned (32), and Stage 2 captured the voices and lived experiences of the proposed target group for a WA older adult drowning prevention programme (33). These narratives added credibility to the findings and supported subsequent strategy selection. Consistent with the literature (45,46), Stages 3 and 4 ensured the evaluation design was fit for purpose and would be appropriate for participants. The research findings from each stage were used to inform the development of a population-wide health promotion water safety programme to target adults (aged 45 years and over across WA), MTRC.
Applying a comprehensive approach
Very few drowning prevention interventions describe and publish in detail the steps undertaken to inform their design and evaluation (1,5). As far as the authors can discern, no peer-reviewed published papers describe stepwise mixed-methods formative research undertaken for an adult-focused drowning prevention intervention in a HIC. Predominantly, the drowning prevention literature focuses on risk factors and drowning burden (1,47). There is a relatively small body of literature on the collective evidence, theory, and survey design underpinning a drowning prevention intervention (5). This finding is also reflected in the broader health promotion literature, which demonstrates a lack of reporting on evaluated interventions (45,48) and limited focus on factors influencing implementation (16).
The broader literature supports the application of a comprehensive health promotion approach, which suggests interventions that incorporate action across the socioecological spectrum (including behavioural and structural strategies) facilitate improvements most effectively (49). In this research, each stage built the evidence to inform a comprehensive health promotion approach to MTRC using a multi-strategy mix. For example, due to the evidence that suggested that the target group were ‘men who boat and fish in company’ and ‘non-drinkers who boat and fish’ (32), population-level behavioural strategies were implemented focusing on communication. Radio messages and print media were developed using vintage graphics and colour themes to resonate with men who boat and fish, for example, ‘Angler Andy liked fishing all day, didn’t wear a lifejacket now he’s drifting away’. Recognising the influence of the environment, in-depth interview participants noted that water conditions prompted the use of safety equipment such as life jackets and observed that surrounding warning signs (33) and socioecological strategies were being developed. These included Old4New, where adults could trade their worn-out or obsolete life jackets for contemporary, new self-inflating jackets. Finally, as the research suggested, grandparents are often called upon to supervise and interact with their grandchildren in backyard and local pools (34), group-level strategy development focused on skill building. For example, Aqua Skills, a water safety and lifesaving skills programme specifically for adults was delivered over four sessions at local aquatic centres (50).
Closing the know–do gap
Timely evidence and knowledge translation actions are vital to decision-making (51). They support the delivery of target-group-appropriate programme strategies and collecting relevant evaluation data. Outcomes can provide helpful insights for programme improvement that also satisfy the requirements of funders and the needs of policymakers (45,48). Yet, despite best efforts and consistent calls for effective strategies to reduce the research–practice pipeline (17,31), a know–do gap continues to exist (31). This is not unique to drowning prevention. However, the recent literature on drowning prevention highlights an imperative to maximise research uptake in practice and policy (1,16). This is to address the magnitude of global fatal and non-fatal drowning and the complexity of factors that underpin drowning as a major public health issue (1).
We reflect here on the role of long-standing partnerships between researchers and practitioners in facilitating evidence-informed decision-making. Such partnerships have been used successfully to increase the evidence base and workforce capacity in other health areas (52). Key features for successful research–practice partnerships, as described in the literature, include time, trust, proximity and reciprocity, generating mutual learning (51,52). For the past decade, the partnership has worked closely to engage in meaningful research, understand the context, and be flexible in our research designs. This reflects a move to what has been described as integrated knowledge translation (iKT) (53): an approach that recognises the active collaboration between researchers and research users in all parts of the research process, including decisions about the methods, involvement in the data collection and tools development and dissemination of research results (31,54). The iKT approach creates knowledge generated throughout the research process to make informed decisions about the programme, the strategy mix and evaluation design composed of evidence-based (i.e. research) and practice-based information (i.e. lived experience) (31,55).
In the context of this study, researchers and practitioners worked alongside each other to ensure that findings were applicable to the end user, an approach implemented at the beginning of the research and subsequently embedded throughout the entire process. The partnership was vital in maximising and accelerating the acceptable transfer of the results to enhance knowledge creation and, ultimately, the design of the MTRC drowning prevention programme.
Strengths and limitations
To the best of the authors’ knowledge, this is the first paper to describe a formative, staged, mixed-methods approach to address drowning prevention in adults. This staged approach used quantitative and qualitative data to inform the design of the MTRC programme and the evaluation instrument used for data collection by the drowning prevention agency in WA. Triangulation of multiple data sets provided a broader and more complete picture (56) of water safety among adults, and the findings directly assisted the drowning prevention agency in delivering a programme that would engage the target audience. In addition, the voices of both those who identify as swimmers and non-swimmers were captured, another strength of the research. The researcher–practitioner partnership employed in this study is also a strength as it involved practitioners from the inception of the research to ensure effective translation and to support a response to a real-world research question. However, there were some limitations: the cluster analysis data only captured incidents in WA, the qualitative interview participants were predominantly from metropolitan WA and no individuals with low socioeconomic status were captured. However, this does align with the profile of those who drown, so may also be deemed a strength. Several other possible limitations were identified. The agency conducted interviews with those aged 45 – 64 years. The pilot survey was distributed by the researchers to the experts via an online link. For some sections of the survey, such as travel and general wellbeing, the context was initially unclear to the experts and therefore, many deemed them not relevant. In relation to the content validity testing, the expert feedback was limited to 11 participants; in future, if CVR is employed, researchers should aim for feedback from 13 panellists for each item, as this would increase the CVR to greater than the critical value (40).
Conclusion
This study presents a comprehensive staged, mixed-methods approach to exploring factors influencing drowning amongst WA’s adults, a real-world concern for public health. We have provided comprehensive insight into the methods, theory and survey instrument design underpinning a first-of-its-kind drowning prevention programme for adults in WA. The research–practice nexus was vital in the design of the MTRC drowning prevention programme and provides a template for those working in drowning prevention. The evidence will help mitigate the lack of literature describing the formative stages underpinning the design of an adult drowning prevention intervention in the Australian context, findings that are important to addressing the implementation gaps in public health.
Footnotes
Acknowledgements
The authors thank the individuals who took the time to participate in this research. The authors also thank the Royal Life Saving Society Western Australia for their contributions. Views articulated in this article are those of the authors and not necessarily those of the funding agency.
Author contributions
JEL, GC and LN conceived the study design. Material preparation, data collection and analysis were performed by MA, JEL, GC, LN and MDB. The first draft of the manuscript was written by JEL and GC. All authors read and approved the final version of the manuscript.
Declaration of conflicting interests
The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Funding
The authors disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This study was supported in part by funding from the Royal Life Saving Society Western Australia Inc. (grant no. CO6283).
