Strengthening the reporting of observational studies in drowning epidemiology (STROBE-D): a protocol for an extension to the STROBE statement

Abstract

Introduction

Drowning remains a substantial global health challenge, and research in this area predominantly relies on observational study designs due to ethical, practical and administrative limitations. Although the Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) statement is available, adherence within the drowning research community has been inconsistent and frequently incomplete. Consequently, essential methodological components and drowning-specific factors are often under-reported or inconsistently described, impeding quality, reproducibility, critical appraisal and evidence synthesis.

Aim

This protocol outlines the development of an extension to the STROBE statement, specifically for reporting observational studies in drowning epidemiology (STROBE-D). The proposed extension aims to provide structured guidance for authors, reviewers and editors, thereby enhancing the clarity, transparency and comparability of reporting practices.

Methods and analysis

The development of the STROBE-D statement will follow established recommendations and methodological frameworks for guideline creation. The process will include a scoping review and initial draft preparation, a two-round modified Delphi process using surveys with multidisciplinary and international expert participation, an expert consensus meeting to finalise the content and manuscript, public comment with interest-holder involvement and a final revision before submission for peer review.

Ethics and dissemination

The Committee on Health Research Ethics in the Region Zealand of Denmark waived the need for ethical approval as this is a consensus study (EMN-2025-09099). Findings will be disseminated through open access peer-reviewed publications, targeted communication through professional networks, conferences and social media platforms. The STROBE-D statement will serve as an extension to the STROBE statement, specifically tailored to observational studies in drowning epidemiology to improve consistency and transparency. Simultaneous publications of the original STROBE-D statement and the accompanying explanation and elaboration paper will be pursued in line with existing best practices.

Strengths and limitations of this study.

• The protocol study for the Strengthening the Reporting of Observational Studies in Drowning Epidemiology statement follows a structured and transparent methodology based on established guidance for reporting guideline development, including a scoping review, Delphi process and consensus meeting.

• Inclusion of a scoping review provides an empirical foundation for identifying gaps in existing reporting practices, rather than relying solely on expert opinion.

• Use of a modified Delphi process with pseudonymised responses reduces the risk of dominance bias and supports balanced input from a multidisciplinary and international expert group.

• The expert group was purposively sampled to ensure content-matter and methodological expertise, and the broader interest-holder perspectives will be included during a public comment phase.

• Although efforts are made to ensure methodological rigour, consensus-based approaches inherently reflect the views of participating experts and may be influenced by selection and participation bias.

Introduction

The current definition of drowning was published in 2005 and has since been endorsed by all major interest-holders: ‘drowning is the process of experiencing respiratory impairment from submersion or immersion in liquid’.^{1 2} Outcomes are classified as death (fatal drowning) or survival, with or without morbidity (non-fatal drowning). In non-fatal drowning, the process of respiratory impairment is stopped before death.³ In 2021, the WHO estimated 300 000 drowning deaths globally with drowning being a major cause of death among children and young people and for people living in low-resource settings.⁴ However, these estimates vary by source.⁵ Contemporary drowning prevention frameworks conceptualise drowning risk as the interaction between hazard, exposure and vulnerability.⁶ This highlights the importance of transparent reporting of factors surrounding drowning incidents, such as contextual characteristics, to strengthen the evidence base for prevention strategies. Despite ongoing efforts in drowning prevention, difficulties remain in gathering consistent and comparable data across settings.7,9 Variations in reporting practices, challenges in assessing severity, differences across socioeconomic and healthcare systems and a range of collection methods complicate comparisons of data and outcomes.⁴ Accurate data collection and transparent reporting of drowning incidents is essential to improving the quality and comparability of future studies and to strengthening research that informs the Drowning Timeline and Drowning Chain of Survival.^{10 11} This includes promoting diverse, multicentre, multinational collaboration.¹²

Observational studies play a vital role in research on drowning incidents, offering essential evidence on risk factors, exposures and outcomes. Given the ethical and practical challenges of conducting experimental studies in drowning research, observational designs provide an alternative for advancing the evidence base in this field. They enable the investigation of naturally occurring events by drawing on various data sources, including lay media reports, records of lifesaving services, non-governmental organisations, civil protection organisations, emergency services, hospital records, national registries and death registries.^{4 13 14} In other areas, such as cardiac arrest, large, combined datasets from observational studies have enabled improved care.¹⁵

However, the unique contextual aspects, including case definition, exposure, rescue and prehospital factors of drowning incidents, may not be adequately addressed by the current Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) statement published in 2007.¹⁶ Preliminary results from an ongoing scoping review indicate that, since 2008, only a minority of published observational studies on drowning incidents have been reported in a manner consistent with the STROBE statement, with frequent omissions including unclear case definitions, incomplete descriptions of exposure circumstances and limited reporting of outcome ascertainment.¹⁷ Developing a STROBE statement extension for reporting of observational studies in drowning epidemiology (STROBE-D) could improve the transparency and consistency by clarifying drowning-specific definitions and contextual characteristics to be reported. This may strengthen evidence synthesis, collaboration and comparison across diverse settings, intervention development and possibly patient outcomes.¹⁸

This study aims to develop the STROBE-D statement to support researchers, reviewers and journal editors in enhancing the clarity and transparency of reporting, thereby improving the quality of observational studies in drowning epidemiology.

Methods and analysis

Study design

The strategy for developing the STROBE-D statement, including checklist and recommendations, is inspired by the Enhancing the QUAlity and Transparency Of health Research (EQUATOR) Network’s methodological framework for guideline development and the strategy for developing reporting guidelines.¹⁹ The EQUATOR Network is an international initiative seeking to improve the quality of scientific publications by promoting transparent and accurate reporting.²⁰ The STROBE-D statement will be developed through a structured, multistep process comprising a scoping review and initial draft preparation, a two-round modified Delphi process using surveys with multidisciplinary and international expert participation, including structured and semistructured questions, an expert consensus meeting to finalise the content and manuscript, public commenting and final revisions prior to submission.^{19 21}

Research question

Which of the existing STROBE items require adaptation and what additional reporting items are necessary to strengthen the reporting of observational studies in drowning epidemiology?

The principal investigators

KB will facilitate and manage all steps of the process in close collaboration with NB and the expert group. Responsibilities for KB and NB include project management, communication with expert group members via email and online meetings, development and piloting of Delphi surveys, data collection and analysis and preparation of the initial draft manuscript. KB and NB will not participate in the consensus-building process as they are responsible for data interpretation.

Selection of expert group members

An international expert group was invited to participate in the development of STROBE-D and to co-author this protocol. Experts were purposively sampled in accordance with EQUATOR Network recommendations to ensure representation across disciplines, interest-holder groups, geographic regions and resource settings.¹⁹ The panel comprises 23 experts from Europe, North America, South America, Africa, Asia and Oceania, including leading drowning researchers, content experts, methodologists, epidemiologists with expertise in drowning epidemiology, prevention, rescue and drowning treatment and journal editors as recommended.¹⁹ This multidisciplinary and geographically diverse composition was intended to maximise the relevance and applicability of STROBE-D across drowning research settings worldwide. All 23 experts are expected to participate throughout the Delphi process. To minimise attrition, participants will receive personalised invitations and reminder emails between rounds. Participation and attrition rates will be reported for each Delphi round. The expert group is responsible for reviewing and editing the protocol, participating in the two-round Delphi consensus-building process, attending a consensus meeting, assisting the principal investigators in incorporating public feedback from the comment period, reviewing and editing manuscript drafts and supporting the guideline’s dissemination, endorsement and adherence.

The development of the STROBE-D statement, including its checklist and recommendations, is a technical and methodological process that demands specialised expertise in epidemiological reporting standards and observational study design. As the primary aim is to produce a statement that is scientifically rigorous, methodologically robust and aligned with the existing STROBE statement, participation is limited to individuals with the requisite expertise.¹⁹ While the broader drowning prevention community, such as lifesavers or emergency medical service (EMS) personnel without formal training in epidemiology or research methodology, can provide valuable practical insights (eg, on feasibility), their input is more appropriately incorporated during the public comment phase rather than the initial development stage.

The initial steps

The strategy for developing the STROBE-D statement is informed by existing paradigms and is divided into five phases.¹⁹ The first phase involves conducting a scoping review of reporting practices in observational studies on drowning epidemiology published after the original STROBE statement in 2007.¹⁷ The protocol for the scoping review has been published previously and describes the database selection, search strategy, eligibility criteria, screening procedures and data extraction methods in detail.¹⁷ Preliminary, unpublished results from this review highlighted the need for a dedicated guideline. Drawing on the findings of the scoping review and previously published STROBE extensions,22,25 the principal investigators will prepare an initial draft of the checklist, incorporating all items from the original STROBE checklist, proposed revisions and additional items.

Premeeting activities

Premeeting activities comprise a two-round, modified Delphi process using surveys and early preparations for the consensus meeting.¹⁹ The Delphi process is a systematic, consensus-building method for gathering expert opinions and achieving agreement when objective information is lacking.²¹ Its purpose is to pseudonymously evaluate the original STROBE checklist, proposed revisions and additional items for inclusion in the STROBE-D statement, as identified during the initial steps by the principal investigators, and to identify any further relevant items before the consensus meeting.²⁶ Data will be collected via online surveys using the Research Electronic Data Capture system.²⁷ Each expert group member will receive a unique participant identification number, which will be used for sending personal reminders and survey links. This method provides pseudonymity, reduces the risk of dominance by any individual or group, and avoids multiple participations. Surveys can be saved at any time, allowing members to access and edit responses until the deadline. Non-respondents will receive weekly reminders until all expert group members have completed the surveys.

Delphi round 1

In the Delphi round 1 survey, the expert group members will review the initial draft of the STROBE-D checklist, provide comments and suggest any additional items they consider relevant. After collecting the responses, the principal investigators will summarise all comments and suggestions. These will be discussed during two online meetings considering differing time zones. Following these meetings, the principal investigators will revise the preliminary STROBE-D statement. The revised statement will then be piloted to allow for further comments before additional revisions are made, and the STROBE-D statement is presented in Delphi round 2. No items will be excluded during this process.

Delphi round 2

The members will be asked to rate the importance of every unique item. Round 2 ratings provide a baseline for each item without excluding any items, serving as a starting point before the consensus meeting and enabling discussions of each item. The question will be phrased as: ‘How important is the following item?’ Ratings will be collected using a 4-point rating scale with the following response options: (1) not important, (2) somewhat important, (3) important and (4) very important. The rating scale will not include a ‘neither important nor unimportant’ option to reduce non-committal responses. All expert group members will have the principal investigators’ contact information to ask questions. The principal investigators will de-identify all questions and answers, as well as the results of the Delphi process, before sharing them with the expert group members.

The consensus meeting

The expert group will convene for a structured consensus meeting to review, refine, and finalise the STROBE-D statement, including detailed discussions of item wording, structure, and overall coherence. The meeting will take place over 2 days, in-person, contingent on securing funding. If funding is unavailable, the consensus process will be conducted virtually through scheduled online meetings and iterative email correspondence, ensuring equivalent opportunities for discussion, revision, and formal approval of the final STROBE-D statement. Approval will require unanimous agreement among the expert group members. The STROBE-D statement will be iteratively revised until unanimous agreement is achieved.

Postmeeting activities

After the consensus meeting, the STROBE-D statement will be published online for public review and comment. Editors of relevant journals and relevant interest-holders, including members of the International Drowning Researchers’ Alliance, the International Life Saving Federation Drowning Prevention Commission and Rescue Commission,^{28 29} the study authors’ research networks, and corresponding authors of all observational drowning studies published since 2008 (identified through the scoping review¹⁷), will be specifically invited to comment on the STROBE-D statement. Invitations will be distributed in English via email, organisational websites, and social media platforms. Public commenting will conclude 8 weeks after initiation, regardless of participation numbers, to ensure timely study progression. Following the public comment period, the principal investigators will prepare the final manuscript draft for critical review and editing by the remaining authors. Multiple, simultaneous publications of the original STROBE-D statement will be sought in high-impact journals that have previously published reporting guidelines.¹⁶ Concurrently, the STROBE-D explanation and elaboration (E&E) paper will be submitted to a relevant journal to facilitate effective dissemination in accordance with best practices.^{19 30 31} The E&E paper will provide a detailed E&E of the recommendations, including illustrative examples of good reporting from observational studies in drowning epidemiology.³²

Postpublication activities

Consistent with existing paradigms,¹⁹ the authors of the STROBE-D statement are committed to actively seeking and addressing feedback and criticism both during and after publication. The group will support implementation and adherence to the STROBE-D statement and promote formal endorsement by relevant journals, organisations, and professional societies. Planned dissemination activities include open online access, targeted communication through professional networks, conferences, and social media platforms.

Discussion

Key results

The STROBE-D statement will extend the STROBE statement to enhance the clarity and transparency of reporting in observational studies on drowning epidemiology, aiming to improve consistency of reporting across studies. This extension will provide structured guidance for researchers, reviewers, and editors, thereby supporting the clarity, transparency, and comparability of drowning reporting practices.¹⁹ Although STROBE-D is primarily intended for observational studies in drowning epidemiology, related fields such as drowning prevention interventions, water rescue operations and aquatic incidents without drowning may share methodological features, and adherence to the STROBE-D statement is recommended.

Key considerations in STROBE-D development

Several methodological and strategic considerations are central to ensuring both the scientific robustness, long-term relevance and impact of the guideline.

First, a scoping review exploring adherence to the original STROBE statement in observational studies in drowning epidemiology represents a critical step in the development of STROBE-D.¹⁷ This process has also been used by the Consolidated Standards of Reporting Trials (CONSORT) group.³¹ Reporting guidelines should address clearly identified deficiencies in the existing literature rather than relying solely on expert opinion. By systematically mapping reporting practices and methodological characteristics in published observational studies on drowning since 2008, the scoping review will provide an empirical basis for identifying gaps, inconsistencies, areas of ambiguity in current reporting and examples of good reporting. This approach ensures that proposed extensions to the original STROBE checklist are grounded in observed reporting challenges specific to drowning epidemiology.¹⁹ While the STROBE-D statement will provide structured guidance for researchers on reporting of observational studies in drowning epidemiology, comparability between studies may also be enhanced by clearly reporting the definitions, terminology frameworks and data templates used. Relevant sources include definition statements,¹ drowning and aquatic incident dictionaries³³ and consensus-based data extraction templates (eg, the Utstein Drowning Template used for reporting of drowning-related resuscitation data³⁴).

Second, the established guideline development process requires sufficient funds to support a face-to-face consensus meeting and associated premeeting and postmeeting activities.¹⁹ This format offers advantages for complex consensus work, including sustained engagement, dynamic discussion and the opportunity to resolve ambiguities in real time. However, the main meeting costs are estimated at $C75 000, and the absence of funding should not compromise methodological rigour or be an obstacle to development of these research reporting guidelines.¹⁹ A structured virtual consensus process comprising scheduled online meetings and iterative written feedback may provide similar opportunities for participation, reflection and formal approval when optimally facilitated. Virtual formats may also enhance inclusivity by reducing financial and logistical barriers to participation, particularly for experts from low-resource settings, where the burden of drowning is highest.⁴ Ultimately, the credibility of STROBE-D will depend more on the transparency and methodological rigour of its development than on whether consensus is achieved face-to-face or online.

Third, public review and open commenting represent an important extension of good practice in guideline development, though this is not explicitly required in the existing recommendations for developers of health research reporting guidelines.¹⁹ Drowning prevention and emergency response involve a wide range of interest-holders, including researchers, clinicians, public health practitioners, rescue organisations, policy makers and community groups. Public review may identify practical barriers to implementation (eg, feasibility), clarify ambiguous wording and ensure that the guideline is relevant across diverse settings.

Finally, the publication of a reporting guideline does not guarantee its uptake or effectiveness in translating to the reduction in drowning morbidity and mortality. Formal endorsement of the STROBE-D statement by relevant journals is essential. This may include recommending authors of drowning studies to submit a completed STROBE-D checklist at the time of manuscript submission (eg, The BMJ states that “For observational studies, use the STROBE checklist and any appropriate extensions”³⁵). Peer reviewers should be encouraged to incorporate the checklist into their assessment of reporting completeness and to recommend its use to authors when it is not required by journals.¹⁹ While few guidelines appear to have been evaluated,³⁶ it is recommended that developers assess the impact of reporting guidelines on research quality, once adopted by the drowning research community.¹⁹ Ideally, evaluation should be conducted independently of the guideline development group to minimise bias.¹⁹ One pragmatic approach is to encourage journals that formally endorse STROBE-D to assess its impact on the completeness and transparency of reporting in published articles.¹⁹ This could involve before-and-after analyses of reporting quality by comparing the results of the ongoing review¹⁷ with those of a future review or by comparing endorsing and non-endorsing journals.

Limitations

Although the development of STROBE-D adheres closely to the recommendations for developers of health research reporting guidelines,¹⁹ several limitations should be acknowledged.

First, while the scoping review will provide an empirical foundation for identifying deficiencies in reporting practices of observational studies in drowning epidemiology, it is inherently descriptive rather than evaluative. It will map what has been reported in observational drowning studies, but cannot fully determine whether omissions reflect poor reporting, methodological weaknesses, or contextual constraints, such as limited data availability, which is well known to occur in drowning research.37,40 Furthermore, the review is subject to various biases and may exclude research that was never submitted or accepted for publication, written in non-English language or research that was published in a journal not indexed in PubMed, Web of Science or Embase.¹⁷ Given that drowning disproportionately affects low-resource settings,⁴ under-representation of studies from these settings and bias against manuscript acceptance may limit the generalisability of identified reporting gaps.⁴¹ Therefore, involving representatives from these settings is paramount.

Second, the composition and size of the expert group may introduce bias. As with all consensus-based processes, although efforts are made to ensure multidisciplinary and international representation, perspectives and opinions are limited to invited experts who agree to participate.⁴² Attention should be given to the risk of disproportionate influence by dominant participants, particularly where a core group of experts contributes to multiple initiatives over time.³⁰ Most reporting guidelines have been developed by international, multidisciplinary groups involving a median of 22 participants,³⁶ similar to the STROBE-D expert group. All members are selected based on their expertise to contribute to this particular guideline and include epidemiologists, methodologists, journal editors and more than a quarter are content experts according to the recommendations for development of reporting guidelines.¹⁹ Some reporting guidelines have used Delphi designs that include participants outside the author group.³¹ To increase response rates and minimise non-response bias, we decided to involve only the expert group in the Delphi process and instead introduce a public comment phase, despite not being mentioned in previous frameworks.¹⁹ However, despite the use of a public comment phase, there remains a risk that certain viewpoints may be underrepresented. Participation in open review processes is often variable and may be skewed toward individuals already engaged in the field. As a result, some interest-holder groups (eg, professional or volunteer lifesavers, data managers or EMS personnel) may not contribute despite being directly affected by the research the guideline seeks to improve. Diminished internet infrastructure, limited awareness of the process, language barriers and resource constraints may further restrict participation from low-resource settings.

Third, STROBE-D focuses on reporting rather than the conduct of research. Improved reporting does not necessarily equate to improved study methods or reduced risk of bias. The guideline is intended to improve transparency and reproducibility, not to replace methodological training or standards for high-quality epidemiological research.⁴³ However, it will help the quality review if all the STROBE-D elements are present in the paper.

Finally, consensus does not necessarily equate to correctness; it represents agreement within a defined group at a particular time and must be evaluated and revised as circumstances change.

In summary, the STROBE-D statement will extend the STROBE statement by incorporating drowning-specific reporting elements tailored to observational studies in drowning epidemiology. This extension is expected to improve consistency of reporting across studies and to provide structured guidance for authors, reviewers and editors to enhance the clarity, transparency and comparability of reports on observational studies in drowning epidemiology.

Ethics and data management

The Committee on Health Research Ethics in the Region Zealand of Denmark waived the need for ethical approval as this is a consensus study (EMN-2025-09099). Data management, processing and protection were approved in compliance with the General Data Protection Regulation by the Region Zealand of Denmark (p-2025-20168). The study is registered on the EQUATOR Network’s website (http://www.equator-network.org).⁴⁴