INTRODUCTION
1. Background
In the field of nursing in Korea, quasi-experimental research has been frequently conducted to evaluate interventions related to practice, education, and policy. At the journal level, the quality management of reporting such studies directly affects the credibility of research and its applicability in clinical practice.
In nursing research, when randomization is not feasible due to ethical or practical constraints, quasi-experimental designs are often utilized. A quasi-experimental design is positioned between the rigor of a true experimental design—which includes at least one control group and one experimental/intervention group with random assignment [1]—and the flexibility of observational studies [2]. Quasi-experimental designs offer a practical method for evaluating intervention effects in real-world clinical and community settings. However, due to the absence of randomization, there are threats to internal validity such as selection bias. Therefore, transparency in design, analysis, and reporting is paramount [3,4]. In nursing and public health intervention studies, the clear reporting of intervention content, implementation context, characteristics of the comparison group, and strategies for bias adjustment plays a crucial role in the interpretation and reproducibility of research findings [3,5]. Incomplete reporting can lead to misunderstandings about the actual effects and limitations of interventions, which in turn can negatively affect the design of subsequent studies and policy decisions [5,6].
In response to these concerns, Des Jarlais et al. [3] published the Transparent Reporting of Evaluations with Nonrandomized Designs (TREND) statement in 2004 with the aim of improving the completeness of reporting for nonrandomized intervention evaluation studies. The TREND statement provides a 22-item checklist covering essential elements of quasi-experimental and nonrandomized research, including study design, detailed descriptions of interventions and comparison conditions, measurement and analysis methods, and bias adjustment strategies, thereby promoting standardization of reporting [7]. Since its introduction, TREND has been widely cited as a benchmark for improving the reporting quality of nonrandomized evaluation studies [5,8].
The Journal of Korean Biological Nursing Science (JKBNS) is a leading domestic academic journal that covers the overall field of biological nursing research, particularly nursing interventions, physiological studies, and educational research. The reporting practices of quasi-experimental designs in this journal reflect the quality of the research as well as the quality of information provided to its readership. However, there has been limited systematic review regarding the extent to which quasi-experimental studies published in this journal adhere to reporting guidelines such as TREND. Therefore, this study aims to analyze quasi-experimental studies published in the JKBNS between 2011 and 2024 in accordance with international reporting guidelines.
The widely used TREND checklist (TREND statement), developed in 2004, was designed to improve the completeness of reporting in nonrandomized or quasi-experimental intervention studies. It emphasizes detailed reporting of interventions and comparison conditions, study design, and strategies for adjusting potential biases [3]. However, there has been increasing discussion in recent years about the scope, limitations, and need for updating the TREND guideline [5,8].
Meanwhile, one of the international efforts to improve inadequate reporting practices is the Template for Intervention Description and Replication (TIDieR) checklist. TIDieR was developed to strengthen the reproducibility and transparency of intervention descriptions by including elements such as “what”, “how”, “who”, and “when and how much”, with a particular focus on describing complex interventions [9]. Accordingly, this study seeks to compare and analyze these two reporting guidelines and propose the guideline deemed more suitable for future application in the field of nursing.
Thus, while reporting guidelines and critical appraisal tools serve different functions in ensuring the transparency and credibility of research reporting, there is a lack of studies that identify which items are vulnerable in quasi-experimental research and compare which guideline or tool can act complementarily. Moreover, in the case of domestic journals, there are limited instances of quantitative assessment and comparison of reporting quality.
Therefore, this study aimed to conduct a multi-layered evaluation of the current status of quasi-experimental study reporting and methodological quality by mapping the TREND checklist with tools such as TIDieR, the Joanna Briggs Institute (JBI) Critical Appraisal Checklist, and Cochrane’s Risk of Bias in Non-randomized Studies of Interventions, Version 2 (ROBINS-I V2). The goal was to provide specific insights to domestic nursing intervention researchers regarding which areas are vulnerable and how reporting and study design should be improved. This effort is expected not only to standardize reporting practices but also to reduce the risk of misattribution and misinterpretation when research findings are applied to actual clinical practice or policy. Ultimately, the results of this study will serve as foundational data to strengthen the quality management system of the JKBNS and explore future directions for the journal’s qualitative development.
METHODS
1. Study design
This study is a methodological review that analyzed the adherence to the TREND reporting guideline among quasi-experimental studies published in the JKBNS from 2011 to 2024.
2. Samples
The data for this study consisted of quasi-experimental pre-post studies published in the JKBNS between 2011 and 2024. The inclusion criteria were original research articles involving human participants that explicitly identified the study design as quasi-experimental in the title, abstract, or methods section. The exclusion criteria included editorials, animal studies, in vitro studies, review articles including meta-analyses, randomized controlled trials, survey studies, methodological studies, retrospective studies, and qualitative studies. Based on these criteria, a total of 91 articles were included in the final analysis (Figure 1, Appendix 1).
3. Instruments
In this study, the TREND checklist was employed to evaluate whether quasi-experimental studies adhered to established reporting standards. To map relevant components of the TREND checklist, additional evaluation tools were utilized, including the TIDieR checklist, the JBI Critical Appraisal Checklist for Quasi-Experimental Studies, and Cochrane’s ROBINS‑I V2.
1) The TREND checklist
In this study, the quality of reporting in nonrandomized experimental studies was systematically assessed using the TREND checklist. Nonrandomized experimental designs are frequently employed in real-world clinical and community settings due to practical and ethical constraints, and they are particularly common in behavioral and public health intervention research. However, the absence of random allocation renders such studies more susceptible to various forms of bias, and inadequate reporting of intervention content and contextual factors can hinder the interpretation and generalizability of the findings. To address these limitations and enhance the completeness of reporting in studies utilizing nonrandomized designs, Des Jarlais et al. [3] (2004) developed the TREND checklist. This tool was adapted from the Consolidated Standards of Reporting Trials (CONSORT) guidelines and revised to reflect the structural features of nonrandomized study designs. The TREND checklist comprises 22 items in total, including 21 core items across five key sections—title and abstract, introduction, methods, results, and discussion—along with a single item of overall evidence. Most of the items include one or more sub-items, each designed to reflect essential elements of transparent reporting across the conventional structure of a scientific manuscript. Key items include participant eligibility criteria, descriptions of the intervention and comparison conditions, methods of assignment, unit of analysis, statistical techniques, and procedures for interpreting results. Particularly important for non-randomized trials, the checklist also addresses intervention fidelity, statistical adjustment for potential confounders, the use of post-hoc analyses, and the reporting of adverse or unintended outcomes [7,10].
Beyond the documentation of study outcomes, the TREND checklist emphasizes the reporting of the theoretical foundation of the intervention, the proposed causal pathway, feasibility of implementation, and participant compliance. Through this comprehensive approach, the tool aims to improve both the interpretability and replicability of nonrandomized intervention studies. In this study, each item in the TREND checklist is evaluated based on the extent to which the corresponding content is reported in the included articles. Reporting is scored as "fully reported" (1 point) when all required information is clearly presented, "partially reported" (0.5 points) when only part of the information is provided, and "not reported" (0 points) when the item is entirely omitted.
2) TIDieR checklist
The TIDieR checklist is a reporting guideline developed to enhance the reproducibility and transparency of intervention research by promoting the detailed description of intervention components and delivery processes. Hoffmann et al. [9] proposed the TIDieR checklist as an extension of Item 5 of the CONSORT 2010 statement for randomized trials and Item 11 of the Standard Protocol Items: Recommendations for Interventional Trials 2013 guideline, which focuses on the description of interventions in study protocols. This tool was specifically designed for use with complex interventions or non-pharmacological treatments that often involve multiple interacting components. Rather than limiting its scope to outcome reporting, TIDieR encourages researchers to provide a comprehensive account of how the intervention was structured and implemented, thereby enabling clinicians and other researchers to understand and potentially replicate the intervention with fidelity [9,11,12]. The checklist comprises 12 items that cover essential elements such as the intervention’s name, theoretical rationale, materials and procedures used, characteristics of the providers, mode of delivery, setting, frequency and intensity of delivery, tailoring, modifications made during the study, and fidelity to the planned delivery. Each item is designed to address core questions related to who delivered the intervention, what was delivered, when, where, how, and to what extent. Notably, the checklist emphasizes reporting elements such as intervention fidelity, contextual factors, whether the intervention was tailored to individual or group needs, and any modifications made throughout the implementation process. This approach moves beyond basic reporting to facilitate a deeper understanding of the nature of the intervention. By doing so, TIDieR supports the assessment of both the internal validity and external applicability of intervention studies. It provides a practical foundation for real-world implementation and future research planning. Each item of the TIDieR checklist is assessed using a three-point scale [13]: “fully reported” (2 points), “partially reported” (1points), and “not reported” (0 points).
3) JBI Critical Appraisal Checklist for Quasi‑Experimental Studies
The JBI Critical Appraisal Checklist for Quasi-Experimental Studies is a standardized tool developed to evaluate the methodological quality of intervention studies that do not use random allocation. This checklist is specifically designed to assess the risk of bias in quasi-experimental research by examining key aspects such as study design, implementation of the intervention, presence of a comparison group, reliability of outcome measurements, and appropriateness of statistical analyses [14]. The checklist comprises nine items, each reflecting a fundamental component of methodological rigor. The items address the following aspects: (1) whether a clear causal relationship is established with appropriate temporal sequencing between cause and effect, (2) whether the comparison groups are similar in their baseline characteristics, (3) whether participants across groups received comparable care aside from the intervention, (4) whether a control group was included, (5) whether outcomes were measured at multiple time points both before and after the intervention, (6) whether follow-up was complete, or, if not, whether the attrition and its potential impact were adequately explained, (7) whether outcomes were measured in the same way across all comparison groups, (8) whether the measurement tools used were valid and reliable, and (9) whether appropriate statistical analysis was conducted. Each item is evaluated using a categorical scale comprising four possible responses: “Yes”, “No”, “Unclear”, and “Not applicable”.
4) ROBINS‑I V2
The Cochrane ROBINS-I V2 tool has been developed to systematically assess the risk of bias in non-randomized studies of interventions when estimating the effectiveness or harm of an intervention [15,16]. It covers study designs, including cohort, case-control, and quasi-randomized designs. It uses the internal validity standards of randomized controlled trials (RCTs) as a benchmark to evaluate structural sources of bias inherent in non‑randomized designs [16]. Compared to its previous version, ROBINS-I V2 refines its evaluation algorithms and signaling question framework, introduces a triage procedure, and incorporates considerations such as time-varying confounding and immortal time bias to enhance both sensitivity and applicability [16]. ROBINS‑I V2 is structured around seven bias domains: (1) confounding; (2) selection of participants; (3) classification of interventions; (4) deviations from intended interventions; (5) missing outcome data; (6) measurement of outcomes; and (7) selection of the reported result [16]. Each domain is assessed through pre‑specified signaling questions with response options including “strong yes”, “weak yes”, “weak no”, “strong no”, or “no information”. Moreover, it incorporates a triage stage to identify results with a critical risk of bias early in the assessment process. Judgments for each domain are classified into one of four levels: low, moderate, serious, or critical. An overall risk of bias is then assigned for each specific outcome by integrating these domain‑level judgments.
4. Data collections
Data were retrieved from the JKBNS official website archive using the keywords “quasi‑experimental”, “non-randomization”, and “experimental”. The search was conducted between July 21 and July 31, 2025. The identified articles were catalogued in an Excel spreadsheet containing author(s), publication year, title, and DOI. From August 1 to August 10, 2025, two independent reviewers (Cho & Kim) examined the titles, study design, and full texts of the retrieved articles. In the first screening stage, studies were selected whose titles included terms such as “experimental study”, “non-randomization”, “quasi*”, “experimental”, or “효과”. In the second stage, abstracts were reviewed to extract those explicitly describing “quasi‑experimental”, “experimental”, or similar designs. During the third stage, articles labeled in the methods section as “quasi‑experimental study” were further examined; those that included descriptions of random assignment in the full text were reclassified as RCTs and excluded. Disagreements over inclusions primarily concerned whether to include single‑group pre‑post studies or comparison studies; after discussion, it was agreed to include them.
5. Data analysis
Data analysis was conducted to systematically summarize the general characteristics and report the quality of the included quasi-experimental studies. It also aimed to compare the conceptual alignment and differences across reporting guideline tools for non-randomised studies. General study characteristics—such as author(s), year of publication, whether reporting guidelines were specified, study design type, participant characteristics, and sample size, intervention methods for experimental and control groups, type of intervention, and dependent variables—were extracted from the full texts and organized using a coding book in Excel. These characteristics were then summarized using frequencies and percentages. Reporting quality was assessed based on the 22 items of the TREND checklist. Each item was coded as “fully reported” (1 point), “partially reported” (0.5 points), or “not reported” (0 points), depending on the degree of adherence. Compliance rates for individual checklist items were calculated as frequencies and percentages. The total and mean scores for each article were used to evaluate the overall level of reporting. Additionally, a conceptual mapping was performed to compare the items in the TREND checklist with those in the TIDieR Checklist, the JBI Checklist for Quasi-Experimental Studies, and Cochrane’s ROBINS-I V2.
6. Ethical considerations
This review was conducted using previously published studies that involved human participants, without the collection of new data or direct involvement with research subjects. Therefore, this study was exempt from Institutional Review Board (IRB) oversight. In adherence to research ethics, all sources of data were properly cited. To reduce potential bias in the selection and evaluation process, two independent reviewers assessed the eligibility of the included articles, and any disagreements were resolved through consensus.
RESULTS
1. Characteristics of quasi-experimental studies published in JKBNS (2011~2024)
This study analyzed the characteristics of 91 quasi-experimental studies published between 2011 and 2024 in the JKBNS. Many studies (n = 50) were published between 2011 and 2015, while only 13 studies were published from 2021 to 2024. Sixty-four studies (70.3%) reported IRB approval, and 24 studies (26.4%) received research funding. Regarding study design, 81 studies (89.0%) were a control group design. In sample size, 43 studies (47.2%) included fewer than 50 participants, and 41 studies (45.1%) had 50 to 99 participants. The most common participant characteristics were included adults and the elderly, aged 18 years or older (n = 39, 42.9%), and patient populations (n = 44, 48.3%). Most interventions applied to experimental groups were non-pharmacological interventions. Education-based interventions were the most common (n = 23, 25.3%), followed by exercise interventions (n = 9) and music therapy (n = 5). Regarding the duration of the interventions, 67 studies (73.5%) implemented programs lasting less than eight weeks. In the number of intervention sessions, 25 studies (27.5%) reported fewer than 24 sessions, while 48 studies did not report specific information on the number of sessions. Post-intervention measurements were most frequently conducted immediately after program completion (n = 77, 84.6%). A variety of outcome variables were assessed, including depression, anxiety, blood pressure, pain, self-efficacy, fatigue, and pulse rate, indicating that the studies evaluated the effects of interventions across both psychosocial and physiological dimensions (Table 1).
2. Evaluation of reporting levels in quasi-experimental studies using the TREND checklist
This study evaluated the extent to which items from the TREND checklist were reported in quasi-experimental studies. A review by individual checklist items revealed that the following were clearly reported in all 91 studies: 1. Title & abstract, 2. Background, 3. Participants, 4. Interventions, 5. Objectives, 6. Outcomes, 10. Unit of analysis, 11. Statistical methods, 13. Recruitment, 14. Baseline data, 16. Numbers analyzed, 17. Outcomes & estimation, 20. Interpretation, 21. Generalizability, and 22. Overall evidence.
Items ‘8. Assignment method’ and ‘15. Baseline equivalence’ were reported in 90 studies, and ‘9. Blinding’ was partially reported with only one study. Items ‘7. Sample size’ and ‘12. Participant flow’ were reported in 88 studies. Item ‘18. Ancillary analyses’ was reported in only one study, and Item ‘19. Adverse events’ were reported in two studies (Table 2).
3. Mapping the TREND checklist to TIDieR, JBI, and ROBINS-I V2 for quasi-experimental studies
Each item of the TREND checklist was mapped to corresponding components of three widely used frameworks in the context of non-randomized studies: the TIDieR checklist, the JBI Critical Appraisal Checklist, and Cochrane’s ROBINS-I V2 tool (Table 3). Among all TREND items, only one—“4. Methods: Interventions”—was found to align with all three tools, indicating its shared emphasis across reporting and quality assessment standards. In contrast, both the JBI and ROBINS-I V2 tools focus predominantly on the methodological rigor and validity of outcome reporting, with no explicit coverage of discussion-related items, which are included in the TREND checklist. In the mapping between the TREND and TIDieR checklists, “1. Title and abstract” corresponded with Item 1 of TIDieR, and “2. Introduction” aligned with Item 2. The TREND item “4. Methods: interventions” was broadly matched with TIDieR Items 3 to 12, which detail various aspects of intervention design, delivery, and implementation fidelity. For the JBI Critical Appraisal Checklist, “5. Methods: objectives” aligned with Item 1, while “2. Methods: participants” matched Item 2. “14. Results: baseline data” corresponded with Items 2 and 3. Additionally, “4. Methods: interventions” aligned with Item 4, and “6. Methods: outcomes” matched Items 5 to 8. The items “15. Results: baseline equivalence” and “16. Results: numbers analyzed” aligned with Items 5 and 8, respectively, while “17. Results: outcomes and estimation” corresponded with Items 7 and 8. Finally, “11. Methods: statistical methods” and “18. Results: ancillary analyses” both matched Item 9 of the JBI checklist.
Regarding the ROBINS-I V2 tool, the following domains were mapped: TREND items 2, 7, 11, 14, and 15 were linked to Domain 1 (Bias due to confounding). Item 3 corresponded to Domain 2 (Classification of interventions). Item 4 aligned with both Domain 3 (Selection of participants) and Domain 4 (Deviations from intended interventions). Item 12 was mapped to Domains 4 and 5 (Missing data). Items 5, 6, 8, and 9 were associated with Domain 6 (Measurement of outcomes), and items 10 and 17~19 corresponded with Domain 7 (Selection of the reported result).
DISCUSSION
The quasi-experimental intervention studies analyzed in this study were generally evaluated as meeting many of the TREND criteria. In other words, from the title to the discussion sections, the overall structure was well established, indicating that most items were reported. However, several critical elements, particularly the reporting of adverse events, were found to be very limited.
Regarding adverse event reporting, previous research on clinical trials and intervention studies found that while 46% of published articles reported adverse events, as many as 95% were identified in unpublished sources. This suggests that when relying solely on published literature, approximately 64% of all adverse events could be omitted from analyses [17]. Such findings indicate that adverse events are often underreported in published research. In the studies analyzed for this research, even when reasons for participant dropout were reported, they were primarily attributed to external factors (e.g., symptom complaints) rather than serious adverse events caused by the intervention itself. This may be explained by the fact that nursing interventions are often non-pharmacological or primarily focused on education. Nonetheless, concerns remain that some actual harms may go unreported due to reproducibility issues or reporting limitations in clinical trials. The Cochrane Manual also emphasizes that reporting only favorable outcomes in research can lead to biased interpretations [18].
Therefore, efforts must be made to ensure that adverse events are consistently reported without omission. In addition, almost none of the studies reviewed reported blinding, with only one study explicitly mentioning it, while the rest did not. This lack of reporting is likely due to the inherent limitations of quasi-experimental designs, where randomization and blinding are often difficult to implement. In addition, the reporting rate for participant flow was low. Among the studies included in this research, there were 22 RCTs and 91 quasi-experimental studies, highlighting a notable imbalance. This difference reflects a limitation in terms of the rigor of the study design. Consequently, it is recommended that future research designs be structured more tightly and, where possible, evolve toward the level of rigor seen in RCTs. In addition, it is recommended to use the TREND checklist when reporting quasi-experimental studies to maintain rigor in reporting aspects other than randomization.
From a methodological perspective, minimizing bias and writing manuscripts in accordance with reporting guidelines can greatly enhance the reliability of future research. Although many studies in this research adhered to reporting guidelines, compliance alone does not fully guarantee internal validity. Rigorous experimental designs and bias-control strategies are also necessary to support the true effectiveness of interventions.
In this study, both reporting guidelines and quality appraisal tools were compared and analyzed. The TREND checklist offers comprehensiveness by requiring reporting across all sections, from the title to the discussion, whereas TIDieR focuses specifically on replication and intervention description. Recently, the use of TIDieR has been increasing, reflecting researchers’ growing need to transparently report “how the intervention was designed and implemented,” particularly in non-pharmacological intervention studies.
In practice, several international nursing journals explicitly require or recommend the use of the TIDieR checklist in their author guidelines. For instance, the Journal of Clinical Nursing and Nursing Open require authors to comply with the TIDieR checklist in conjunction with the CONSORT guidelines when submitting manuscripts [19,20]. Similarly, the International Journal of Nursing Sciences and International Journal of Nursing Studies Advances emphasize adherence to TIDieR to ensure reproducibility in intervention studies [21,22]. Furthermore, the Journal of Emergency Nursing strongly recommends following the TIDieR guidelines and checklist when submitting manuscripts focused on intervention development [23]. This trend indicates that TIDieR is gradually becoming an essential reporting standard in international journals, representing an evolving academic consensus aimed at enhancing the transparency and credibility of intervention reporting.
In contrast, many domestic nursing journals remain at the level of recommending general compliance with comprehensive reporting guidelines, such as those provided by the EQUATOR Network [24], without requiring the TIDieR checklist as a mandatory submission item. Therefore, to elevate the quality of intervention reporting by domestic researchers to meet international standards, it is necessary for domestic journals to gradually adopt TIDieR. This will ultimately improve the reproducibility of intervention studies and enhance their clinical applicability. Although there are various reporting guidelines and critical appraisal tools, some have pointed out that the mere publication of reporting guidelines does not automatically lead to improvements in reporting quality [25]. Therefore, it is also necessary to explore strategies for enhancing the quality of manuscripts through the effective application of reporting guidelines.
Ultimately, in applied disciplines such as nursing, it is important to strike a balance between the TREND approach, which requires comprehensive and detailed reporting from start to finish, and the TIDieR approach, which emphasizes the reproducibility of interventions. Strictly adhering to every single reporting item may impose a heavy burden on researchers, so a more practical alternative could be the gradual introduction of reporting guidelines focused on core intervention-related items. For example, certain items such as mid-trial stopping plans or documentation of fidelity changes may not be highly relevant to non-pharmacological intervention studies and could therefore be considered as optional reporting elements.
This study was limited to evaluating publications reported in JKBNS. If study designs or intervention details were not fully reported by the researchers, it may have been difficult to accurately assess the methodological quality. Furthermore, given that most interventions in nursing are non-pharmacological and have a low likelihood of causing adverse effects, the omission of adverse event reporting may lead to an underestimation of potential risks. Future studies could address this limitation by directly contacting researchers about selected studies or obtaining original materials (e.g., protocols, supplementary documents) to supplement missing information and then conducting a reassessment.
CONCLUSION
This study explored the complementary use of reporting guidelines and critical appraisal tools in quasi-experimental research and examined the reporting quality of studies, focusing on domestic nursing journals. The findings revealed that quasi-experimental studies reported in JKBNS generally demonstrated a high level of reporting quality. The TREND reporting guideline contributed to enhancing the transparency and reproducibility of research; however, certain items were found to be more precisely evaluated using critical appraisal tools. Therefore, future studies should consider the combined use of various reporting guidelines and critical appraisal tools to further improve the reliability and validity of research.
PDF Links
PubReader
ePub Link
XML Download
Full text via DOI
Download Citation
Print
