Learning From Diagnostic Success in Viet Nam: Safety-II, Adaptive Practices and Organizational Resilience

MEASUREMENT, LEARNING & IMPROVEMENT

11/10/202619 min read

Learning From Diagnostic Success in Viet Nam: Safety-II, Adaptive Practices and Organizational Resilience

Consider a patient arriving at a specialist clinic with an incomplete referral record. The clinician cannot access all the previous investigations, and the patient is unsure why further assessment was requested.

Before proceeding, a staff member helps retrieve the missing information. The clinician reviews the available findings, clarifies an important detail with the patient, and seeks a colleague's opinion about an unresolved concern. The team reaches a reasoned assessment, discusses the remaining uncertainty, and agrees on appropriate next steps.

From the patient's perspective, the consultation may appear routine. The medical record may show that the necessary assessment was completed, without fully capturing the coordination that made it possible.

Yet several actions contributed to the outcome. What enabled the team to recognize the missing information? Was specialist advice readily available, or did the clinician depend on an informal professional relationship? Would the same process have worked during a busy shift or when experienced staff were unavailable?

This hypothetical situation illustrates an important opportunity in diagnostic safety. Healthcare organizations can learn from failures, but they can also learn by examining how appropriate diagnostic care is achieved under everyday conditions.

This perspective is closely associated with Safety-II and resilience engineering. Rather than concentrating exclusively on what went wrong, it examines how healthcare professionals, patients, and organizations respond to uncertainty and changing circumstances to achieve acceptable outcomes.

For healthcare organizations in Viet Nam, the value of this approach lies in understanding the conditions that support reliable diagnostic work and identifying how those conditions can be strengthened.

Why Diagnostic Success Deserves Closer Examination

Efforts to improve diagnostic safety have understandably focused on missed, delayed, or incorrect diagnoses. Investigating these events remains essential for identifying preventable harm and understanding weaknesses in healthcare systems.

However, concentrating mainly on unfavorable outcomes provides an incomplete picture of how diagnosis works in practice.

Every day, healthcare professionals make diagnostic decisions despite incomplete clinical histories, evolving symptoms, competing priorities, and differences in available resources. They obtain additional information, reconsider initial interpretations, communicate concerns, and adjust their plans as circumstances change.

Some of these adjustments are supported by established processes. Others depend on professional experience, teamwork, or resources that may not be consistently available.

When the outcome is favorable, the actions that helped achieve it may receive little attention. Conventional incident reporting systems primarily identify problems rather than explain how everyday clinical work succeeds.

The absence of a reported error tells us relatively little about why a diagnostic process functioned effectively. Equally, a correct diagnosis does not necessarily mean that every step was appropriate. A favorable conclusion may sometimes follow an incomplete assessment, an inefficient process, or circumstances that happened to work in the patient's favor.

Conversely, a carefully conducted diagnostic evaluation may not immediately establish a definitive diagnosis because the available evidence is insufficient.

Learning from diagnostic success therefore involves more than identifying cases in which the final diagnosis was correct. It requires examining how decisions were made, what information was available, how uncertainty was managed, and whether the process supported timely, safe, and patient-centered care.

This distinction matters because a healthcare organization cannot assume that a process is dependable simply because it has repeatedly produced favorable outcomes.

If success depends on one highly experienced professional recognizing problems that others might miss, the organization should understand what enables that professional's response and whether similar capabilities can be supported more consistently.

Understanding Safety-I and Safety-II

Safety-I and Safety-II provide different but complementary perspectives on healthcare safety.

Safety-I is commonly associated with identifying adverse events, investigating contributing factors, and developing safeguards intended to prevent similar problems. In diagnostic care, this includes examining missed diagnoses, communication failures, inappropriate testing, and opportunities for earlier recognition of disease.

Safety-II focuses on a different question: how do people and organizations achieve acceptable outcomes when everyday working conditions are variable and sometimes difficult?

Patients do not present with identical symptoms. Clinical information arrives at different times, priorities change, and available resources may vary throughout the day. Healthcare teams must respond to this variability while maintaining appropriate standards of care.

Safety-II examines the adjustments, interactions, and capabilities that help clinical work succeed under these conditions.

Importantly, Safety-I and Safety-II are perspectives on understanding safety rather than specific clinical interventions. Applying a Safety-II perspective does not, by itself, establish that a healthcare activity is safe or that an improvement has been achieved.

The U.S. Agency for Healthcare Research and Quality (AHRQ), in its 2025 publication Exploration of Foundational Terminology and Paradigms for Improving Diagnosis, discusses Safety-I and Safety-II as different approaches to understanding diagnostic quality and safety.

The publication recognizes that studying diagnostic successes, failures, and ordinary working conditions can provide insights into how healthcare systems support appropriate diagnosis.

The Organisation for Economic Co-operation and Development's 2025 report, The Economics of Diagnostic Safety, also discusses Safety-II, complexity, resilience, and learning from successful diagnostic processes as part of the broader consideration of diagnostic safety.

A particularly relevant example comes from the University of California, San Francisco's Coordinating Center for Diagnostic Excellence (UCSF CoDEx). Its strategic plan explicitly identifies the use of both Safety-I and Safety-II, including examining diagnostic successes and failures in real-world situations, as a principle for advancing diagnostic excellence.

These developments reflect growing interest in understanding successful diagnostic performance alongside established approaches to preventing errors.

However, Safety-II should not be interpreted as a replacement for clinical standards, incident investigation, professional accountability, or established patient safety safeguards.

Certain diagnostic activities require strict adherence to validated procedures. Specimen identification, laboratory quality controls, and other essential safety requirements cannot be modified simply because an alternative approach appears convenient.

Other situations require appropriate professional flexibility when clinical circumstances cannot be fully anticipated.

The challenge is understanding where standardization is necessary, where adaptation contributes to appropriate care, and how healthcare organizations can support both.

What Adaptive Practices Reveal About Diagnosis

Adaptation is a routine part of clinical work, but it is not necessarily a departure from good practice.

A clinician may question an initial interpretation because new information does not fit the suspected diagnosis. A nurse may recognize a change in a patient's condition that warrants reassessment. A radiologist or laboratory professional may request clarification when a diagnostic finding appears inconsistent with the clinical question.

In each situation, the professional recognizes that the original understanding or planned process may be insufficient.

Resilience engineering commonly describes four related capabilities: anticipating, monitoring, responding, and learning.

Anticipation involves recognizing possible developments and preparing for them before difficulties arise. Monitoring means remaining attentive to changes in clinical conditions, available information, and the operating environment. Responding involves making appropriate adjustments when circumstances require a different course of action. Learning involves examining experience to understand what should be maintained or improved.

These capabilities provide a useful way to study diagnostic work. However, identifying them in a clinical encounter does not automatically demonstrate that a particular decision was correct or that an adaptation improved patient safety.

Consider the process of obtaining a second professional opinion.

A clinician may recognize that the available findings do not adequately explain a patient's condition and seek advice from another specialist. This additional input may help clarify the diagnostic assessment.

However, if timely specialist advice is available only because the clinician has a personal relationship with a colleague, the favorable outcome may also reveal a weakness in the organization's consultation arrangements.

The individual adaptation may have helped the patient, but the underlying process may not be dependable for others.

A similar situation can arise when an experienced staff member routinely identifies incomplete referrals before consultations. The person's initiative may help prevent unnecessary delays, but the organization still needs to understand why referral information is frequently incomplete.

Such situations reveal both what worked and why additional effort was necessary.

This is particularly relevant when healthcare organizations rely on informal workarounds to maintain continuity of care.

Some workarounds reflect appropriate professional judgment in unexpected circumstances. Others may introduce risks, bypass essential safeguards, or conceal weaknesses that should be addressed.

The purpose of studying adaptation is to understand which responses support appropriate care, under what circumstances they are useful, and how their benefits can be sustained without creating additional risks.

From Individual Effort to Organizational Resilience

Diagnostic reliability depends on more than the knowledge and experience of individual clinicians.

Healthcare organizations influence whether professionals can obtain essential information, access diagnostic services, communicate with colleagues, receive specialist support, and respond appropriately when routine processes become difficult.

Organizational resilience broadly concerns the capacity to adapt to changing circumstances while maintaining essential functions and acceptable performance.

In diagnostic care, this may involve dependable arrangements for obtaining additional expertise, mechanisms for reconsidering uncertain diagnoses, and coordination between departments when new findings require a different clinical response.

However, resilience should not be confused with the ability of employees to compensate indefinitely for organizational limitations.

A hospital may continue delivering appropriate care during periods of operational pressure because its staff work exceptionally hard. That effort may benefit patients in the short term, but it does not necessarily demonstrate that the underlying system is reliable or sustainable.

Repeated dependence on extraordinary individual effort can contribute to fatigue, inconsistent performance, and additional safety risks.

Understanding diagnostic success therefore requires attention to the conditions under which professionals work, including workload, staffing, information availability, and organizational support.

An important concept in resilience research is the distinction between work-as-imagined and work-as-done.

Work-as-imagined describes how activities are expected to occur according to policies, protocols, process maps, and organizational assumptions. Work-as-done describes how those activities are actually performed under everyday conditions.

A referral pathway, for example, may assume that the relevant clinical information is available before a specialist consultation. In practice, records may be incomplete, previous investigations may be difficult to retrieve, or the patient's condition may have changed since the referral was made.

Understanding how professionals manage these differences can reveal improvement opportunities that may not be apparent from reviewing written procedures alone.

Nevertheless, differences between planned and actual practice should be examined critically. Some reflect appropriate flexibility, while others may represent unsafe workarounds or failures to follow essential requirements.

The patient's experience adds another important perspective.

A diagnostic process may appear efficient to the healthcare team while requiring the patient or family to make repeated visits, retrieve records independently, or coordinate communication between clinical services.

Examining how care is experienced by patients, alongside how it is planned and delivered, can help organizations recognize burdens that might otherwise remain unnoticed.

A more dependable healthcare system should support appropriate professional adaptation while reducing unnecessary reliance on individual efforts to overcome persistent organizational weaknesses.

What Recent International Research Shows

Research applying Safety-II specifically to diagnostic care has developed in recent years. Publications from 2025 and 2026 provide examples of how diagnostic success and adaptive practices can be studied, although their findings require careful interpretation.

In July 2025, Taro Shimizu published Learning From What Goes Right: A Safety-II Framework for Improving Diagnosis at the Point of Care, subsequently included in the May 2026 issue of Diagnosis.

The article introduced SIDER, a structured approach for reflecting on challenging diagnostic cases that were successfully resolved.

SIDER consists of five elements: Specification, Ishikawa diagram, Driver diagram, Engaging the patient and the team, and Reflection.

The framework encourages clinicians to identify an instructive case, examine factors associated with the outcome, consider potential improvement strategies, incorporate relevant patient and team perspectives, and reflect on lessons for future practice.

Its contribution is to provide a structured way of examining diagnostic success rather than relying solely on informal recollection.

However, SIDER was introduced as a proposed reflective framework. Its publication does not establish that routine implementation improves diagnostic accuracy, reduces patient harm, or produces sustained organizational improvement. Further evaluation is needed to determine whether and under what conditions it offers measurable benefits.

In June 2026, Kamei and colleagues published Learning From What Went Right: A Safety-II Application of the SIDER Protocol to a Case of Occult Breast Cancer.

The report examined a diagnostically challenging case involving a 56-year-old woman with progressive symptoms and initially nondiagnostic imaging investigations.

Despite the absence of an identifiable primary breast lesion on the initial investigations, the clinical team retained breast malignancy as a possible diagnosis because of the overall clinical presentation.

Diagnostic uncertainty was discussed with the patient, supporting continued follow-up. Further investigation and an axillary lymph node biopsy subsequently established the diagnosis of occult breast cancer.

The authors used SIDER to examine how continued consideration of a low-probability but clinically important diagnosis, patient engagement, and collaboration across specialties contributed to diagnostic progress.

The case illustrates how structured reflection on a successful diagnostic process may identify potentially useful practices. Nevertheless, an individual case cannot establish that the same approach will consistently produce similar outcomes in other patients or healthcare settings.

A different application of Safety-II was reported by Roos and colleagues in an action research study conducted in an emergency department in the Netherlands.

Published online in October 2025 and included in the May 2026 issue of Diagnosis, the study examined variation in everyday diagnostic practice.

The researchers observed 40 diagnostic processes and used repeated cycles of observation, analysis, and collaborative discussion to identify opportunities for improvement.

Their work led to five practice changes: a template for documenting emergency department preannouncements, improved access to digital information resources for residents, face-to-face supervision by internal medicine consultants during office hours, blood sampling at triage, and the addition of lipase to standard emergency department blood tests.

The authors reported adoption rates ranging from 67% to 100% in observed cases following implementation.

These findings demonstrate how examining everyday diagnostic work can lead to operational changes without relying exclusively on the identification of diagnostic errors.

However, the study was conducted in a single healthcare setting. Adoption of the changes should not be interpreted as proof that diagnostic accuracy improved or that patient harm was reduced.

Another important contribution appeared in August 2026, when Cifra and colleagues published Beyond Diagnostic Error: An Ethnography of Resilient Behaviours and System Adaptations in Paediatric Outpatient Diagnosis in BMJ Quality & Safety.

The researchers observed 33 outpatient visits involving children with medical complexity across three academic health systems in the United States.

They identified 258 instances of resilience-promoting behaviors and organizational properties across 31 of the 33 visits.

Of these, 178 involved activities by patients, families, and clinicians, while 80 reflected organizational properties relevant to diagnostic resilience.

Frequently observed characteristics included anticipating future needs, monitoring clinical developments, maintaining robustness through comprehensive or redundant diagnostic processes, and extending available resources to meet clinical needs.

The study is notable for examining actual clinical activities and incorporating the perspectives of patients, families, and healthcare professionals.

It provides insight into how individuals and organizations manage diagnostic complexity. However, the research was qualitative and involved a specific pediatric population.

The frequency of observed resilient behaviors does not establish that they caused improvements in diagnostic accuracy or patient outcomes.

Taken together, these publications suggest that structured reflection, direct observation, and examination of adaptive practices can help organizations better understand how diagnostic care functions.

They also demonstrate why methodological caution is necessary.

Conceptual frameworks, individual case reports, qualitative studies, and single-center improvement research can generate valuable insights. They do not independently establish the clinical effectiveness of Safety-II interventions across different healthcare environments.

This distinction is consistent with concerns discussed by Verhagen and colleagues in their 2022 article, The Problem With Making Safety-II Work in Healthcare, which highlighted difficulties in translating Safety-II concepts into practical interventions supported by evidence of effectiveness.

For healthcare organizations considering these approaches, the available research provides a basis for carefully designed learning and improvement activities, not a guarantee of better diagnostic outcomes.

Defining and Evaluating Diagnostic Success

Studying successful diagnostic care creates methodological challenges that differ from those encountered when investigating diagnostic errors.

One of the most important is deciding what should count as success.

A clinician may reach the correct diagnosis after unnecessary investigations or avoidable delays. A patient may receive an accurate diagnostic conclusion but experience poor communication or significant difficulty obtaining follow-up care.

Conversely, a carefully conducted assessment may not immediately establish a definitive diagnosis because the evidence remains insufficient.

These examples illustrate why diagnostic success cannot be defined solely by whether the final diagnostic label was correct.

A meaningful evaluation should consider whether decisions were appropriate given the information available at the time, how uncertainty was handled, whether communication supported patient understanding, and whether the process avoided unnecessary risk or burden.

Hindsight presents another challenge.

Once the final diagnosis is known, earlier decisions may appear more clearly justified than they were when the clinician was working with limited information.

A retrospective review may attribute success to a particular adaptation even though other factors, including clinical circumstances or chance, contributed to the outcome.

Reviewers should therefore distinguish between what participants actually knew and did at the time and what later interpretation suggests may have been important.

Selection bias also deserves attention.

Unusual cases in which a difficult diagnosis was eventually established may be particularly memorable. These cases can provide valuable learning opportunities, but they may not represent ordinary diagnostic practice.

An organization that studies only exceptional successes may overestimate the importance of individual expertise while overlooking the less visible systems that support routine care.

Where feasible and appropriate, examining everyday clinical work through observation or prospective review can complement the analysis of selected success stories.

The purpose is to understand how diagnostic processes function, identify credible opportunities for improvement, and evaluate whether proposed changes make care more dependable.

Applying Learning From Diagnostic Success in Viet Nam

Healthcare organizations in Viet Nam operate across a wide range of clinical and organizational environments.

Diagnostic services are delivered in primary care facilities, general hospitals, specialized hospitals, outpatient clinics, and other authorized healthcare settings.

Differences in service capacity, workforce arrangements, information systems, and access to specialist expertise can influence how diagnostic care is organized.

These variations make the study of everyday diagnostic work relevant, but international Safety-II findings should not be assumed to describe actual practices or outcomes in Vietnamese healthcare organizations.

There is insufficient locally established evidence to conclude that particular Safety-II adaptations are widespread or effective across Vietnamese clinical settings.

A reasonable starting point would be to select a clearly defined diagnostic process and examine how it functions under ordinary working conditions.

For example, a hospital might review a referral pathway involving patients who require assessment by several clinical services.

Rather than examining only cases in which an error or delay occurred, the organization could also consider appropriately completed assessments in which the team encountered difficulties obtaining information, coordinating care, or managing uncertainty.

The review might explore how missing information was recognized, how clinical concerns were communicated, when additional expertise became available, and which arrangements helped professionals make appropriate decisions.

Conversations with the professionals involved could reveal whether these actions reflected established procedures or informal adaptations.

Patient perspectives may provide additional information, particularly when patients or families had to coordinate appointments, retrieve medical records, or seek clarification about the next steps themselves.

Such activities should be approached with an open mind.

An apparently successful process may reveal an effective practice worth supporting more consistently. It may also expose weaknesses in referral documentation, communication arrangements, or access to specialist advice.

Both findings can contribute to improvement.

The review should not assume that every adaptation was beneficial or that every difference from a written procedure represents a problem. Instead, it should examine what made the response appropriate in the particular circumstances and whether the organization can provide more dependable support.

These learning activities should be integrated into existing quality and patient safety processes where feasible.

They should not create unnecessary documentation requirements or place additional expectations on professionals to compensate for persistent organizational deficiencies.

A credible learning environment should allow healthcare workers to discuss difficulties, trade-offs, and limitations without pressure to present only favorable experiences.

Turning Observations Into Meaningful Improvement

Learning from diagnostic success becomes valuable when the resulting insights lead to changes that can be examined and evaluated.

However, identifying an adaptation that appeared effective does not mean it should automatically become standard practice.

Some adjustments work only under particular conditions. Others depend on resources, expertise, or patient characteristics that may not be present elsewhere.

Before introducing a practice more widely, healthcare organizations should understand why it was considered useful, what risks it may introduce, and whether it is consistent with applicable professional requirements.

A locally appropriate improvement might involve clarifying referral information requirements, making clinical records more accessible, strengthening coordination between services, or establishing a more dependable process for obtaining additional diagnostic advice.

Where a proposed change affects existing workflows, small-scale testing may be appropriate before broader implementation.

Evaluation should focus on whether the intended process becomes more reliable, rather than simply counting how many professionals adopt the change.

For example, an intervention intended to improve access to specialist advice might be evaluated through the timeliness and appropriateness of consultations, clarity of clinical responsibility, and feedback from professionals and patients.

Balancing measures are also important.

Additional professional reviews may create new delays or increase workloads if they are used indiscriminately. More extensive documentation may improve access to information while consuming time that could otherwise be devoted to patient care.

Organizations should therefore consider both intended benefits and unintended consequences.

There is no universally validated set of Safety-II indicators that can establish the diagnostic resilience of every healthcare organization.

Counts of successful cases, observed adaptations, or completed reflection sessions should not be treated as direct measures of diagnostic accuracy or patient safety.

An increase in recorded adaptive behaviors may reflect more intensive observation rather than a genuine improvement in clinical performance.

The more meaningful question is whether a credible weakness has been addressed and whether the resulting change produces benefits that can reasonably be demonstrated.

This requires combining professional insight, appropriate data, patient perspectives where relevant, and continued evaluation of implementation effects.

Legal, Ethical and Professional Considerations in Viet Nam

Activities intended to study diagnostic practices and improve healthcare quality must operate within Viet Nam's applicable legal and professional framework.

The Law on Medical Examination and Treatment No. 15/2023/QH15, as applicable and amended, establishes relevant requirements concerning professional responsibilities, healthcare quality, patient rights, and medical incident prevention.

Article 57 addresses quality standards for healthcare facilities, while Article 58 governs quality assessment and certification. Article 71 establishes responsibilities relating to medical incident prevention, including identification, reporting, analysis, and measures intended to prevent recurrence.

These provisions provide an important legal context for organizational quality and patient safety activities.

However, they do not establish Safety-II as a separate mandatory diagnostic safety framework in Viet Nam.

Healthcare organizations may consider Safety-II principles as part of evidence-informed professional learning and quality improvement, provided that implementation remains consistent with applicable laws, clinical standards, and institutional responsibilities.

If an internal learning activity identifies an actual or potential medical incident, the organization must assess and fulfill any applicable reporting, management, and professional obligations.

Relevant provisions of Ministry of Health Circular No. 43/2018/TT-BYT, to the extent they remain in force, provide additional requirements concerning medical incident prevention, reporting, analysis, and organizational response.

A review intended to learn from successful practice cannot be used to avoid mandatory incident reporting or other legal responsibilities.

The use of patient information requires particular attention.

The Law on Personal Data Protection No. 91/2025/QH15 and Decree No. 356/2025/ND-CP, both effective from January 1, 2026, form part of the current legal framework governing personal data processing in Viet Nam.

Decree No. 356/2025/ND-CP replaced the earlier Decree No. 13/2023/ND-CP.

Activities involving medical records, clinical observations, patient interviews, or other identifiable health information may require safeguards concerning lawful access, permitted processing purposes, confidentiality, information security, data retention, and disclosure.

Using aggregated or appropriately de-identified information may reduce privacy risks. However, removing direct identifiers does not automatically eliminate legal obligations if individuals can still be identified.

Additional considerations arise when findings from internal quality improvement are intended for research, external collaboration, academic presentation, or publication.

Internal quality improvement and research involving human participants may be subject to different requirements. The applicable classification depends on the activity's actual purpose, design, methods, and governing rules, rather than simply how the organization describes it.

Where relevant, the responsible institution should determine whether ethics review, institutional authorization, consent, or additional legal safeguards are required.

Healthcare organizations should also avoid assuming that internal learning discussions or review records are automatically protected from disclosure or exempt from applicable legal processes.

Professional learning must respect the responsibilities and rights of healthcare workers. Studying adaptive practices should not become a mechanism for assigning blame without appropriate assessment or for expecting individuals to compensate indefinitely for inadequate resources.

International Safety-II concepts and research findings are educational and improvement resources. They do not replace Vietnamese clinical regulations, professional accountability, or authorized institutional decision-making.

Building the Capacity to Learn From What Goes Right

An important contribution of Safety-II is the recognition that diagnostic success deserves examination, not simply recognition.

When an appropriate diagnostic decision is achieved under challenging circumstances, the outcome may reveal something about the professional judgment, teamwork, information systems, and organizational arrangements that supported the process.

Yet it may also reveal vulnerabilities that remain hidden because the immediate outcome was favorable.

Healthcare organizations need to understand both.

A useful learning process allows clinicians and other professionals to explain what actually happened, what information was available, which difficulties they encountered, and why they made particular decisions.

It should also consider the patient's experience and examine whether the conditions supporting success are consistently available.

The emerging international literature offers several ways to approach this work, including structured reflection through SIDER, observation of everyday diagnostic processes, and qualitative examination of professional and organizational resilience.

These approaches may contribute to professional education and organizational learning. However, further research is needed to determine their effects on diagnostic performance, patient outcomes, and sustained healthcare system reliability.

For healthcare organizations in Viet Nam, a practical opportunity lies in incorporating carefully selected learning-from-success activities into existing multidisciplinary discussions, professional education, and quality improvement programs.

Such activities should begin with clearly defined questions, remain attentive to local circumstances, and be evaluated before wider adoption.

Learning from diagnostic success is not about proving that healthcare professionals consistently make the right decisions. It is about understanding what helps them make appropriate decisions under difficult conditions and how healthcare organizations can make those conditions more dependable.

A reliable diagnostic system should not depend entirely on exceptional individuals, informal workarounds, or favorable circumstances.

It should support healthcare professionals in obtaining necessary information, recognizing changes in clinical conditions, seeking appropriate expertise, communicating uncertainty, and arranging reassessment when needed.

Preventing diagnostic failures remains essential. Understanding how diagnostic care succeeds provides an additional perspective on the capabilities that make safe and effective care possible.

For Viet Nam, the opportunity is to develop a better understanding of how diagnostic work is actually performed, identify the organizational conditions that support appropriate care, and evaluate improvements that can be sustained.

Ultimately, diagnostic excellence is not defined only by the absence of error. It also depends on creating conditions in which accurate, timely, appropriate, and patient-centered diagnosis can be achieved more consistently.

Evidence and Policy Context

This article draws on international research and professional guidance concerning diagnostic safety, Safety-II, resilience engineering, adaptive clinical practices, and organizational learning.

Foundational sources include the National Academies of Sciences, Engineering, and Medicine's Improving Diagnosis in Health Care (2015); the U.S. Agency for Healthcare Research and Quality's Exploration of Foundational Terminology and Paradigms for Improving Diagnosis (2025); and the Organisation for Economic Co-operation and Development's The Economics of Diagnostic Safety (OECD Health Working Papers No. 176, 2025; DOI: 10.1787/fc61057a-en).

The discussion also considers the UCSF Coordinating Center for Diagnostic Excellence Strategic Plan and the U.S. Centers for Disease Control and Prevention's Core Elements of Hospital Diagnostic Excellence, updated in February 2026.

These resources provide broader context for diagnostic quality and safety improvement but do not establish the clinical effectiveness of Safety-II interventions.

Selected Research Publications

Shimizu T. Learning From What Goes Right: A Safety-II Framework for Improving Diagnosis at the Point of Care. Diagnosis. Published online July 17, 2025; 2026;13(2):181–186. DOI: 10.1515/dx-2025-0086.

Kamei T, Shimada Y, Shimura Y, et al. Learning From What Went Right: A Safety-II Application of the SIDER Protocol to a Case of Occult Breast Cancer. Diagnosis. Published online June 26, 2026. DOI: 10.1515/dx-2026-0057.

Roos R, Zwaan L, Maranus M, et al. A Safety-II Approach to Learn From Practice Variation in the Diagnostic Process in the Emergency Department: An Action Research Study. Diagnosis. Published online October 13, 2025; 2026;13(2):274–284. DOI: 10.1515/dx-2025-0087.

Cifra CL, Melton K, Walsh KE, et al. Beyond Diagnostic Error: An Ethnography of Resilient Behaviours and System Adaptations in Paediatric Outpatient Diagnosis. BMJ Quality & Safety. Published online August 11, 2026. DOI: 10.1136/bmjqs-2026-020175.

Verhagen MJ, de Vos MS, Sujan M, Hamming JF. The Problem With Making Safety-II Work in Healthcare. BMJ Quality & Safety. 2022;31(5):402–408. DOI: 10.1136/bmjqs-2021-014396.

These publications differ in their research designs and evidentiary strength. Conceptual frameworks, qualitative observations, action research, and individual case reports can provide valuable insights, but they do not independently establish the clinical effectiveness of Safety-II interventions.

As of November 10, 2026, the evidence reviewed in this article does not establish that implementing Safety-II or the SIDER framework consistently reduces diagnostic errors or improves patient outcomes across healthcare settings.

The Vietnamese legal and policy context includes the Law on Medical Examination and Treatment No. 15/2023/QH15, as applicable and amended; relevant provisions of Ministry of Health Circular No. 43/2018/TT-BYT that remain in force; the Law on Personal Data Protection No. 91/2025/QH15; and Decree No. 356/2025/ND-CP.

Safety-II is presented as an emerging, evidence-informed perspective for professional learning and organizational improvement. It is not a mandatory Vietnamese diagnostic safety standard, a validated accreditation requirement, or a substitute for applicable legal and professional obligations.

Professional Scope and Disclaimer

This article is published by Diagnostic Excellence Vietnam (DxEx.vn) for professional education, knowledge-sharing, and general information. It examines learning from diagnostic success, Safety-II, adaptive clinical practices, and organizational resilience from the perspectives of healthcare quality, patient safety, evidence appraisal, and systems-based improvement.

The content does not constitute individualized medical advice, clinical diagnosis, treatment recommendations, mandatory clinical practice guidelines, legal advice, or legally binding interpretations of Vietnamese regulations.

Diagnostic Excellence Vietnam (DxEx.vn) is not a healthcare provider, regulatory authority, or accreditation or certification body.

Discussion of Safety-II, SIDER, resilience engineering, and international research does not constitute endorsement, validation, or certification of any specific method, clinical practice, or healthcare organization.

Clinical decisions, professional accountability, medical incident reporting, quality improvement activities, patient information management, personal data processing, research, and organizational implementation remain the responsibility of appropriately qualified healthcare professionals and authorized healthcare organizations.

All activities must comply with applicable Vietnamese laws, professional standards, institutional requirements, and individual patient circumstances.

International recommendations and emerging research findings should be critically evaluated and adapted to the relevant clinical and organizational context before implementation.