Diagnostic errors, defined as a diagnosis that is missed, wrong, or delayed, as detected by some subsequent definitive test or finding [3], are gradually receiving significant attention from researchers and the authorities. The US Institute of Medicine Committee on Diagnostic Error in Health Care has recently defined diagnostic error as "the failure to (a) establish an accurate and timely explanation of the patient’s health problem(s) or (b) communicate that explanation to the patient" [4]. On the other hand, harm due to diagnostic errors i.e., misdiagnosis-related harm, can be defined as ‘preventable harm that results from the delay or failure to treat a condition actually present (when the working diagnosis was wrong or unknown) or from treatment provided for a condition not actually present’ [5]. With the introduction of Consumer Protection Act, and the increasing frequency of litigations related to wrong diagnosis or treatment, there is a need for training clinicians to reduce such errors.

Why these errors occur and how to reduce them is a hotly debated issue. It has been suggested that the focus of clinical teaching is mostly on ‘diagnosis’, and not as much on ‘misdiagnosis’ or errors of diagnosis and ways to avoid them. The elucidation of various theories on the cognitive process behind decision-making, and recognition of variety of ways to teach clinical reasoning to students have now enabled us to address the cognitive aspects of diagnostic reasoning [6]. Various mechanisms for reducing diagnostic errors [7,8], and teaching and assessing reduction of diagnostic errors [8,9] have been suggested. Electronic decision-support website or mobile applications (Apps) [10,11], written case-scenarios for teaching diagnostic errors, electronic trigger tools (electronic algorithms that identify potential adverse events by searching electronic health records and flagging specific occurrences) [12,13], simulated patients [14] etc. are increasingly being used as strategies to teach about and reduce diagnostic errors.

Although exact prevalence data are not available, and diagnostic errors are difficult to document reliably, these have been estimated to be in the range of 5-15% [15,16]. In a　survey of 726 pediatricians, more than half had made a diagnostic error at least once- or twice-a-month (the frequency was 77% among trainees); and nearly half reported diagnostic errors that harmed patients at least once- or twice-a-year [17]. Data from other countries also shows similar trends [18-21]. Incidence of medical errors in India is not documented, but probably these occur frequently [22]; however, they are usually under-reported, overlooked and remain a controversial area [23].

In addition to being highly prevalent, diagnostic errors also have important implications for both the patients and the physicians. Diagnostic errors are important causes of both adverse events (second largest cause) [24] and also malpractice suits against hospitals (second leading cause) [25]. Diagnostic discrepancy has been noted even in 20% cases of autopsy [26]. They may also cause harm to patients by inappropriate treatment, delayed treatment, or even harmful treatment. In addition, they may have psychological or financial repercussions, and lead to medical malpractice claims [4]. There is thus an urgent need to teach students to address and respond to diagnostic errors, and by extension, overall improvement in diagnosis.

Before we address the various mechanisms to teach reduction of diagnostic errors, two issues need to be deliberated upon viz., how physicians make diagnosis, and what are the causes of diagnostic errors?

The skills required during a clinical encounter, including history-taking, physical-examination, synthesizing data, and presenting this information, represent an amalgamation of multiple cognitive and psychomotor skills and behaviors. The Dual Processing theory is one of the widely accepted theories and proposes that clinicians use two types of cognitive processes to arrive at a diagnosis; System I and System II approaches [27]. System I approaches are faster, more intuitive and based on past experience and pattern-recognition, whereas system II approaches are analytical and more deliberate [6]. A beginner primarily uses system II approach; whereas experts primarily use the system I approach, though they may switch between the approaches depending upon the complexity of the case [28]. Further details of these processes have been previously published [6].

Clinicians routinely use a lot of thumb-rules or heuristics, whether explicitly stated or self-developed [29]. Although, these are essential if multiple critical decisions need to be made in a situation of constraints of time and information, a corollary of this process is a fixed proportion of error, due to inherent biases associated with each of these heuristics [30]. The principle that it is more likely for a single disease to explain multiple symptoms than for multiple diseases to occur, is called as the Occham’s razor [31]. Although infrequently acknowledged, this is one of the most frequently relied upon argument for verifying a diagnosis. This is also accompanied frequently by the tendency to be biased by our initial hypothesis and thus paying minimal attention to other possibilities, also called as Premature closure [32]. This may sometimes just be due to overconfidence, but may occasionally be due to confirmation bias - a tendency to favor evidence that confirms our hypothesis than evidence favoring alternate diagnosis [30]. Clinicians frequently use diagnostic tests to rule-in or rule-out clinical diagnosis, but mostly intuitively, without using an objective method. Bayes’ theorem is a method [33] to modify the probability of a particular diagnosis after the results of a diagnostic test. However, it is infrequently used in clinical practice, despite the information on pre- and post-test probability being available for many of the tests in use. Although infrequently taught, in practice, physicians use system I processes more often.

The deliberate diagnostic process consists of four sequential activities, which sometimes occur so rapidly in experienced clinicians so as to appear as if occurring simultaneously [30]. These include the initial recognition of the abnormal data, abnormal physical findings or radiological abnormalities (Perception); followed by generation of a clinical framework for analysis or a list of highly likely diagnosis (Hypothesis-generation). These initial diagnoses are then evaluated in the light of available test results (Data-interpretation). This is then followed by the confirmation of the working diagnosis (Verification). There are a variety of steps in this cognitive process of diagnosis, problems in either of which may lead to diagnostic errors.

Diagnostic errors have been categorized as No-fault errors, System errors, and Cognitive errors. A schematic representation of these categories, with various examples [30] is shown in Fig. 1. Due to its etiological underpinnings, this is an important differentiation as it also guides us towards preventive strategies to address each group. We also need to appreciate that in addition to multiple system errors, cognitive factors may co-exist, and an interaction between the two may be at the root of many cases of diagnostic errors [34]. Errors have been reported to be more likely to occur when the level of uncertainty is high, there is unfamiliarity with the patient, there are co-morbid disorders confusing the clinical picture, and if atypical or non-specific manifestations are the presenting feature [35].

Fig. 1 Three types of diagnostic errors in the healthcare system.

No-fault errors: These are cases where due to the unique illness- or patient-characteristics, making the correct diagnosis with the current state of medical knowledge is not likely to be routine [35,36]. This may be due to a rare disease, an atypical presentation, a non-compliant patient, and a new disease or an as yet unrecognized manifestation.

System errors, on the other hand, can be ascribed to the deficient functioning of the healthcare system, rather than on the patient-doctor interaction. These reflect latent flaws in the system [30], whether they are due to technical failure or failure of the organizational policies and protocols. These range from those affecting the individual diagnostician e.g., stress or workload, to those affecting all beneficiaries/stakeholders of that system e.g., defective communication and weak policies.

Cognitive errors: are those in which the problem is with the individual diagnostician and include inadequate knowledge or faulty data-gathering, inaccurate clinical reasoning, or faulty verification [9,35,36].

Although many complex constructs to explain diagnostic errors exist, most clinicians are unaware of the cognitive and system-based underpinnings of this phenomenon [7]. Whether some or all diagnostic errors can be prevented is not known presently, with experts even questioning the goal of zero diagnostic errors [30]. Still, it stands to reason that all attempts should be made to increase diagnostic reliability and decrease diagnostic errors. An expert committee of the Institute of Medicine in US has outlined eight goals to reduce diagnostic errors and improve diagnosis [4]. Two of these are primarily focused on teaching-learning activities viz., "Enhance health care professional education and training in the diagnostic process" and "Research on the diagnostic process and diagnostic error"; in addition to the process of reflection addressed by the goal: "Develop and deploy approaches to identify, learn from, and reduce diagnostic errors and near missed in clinical practice" [4].

One major problem with addressing diagnostic errors is the multiplicity of the underlying factors; an average of six contributory factors per error were identified in an internal medicine department, with almost two-third being system-related ones [34]. Strategies to reduce diagnostic errors may focus on two inter-related aspects viz., Educational activities (teaching–learning activities to teach avoidance of diagnostic errors), and System-related or Managerial strategies, which can be used in practice to address some aspects of diagnostic errors. In this communication, the focus would be on the educational interventions to reduce diagnostic errors. Educational interventions can be used to address both system-related and cognitive root causes [36,37]. Although, it may appear intuitive that only cognitive errors would be amenable to teaching-learning strategies and it may not be possible to address system-errors, educational strategies have been suggested for both. Cognitive interventions to reduce diagnostic errors have been classified into three mutually exclusive categories by some experts [36-38]: (a) Interventions to increase clinicians’ knowledge and experience; (b) Interventions to improve clinical reasoning and decision-making skills; and (c) interventions that assist clinicians with tools or access to other clinicians or experts.

Two educational strategies have been suggested for reducing diagnostic errors viz., efforts to directly improve cognition, and an indirect system-related approach. However, we need to be aware that cognitive shortcomings can always undermine potential improvements from system changes, and thus educational efforts to directly improve cognition are the lynchpins for reducing diagnostic errors. Moreover, cognitive shortcomings play a dominant role in contributing to diagnostic error [36-38], and thus need to be primarily addressed.

The primary responsibility for training and teaching the diagnostic process and provide more opportunities for learning the same, is on medical colleges and clinical departments, so as to develop highly competent diagnosticians [36,37].

Although the normative approach to diagnosis (consisting sequentially of differential diagnosis, narrowing to a limited number possibilities, and Bayesian probability evaluation) is diligently taught and re-taught in medical school, it is infrequently used in clinical practice, and in fact may not work if the case is just too complicated [37-40]. The current paradigm of the diagnostic process by a knowledgeable and experienced physician is called as dual-process model, part automatic and part conscious [36-40]. The system 1 or the subconscious, rapidly acting automatic system solves the common problems; whereas, the conscious, rational and analytical cognition (System II) comes into play when problems are complicated and not amenable to automatic decision-making, and also to check on the solutions to the easy problems. This process of diagnosis needs to be taught to trainees so that they know what are its shortcomings and ways to overcome these. The traditional model of medical education aiming for initial building up of the basic sciences theoretical knowledge, followed by clinical exposure and application of the knowledge, has been replaced completely by a Problem-based learning curriculum, where all teaching-learning activities are in the setting of a clinical problem [40,41].

A framework for considering educational interventions for reducing diagnostic errors has been suggested by Graber [36,37], wherein the diagnostician progresses from the basic level (using deductive reasoning) to intermediate level (using heuristics) to expert level skills, with increasing reliability and accuracy, in addition to decreased cost and effort (Fig. 2). Two possible educational strategies have been suggested based on this framework viz., by improving expertise, and by conscious, reflective review. As Croskerry [38-42] has suggested, this allows a cross-check of the solutions provided by the subconscious, System I.

Fig. 2 An inverted pyramid model of diagnostic proficiency.

Enhancing expertise is the apparently easier strategy to improve the diagnostic reliability and thereby reduce diagnostic errors. Students need to be trained for being better thinkers by teaching discriminative skills and providing more examples and repetitions. This improves their clinical-decision making skills. However, it needs long-term sustained efforts, changes in the educational curriculum, and application over the whole duration of the course.

A satisfactory skill-set coupled with an adequate knowledge base gives the ability to recognize patterns in disease presentations and make reliable diagnoses [42,43]. For the undergraduates, it will primarily involve acquiring knowledge and the skills required to apply this for diagnosis i.e., history-taking and examination skills. It has been the experience of many, and also published literature [43,44], that these skills are infrequently achieved. Failure to gather information through history, physical examination or chart-review was the most common system factor reported, according to a physician survey in US [17]. Although postgraduates already have a wide range of competencies, still further development is possible, though difficult. It requires extensive practice combined with formative feedback [36,37]. We need to recognize these principles and use the opportunity to teach superior diagnostic skills through opportunities to practice on a wide-range of diseases with all their spectrum of presentations – definitely a tall order in the limited period residents are in training. Simulated patients [44,45], especially for uncommon diseases or unusual presentations, may be helpful in this regard.

Encouraging sub (or so-called ‘super’)-specialization is another way to make available expert diagnosticians to diagnose diseases in a focused filed, as their knowledge-base and skill set are still higher. However, in the absence of a trained person making an initial diagnosis and referring correctly to the appropriate specialist, and given the unequal distribution of expertise in our scenario, ensuring access to primary-care by a basic doctor may be a better goal than this approach.

Conscious reflective review: In the short term, conscious reflective review may gainfully be employed to improve clinical diagnosis and reduce diagnostic errors. The subconscious problem-solving, utilized to make diagnosis in a large number of cases, is characterized by automatic actions [38-40]. For effective use of reflections, the user needs to be taught and made aware of these and other pitfalls in the diagnostic process, though the benefits of reflections for improving diagnosis have also been questioned [45-48].

Students must be encouraged and provided an opportunity to reflect on their diagnostic approach, and think about what they could be missing. Morbidity and mortality conferences are a good place to do this [6]. Residents can be asked to reflect on their delivery of patient care based on actual case records (case based discussions) or on the discharge papers of patients (chart-stimulated recall) [6].

Use of a checklist to have a review of the diagnostic process and ensure a feedback loop has been suggested [36,37]. Using a checklist has a long history in other high-pressure multiple decision-making settings like aviation, and can be used to benefit the process of diagnosis. A checklist ensures that no step is overlooked and the same steps are followed by all team members; one with an inbuilt feedback loop has been suggested recently [36, 37, 48,49]. The same has also been detailed in Fig. 1.

Metacognitive training: Metacognition, the ability of being aware of one’s thinking process, is apparently a unique human ability. One of the applications of this process is active open-mindedness, and has been shown to lead to ‘better’ decisions [30]. Training strategies for the same have been described and can be used to improve the cognitive process for diagnosis-making.

To incorporate these principles in to day-to-day clinical teaching, clinical case-presentations and case-based discussions are important opportunities. Structured case presentation models like SNAPPS (Summarize the case, Narrow the differentials, Analyze the differentials, Probe the preceptor about uncertainties, Plan management for the patient, Select case related issues for self-study) and One-Minute Preceptor (OMP) provide opportunities to get knowledge, probe students’ reasoning and provide opinion [6]. The opportunity to describe one’s clinical reasoning and being guided to the correct diagnosis provides important insight to the students regarding the diagnosis-making process. Such a metacognitive approach may help learners to recognize the need to slow down and avoid errors that occur due to premature closure of reasoning [49].

Majority of these inputs are administrative in nature and have only a small educational component.

Clinical Guidelines and Clinical decision-support systems: Clinical guidelines, and their extension, clinical decision-support systems, are likely to be helpful in reducing diagnostic errors [36,37]. Guidelines, if developed by considering appropriate prevalence rates and probability estimates, with efforts to minimize errors due to heuristics, may help to improve diagnosis in the clinical setting [50,51]. The problems with clinical guidelines are many, including clinicians being unaware of their existence or unwilling to follow them, not being available to all levels of healthcare, and quite a few of these themselves being heuristic in nature. Teaching about the evidence-based medicine, and consensus-building for synthesizing research evidence to generate guidelines will help in ensuring development of clinically relevant guidelines that are widely-accepted. Guidelines have in fact, been shown to be ineffective in reducing diagnostic errors, though effective in reducing errors in treatment and preventive settings [51,52].

Fig. 3 Steps in improving the reliability of diagnosis (developed from Graber, 2009 [24]).

Use of technology in health-management and patient care, the so called Health-information technology (HIT) approaches, are one of the most rapidly expanding areas in the field of patient-safety. These tools and algorithms can be characterized as those assisting: (i) information gathering; (ii) information organization and display; (iii) differential diagnosis generation; (iv) weighing of diagnoses; (v) generation of diagnostic plan; (vi) access to diagnostic reference information; (vii) facilitating follow-up; (viii) screening for early detection in asymptomatic patients; (ix) collaborative diagnosis; and (x) facilitating diagnostic feedback to clinicians [52,53]. Recognizing the importance of information technology for the doctor of today, the General Medical Council in UK advocates that medical graduates should be able to "make effective use of computers and other information systems, including storing and retrieving information" [53], something which the MCI document overlooks [2]. A recent systematic review showed that HIT is efficacious in improving quality and efficiency of medical care [54]; though, electronic health records and electronic decision-support systems were the two of the HIT modalities most commonly used. There is an urgent need to incorporate these in the health management system, and train students in their appropriate use for a patient-centric healthcare.

Clinical decision-support systems directly incorporate computer-based decision system into the clinical practice of medicine at the bedside. These are gradually being used more widely, but the evidence of efficacy for clinical diagnosis is sketchy, though they have been shown to improve patient outcomes and provider behavior related to drug-dosing, preventive care and general process of care [54,55]. One reason for this could be the poor quality of input into the system [30], thereby limiting the functionality of the system, which could either be due to time constraints preventing complete input of data, or biased data due to the initial hypothesis being considered, or sufficient input data has not been collected.

Improving perception: The method of presenting data has a major impact on our ability to pick up abnormalities [30]. Presenting data in such a way that abnormal findings stand out e.g., highlighting (or using bold typeface) the abnormal results in a report, and presentation as graphs/figures may be more helpful in identifying an abnormal result from a series of normal results. SMS-based or other app-based reminders for abnormal results or need for follow-up a patient may also reduce diagnostic errors.

Expert-opinion: Non-availability of appropriate experts is a recognized reason for missed findings leading to diagnostic errors [55,56], and has been reported even from emergency-rooms in developed countries. It is likely to be more common in the developing country scenario. Although more training may appear a low-cost and direct method, its sustainability is questionable. As far as the educational aspects are considered, improved supervision of trainees [30] and posting of residents of varying seniority together are ways of making expert-opinion available on call, and possibly reducing diagnostic errors.

An extension of this approach, especially in non-urgent situations will be the concept of second opinion. These have been proven effective in decreasing medication errors and can also gainfully be utilized in diagnostic errors whether for diagnostic tests or for complex or difficult clinical diagnoses [57,58].

Given that system errors affect all persons and activities concerned with patient-care, addressing these is likely to reduce diagnostic errors for all diagnosticians working in the system, and also have a long-term effect [30].

A major educational strategy for system-related errors could be addressing communication-related deficits including communication between patient and healthcare personnel, and between various members of the healthcare team (among physicians, between physicians and other team members). This should also address the process of handing- and taking-over of patients’ records at the time of shift changes (‘handoffs’ or ‘sign-outs’), an issue that is especially relevant in the current scenario of restricted work hours of residents [59,60]. Improper handoffs may lead to insufficient clinical or test-related information being communicated, which may lead to increased opportunity for diagnostic errors [59].

Improvements in verbal and written communication between health care providers during patient handoffs can reduce injuries due to medical errors by up to 30 percent [61]. A study on a system of bundled communication and training tools for handoff of patient care between providers significantly increased patient safety without significantly burdening existing clinical workflows. Moreover, physicians in a study ranked access to electronic health records and following up with patients as effective strategies in preventing diagnostic errors [17].

There is ample evidence to demonstrate that communication skills can be increased through directed training [62]; thus making these a promising target for educational interventions. However, the current Attitudes and Communications (ATCOM) module of MCI is focusing primarily at doctor-patient communication [63], and needs to be expanded to also include communication between healthcare workers. Improving communication skills has been shown to reduce diagnostic errors due to communication-related system errors [61,64]. New tools to improve communication like Situation, Background, Assessment, and Recommendation (SBAR) technique (a framework for effective communication among members of the health care team) [65], and techniques for improved hand-off [66] can be included into training curricula of nursing staff and residents and serve as an opportunity to address diagnostic errors by educational intervention. Training of emergency medicine residents in SBAR [67] has been used to improve communication during hand-offs. Team training, is another educational intervention that may improve communication between the healthcare team and reduce system-related diagnostic errors [36,37]. Most of these educational interventions could gainfully be incorporated during the internship and residency training.

As primarily these are due to the shortcomings of current medical knowledge and technological abilities of the diagnostic armamentarium, there is likely to be gradual reduction in them with advancement of medical knowledge and technological progress, which will allow recognition of newer diseases and also permit diagnosis at early stages of the disease.

However, there is certain inevitability about these errors and they are unlikely to be eliminated, e.g., atypical presentations, early non-specific manifestations, emerging diseases (new diseases, new pathogens, or side-effects of newer drugs etc.) may continue to lead to diagnostic errors, and medical science will need to continuously evolve to catch up with the utopia of nil diagnostic errors. Patient-specific factors include both non-compliance (due to a variety of reasons [67,68]), and the patients’ appreciation and description of their symptoms (which may vary depending on their socio-cultural or educational background, and their belief-system). Such training in communication and coordination of care would go a long way in reducing this group of errors.

Educational intervention can be aimed at improving cognition directly by training better history-takers (able to tease out hidden aspects of history from the patient), and better communicators (thereby ensuring compliance with investigations, treatment and follow-up). The students need to be taught, in addition, to practice worst-case scenario medicine (always consider the worst possible diagnosis when faced with a particular symptom), and ask what (consider what symptoms or findings cannot be explained by the diagnosis) and why (thinking about the reason why some problem occurred, especially for an acute exacerbation of a chronic disorder) questions [7]. Addressing the culture of non-performance and tolerance of errors also needs to be addressed by educational interventions; departmental and Institutional mortality reviews and difficult-case meetings should aim to address these.

Improving the diagnostic process is not only possible, but also represents a moral, professional, and public health imperative [4]. Although, the individual physicians may be more comfortable in discussing the ‘system-faults’ rather than cognitive errors, which are apparently more personal, it is the cognitive aspect that is more amenable to educational interventions. Addressing diagnostic errors systematically will require a comprehensive approach; developing formal curricula to educate trainees about misdiagnoses is one such [68,69]. We hope that individual institutions and teachers will take the onerous responsibility of carrying out teaching-learning activities for reduction of diagnostic error, till such time professional or regulatory bodies develop such curricula.

1. Athreya BH. Handbook of Clinical Skills. World Scientific, Singapore; 2010.

2. Medical Council of India. Vision 2015. MCI, Delhi; 2011.p.15-16. Available from: http://www.mciindia.org/tools/ announcement/MCI_booklet.pdf. Accessed January 25, 2016.

3. Graber　M. Diagnostic errors　in medicine: a case of neglect. Jt Comm J Qual Patient Saf. 2005;31:106-13.

4. Committee on Diagnostic Error in Health Care, Institute of Medicine. Improving Diagnosis in Health Care. September, 2015. Available from: https://iom.nationalacademies.org/~/media/Files/Report%20Files/2015/Improving-Diagnosis /Diagnostic Error_ReportBrief.pdf. Accessed December 12, 2015.

5. Newman-Toker　DE,　Pronovost PJ. Diagnostic errors–the next frontier for　patient safety. JAMA.　2009; 301:1060-2.

6. Modi JN, Anshu, Gupta P, Singh T. Teaching and assessing clinical reasoning skills. Indian Pediatr. 2015; 52:787-94

7. Trowbridge RL. Twelve tips for teaching avoidance of diagnostic errors. Med Teach. 2008;30:496-500.

8. Croskerry P. The importance of cognitive errors in diagnosis and strategies to minimize them. Acad Med. 2003;78:775-80.

9. Bordage G. Why did I miss the diagnosis? Some cognitive explanations and educational implications. Acad Med. 1999;74(10 suppl):S138-S143.

10. McLeod W, Eidus R, Stewart E. Clinical decision support: Using technology to identify Patients’ Unmet Needs. Fam Pract Manag.　2012;19:22-8. Available from: http://www.aafp.org/ fpm/2012/0300/p22.html. Accessed January 27, 2016.

11. Apps: Bloice MD,　Simonic KM,　Holzinger A. Casebook: a virtual patient iPad application for teaching decision-making through the use of electronic health records. BMC Med Inform Decis Mak.　2014;14:66.

12. Murphy DR, Wu L, Thomas EJ, Forjuoh SN, Meyer AN, Singh H. Electronic trigger-based intervention to reduce delays in diagnostic evaluation for cancer: a cluster randomized controlled trial. J Clin Oncol. 2015;33:3560-7.

13. Singh H,　Giardina TD,　Forjuoh SN,　Reis MD,　Kosmach S,　Khan MM,　et al. Electronic health record-based surveillance of diagnostic errors in primary care. BMJ Qual Saf.　2012;21:93-100.

14. Olsen DE,　Sticha D. Interactive simulation training: computer simulated standardized patients for　medical diagnosis. Stud Health Technol Inform.　2006;119:413-5.

15. Berner ES,　Graber ML. Overconfidence as a cause of diagnostic error in medicine. Am J Med. 2008;121 (5 Suppl):S2-23.

16. Singh H, Giardina TD, Meyer AN, Forjuoh SN, Reis MD, Thomas EJ, et al. Types and origins of diagnostic errors in primary care settings. JAMA Intern Med. 2013;173:418-25.

17. Singh H, Thomas EJ, Wilson L, Kelly PA, Pietz K, Elkeeb D, et al. Errors of diagnosis in pediatric practice: a multi-site survey.　Pediatrics. 2010;126:70-9.

18. Ordi J, Ismail MR, Carrilho C, Romagosa C, Osman N, Machungo F, et al. Clinico-pathological discrepancies in the diagnosis of causes of maternal death in sub-saharan Africa: retrospective analysis. PLoS Med. 2009;6:e1000036.

19. Herman AA, Stein DJ, Seedat S, Heeringa SG, Moomal H, Williams DR. The South African Stress and Health (SASH) study: 12-month and lifetime prevalence of common mental disorders.　South African Med J. 2009;99:339-44.

20. Cheng XJ1,　Cheng Q,　Xu LZ,　Zhao HQ,　Zhao Z,　Wang W,　et al. Evaluation of multiple sclerosis diagnostic criteria in Suzhou, China—risk of under-diagnosis in a low prevalence area. Acta Neurol Scand.　2010;121:24-9.

21. Kotovicz F, Mauad T, Saldiva PHN. Clinico-Pathological Discrepancies in a General University Hospital in São Paulo, Brazil.　Clinics (Sao Paulo, Brazil). 2008;63:581-88. Available from: http://www.ncbi.nlm.nih.gov/pmc/articles/PMC2664713/. Accessed January 26, 2016.

22. Prashanth LK, Taly AB, Sinha S, Arunodaya GR, Swamy HS. Wilson’s disease: diagnostic errors and clinical impli-cations. J Neurol Neurosurg Psychiatry. 2004;75:907-09.

23. Parikh PM, Nair V, Malhotra H, Ranade AA, Bhattacharyya GS. Medical Errors – Aiming to improve the Art and Science of Healthcare. Medicine Update 2008.p.873-81. Available from: http://www.apiindia.org/pdf/medicine_ update_2008/chapter_115.pdf. Accessed January 26, 2016.

24. Leape LL, Brennan TA, Laird N, Lawthers AG, Localio AR, Barnes　BA, et　al. The nature of adverse events in hospitalized patients. Results of the Harvard Medical Practice Study II. N Engl J Med. 1991;324:377-84.

25. Bartlett EE. Physicians’ cognitive errors and their liability consequences. J Healthcare Risk Manage. 1998;18:62-9.

26. Tai DYH, El-Bilbeisi H, Tewari S, Mascha EJ, Wiedermann HP, Arroliga AC. A study of consecutive autopsies in a medical ICU: a comparison of clinical cause of death and autopsy diagnosis. Chest. 2001; 119:530-6.

27. Croskerry P. The theory and practice of clinical decision making. Can J Anesth. 2005;52:R1-R8.

28. Norman G. Dual processing and diagnostic errors. Adv Health Sci Educ Theory Pract. 2009;14:37-9.

29. Elstein AS. Heuristics and biases: selected errors in clinical reasoning. Acad Med. 1999;74:791-4.

30. Graber M, Gordon R, Franklin N. Reducing diagnostic errors in medicine: what’s the goal? Acad Med. 2002;77:981-92.

31. Wardrop D. Ockham’s Razor: sharpen or re-sheathe? J R Soc Med. 2008: 101: 50-51.

32. Voytovich AE, Rippey RM, Suffredini A. Premature closure in diagnostic reasoning. J Med Educ. 1985;60: 302-7.

33. Herrle SR, Corbett EC, Fagan MJ, Moore CG, Elnicki M. Bayes’ theorem and the physical examination: probability assessment and diagnostic decision-making. Acad Med. 2011;86:618-27.

34. Graber ML, Franklin N, Gordon R. Diagnostic error in internal medicine. Arch Intern Med. 2005;165:1493-9.

35. Kostopoulou O, Delaney BC, Munro CW. Diagnostic difficulty and error in primary care - a systematic review. Fam Pract. 2008;25:400-13.

36. Kassirer JP, Kopelman RI. Cognitive errors in diagnosis: instantiation, classiûcation, and consequences. Am J Med. 1989;86:433-41.

37. Graber ML. Educational strategies to reduce diagnostic error: can you teach this stuff? Adv Health Sci Educ. 2009;14:63-9

38. Graber ML, Kissam S, Payne VL, Meyer AN, Sorensen A, Lenfestey N, et al. Cognitive interventions to reduce diagnostic error: a narrative review. BMJ Qual Saf. 2012;21:e535-57.

39. Wears RL. What makes diagnosis hard? Adv Health Sci Educ Theory Pract.　2009;14 (Suppl 1):19-25

40. Croskerry, P. Clinical cognition and diagnostic error: applications of a dual process model of reasoning. Adv Health Sci Educ. 2009; 14 (Suppl 1):27-35.

41. Diemers AD,　van de Wiel MW,　Scherpbier AJ,　Baarveld F,　Dolmans DH. Diagnostic reasoning and underlying knowledge of students with preclinical patient contacts in PBL. Med Educ.　2015;49:1229-38.

42. Croskerry, P. Cognitive forcing strategies in clinical decision making. Ann Emergency Med. 2003; 41:110–20.

43. Bowen JL. Educational strategies to promote clinical diagnostic reasoning. NEJM. 2006; 355:2217-25.

44. Hauer KE,　Teherani A,　Kerr KM,　O’Sullivan PS,　Irby DM. Student performance problems in　medical　school clinical skills assessments. Acad Med.　2007;82(10 Suppl):S69-72.

45. Murray DJ,　Freeman BD,　Boulet JR,　Woodhouse J,　Fehr JJ,　Klingensmith ME. Decision making in trauma settings: simulation to improve diagnostic skills. Simul Healthc. 2015;10:139-45.

46. Croskerry P1,　Petrie DA,　Reilly　JB,　Tait G. Deciding about fast and slow decisions. Acad Med.　2014;89:197-200

47. Monteiro SD,　Sherbino J,　Patel A,　Mazzetti I,　Norman GR,　Howey E. Reflecting on　diagnostic errors: taking a second look is not enough. J Gen Intern Med.　2015; 30:1270-4

48. Schiff GD. Minimizing diagnostic error: The importance of follow-up and feedback. Am J Med. 2008;121:S38-S42.

49. Moulton CE, Regehr GE, Mylopoulos M, MacRae HM. Slowing down when you should: A new model of expert judgment. Acad Med. 2007;82:S109-S116.

50. Garfield FB, Garfield JM. Clinical judgement and clinical practice guidelines. Int J Technol Assess Health Care. 2000;16:1050-60.

51. Klassen TP, Ropp LJ, Sutcliffe T, Blouin　R, Dulberg C, Raman S, et al. A randomized controlled trial of radiograph ordering for extremity trauma in a pediatric emergency department. Ann Emerg Med. 1993;22:1524-9.

52. El-Kareh R, Hasan O, Schiff GD. Use of health information technology to reduce diagnostic errors. BMJ Qual Saf. 2013;22:ii40-51.

53. General Medical Council. Tomorrow’s Doctors: Outcomes for Graduates, 2015, p. 7. Available from: http://www.gmc-uk.org/Outcomes_for_graduates_Jul_15.pdf_61408029. pdf. Accessed May 11, 2016.

54. Chaudhry B, Wang J, Wu S, Maglione M, Mojica W, Roth E, et al. Impact of health information technology on quality, efficiency, and costs of medical care. Ann Intern Med. 2006;144:742-52.

55. Hunt DL, Haynes RB, Hanna SE, Smith K. Effects of computer–based clinical decision support systems on physician performance and patient outcomes: a systematic review. JAMA. 1998;283:2816-21.

56. Kripalani S, Williams MV, Rask K. Reducing errors in the interpretation of plain radiographs and computed tomography scans. In: Shojania KG, Duncan BW, McDonald KM, Wachter RM (eds). Making Health Care Safer. A Critical Analysis of Patient Safety Practices. Rockville, MD: Agency for Healthcare Research and Quality, 2001.

57. Kronz JD, Westra WH, Epstein JI. Mandatory second opinion surgical pathology at a large referral hospital. Cancer. 1999;86:2426-35.

58. Payne VL,　Singh H,　Meyer AN,　Levy L,　Harrison D,　Graber ML. Patient-initiated second opinions: systematic review of characteristics and impact on diagnosis, treatment, and satisfaction. Mayo Clin Proc.　2014;89:687-96.

59. Riesenberg LA, Leitzsch J, Massucci JL, Jaeger J, Rosenfeld JC, Patow C, et al. Residents’ and attending physicians’ handoffs: a systematic review of the literature. Acad Med 2009;84:1775-87.

60. DeRienzo CM, Frush K, Barfield ME, Gopwani PR, Griffith BC, Jiang X, et al. Handoffs in the era of duty hours reform: a focused review and strategy to address changes in the Accreditation Council for Graduate Medical Education Common Program requirements. Acad Med. 2012;87: 403-10.

61. Starmer AJ, Spector ND, Srivastava R, West DC, Rosenbluth G, Allen AD, et al. for the I-PASS Study Group. Changes in medical errors after implementation of a handoff program.　NEJM. 2014; 371:1803-12D.

62. Modi JN, Anshu, Chhatwal J, Gupta P, Singh T. Teaching and assessing communication skills during undergraduate training. Indian Pediatr. 2016:53:.

63. Medical Council of India. Sensitization Program on Attitude and Communication Module (AT-COM). MCI decisions on MET. 2015. Available from: http://www.mciindia.org/fdp/MCI_decisions_on_MET.pdf. Accessed October 3, 2015.

64. Singh H, Naik AD, Rao R, Petersen LA. Reducing diagnostic errors through effective communication: harnessing the power of information technology. J Gen Intern Med. 23:489-94.

65. WHO Collaborating Centre for Patient Safety Solutions. Communication during patient handovers. Patient Safety Solutions 2007; 1: 3. Available from: http://www.who.int/patientsafety/solutions/patientsafety/PS-Solution3.pdf. Accessed January 21, 2016.

66. Kemp CD, Bath JM, Berger J, Bergsman A, Ellson T, Emery K, et al. The top 10 list for a safe and effective sign-out. Arch Surg. 2008;143:1008-10.

67. Tews MC, Liu JM, Treat R. Situation-Background-Assessment-Recommendation (SBAR) and emergency medicine residents’ learning of case presentation skills.　J Grad Med Educ. 2012;4:370-73.

68. Jin J, Sklar GE, Min Sen Oh V, Chuen Li S. Factors affecting therapeutic compliance: A review from the patient’s perspective.　Therapeut Clin Risk Manag. 2008;4:269-286.