Search

Background: Large language models such as GPT-4 (Generative Pre-trained Transformer 4) are being increasingly used in medicine and medical education. However, these models are prone to "hallucinations" (ie, outputs that seem convincing while being factually incorrect). It is currently unknown how these errors by large language models relate to the different cognitive levels defined in Bloom's taxonomy., Objective: This study aims to explore how GPT-4 performs in terms of Bloom's taxonomy using psychosomatic medicine exam questions., Methods: We used a large data set of psychosomatic medicine multiple-choice questions (N=307) with real-world results derived from medical school exams. GPT-4 answered the multiple-choice questions using 2 distinct prompt versions: detailed and short. The answers were analyzed using a quantitative approach and a qualitative approach. Focusing on incorrectly answered questions, we categorized reasoning errors according to the hierarchical framework of Bloom's taxonomy., Results: GPT-4's performance in answering exam questions yielded a high success rate: 93% (284/307) for the detailed prompt and 91% (278/307) for the short prompt. Questions answered correctly by GPT-4 had a statistically significant higher difficulty than questions answered incorrectly (P=.002 for the detailed prompt and P<.001 for the short prompt). Independent of the prompt, GPT-4's lowest exam performance was 78.9% (15/19), thereby always surpassing the "pass" threshold. Our qualitative analysis of incorrect answers, based on Bloom's taxonomy, showed that errors were primarily in the "remember" (29/68) and "understand" (23/68) cognitive levels; specific issues arose in recalling details, understanding conceptual relationships, and adhering to standardized guidelines., Conclusions: GPT-4 demonstrated a remarkable success rate when confronted with psychosomatic medicine multiple-choice exam questions, aligning with previous findings. When evaluated through Bloom's taxonomy, our data revealed that GPT-4 occasionally ignored specific facts (remember), provided illogical reasoning (understand), or failed to apply concepts to a new situation (apply). These errors, which were confidently presented, could be attributed to inherent model biases and the tendency to generate outputs that maximize likelihood., (©Anne Herrmann-Werner, Teresa Festl-Wietek, Friederike Holderried, Lea Herschbach, Jan Griewatz, Ken Masters, Stephan Zipfel, Moritz Mahling. Originally published in the Journal of Medical Internet Research (https://www.jmir.org), 23.01.2024.)

Összefoglaló. Bevezetés: A pszichoszomatikus medicina területéhez tartozó betegek ellátása rossz hatásfokú. Ez egyaránt jelentős terhet jelent a betegek és az egészségügy számára. Célkitűzés: A pszichoszomatikus tünetek és betegségek megjelenésének és ellátási ajánlásainak vizsgálata az irányadó hivatalos szakmai anyagban. Módszer: Az adatgyűjtés időpontjában elérhető 531 Szakmai irányelvből annak a 134 dokumentumnak a részletesebb elemzése, amelynek témájában pszichoszomatikus tünetek/betegségek gyakoribbak. Ebből 39 egyértelműen pszichoszomatikus vonatkozású anyag tartalmi vizsgálata, továbbá a Szakmai Kollégiumok listáján szereplő három pszichiátriai szakterület összes (26) szakmai irányelvének hasonló szempontú értékelése. Eredmények: A megvizsgált irányelvek töredéke tartalmaz csupán a pszichoszomatikus tünetképzésre, illetve kezelésre, megelőzésre vonatkozó megállapítást. A releváns témájú irányelvekben szereplő diagnosztikai utalások aránya: direkt 10,25, indirekt forma 23,7, a terápiás ajánlások között direkt 25,64, indirekt forma 15,38%. A megelőzésre vonatkozóan alig (5,12%-ban) szerepel említés, a pszichés betegségek egyidejű fennállásának esélyére is mindössze 10,25%-ban történik megállapítás. A szorongásos zavarok befolyásáról kevesebb, mint minden harmadik irányelv számol be (30,76%). A pszichiátriai irányelvek között pszichoszomatikus betegségekre vonatkozóan nincs célzott szakmai irányelv. Következtetés: A pszichoszomatikus medicina a hatékony betegellátás követelményeit meghatározó szakmai irányelvekben elégtelen arányban jelenik meg. Okait szemléleti zavarokban, a pszichiátriai szakmai konszenzus és a szakmaközi együttműködés hiányosságaiban kell keresni. Mind a vonatkozó kutatásokban, mind az irányelvek fejlesztése során erőfeszítéseket kell tenni a fentiek miatt kallódó nagy betegpopuláció hatékonyabb ellátása érdekében. A szerző javaslatot tesz a szükséges konkrét lépésekre. Orv Hetil. 2021; 162(7): 252-261., Introduction: The care of patients with psychosomatic disorders has bad efficacy. It beards hard both for the patients and the health care providers., Objective: Review of presentrecommendations for psychosomatic symptoms and illnesses in the recent Hungarian medical protocols., Method: A detailed evaluation of 134 medical protocols from altogether 531 protocols with closer connection with psychosomatic disorders were carried out, their diagnostic and therapeutic recommendations were analyzed. Further review of 39 (of the 134)directly related to psychosomatics was done. Investigation of psychiatric protocols for the same aspects was carried out, as well., Results: Only a small part of the protocols contain recommendations for psychosomatic disorders. Ratios of their presence: diagnostic - direct 10.25, indirect 23.7, therapeutic - direct 25.64, indirect 15.38%. Efforts for prevention are only in 5.12% of the materials and 10.25% of the protocols mention a possible dual nature of the given clinical manifestation. Anxiety disorders occur only in 30.76%. Psychiatric protocol directly aiming psychosomatic problems is lacking., Conclusion: Presence of psychosomatic medicine in clinical guidelines defining the conditions of successful care of patients is practically minimal. Possible explanations: confusion in approaches, lack of consensus among the psychiatric institutions, and deficiency of interdisciplinary cooperation. Both in relevant clinical research and in development of medical protocols common additive efforts are necessary for achieving a more successful care of high numbers of patients presently neglected because of the mentioned causes. Author advises some concrete steps for the solution. Orv Hetil. 2021; 162(7): 252-261.

The psychiatric, psychosocial, and existential/spiritual pain determined by chronic medical disorders, especially if in advanced stages, have been repeatedly underlined. The right to approach patients as persons, rather than symptoms of organs to be repaired, has also been reported, from Paul Tournier to Karl Jaspers, in opposition and contrast with the technically-enhanced evidence-based domain of sciences that have reduced the patients to 'objects' and weakened the physician's identity deprived of its ethical value of meeting, listening, and treating subjects. The paper will discuss the main psychosocial and existential burden related to chronic and advanced medical illnesses, and the diagnostic and therapeutic implications for a dignity preserving care within a person-centred approach in medicine, examined in terms of care of the person (of the person's whole health), for the person (for the fulfilment of the person's health aspirations), by the person (with physicians extending themselves as total human beings), and with the person (working respectfully with the medically ill person).

The essay deals with the relationship between the scientific and humanistic social scientific aspects of medical concept formation and science. The first part focuses on methodological differences associated with the two forms of knowledge, emphasizing the distinction between quantitative and qualitative methods. The differences are traced back to type of phenomenon or object researched. On the basis of the special character of human beings and their illnesses it is then argued that scientific medicine necessarily will have to transcend the divide between the natural sciences on the one hand and the humanities and social sciences (the "human sciences") on the other. The unity in medicine is tied to its practice--the art of medicine as carried on in the clinical meeting between doctor and patient.

The writer criticizes the unifying models such as the classical, nineteenth century Psychiatry which makes of Psychiatry a mere branch of Medicine or Psychosomatic Medicine which imposes a "psychodynamic" pattern for the whole Medicine. He therefore recommends a dual and simultaneous approach destined to decipher two series of different signals and to meet somewhat different situations and decisions. This dual approach supposes a double training; it applies to the patient seen for the first time, but also to the patient in treatment who can either unveil significant symptoms of an underlying somatic disease or become somatically ill, with no relation with his psychiatric symptoms. Finally, a similar attitude of vigilance should be observed as far as supervision of chemotherapy is concerned.