[PD] Basics of Clinical Reasoning.pdf

PHYSICAL DIAGNOSIS 08/24/2024. MOD 3: BASICS...

PHYSICAL DIAGNOSIS 08/24/2024. MOD 3: BASICS OF CLINICAL REASONING Dr. Jose Ramon Pascual III Trans Group/s: 1B I. CLINICAL REASONING The rational process by which a decision is made regarding various aspects of patient care, specifically the prevention, diagnosis, and management of disease. TYPES OF CLINICAL REASONING based on the Dual Process Theory of Reasoning NON-ANALYTIC ANALYTIC (SYSTEM 1) (SYSTEM 2) Fast Slow Intuitive Deliberate Low cognitive load High cognitive load Used for standard Used for difficult cases: cases: common CPC, EBM, M and M bedside, outpatient diagnosis Model of Diagnostic Process. Default process: Works best when dealing works best for most with unusual situations or 1. PATTERN MATCHING/SCRIPT-BASED (or SYSTEM 1) other conditions diagnostic/therapeutic REASONING STRATEGIES equipoise Expert clinicians use pattern matching and script-based reasoning strategies also known as Proper clinical reasoning requires using both System 1 reasoning strategies more often than the systems. hypothetical deductive strategies. ○ It implies that both systems have to be taught and Experts using System 1 reasoning are usually able to practiced. make a correct diagnosis faster and with less effort than novices. A. MODEL OF DIAGNOSTIC PROCESS Data gathered is used to generate a problem 2. HYPOTHETICO-DEDUCTIVE (or SYSTEM 2) representation. REASONING STRATEGIES The scripts are retrieved until a match is found between Novices, on the other hand, defaulted a slower the problem representation and the script. Hypothetico-Deductive strategy or System 2 strategy ○ If there is a match, the clinician will proceed to the primarily because: action plan. ○ The script requires some experience and they ○ If there are no suitable scripts, a hypothetical cannot use what they do not have; and deductive strategy is chosen wherein relevant ○ This is how most novices are taught. differentials are generated and are either ruled in or While the Hypothetico-Deductive strategy is theoretically ruled out preferably using evidence-based medicine more thorough, it is much slower and is relatively procedures. cognitively taxing to the user. ○ Both experts and novices utilize the pattern matching and the hypothetical deductive strategies. ○ The preference for one strategy over the other is what differentiates an expert from a novice. Note also that failure of a treatment or action plan warrants further data gathering. Pattern Matching / Script Based Reasoning vs. Hypothetico-Deductive Reasoning. REASONING STRATEGIES PATTERN MATCHING/ HYPOTHETICO- Physical Diagnosis - Mod 3 Basics of Clinical Reasoning 1 of 10 The use of trans, practice questions, and evals ratio must be used discreetly and social media/public exposure of the aforementioned shall be strictly prohibited.  SCRIPT-BASED DEDUCTIVE (SYSTEM 1) (SYSTEM 2) Faster Slower Favored by Experts Favored by Novices More efficient Taxing II. COGNITIVE LOAD THEORY One way to understand reasoning is to look at it through the lens of cognitive load theory. For any practitioner, to reason properly, he must be able to call upon a large body of knowledge stored in Steps in Hypothetico-Deductive Reasoning. memory. Novices use this strategy because there is no A. ATKINSON-SHIFFRIN THREE STAGE MODEL OF alternative. HUMAN MEMORY Theoretically, more thorough than pattern matching in practice, hypothetical deductive reasoning imposes a significant cognitive load in working memory to the point it may overwhelm the working memory. Analogy: Amount of random access memory (RAM) in laptops ○ Most budget laptops in 2024, have 8gb of RAM which is sufficient for running office and browsers however, opening more than ten tabs, video or imaging editing software, or 3D gaming, you will begin to notice sluggishness in your system. ○ This is the reason why computer system Atkinson-Shiffrin Three Stage Model of Human Memory. compromises for video editing or gaming have a minimum of 16gb of RAM. 1. SENSORY MEMORY ○ Unfortunately, humans there is no way to open up Receives information in the form of sounds (echoic) the cranium and pop-in few chips to increase the and images (iconic) from the sensory organs. working memory capacity It has a large but very volatile capacity that lasts around 2500 ms. Information received from sensory memory will then be transmitted to working memory for processing. 2. WORKING (or SHORT-TERM) MEMORY Process the working and brain elements known as chunks. Has limited capacity ○ You can cram only 7 +/- 2 chunks, and can hold into that memory for around 15s to 30s without rehearsals, which makes working memory as bottleneck in learning Cognitive load of multiple differential diagnosis in hypothetico-deductive reasoning. 3. LONG-TERM MEMORY 1. EXCESS COGNITIVE LOAD CAN FILL-UP WORKING Work around the limitations of working memory. MEMORY It has a theoretically infinite capacity. It can hold information for a very long time. This can either stop the reasoning process or slow it down. B. HYPOTHETICO-DEDUCTIVE REASONING STRATEGY Reasoning errors, as a result of working memory The following are the steps in this strategy: congestion, can also occur. 1. Acquire Data This is not the natural or preferred way by which 2. Interpret Data humans carry out reasoning. 3. Make a hypothesis and differential diagnosis 4. Test hypothesis: can be an additional physical III. SCRIPT THEORY examination, laboratory, or physiologic test, or A schema with a particular order or sequence (e.g. imaging diagnostics. driving a car script). 5. Diagnosis: reached when there is enough clinical Schema is an organized set of knowledge for specific data, if not additional data shall be collected and the items (e.g. Trece Martires is the capital of Cavite). cycle begins anew. A. SCRIPTS A set of interconnected concepts that allows individuals to make predictions about how a particular event or sequence of events is likely to play out — a mental model. Physical Diagnosis - Mod 3 🏠 Basics of Clinical Reasoning 2 of 10 The use of trans, practice questions, and evals ratio must be used discreetly and social media/public exposure of the aforementioned shall be strictly prohibited.  Our brains interpret the world by comparing the attributes of the mental models to the features of an actual scene. ○ Check for consistencies and discrepancies, patterns and irregularities When a perfect or close-fit exits between the attributes of a mental script and the details of the real world scene, interpretation occurs quickly, automatically, and effortlessly in the form of instance or pattern recognition Pattern-Recognition/Script-Based Reasoning Strategy of experts. Experts are still all too human; they do not have an expanded working memory nor are they significantly more intelligent. ○ What they have is a large amount of organized packets or schemata of knowledge included previously in working memory, and now currently stored in an easily retrievable form in long-term memory. The working memory of experts is not overwhelmed Sample application of Script Theory. by information because all they need to store are the processes used to retrieve an organized schemata When there is a mismatch between predicted and stored in long-term memory. actual data, a individual initiates a search for an ○ This is how expert clinicians can make an explanation of the anomalous information accurate diagnosis after spending less than 15 ○ Demands slower, more laborious processing minutes with a patient. than that required to make sense of routine or Intrinsic load is devoted to retrieving previously typical situations encoded and stored packets of organized information. ○ The richer your library of illness and management scripts stored in your long-term C. ILLNESS SCRIPTS memory, the less likely you will need to perform Script utilized for the diagnosis of a disease laborious processing. Schemata or framework in clinical medicine Chunks of encapsulated information in a clinician’s long-term memory ○ Affected by the clinician’s knowledge based in experience ○ Vary in some degree from one clinician to another Some clinicians may have similar illness scripts, but will never be identical COMPONENTS OF ILLNESS SCRIPT 1 Enabling or causal factors Time and epidemiology Mismatch between the contact (dog) and the mental model Physical disease Pathophysiology 2 of a cat. mechanism or faults Resulting signs and Clinical manifestations The more detailed your metal model is, the better you 3 symptoms will be at recognizing potentially deleterious or fatal situations ○ A more developed illness script will tell you that a COMPONENTS OF ILLNESS SCRIPT lion is potentially hazardous to your health if you were to make contact with it, even if it matches the 1 Epidemiology Who gets the disease? mental model of a house cat. 2 Pathophysiology Mechanism of disease/injury What are the findings on history, Clinical 3 physical examination, laboratory Manifestations and imaging studies? How does this disease present in 4 Time Course time and what is the natural history of this disease? 1. SAMPLE ILLNESS SCRIPT: SEPTIC ARTHRITIS An example of a developed mental model. ILLNESS SCRIPT FOR SEPTIC ARTHRITIS B. PATTERN-RECOGNITION/ SCRIPT-BASED REASONING STRATEGY IV drug abusers, penetrating 1 Epidemiology trauma to joint Physical Diagnosis - Mod 3 🏠 Basics of Clinical Reasoning 3 of 10 The use of trans, practice questions, and evals ratio must be used discreetly and social media/public exposure of the aforementioned shall be strictly prohibited.  2 Pathophysiology Bacterial infection of joint 4 days = acute 2 Time (semantic qualifier — acute) Clinical Swollen knee, synovial WBC 3 Young Manifestations count >100,000/mm3 Epidemiology Male Presents acutely, natural history 3 (predisposing Basketball 4 Time Course includes sepsis or loss of factors) Player cartilage. Otherwise well Right knee pain and swelling Clinical IV. PROBLEM REPRESENTATION Pop in the right knee 4 Manifestations Rephrasing of the patient information Pain on standing/playing (consequences) Concise synthesis and summary of the patient story Buckling sensation Abstract Whenever we make a problem representation, we 1. ORIGINAL DATA GATHERED VS PROBLEM organize the data or knowledge in the same way we REPRESENTATION organize the scripts. ○ This makes it easier to match what you see in the Original Data Gathered patient to the illness script or the mental model of the disease in our mind. A 20-year-old man presents to your office complaining of right knee pain after playing basketball four days prior to COMPONENTS OF PROBLEM REPRESENTATION consultation. During the game, as he came down after jumping for a rebound, he twisted his extended knee while 1 Pathophysiology (faults) another player fell on the back of his calf. He remembers hearing a pop in the area of the knee and severe pain 2 Time upon standing, preventing him from playing in the remainder of the game. While on the bench, he noticed 3 Epidemiology (predisposing factors) that the pain improved, but his knee remained swollen. He 4 Clinical Manifestations (consequences) iced the knee and was able to put some weight on it later that day. Today, putting all of his weight on the knee makes it feel as if it will buckle. V. SEMANTIC QUALIFIERS Paired opposing descriptors (adjectives or adverbs) Problem Representation that can be used to systematically compare and contrast considerations. The patient is an otherwise well young, adult, male For example: basketball player who presents acutely with a painful, ○ For timing: 2 days becomes acute, more than 3 swollen and unstable knee after twisting the involved weeks becomes chronic. extended knee over a fixed foot during a game. ○ For severity: mild, severe ○ For location: proximal, distal; axial, appendicular ○ Worsening — improving B. PATTERN-MATCHING THE PROBLEM REPRESENTATION WITH THE ILLNESS SCRIPT Comparing the contents of the various categories of a VI. CASE patient problem representation with the illness script A 20-year-old man presents to your office complaining of stored from memory right knee pain after playing basketball four days prior to consultation. During the game, as he came down after jumping for a rebound, he twisted his extended knee while PATTERN-MATCHING THE PROBLEM another player fell on the back of his calf. He remembers REPRESENTATION WITH THE ILLNESS SCRIPT hearing a pop in the area of the knee and severe pain upon standing, preventing him from playing in the remainder of the PROBLEM ILLNESS SCRIPT: game. While on the bench, he noticed that the pain REPRESENTATI ACL improved, but his knee remained swollen. He iced the knee ON and was able to put some weight on it later that day. Today, putting all of his weight on the knee makes it feel as if it will TIME Acute Acute buckle. EPIDEMIOLO Young adult Adult, athletic A. PROBLEM REPRESENTATION GY male, basketball activities with It is relatively easy to mind the case for information to player pivoting/cutting assign the chunks of data to the various categories required for problem representation. CLINICAL Pain, swelling, Pain, swelling, More concise without sacrificing key and discriminating MANIFESTATI instability anterior features ONS translational Uses proper medical terminologies and semantic instability of tibia qualifiers that allows us to cut down extraneous data relative to femur hence reducing the cognitive load. PATHO- Twisted Anterior shear force PROBLEM REPRESENTATION: ACL PHYSIOLOGY extended knee on proximal tibia on fixed foot with knee in Pathophysiology Twisted knee extension with or 1 (faults) without internal Physical Diagnosis - Mod 3 🏠 Basics of Clinical Reasoning 4 of 10 The use of trans, practice questions, and evals ratio must be used discreetly and social media/public exposure of the aforementioned shall be strictly prohibited.  B. DIAGNOSIS PRE-TEST PROBABILITY rotation moments Clinical decision-making is enhanced by evidence-based medicine VII. PRIORITIZATION OF DIAGNOSIS Tier system is developed by Meadow and Lucey to The extent of the match between the patient’s illness objectively translate the categorical probabilities into a script and the physician’s illness script determines numerical pre-test probability priority In the sample case, the pre-test probability for an ACL DIAGNOSIS PRE-TEST PROBABILITY injury is pretty high, probably a tier 1 diagnosis. CATEGORI NUMERIC A. COMPARISON OF THE REASONING STRATEGIES DIAGN CAL AL CHARACTERISTICS OSIS PROBABIL PROBABI 1. PATTERN MATCHING/SCRIPT-BASED (or SYSTEM 1) ITY LITY REASONING STRATEGIES Patient’s illness script Very Likely >90% Experts are usually able to make a diagnosis faster (patient probability and with less effort using pattern matching because presentation) they have a lot of validated illness scripts residing in Tier 1 matches the organized packets inside their long term memory. physician’s illness Likely 67-90% If the pattern match is anything less than 90% or very script almost likely (i.e. likely, uncertain, or unlikely) downshift to a perfectly hypothetico-deductive strategy may be necessary to avoid cognitive biases and diagnostic error. Patient is missing key features of the disease or the Tier 2 disease does not Uncertain 34-66% explain prominent features of the patient presentation Unlikely 10-33% Single or Few Tier 3 Very Features Match 90% No test necessary Flow chart of Hypothetico-deductive (or System 2) Tier 1 No test or few tests Reasoning Strategies. Likely 67-90% needed Red Box: marks the intuitive steps taken in the hypothetico-deductive method Tier 1e Uncertain Use test with high (emerge but must be sensitivity (99%) to ncy — ruled out to rule out (“SnOUT”) Physical Diagnosis - Mod 3 🏠 Basics of Clinical Reasoning 5 of 10 The use of trans, practice questions, and evals ratio must be used discreetly and social media/public exposure of the aforementioned shall be strictly prohibited.  more avoid according to their pathophysiology, and predict fatal harm/damag possible clinical manifestations from there disease e For example, in a patient with an erythematous and s) blistered leg, you can consider the following conditions that may present in the same way: Tier 2 Uncertain 34-66% Use test with LR > 2 ○ Cellulitis ○ Necrotizing Fasciitis Tier 3 Use test with high ○ Thrombosis (not Unlikely 10-33% specificity (99%) to necessa rule in (“SpIN”) C. CATEGORY CHASE (CONGENITAL OR ACQUIRED) rily The category chase of various etiologies may also be Drop the Differential needed but this may also be the most taxing of all the life-thre Very