One of the most common strategies used for medical decision-making reflects the scientific method of formation of hypotheses followed by tests. Diagnostic hypotheses are confirmed or refuted on the basis of tests. A hypothesis set up a hypothesis includes the identification of the most important diagnostic possibilities (differential diagnosis) which could be responsible for the clinical problem of the patient. The patient’s chief complaint (eg. As chest pain) and basic demographic data (age, gender, ethnicity) are the starting points for the differential diagnosis, which is usually generated by pattern recognition. Each item in the list of possibilities is ideally associated with an estimated probability (see Table: Hypothetical differential diagnosis and pre-test and post-test probabilities for a 50-year-old patient who has chest pain, smokes cigarettes and diabetes and hypertension has) the correct diagnosis to be (pre-test probability. doctors often use vague terms such as “likely”, “unlikely” and “it can not be excluded” to describe the likelihood of disease. Both physicians and patients understand these nonspecific expressions often wrong;. if it is possible, an accurate statistical terminology should be used mathematical (statistical) calculations can help physicians in making clinical decisions, and even if there are no exact figures, they help clinical probabilities and logical relationships better to define and hyp narrow othetischen options. Probability and odds The probability that a disease (or event) occurs in a patient whose clinical situation is not specifically known, corresponds to the frequency of this disease (or this event) in the population. The probability is ranked on a scale of 0.0 (impossible) to 1.0 (safe) and is often expressed as a percentage of 0-100%. A disease which occurs in 2 out of 10 patients, has a probability of 2 to 10 (0.2, 20%). Rounding down to 0 at a very low probability (as is sometimes implied in clinical considerations happens) exclude the possibility of disease from completely, but may result in quantitative methods to wrong conclusions. The chance is the ratio of affected patients to non-affected patients is (d. E. The ratio of disease no disease). Thus, a disease which occurs in 2 out of 10 patients (probability of 2 to 10), an expected probability of 2 to 8 (0.25, often expressed as 1 to 4). Expected value (chance) (?) and probability (p) can be prepared according to the formula ? = p (1- p) or p = ?: (1 + ?) converted into each other. Clinical Calculator: rates based on probability of clinical calculator: probability based on quota checking of hypotheses The initial differential diagnosis is based on the main appeal and the demography is very large, as a rule, so that the clinician first examine the hypothetical possibilities during the history and physical examination, by asking questions or making specific studies to support the diagnosis or disprove. For example, increasing pain in the legs and swollen, pressure-sensitive legs during the examination in a patient with chest pain, the probability of pulmonary embolism. If the history and physical examination form a clear pattern, a diagnosis is made. Diagnostic tests are used when uncertainty after the history and physical examination are made, especially if the illness is serious or dangerous or costly treatments could be the result. The test results modify the probabilities of the various diagnoses (post-test probability). For example, shows Hypothetical differential diagnosis and pre-test and post-test probabilities for a 50-year-old patient who has chest pain, smokes cigarettes and diabetes and high blood pressure, like the additional findings about pain and swelling in the legs, and a normal ECG modify the diagnostic probabilities and radiograph in a patient: while the probability of pulmonary embolism increases the likelihood of acute coronary syndrome, of dissecting aneurysm and pneumothorax decreases. These changes in the probability lead to additional tests (in this example probably chest CT angiography), which further modify the post-test probability (see Table: Hypothetical differential diagnosis and pre-test and post-test probabilities in a 50-year-old patient who has chest pain, smokes cigarettes and diabetes and hypertension has), and in some cases confirm the diagnosis or disprove. It may seem intuitive that the sum of the probabilities of all diagnostic capabilities should be nearly 100%, and that a single diagnosis of a complex set of symptoms and signs can be derived. However, the application of this principle can be that the best explanation for a complex situation involves a single cause (often called Occam’s razor called) Doctors mislead. Rigid application of this principle discounts the possibility that a patient may have more than one active disease. For example, it can be assumed in a patient with dyspnea and known COPD that deterioration of COPD is present, but in fact he is suffering from a pulmonary embolism. Hypothetical differential diagnosis and pre-test and post-test probabilities for a 50-year-old patient who has chest pain, smokes cigarettes and diabetes and hypertension has diagnostic pre-test probability post-test probability I (Additional findings from pain in the legs, swelling, normal ECG and chest x-ray) post-test probability II (Other symptoms of a segmental defect on the breast in CT angiography and inconspicuous serum troponin I levels) Acute coronary syndrome 40% 28% 1% ST-segment elevation MI 20% <1% <1% chest wall pain 30% 20% <1% pulmonary embolism 5% 50% 98% divided thoracic aortic aneurysm <3% <1% <1% Spontaneous pneumothorax <2% <1% <1% probability estimates and the investigation threshold even if the diagnosis is uncertain, tests are not always useful. Tests should only be carried out if the results influence treatment. If the disease pretest probability is above a certain threshold, the treatment is guaranteed (treatment threshold) and a test is not displayed. Under the treatment threshold, a check is displayed only when a positive test result would lift the post-test probability over the treatment threshold. The cheapest pre-test probability at which this can be done depends on the test criteria and is called the inspection threshold. Tests for threshold values ??are discussed in more detail elsewhere. Probability estimates and the treatment threshold The disease probability at which treatment is indicated and no further testing is more justified is called a treatment threshold. The above-mentioned hypothetical example of a patient with chest pain converges on an almost certain diagnosis (98% probability). If the diagnosis of a disease is safe, the decision is a simple statement for a treatment, that treatment has a benefit (compared to no treatment, and taking into account the negative effects of the treatment). If the diagnosis has a certain degree of uncertainty, as is almost always the case, the decision to treat must also consider the benefits of treatment of a patient against the risk of treating a misdiagnosed or healthy people. The balance between benefit and risk also includes financial and medical consequences. The decision is then based on the balance between the likelihood of the disease and the size of the benefit and risk. This balance determines how the doctor attaches the treatment threshold. Tips and risks when some uncertainty exists regarding the diagnosis, decision to treat must take into account the benefits of the treatment of a patient against the risk of treating a misdiagnosed or healthy people. Based on this conception doctors accept possibly a high diagnostic uncertainty and initiate treatment if the benefit of treatment is very high and the risk is very low (such. as by the gift of a safe antibiotic for a patient with diabetes, may have a life-threatening infection.) - even if the probability of infection (for example, 30% is relatively low - variation of the threshold treatment with the risk of treatment).. However, if the risk of treatment is very high (such. as at a pneumonectomy for lung cancer possible), want doctors have a very safe diagnosis and arrange treatment only if the probability of cancer is very high, eg. B.> 95% (variation of the treatment threshold with the risk of treatment.). Note that the treatment threshold is not necessarily equal to the probability that the disease could be confirmed or ruled out. It is simply the point where not to treat the risk is greater than the risk of treatment. Varying the treatment threshold with the risk of treatment. Horizontal lines represent the post-test probability. Quantitatively, the treatment threshold (TT) can be the point define where the probability of the disease (p) Multiplies with the advantage of treating a person with a disease (B) equals the probability of no disease (1 – p) multiplied by the risk is the treatment of a person without the disease (R). Thus, (TT) is for the treatment threshold: = (1 – p) p × B × R Solving for p, this equation reduces to: p = R: (B + R) From this equation it can be seen that if B ( benefits) and R (risk) are the same, the treatment threshold 1: (1 + 1) = 0.5, which means that if the probability of the disease is> 50%, the doctor would treat, and when the probability <50 %, the doctor would not treat. For a clinical example, a patient may be considered with chest pain. How high clinical probability of acute myocardial infarction (MI) should be that thrombolysis therapy would have to be made if the only risk would be an early mortality? Suppose that it is the case that the mortality rate is 1% due to intracranial hemorrhage with thrombolysis, then R is with 1% lethality an erroneous treatment of a patient who has no MI. If the mortality is reduced in a patient with MI 3% when it is subjected to thrombolytic therapy, then B is 3%. Then the treatment threshold 1: (3 + 1) or 25%; then treatment should be started when the probability of an acute MI is> 25%. Alternatively, the equation can be rearranged to show that the treatment is the point at which the chance of a disease p (1 p) equal to the risk-benefit ratio (R: B). It is the same numerical result as in the example described above, with the TT equal to the chance of the benefit-risk ratio (1 to 3). A chance of 1 to 3 corresponds to the probability previously obtained of 25% (see probability and Opportunities). Limitations of quantitative decision methods Quantitative clinical decision appears to be accurate, but because many elements in the calculations are often missing (if they are known at all), this method can be used successfully only in cases with very well-known and researched clinical Paramtern.

- Home
- /
- Clinical Strategies Of Decision
- /
- Uncategorized
- /
- Clinical Strategies Of Decision