Medical diagnostic
In medicine, diagnosis or clinical propaedeutics is the procedure by which a disease, disease entity, syndrome, or any state of health or disease is identified (the "state of health& #34; is also diagnosed).
In terms of medical practice, diagnosis is a clinical judgment about the psychophysical state of a patient, whether animal or human, it represents a manifestation in response to a demand to determine such state, the only one that can indicate a diagnostic process or management of the patient is the Professional (clinical doctor who carries out the reception of the patient and performs emergency treatment according to the clinical cases of the patient).
To diagnose is to name the suffering of the patient; is to assign a "label".
Diagnostic process
Clinical diagnosis requires taking into account the two aspects of logic, that is, analysis and synthesis, using various tools such as anamnesis, clinical history, physical examination, and complementary examinations.
The medical diagnosis establishes from symptoms, signs and the findings of complementary explorations, what disease a person suffers from. Generally, a disease is not related in a biunivocal way with a symptom, that is, a symptom is not exclusive to a disease. Each symptom or finding in an examination presents a probability of occurrence in each disease.
Bayes' theorem helps diagnose a disease from the symptoms and other findings presented by the patient if the diseases are mutually exclusive, their prevalence is known and the frequency of appearance of each symptom in each disease. Depending on the prevalence of each disease in each population, the same set of symptoms or syndrome may produce a different diagnosis in each population, that is, each syndrome may be caused by a different disease in each population.
The calculation of the post-test probability is called Bayes' theorem and is a good model of how to proceed in the clinic, since the previous information modifies the probabilities of our diagnoses, making them more or less plausible our hypotheses. In other words, Bayes' theorem provides us with the mathematical tool to calculate how a diagnostic test changes the pretest probability to a new posttest value.
Example: A patient suspected of lupus erythematosus, with a pre-test probability of 20%, or 0.2. This pre-test probability depends on the prevalence of the disease in a given population, to which the value established by clinical judgment is added. If we request the DNA test, and the result is positive, the post-test probability that the patient suffers from lupus erythematosus is now 89.7%. On the other hand, if the test is negative, the post-test probability is reduced to 7.2%.
Diagnostic tools
- Anamnesis: It is the information provided by the patient during the clinical interview useful to analyze their clinical situation.
- Symptoms: These are the negative physical subjective experiences that the patient refers, collected by the doctor in anamnesis during the clinical interview, with a medical language, that is understandable to all doctors. For example, patients with the feeling of lack of air or uncomfortable and unpleasant perception of breathing (dyspnea), call it ahogo, anguish, fatigue, easy fatigue...
- Signs: These are the objective findings that the doctor detects by observing the patient, for example tachypnea to more than 30 breaths per minute. Medical semiotic or clinical semiology is the part of medicine that treats signs of disease from the point of view of diagnosis and prognosis.
- Physical or Semiotechnical Exploration: It consists of various maneuvers performed by the doctor about the patient, the main ones being inspection, palpation, percussion, smell and auscultation, with which more specific clinical signs are obtained.
All the symptoms reported in the anamnesis and the signs observed in the physical examination are noted in the patient's medical record.
Generally, the signs and symptoms define a syndrome that can be caused by several diseases. The doctor must formulate a hypothesis about the diseases that may be causing the syndrome and to verify the accuracy of the hypothesis, he requests additional examinations.
Types of complementary examinations
Complementary examinations confirm or rule out a specific disease, before starting treatment. Sometimes they do not offer any type of useful information, especially when they are requested without any type of criteria or there is no differential diagnosis.
- Laboratory tests: It consists of the analysis, usually biochemical of different body fluids, the most common being the blood.
- Diagnostic techniques by image: Like ultrasound, simple x-ray, CT, MRI, or PET.
- Endoscopic techniques:
- Fibrobroncoscopia
- Collonoscopia
- Gastroscopia
- Colposcopia
- Toracoscopia
- Laryngoscopy
- Biopsy.
Types of diagnosis
- Clinical or individual diagnosis. It is the total emitted from the contrast of all the mentioned and the personal conditions of the patient. All these factors qualitatively and quantitatively determine the clinical picture, so that this can be different even if the morbid entity is the same. This refers to the maxim “there are no diseases, but sick”.
- Diagnosis of certainty. It is the diagnosis confirmed through the interpretation and analysis of complementary methods.
- Differential diagnosis. Knowledge to which is raised after the comparative critical assessment of their most common manifestations with those of other diseases.
- Ethological diagnosis. It determines the causes of the disease; it is essential for the diagnosis of certainty of many diseases.
- Generic diagnosis. Determine whether or not the subject is ill. Prosecution problems may arise, as there is a possible simulation and neurosis and histeria, which constitute true diseases.
- Heroic diagnosis. When the diagnosis becomes an obsession, an absurd one, a kind of mental and professional imposition, that is, an extreme diagnosis in the sense of "self," and unnecessary.
- Injurious, anatomical or topographic diagnosis. It is the location and identification of the lesions in the different organs and tissues.
- Theological diagnosis. It is the specific determination of the disease.
- Patogenetic diagnosis. It consigns the mechanisms that produce the disease by the action of the causes and the organic reaction.
- A presumptive diagnosis. It is the one that the professional considers possible based on the data obtained in anamnesis and physical examination.
- Syndrome and functional diagnosis. Syndromes are set of signs and symptoms with a common development; e.g., Ictric syndrome (yellow skin, lit urine, faeces, etc.). Although sometimes no further progress can be made, it allows a partial pathogenetic diagnosis, but allows for functional treatment.
- Symptom diagnosis. It aims to identify the disease through symptoms. Generally an isolated symptom does not give a precise indication of the disease, since it can be the same as many of them.
Diagnostic issues
- Late diagnosis
- Absent diagnosis
- Generic diagnosis
- Inaccurate Diganognosis
- Fashion diagnosis
- Overdiagnosis
- Labeling
- Diagnostic Cascade
Characteristics of a diagnostic test
A diagnostic test is considered to be good when it offers positive results in patients and negative results in healthy patients, with the smallest possible range of error. Therefore, the conditions that must be required in a diagnostic test are mainly three:
- Validez: It is the degree to which a test measures what is supposed to measure, that is, the frequency with which the results obtained with this test can be confirmed by other more complex and rigorous. The parameters that measure the validity of a diagnostic test are sensitivity and specificity.
- Reproductivity: It is the ability of a test to deliver the same results when your application is repeated in similar circumstances. Reproductivity is determined by the biological variability of the observed fact, introduced by the observer himself and the derivative of the test itself.
- Security: It is the certainty that a test will predict the presence or absence of disease in a patient. Security is determined by the predictive value of a positive or negative result, i.e., the probability that a positive test the patient is really sick.
Validity of a diagnostic test
The simplest case is that of a dichotomous test, which classifies each patient as healthy or sick, depending on whether the test result is positive or negative. Thus, a positive result normally corresponds to the presence of the studied disease and a negative result to its absence.
In general, we tend to work with a heterogeneous population of patients, so that the data obtained allows them to be classified into four groups, which are usually represented in a 2X2 table. In it, the result of the diagnostic test (in rows) is compared with the real state of the patients (in columns) or, failing that, the result of the reference or gold standard test that we are going to use The test result can be positive or negative, but these can be correct or incorrect, leading to four types of results: true positives, true negatives, false positives, and false negatives.
| Test result | Sick | Sano |
|---|---|---|
| Positive | Positive truths (PV) | Positive False (FP) |
| Negative | Negative (NF) | Negative truths (NV) |
What determines the validity of the test used will be the calculation of the sensitivity and specificity values.
- Sensitivity: indicates the likelihood that a patient is really ill being the result of the positive test. Therefore, it is the ability of a test to actually detect the presence of disease. Sensitivity can be estimated as the proportion of sick patients who obtained a positive outcome in the diagnostic test, i.e. the proportion of true positives, or of patients diagnosed, regarding the total number of patients in the study population. Then:
- Sensitivity=(VP)/(VP+FN)
- Specificity: is the probability that a patient is really healthy after obtaining a negative test result. It is the ability of a test to detect the absence of disease. Thus, specificity can be estimated as the proportion of healthy patients who obtained a negative result in the diagnostic test, i.e., the proportion of true negatives, or recognized as such, regarding the total number of healthy in the population. In this way:
- Specificity=(VN)/(VN+FP)
The ideal is to work with diagnostic tests with high sensitivity and specificity, exceeding at least 80% in both cases. However, this is not always possible. In general, a very sensitive test will be especially suitable in those cases in which failure to diagnose the disease could be fatal for the patients, or in diseases in which a false positive does not cause serious psychological or economic problems for the patient.
On the other hand, tests with a high specificity are necessary in serious diseases, but without available treatment that makes them curable, when there is great interest in knowing the absence of disease or when diagnosing a patient with a disease, being false positive, can lead to serious consequences, whether physical, psychological or economic.
Effectiveness of a diagnostic test
Both sensitivity and specificity provide information about the probability of obtaining a particular result (positive or negative) based on the true condition of the patient with respect to the disease. However, when a patient undergoes a test, the doctor lacks a priori information about his true diagnosis, and rather the question is posed in the opposite direction: in the face of a positive or negative result in the test, which is the probability that the patient is really sick or healthy? The parameters that provide this information (post-test probability) to the physician are the so-called predictive values.
The predictive values will depend on the prevalence of the disease in the study population. It will therefore be a value that cannot be extrapolated to different populations. There are two types of predictive value:
- Positive predictive value: indicates the likelihood that the patient suffers from the disease after obtaining a positive result in the test. Therefore, the positive predictive value (PPV) can be estimated as the proportion of true positives in respect of the total positive results obtained in the test, i.e. the actual number of patients in respect of all the results that indicate the presence of disease. So:
- VPP=(VP)/(VP+FP)
- Negative predictive value: is the likelihood that the patient does not suffer the disease after obtaining a negative result in the test. In this way, the negative predictive value (PNV) can be estimated as the proportion of true negatives regarding the total negative results obtained in the test, i.e. the actual number of healthy patients with respect to all results indicating absence of disease. Then:
- VPN=(VN)/(VN+FN)
Choosing a diagnostic test
To correctly choose between two or more diagnostic tests, statistical parameters can be used. The so-called "Operational Characteristic of the Receiver" or ROC curve. The ROC curve is a representation that compares the sensitivity of the test with the parameter (1-Specificity), thus assuming a global measurement independent of any established cut-off point.
The most commonly used indicator parameter is the "area under the curve" (AUC). It is an index whose value is between 0.5 and 1; being 1 the value that determines a perfect diagnosis, and 0.5 a test without diagnostic discriminatory capacity.