Simplifying data - bpums

Simplifying data - bpums

DIAGNOSIS Diagnosis A diagnostic test is ordinarily understood to mean a test Performed in laboratory ,but the principles discussed apply equally well to clinical information obtained from History, physical examination ,and imaging procedures. They also apply when a constellation of findings serves as a diagnostic test. 2 Simplifying data

clinicians generally reduce the data to simpler from to make them useful in practice. Example :heart murmurs More often. Complex data are reduced to a simple dichotomy(for example. Present/absent, abnormal/ normal, or diseased/well)

Exampel :blood pressure 3 the accuracy of a test result The doctors certnainty or uncertainty about a diagnosis has been expressed by using terms such as rule out or possible before a clinical diagnosis

A simple way of looking at the relationships between a tests resulst and true diagnosis is shown in next figure 4 The accuracy of a test result The doctors certnainty or uncertainty about a diagnosis has been expressed by using terms such as rule out or possible before a clinical diagnosis

5 the gold standard More often, one must turn to relatively elaborate, expensive, or risky tests to be certain whether the disease is present or absent

Sometimes the standard of accuracy is itself a relaltively simple and inexpensive test such as a throat culture for group A streptococus 6 For disease that are not-limited and ordinarily become overt in matter of few years after they are first suspected,the resulsts of follow up can serve as a gold standard such as cansrs and coronic diseases

Clinicians and patients prefer simpler tests to the rigorous gold standard 7 Lack of information on negative test The goal of all clinical studies

Most information about the value of a diagnostic test is obtained from clinical,and not reaserch Data on the number of true negative versus false negetives generated by a test tend to be much less complete in the medical literature than data collected about positive test resultes 8

Lack of objective standards for disease For some conditions,there are simply no hard and fast criteria for diagnosis : angina pectoris The validity of laboratory test is established by comparing its results to a clinical diagnosis based on a careful history of symptoms and a physical examination

9 Consequences of imperfect standards physician must choose as their standard of validity another test that admittedly is imperfect,but is considered the best avalible

If a new test is compaired with an old standard test,the new test may seem worse even thought it is actually better 10 Sensitivivity and specifity 11 12 Definition

Sensitivity is defined as the proportion of people with the disease who have a positive test for the disease.A sensitive test will rarely miss people with the disease Specificity is the proportion of people without the diseases

who have negative test. A specificity test will rarely misclassify people as having the disease when they do not.

13 Use of sensitive test when there is an important penalty for missing a disease such as TB, syphilis; hodgkindisease when a great many possibilities are being considered; to reduce the number of possibilities.

A highly sensitive test is most helpful to the clinician when the test result is negative. 14 Use of specific test Spesific test useful to confirm for rule out diagnosis that has been suggested by other data

Before patients are subjected to cancer chemotherapy In sum, a highly specific test is most helpful when the test result is positive 15 Trade_off between sensitivity and specificity

16 The ROC curve Another way to express the relationship between sensitivity and specificity for a given test is to construct a curve, called receiver operator characteristic(ROC) curve

The ROC curve shows how severe the trade-of between sesitivity and spesitivity is for a test and can be used to help decide where the best cutoff point should be ROC curves are particularly valuable ways of comparing alternative tests for the same diagnosis. The overall accuracy of a test can be described as the area under

the ROC curve 17 18 19 Establishing sensitivity and specificity At the crudest level, Sensitivity and specificity

may be inaccurately because an improper gold standard has be chosen . Two other issue related to the selection of diseased and nondiseased patients can profoundly affect the determination of sensitivity and specificity Spectrum/bias

20 Spectrum of patient Difficulties may arise when the patients used to describe the testproperties are somehow different from those to whom the test will be applied in clinical practice Example: CEA in diagnosing colorectal cancer

The sensitivity and specificity of test are independent of the prevalence of diseased(in theory) Several characteristics of patient,such as stage and severity of

disease may be related to the sensitivity and specificity of a test and to the prevalence(in practice) 21 22 Bias The sensitivity and specificity of a test should be established independently of the means by which the true diagnosis is established

All the biases discussed tend to increase the agreement between the test and the gold standard. That is, they tend to make the test seem more useful than it actually is 23 Chance Values for sensitivity and specificity are usually estimated from observations on relatively small samples of people of peaple

with and without the disease of interest Reported value for sensitivity and specifity shoud not be taken too literally if a small number of patients is studied 24 25 Definitions

The probability of disease,given the results of a test is called the predictive value of the test Positive predictive value is the probability of disease in a patient with a positive test Negative predictive value is the probability of not having the disease when the test result is negative

Predictive value : if my patient test result is positive,what are the chance that my patient does have the diseases? 26 Determinants of predictive value The predictive value of a test is not a property of the test alone

It is determined by the sensitivity and specificity of the test and the prevalence of disease in the population being tested 27 The more sensitive a test is, the better will be its negative predictive value (the more confident the clinician can be that

a negative test result rule out the disease being sought) The more specific the test is the better will be its positive predictive value 0 28

Positive results even for a very specific test,when applied to patients with a low likelhood of having the disease will be largely false positive.similarly,negative results Therefore as the prevalence of disease in a population approaches zero,the positive predictive value of a test also approaches zero 29

As prevalence approache 100% negative predictive value approaches zero 30 Estimating prevalence Because prevalence is such a powerful determinant of how useful a diagnostic test will be, clinicians must consider the

probability of diseases before ordering a test Although the resulting estimate of prevalence are not likely to be very precise,using them to be more accurate than implicit judgment 31 Increasing the prevalence of disease before testing

Diagnostic tests are most helpful when the presence of disease is neither very likely nor very unlikely There are several ways in which the probablity of a disease can be increased before using a diagnosis test 32

Referral process Referral to teaching hospital wards,clinics, and emergency departments increases the chance that significant disease will underlie patientcomplains Diagnostic evaluations may need to be adjusted to suit the specific situation

33 Selected demographic groups Physician can increase the yield of diagnostic tests by applying them to demographic group known to be at higher risk for a disease

a sickle cell test in african american 34 Specifics of the clinical sitution They are clearly the strongest influence on the order tests . Symptoms,signs,disease risk factors all raise or lower the

probability of finding a disease The value of applying diagnostic tests to persons more likely to have a particular illness is intuitively obvious to most doctors The less selective the approach,the lower the prevalence of the disease is likely to be and the lower will be the positive predictive value of the test

35 Implication for interpreting the medical literature Published description of diagnostic tests often in clude,in addition to sensitivity and specificity,some conclusion about predictive value

The data for these publications are often gathered in university teaching hospitals where the prevalence of serious disease is relatively high Statements about predictive value in medical literature may be misleading when the test is applied in less highly selected settings 36 Likelihood ratio

Likehood ratios are an alternative way of describing the performance of a diagnostic test that can be used to calculate the probability of diesease after a positive or negative test 37 odds

Probability used to express sensitivity, specifity and predictive valuertion It is the proportion of people in whom a particular characteristic Odd is the ratio of two probabilities 38

Definitions The likelihood ratio for a particular value of a diagnostic test is defined as the probability of that test result in people with the disease divided by probability of the result in people without disease If a test yields dichotomous results two types of likelihood ratio describe its ability to discriminate

between disease and no diseased people 39 It can accommodate the common clinical practice of putting more weight on extremely high test results than on borderline ones when estimating the probability that particular disease is present

It is particularly well suited for describing the overall odds of disease when a series of diagnostic tests is used 40 Use of likelihood ratios Pretest odds contain the same information as prior the same as sensitivity/specifity and post test odds the same as positive predictive value The main advantage of likelihood ratio is possible to go beyond the simple and

clumsy classification of a test result as either abnormal or normal 41 02/24/2020 42 02/24/2020 43

02/24/2020 44 Multiple tests Because clinicians commonly use imperfect diagnostic tests with less than 100% sensitivity and specificity ,a single test results in probability of disease that is neither vary high nor vary low

The physician is ordinarily bound to raise or lower the probability of disease substantially in such situations When multiple different test are performed and all are positive or all are negative ,the interpretation is straighforward 45

Multiple diagnostic tests can be applied in two basic ways Multiple diagnostic can be used in parallel testing and a positive result of any test is considered evidence for disease They can be done in serial testing with the decision

to the next test it based on the previous test and diagnostic process is stopped with a negative result 46 47 Parallel testing Multiple tests in parallel generally increase the sensitivity and negative predictive value above of each individual test By using the tests in parallel the net effect is a more sensitive diagnostic strategy

48 49 Clinical prediction rules 02/24/2020 50 Serial testing Physician most often use serial testing

strategies where rapid assessment of patient is not required When some of tests are expensive or risky It leads to less laboratory use than parallel testing Maximizes specificity and positive predictive value but lowers sensitivity and negative predictive value

51 Serial likelihood ratios As reach test done, the posttest odds of one became the pretest odds for the next in the end a new probability of disease is found that takes into account the information contributed by all the tests in the series 52


Recently Viewed Presentations


    PHMSA RegionalOffices-4 - Presenter. 2016-08-11 18:29:49--------------------------------------------Western Region Offices- Hazmat Office is based in Ontario ...
  • History of Evolution Theory - Quia

    History of Evolution Theory - Quia

    Constantia Arial Calibri Wingdings 2 Wingdings Flow 1_Flow 2_Flow 3_Flow History of Evolution Theory Evolution Darwin Influential Scientists James Hutton (1795) Charles Lyell (1833) Jean-Baptiste Lamarck (1809) Example Thomas Malthus (1798)
  • PowerPoint プレゼンテーション -

    PowerPoint プレゼンテーション -

    Charge (uC) Chargingvoltage(V) Threshold specified in SAE J1772. Question. Is there any specific rationale for the threshold in SAE J1772? Why is it considered to be safe?
  • Types of Caves Cynthia Sandeno National Cave and

    Types of Caves Cynthia Sandeno National Cave and

    By capturing bats in the summer and putting small transmitters on their backs, we have been able to track them to caves that we did not know existed. And, with the emergence of WNS, we have also discovered new caves...
  • Age of Early European Explorations & Conquests

    Age of Early European Explorations & Conquests

    Each new slide will start with a Roman Numeral and a Title. Through the slide I will use bulleted points. What is a bullet point? Most of the time the bullet point will be a word or short phrase that...
  • The Modern Period in British Literature - Winthrop

    The Modern Period in British Literature - Winthrop

    The Modern Period in British Literature ~1901 to ~1939 ... past It's hard to say what genres are typical The short story and the novel The critical essay The manifesto The imagist poem A kind of narrative poem Remember: "free...
  • Chapter title - California State University, Sacramento

    Chapter title - California State University, Sacramento

    The payback method focuses on the payback period, which is the length of time that it takes for a project to recoup its initial cost out of the cash receipts that it generates. When the net annual cash inflow is...
  • The Endocrine System

    The Endocrine System

    Describe the role of the endocrine glands in maintaining homeostasis. Name and describe the functions of the primary hormones secreted by each of the endocrine glands. Recognize, define, spell, and pronounce terms relating to the pathology and the diagnostic and...