Cadmore media player playing audio Diagnostic Tests: Toshi A. Furukawa, MD, PhD, discusses diagnostic tests.
Audio Player
Accessibility
Using forms mode, you can utilize keyboard shortcuts to make usability more efficient.
Space bar will play and pause the audio. Pressing escape will close all displays and pressing it again will jump to the end of the player.
You can press A to open the Keyboard Shortcut Dialog which also provides an audio description of the player and all of its features.
Transcript
Result Count
of
Search
Segments
Result Count
of
Search
Name:
Diagnostic Tests: Toshi A. Furukawa, MD, PhD, discusses diagnostic tests.
Description:
Diagnostic Tests: Toshi A. Furukawa, MD, PhD, discusses diagnostic tests.
Thumbnail URL:
https://cadmoremediastorage.blob.core.windows.net/72da52ca-acd3-44e6-bf10-4e6f7dc16884/thumbnails/72da52ca-acd3-44e6-bf10-4e6f7dc16884.jpg?sv=2019-02-02&sr=c&sig=1BRO9y4L0Xv7qkx5U2cfThnX4dwwXr4fVRIEvOM2Iok%3D&st=2025-01-03T06%3A31%3A33Z&se=2025-01-03T10%3A36%3A33Z&sp=r
Duration:
T00H16M10S
Embed URL:
https://stream.cadmore.media/player/72da52ca-acd3-44e6-bf10-4e6f7dc16884
Content URL:
https://cadmoreoriginalmedia.blob.core.windows.net/72da52ca-acd3-44e6-bf10-4e6f7dc16884/6830499.mp3?sv=2019-02-02&sr=c&sig=VEmdVfy7JS%2Bs%2Bo%2F6mXS5CVPVfncTWshcOWnnyhf3%2F2U%3D&st=2025-01-03T06%3A31%3A33Z&se=2025-01-03T08%3A36%3A33Z&sp=r
Upload Date:
2022-02-28T00:00:00.0000000
Transcript:
Language: EN.
Segment:0 .
>> I'm Joan Stephenson, Editor of JAMA's Medical News and Perspectives section. Today I have the pleasure of speaking with Dr. Toshi Furukawa about an important subject for clinicians, diagnostic tests. This is a topic discussed in Chapter 16 of Users' Guides to the Medical Literature, a chapter coauthored by Dr. Furukawa. Welcome back to the podcast, Dr. Furukawa, and please introduce yourself to our listeners. >> Hi. My name is Toshi Furukawa. I am professor of Behavioral Medicine and Clinical Epidemiology at Kyoto University School of Public Health in Japan.
My background is psychiatry, and my current activities mainly consist in teaching and research in clinical epidemiology, mental health and cognitive behavior therapy. >> Dr. Furukawa, what is a diagnostic test, and what is this chapter on diagnostic tests about? >> That's a good question to start with. There are studies about diagnoses that is about disease entities. In this chapter on diagnostic tests we presuppose such a disease entity.
In other words, we presuppose that there is a gold standard diagnosis, or at least something that we can consider to be a gold standard, and we will examine how well a diagnostic test can detect this reference standard diagnosis in the patient you see. A diagnostic test in this context can be anything from an item in the patient's history, any physical examination, a lab test or an x-ray, and so on.
>> When reading a study that reports on the accuracy of a diagnostic test, what steps should clinicians take to interpret the study results? >> As with all types of clinical questions we may have, the Users' Guide is divided into three parts; part one to ask if the study results are valid, part two to ask what the study results are, and part three to ask how you can apply the results to your patient.
There are four validity questions in part one to ask if the study results are valid. First, the participating patients present a diagnostic dilemma. Second, did investigators compare the test to an appropriate independent reference standard. Third, were those interpreting the test and the reference standard blind to the other results. Fourth, did investigators perform the same reference standard to all patients regardless of the result of the test under investigation.
These four questions can be grouped into two. The first group consists of the first question only and concerns the nature of the participants of the study. It is easy to see that, unless the participating patients present a diagnostic dilemma. In other words, unless the study is done with patients for whom you are at first unsure of the true diagnoses, the results are unlikely to be valid. The second group of validity questions consists of the second, third and fourth checkpoints, and they all concern the reference standard diagnosis, whether there is any appropriate reference standard diagnosis, and if the comparison between the diagnostic tests in question and the reference diagnosis is fair.
>> How might design-related bias present in a study of diagnostic tests? >> Well, that is a bit to explain in a podcast without any visual aid, but let me try. The end results of a diagnostics test study can be represented by a two-by-two table with the test results either positive or negative on the left, and the true reference diagnosis either present or absent on the top. In another podcast in this JAMAevidence series, Dr. Romanyaska [phonetic] explained that a two-by-two table is all we need in understanding the dichotomous results of a clinical trial, and so it is essentially with a diagnostic test study, all four numbers in this two-by-two table must be accurate.
Let's now look at the first validity question. Did participating patients present a diagnostic dilemma? In many studies of a diagnostic test patients with a particular diagnosis are compared with so-called healthy controls. You must then ask yourself did participating patients present a diagnostic dilemma, and your answer is no. Right. Such a study is an invalid diagnostic test study and will not serve you in using the test with your patients.
Why so is easy to see in the two-by-two table. In such a case control design typical patients with a disease and typically healthy subjects without a disease are represented. However, people with a disease but without typical manifestations, for example, those at early stage of a developing disease and people without that particular disease, but probably with a related disease and, hence, with confusing manifestations, are underrepresented.
In the two-by-two table, the cell for the disease present and test positive people, and the cell for the disease absent and the test negative people will be bigger. While the cell for the disease present, but test negative people, and the cell for the disease absent, but test positive people will be smaller. The four numbers in the two-by-two table and any calculations thereof, will then be unlikely to be reflecting the truth.
>> What is verification bias and in what way can it potentially influence study results? >> The so-called verification bias, also known as workup bias, concerns the way the reference standard diagnosis is arrived at. The second group of validity checkpoints in part one, which I referred to, says that we must have a reference standard diagnosis and that diagnosis must be done independently from the test in question.
If the people making the definitive diagnosis knew the results of the initial test, knowingly or unknowingly, they would classify more people in the cell representing the test positive and the disease present people, or in the cell representing the test negative and disease absent people. The two-by-two table would then be inaccurate. Likewise, if the people do not perform the definitive diagnosis to some of the test negative people, the cells representing such people will be smaller and the two-by-two table would be incomplete.
In any of these circumstances, again, the four numbers in the two-by-two table and any calculations thereof will be unlikely to be reflecting the truth. >> What role do likelihood ratios play in interpreting diagnostic test results? >> If you can answer yes, or almost yes, to all the four validity questions that we have just examined, you can now be confident of the performance of the diagnostic test in question, vis-a-vis the reference diagnosis, and we can move on to part two of the Users' Guide looking at what the results are.
At this point you may wish to listen to another podcast, "The Process of Diagnosis" by Dr. Gordon Guyatt, if you have not done so already. To summarize the points very, very briefly, the diagnostic process is how you move from a pre-test probability through the test result to a post-test probability for a range of differential diagnoses for your patient until you reach below the test threshold or above the treatment threshold.
The study of differential diagnosis will give you the test probability, and the study of a diagnostic test will provide information necessary to move from a pre-test to a post-test probability. The most convenient summary number to move from a pre-test probability to a post-test probability is a so-called likelihood ratio. I would like to refrain from how to calculate likelihood ratios from a two-by-two table here, so as not to sound any more technical and cumbersome, but let me show you an example.
This JAMAevidence home page has prepared very convenient calculators to save you the hassles of calculations. I suppose you are connected to the internet while listening to me. You can leave my podcast on while browsing other pages. Please click and go back to JAMAevidence [inaudible]. Have you done that? Then click Calculators tab at the top of the page.
Then you will find several convenient calculators for therapy, harm, and diagnosis. Please click on Diagnosis Two-By-Two Table Calculator Giving Cell Values under Diagnosis. Can you now see a two-by-two table toward the right of the screen? You enter the four numbers and then LR plus values, positive likelihood ratio, and LR minus, negative likelihood ratio, are calculated automatically and instantaneously.
LR plus is the likelihood ratio you use when the test result is positive. LR minus is the one you use when the test result is negative. How do I use it? Well, please click on Likelihood Ratio Nomogram. If you're on that page, let's try together. Let's imagine you examine an adult patient with abdominal pain.
You suspect appendicitis and perform a varietal test. LR plus for this test is 3.1, and LR minus is 0.26. Let's say your pre-test probability for a particular patient is high at 50% and his test was positive. Then you enter 50% in the left red cell of the nomogram, and 3.1, which is LR plus, in the middle blue cell in the nomogram, and you can get the post-test probability of 76% in the right green cell.
Or if the test was negative, you enter LR minus 0.26 in the middle blue cell, and the post-test probability now goes down to 21%. Or if your pre-test probability of another patient is rather low at 10%, then the post-test probability, if a varietal test is positive, will be 26%, and if negative, it will be 3%.
You can play around with several different possibilities later. In summary, the likelihood ratio is a summary index of a two-by-two table obtained from a study of a diagnostic test and it converts the pre-test probability into a post-test probability. >> After reviewing a study on a particular diagnostic test how can clinicians determine if the study results are applicable to patients in their clinical setting?
>> That's the crucial question. Test properties may change with a different mix of disease severity or with a different distribution of competing conditions. If you practice in a setting similar to that of the study, and if the patient under consideration meets all the study eligibility criteria, you can be confident that the results are applicable. If not, you must make a judgment. As with therapeutic interventions, you should ask whether there are compelling reasons why the results should not be applicable to your patients.
Also, if you can find an overview that summarizes the results of a number of studies, it can help resolve this issue of generalized ability. >> Is there anything else you would like to tell our listeners about diagnostic tests? >> Yes. In this podcast I have summarized a diagnostic test result in a two-by-two table and gave the explanations accordingly for the sake of explanation. However, in the real world the diagnostic test results may not come in a dichotomous, positive/negative format, but rather, in a positive, inconclusive or unclear negative trichotomy, or even more often, on the continuous scale ranging from, say, zero to several hundred.
These results can also be accommodated in the likelihood ratio approach if we use the so-called multilevel likelihood ratios, or stratum specific likelihood ratios. Sound complicated or daunting? In fact, they are very intuitive. I hope you will read through Chapter 16 of the Users' Guide to learn more. One more piece of information. The JAMAevidence home page has a whole series of very valuable information regarding rational clinical examinations.
There they provide the likelihood ratios for many clinical examinations, and also in many cases the pre-test probabilities of their target conditions. After listening to this podcast, I hope you can now make better and more use of these pages. Thank you for listening through this rather long podcast. >> Thank you, Dr. Furukawa, for this helpful look at diagnostic tests. Additional information about this topic is available in Chapter 16 of Users' Guides to the Medical Literature.
This has been Joan Stephenson of JAMA talking with Dr. Toshi Furukawa about diagnostic tests for JAMAevidence.
Segment:0 .
>> I'm Joan Stephenson, Editor of JAMA's Medical News and Perspectives section. Today I have the pleasure of speaking with Dr. Toshi Furukawa about an important subject for clinicians, diagnostic tests. This is a topic discussed in Chapter 16 of Users' Guides to the Medical Literature, a chapter coauthored by Dr. Furukawa. Welcome back to the podcast, Dr. Furukawa, and please introduce yourself to our listeners. >> Hi. My name is Toshi Furukawa. I am professor of Behavioral Medicine and Clinical Epidemiology at Kyoto University School of Public Health in Japan.
My background is psychiatry, and my current activities mainly consist in teaching and research in clinical epidemiology, mental health and cognitive behavior therapy. >> Dr. Furukawa, what is a diagnostic test, and what is this chapter on diagnostic tests about? >> That's a good question to start with. There are studies about diagnoses that is about disease entities. In this chapter on diagnostic tests we presuppose such a disease entity.
In other words, we presuppose that there is a gold standard diagnosis, or at least something that we can consider to be a gold standard, and we will examine how well a diagnostic test can detect this reference standard diagnosis in the patient you see. A diagnostic test in this context can be anything from an item in the patient's history, any physical examination, a lab test or an x-ray, and so on.
>> When reading a study that reports on the accuracy of a diagnostic test, what steps should clinicians take to interpret the study results? >> As with all types of clinical questions we may have, the Users' Guide is divided into three parts; part one to ask if the study results are valid, part two to ask what the study results are, and part three to ask how you can apply the results to your patient.
There are four validity questions in part one to ask if the study results are valid. First, the participating patients present a diagnostic dilemma. Second, did investigators compare the test to an appropriate independent reference standard. Third, were those interpreting the test and the reference standard blind to the other results. Fourth, did investigators perform the same reference standard to all patients regardless of the result of the test under investigation.
These four questions can be grouped into two. The first group consists of the first question only and concerns the nature of the participants of the study. It is easy to see that, unless the participating patients present a diagnostic dilemma. In other words, unless the study is done with patients for whom you are at first unsure of the true diagnoses, the results are unlikely to be valid. The second group of validity questions consists of the second, third and fourth checkpoints, and they all concern the reference standard diagnosis, whether there is any appropriate reference standard diagnosis, and if the comparison between the diagnostic tests in question and the reference diagnosis is fair.
>> How might design-related bias present in a study of diagnostic tests? >> Well, that is a bit to explain in a podcast without any visual aid, but let me try. The end results of a diagnostics test study can be represented by a two-by-two table with the test results either positive or negative on the left, and the true reference diagnosis either present or absent on the top. In another podcast in this JAMAevidence series, Dr. Romanyaska [phonetic] explained that a two-by-two table is all we need in understanding the dichotomous results of a clinical trial, and so it is essentially with a diagnostic test study, all four numbers in this two-by-two table must be accurate.
Let's now look at the first validity question. Did participating patients present a diagnostic dilemma? In many studies of a diagnostic test patients with a particular diagnosis are compared with so-called healthy controls. You must then ask yourself did participating patients present a diagnostic dilemma, and your answer is no. Right. Such a study is an invalid diagnostic test study and will not serve you in using the test with your patients.
Why so is easy to see in the two-by-two table. In such a case control design typical patients with a disease and typically healthy subjects without a disease are represented. However, people with a disease but without typical manifestations, for example, those at early stage of a developing disease and people without that particular disease, but probably with a related disease and, hence, with confusing manifestations, are underrepresented.
In the two-by-two table, the cell for the disease present and test positive people, and the cell for the disease absent and the test negative people will be bigger. While the cell for the disease present, but test negative people, and the cell for the disease absent, but test positive people will be smaller. The four numbers in the two-by-two table and any calculations thereof, will then be unlikely to be reflecting the truth.
>> What is verification bias and in what way can it potentially influence study results? >> The so-called verification bias, also known as workup bias, concerns the way the reference standard diagnosis is arrived at. The second group of validity checkpoints in part one, which I referred to, says that we must have a reference standard diagnosis and that diagnosis must be done independently from the test in question.
If the people making the definitive diagnosis knew the results of the initial test, knowingly or unknowingly, they would classify more people in the cell representing the test positive and the disease present people, or in the cell representing the test negative and disease absent people. The two-by-two table would then be inaccurate. Likewise, if the people do not perform the definitive diagnosis to some of the test negative people, the cells representing such people will be smaller and the two-by-two table would be incomplete.
In any of these circumstances, again, the four numbers in the two-by-two table and any calculations thereof will be unlikely to be reflecting the truth. >> What role do likelihood ratios play in interpreting diagnostic test results? >> If you can answer yes, or almost yes, to all the four validity questions that we have just examined, you can now be confident of the performance of the diagnostic test in question, vis-a-vis the reference diagnosis, and we can move on to part two of the Users' Guide looking at what the results are.
At this point you may wish to listen to another podcast, "The Process of Diagnosis" by Dr. Gordon Guyatt, if you have not done so already. To summarize the points very, very briefly, the diagnostic process is how you move from a pre-test probability through the test result to a post-test probability for a range of differential diagnoses for your patient until you reach below the test threshold or above the treatment threshold.
The study of differential diagnosis will give you the test probability, and the study of a diagnostic test will provide information necessary to move from a pre-test to a post-test probability. The most convenient summary number to move from a pre-test probability to a post-test probability is a so-called likelihood ratio. I would like to refrain from how to calculate likelihood ratios from a two-by-two table here, so as not to sound any more technical and cumbersome, but let me show you an example.
This JAMAevidence home page has prepared very convenient calculators to save you the hassles of calculations. I suppose you are connected to the internet while listening to me. You can leave my podcast on while browsing other pages. Please click and go back to JAMAevidence [inaudible]. Have you done that? Then click Calculators tab at the top of the page.
Then you will find several convenient calculators for therapy, harm, and diagnosis. Please click on Diagnosis Two-By-Two Table Calculator Giving Cell Values under Diagnosis. Can you now see a two-by-two table toward the right of the screen? You enter the four numbers and then LR plus values, positive likelihood ratio, and LR minus, negative likelihood ratio, are calculated automatically and instantaneously.
LR plus is the likelihood ratio you use when the test result is positive. LR minus is the one you use when the test result is negative. How do I use it? Well, please click on Likelihood Ratio Nomogram. If you're on that page, let's try together. Let's imagine you examine an adult patient with abdominal pain.
You suspect appendicitis and perform a varietal test. LR plus for this test is 3.1, and LR minus is 0.26. Let's say your pre-test probability for a particular patient is high at 50% and his test was positive. Then you enter 50% in the left red cell of the nomogram, and 3.1, which is LR plus, in the middle blue cell in the nomogram, and you can get the post-test probability of 76% in the right green cell.
Or if the test was negative, you enter LR minus 0.26 in the middle blue cell, and the post-test probability now goes down to 21%. Or if your pre-test probability of another patient is rather low at 10%, then the post-test probability, if a varietal test is positive, will be 26%, and if negative, it will be 3%.
You can play around with several different possibilities later. In summary, the likelihood ratio is a summary index of a two-by-two table obtained from a study of a diagnostic test and it converts the pre-test probability into a post-test probability. >> After reviewing a study on a particular diagnostic test how can clinicians determine if the study results are applicable to patients in their clinical setting?
>> That's the crucial question. Test properties may change with a different mix of disease severity or with a different distribution of competing conditions. If you practice in a setting similar to that of the study, and if the patient under consideration meets all the study eligibility criteria, you can be confident that the results are applicable. If not, you must make a judgment. As with therapeutic interventions, you should ask whether there are compelling reasons why the results should not be applicable to your patients.
Also, if you can find an overview that summarizes the results of a number of studies, it can help resolve this issue of generalized ability. >> Is there anything else you would like to tell our listeners about diagnostic tests? >> Yes. In this podcast I have summarized a diagnostic test result in a two-by-two table and gave the explanations accordingly for the sake of explanation. However, in the real world the diagnostic test results may not come in a dichotomous, positive/negative format, but rather, in a positive, inconclusive or unclear negative trichotomy, or even more often, on the continuous scale ranging from, say, zero to several hundred.
These results can also be accommodated in the likelihood ratio approach if we use the so-called multilevel likelihood ratios, or stratum specific likelihood ratios. Sound complicated or daunting? In fact, they are very intuitive. I hope you will read through Chapter 16 of the Users' Guide to learn more. One more piece of information. The JAMAevidence home page has a whole series of very valuable information regarding rational clinical examinations.
There they provide the likelihood ratios for many clinical examinations, and also in many cases the pre-test probabilities of their target conditions. After listening to this podcast, I hope you can now make better and more use of these pages. Thank you for listening through this rather long podcast. >> Thank you, Dr. Furukawa, for this helpful look at diagnostic tests. Additional information about this topic is available in Chapter 16 of Users' Guides to the Medical Literature.
This has been Joan Stephenson of JAMA talking with Dr. Toshi Furukawa about diagnostic tests for JAMAevidence.
Form Title Download
Media
Transcript
Form Title Bookmark
Entire Media
Segment(s)
Custom Clip
Start At:
End At:
Form Title Share
Entire Media
Segment(s)
Custom Clip
Start At:
End At:
Form Title Accessibility
The cadmore media player is a professional video and audio player experience designed to deliver media as well as relevant transcript, segmentation, and metadata. On the screen is a traditional html audio player with buttons such as play, pause, forward, captioning, language and play speed selection and more. Adjacent to the audio player are generally tabs, one of which shows a scrolling transcript of the audio and the other shows any segmentation of the media, sometimes referred to as chapters. At the top of the audio player is a menu bar containing an Explore button and Tools button. The Explore button allows you to dig deeper into the audio by opening up an explore dialog which contains an about tab which contains rich metadata about the media, a segments tab which contains further information about the audio, and a related tab which contains links to other media related to this media. The tools button opens a menu with options for sharing the media on social media, creating bookmarking links, creating embed codes for your website, generating citations, and more.
Keyboard Shortcuts
Keyboard shortcuts only work while you are in forms mode. By entering forms mode, you will be able to quickly access portions of the media player with single key commands.
Spacebar: Play/Pause Toggle
E: Explore Menu,
O: Tools Menu
T: Transcript Tab,
S: Segments Tab
F: Forward 15,
R: Rewind 15
L: Languages,
P: Playspeed
M: Mute Toggle,
V: Volume Bar
N: Video Only Toggle,
B: Full Screen Toggle
C: Captions Toggle
Escape: Closes dialogs and menus
Z: Next Transcript Chapter
X: Previous Transcript Chapter
Q: Next Transcript Paragraph
W: Previous Transcript Paragraph
Form Title Embed Media
Entire Media
Segment(s)
Custom Clip
Start At:
End At:
Embed Size
-
You Size - Responsive
-
We Size - Responsive
-
Picture Overlay Popup
-
400 x 225
-
512 x 288
-
400 x 700
-
768 x 1024
-
1024 x 512
-
1280 x 608
Form Title Cite
No citation provided.
Info
Segments
Related
Thumbnails
Accessibility
Using forms mode, you can utilize keyboard shortcuts to make usability more efficient.
Space bar will play and pause the media.
You can press A to open the Accessibility Dialog.
This describes many other keyboard shortcuts for media controls.
Exit Clip Mode
The video is currently set-up to show just a clip from 0:00 to video end. You have requested to see a portion of the media outside of the clip.
Click OK to switch to the full media or click if you would like to stay inside the clip. You can always go back to the clip with Tools and Clip Mode