Evidence forms the basis of modern medicine. Clinical research provides us with this evidence, guiding health professionals towards solutions to problems that they face in daily practice. Transferring existing problems in medical practice to a research setting is a challenging process that requires careful consideration. The practice of clinical epidemiology aims to address this through the application of established approaches for research in human populations, while at all times focussing on the problem at hand from a clinical perspective. This course teaches the principles and practice of clinical epidemiology, drawing on real problems faced by medical professionals and elaborating on existing examples of clinical research. Medical researchers will lean how to translate real clinical problems into tangible research questions for investigation, gaining insight into some of the most important considerations when designing an epidemiological study along the way. Core concepts will be introduced along four key themes: diagnosis, prognosis, treatment and etiology. Followers of this course will develop their understanding of the topics addressed through lectures from experts, peer interaction and review assignments.
An Introduction to Diagnostic Research
Loading...
來自 Utrecht University 的課程
Clinical Epidemiology
134 個評分
從本節課中
Diagnostic Research
This week we begin at the start of the DEPTh model: Diagnosis and diagnostic research. Through a series of lectures, you will come across familiar challenges in the area of diagnosis and discover how these problems can be addressed by clinical research.
與講師見面
Diederick E. GrobbeeProfessor
Clinical Epidemiology
Arno W. HoesProfessor
Clinical Epidemiology
Daily clinical practice is filled with uncertainty.
From the moment a patient walks through the office door,
clinicians are faced with a number of questions.
The most fundamental of which being, what's wrong with this patient?
While every clinician will remember unusual cases where reaching a diagnosis
proved difficult.
Common clinical problems can also present a real diagnostic challenge.
A woman enters your primary care clinic with her baby daughter of six months.
Over the last couple of weeks, she's developed a rash on her arms and
seems to have been suffering from belly cramps now and again.
Her mother suspects it might have started because she recently introduced cow milk
into her daughter's diet.
You're faced with the challenge of determining whether or
not your patient has a cow milk allergy.
You start by taking a detailed history and doing a physical examination.
You find a mild rash on both arms and the belly and no other abnormalities.
Based on the information you have so far, the diagnosis is far from clear.
You decide to do further tests in order to confidently reach a diagnosis.
One option would be to refer your patient to an allergy specialist for
an oral food dechallenge rechallenge test which may give you conclusive evidence.
Alternatively you could decide to do this test yourself.
During such a test, cow milk is first removed from the baby's diet and
then carefully re-introduced while closely monitoring the symptoms.
Anyway, this kind of testing requires weeks and it's quite burdening.
So you draw the not use it unless absolutely necessary.
On the other hand,
you don't want to diagnose the baby with having cow milk allergy and consequently
adapt to a diet unless you're confident enough about the presence of the disease.
It then occurs to you that you recently saw some research about potential block
biomarkers that could be used to identify cow milk allergy in children.
But it's not clear whether these markers will be accurate enough for
you to reach a firm conclusion that back the presence or absence of the disease.
Uncertainty now lies with the test or
combination of tests that you should use in this situation.
You're in a position where you need to balance a need for
diagnostic accuracy with considerations for your patient.
And what you really need is clearer scientific evidence about, for
example, the diagnostic value of symptoms and signs
of these new blood biomarkers, and of the dechallenge rechallenge test.
This will help you choose the most appropriate diagnostic work up.
We'll come back to this example later.
It's often the case that you have an idea of what might be wrong with your patient,
but you need a means of improving the chance of making a correct diagnosis.
It's at this point that you might consider asking further question to your patient or
running additional test to confirm your suspicions.
First, there's the need for accuracy in your predictions.
Another important consideration, is the invasiveness or
risks involved with conducting certain tests.
Sometimes a test is extremely accurate but if it'll be a large burden for
the patient it might be best to use it as a last resort, for use only when other
tests failed to provide enough information to confirm or exclude a diagnosis.
Costs and availability are also important factors for
consideration for both clinicians and patients.
As with invasive tests, expensive or time consuming tests may be a serious burden.
Although the ultimate aim is to be 100% sure about the presence or absence of
a specific diagnosis, it's often difficult to reach a conclusion with such certainty.
Consequently, the diagnostic process in clinical practice deals with uncertainty.
We're dealing with a chance or probability that our patient has a certain disease.
Probabilities are an implicit part of diagnosis and
practice and clinicians apply diagnostic tests to make sure
that probability is high enough to decide that the diagnosis is present.
Alternatively, diagnostic tests can be used to reduce that probability to
such an extent that this diagnosis can be excluded.
What we, therefore, need is evidence,
not only to guide us towards the best combination of tests for our needs but
also, evidence based guidance of how to convert the results of these tests into
a probability of disease presence in our patient.
We can collect this kind of evidence by conducting clinical research,
specifically, diagnostic research.
Ultimately, diagnostic research aims to guide clinicians
towards the most efficient diagnostic workup by creating tools
known as prediction rules or risk scores, that can be used to translate patient and
test information into probabilities of disease presence.
This will provide clinicians with evidence of how accurate their diagnosis is likely
to be and what the chances of them having made an error.
As with all branches of clinical epidemiology, diagnostic research involves
collecting information about a large group of patients as a part of a clinical study
which is then used to form evidence that can be used to help clinicians and
ultimately patients in the future.
And as we discussed last week, the first step in any kind of
clinical research should be to clearly identify the clinical problem.
And then translate this into a workable research question.
Let's go back to our patient suspecting of having a cow milk allergy.
Our problem is that we're not sure whether our patient actually has a cow
milk allergy.
And we'd like to know whether a set of less burdening tests including symptoms,
signs, and blood tests could be used
instead of performing the food challenge rechallenge test.
In this case it's especially important that our diagnostic prediction is accurate
as we do not want to risk making a misdiagnosis, which could either result
in future allergic reactions or lead to unnecessary dietary adaptations.
So for this problem, an appropriate research question could be,
which combination of less invasive tests
best predicts the presence of cow milk allergy in young children?
In the coming lectures,
we're going to discuss how we can conduct a clinical epidemiological study
to address this question as well as many of the key considerations
that need to be made when designing a diagnostic study such as this one.
