Class notes
From Clinical Trials To Big Data Research - full lecture notes.
- Course
- Institution
This document covers all lectures (except for the introduction) regarding the elective From Clinical Trials to big data research.
[Show more]
Seller
Follow
esrapolat85
Reviews received
Content preview
From clinical trials to big dataLecture 1 – introduction – 1-02
Lecture 2 – diagnostic and prognostic research 1-02
Prediction research → diagnostic and prognosis research
Challenges in clinical practice
- What is wrong with this patietnt?
o Diagnostic knowledge
- Why is this patient ill?
o Etiologic knowledge
- What will happen if I don’t intervene?
o Prognostic knowledge
- What will be the effects of an intervention?
o Therapeutic (influencing prognostic!) knowledge
- DEPTh model → Diagnostic Etiologic Prognostic Therapeutic model
Influenca H1N1 pandemic
→ het kost veel geld. Als er een te kort is aan de hoeveelheid vaccinaties, moet je kiezen aan wie je wel
een vaccinatie geeft en aan wie niet. Wat je dan moet doen, is een predictive chance opstellen:
RISK-PREDICTION role: the higher the number of points, the higher the chance that in the coming epidemic
you will die or be hospitalized.
How do you determine the amount of points for a specific disease? → mathematically based on the
regression model with an estimate → betas associated with the determinance → the higher the beta, the
higher the points.
,y-axis: probability of the disease, estimated. Low probability → no treat.
In between: continue diagnostic tests to change probability; disease uncertain
80% of complication → start treatment
Where the points A and B are depend on the prognosis of the disease; whether it can quickly lead to a
complication. E.g. corona can go very fast, so we have to aggresively treat.
It also depends on the benefits (how sure it helps), costs and side effects of the treatment. Cancer
treatments are expensive, so we have to be really sure that we need to start the treatment. We are not
going to treat if there is a probability of complication which is lower than 15%.
Benefits and side effects next available test.
Diagnostic research
Diagnosis in practice = basis medical care
- Directs (medical drug) treatment
- Indication of a patient’s prognosis
Scientific diagnostic research should serve practice
- Describe (follow) practice in study question, design and analysis
- Test should follow clinical practice (availability)
- Diagnostic outcome should be measured by a golden standard → definitely yes or definitely no, the
research we are interested in.
Making a diagnosis starts with patients having a clinical problem (sign and symptoms)
- Symptom/ sign → suspected of having particular disease
- Helps the medical doctor to rule out/ in a certain disease
o Young female with swollen, red leg → DVT?
o Women with palpable node in breast → breast cancer?
o Child with neck stifness → meningitis?
o Sore throat and fever, dec 2020 → corona?
Symptoms and signs are important in the diagnosis.
Diagnosis practice
- Young man with acute cough and fever → what disease could it be?
- Possible diseases → differential diagnosis
o Upper respiratory tract infection
o Bronchitis
o Pneumonia → needs direct treatment
o Pulmonary embolism
o Other? → we must be sure that it is not the diases.
- Patient with differential diagnoses (DD) and target diagnosis (TD)
- TD = one not to miss (most severe or probable)
, - In our case TD = pneumonia (or corona)
- Before any test is done, we would like to know the chance that they have pneuminia. The prior
probability =
o Pr(pneumonia)= Pr(TD)
▪ Often Pr(TD) too low to treat everybody and too high to send a patient home. We
want to be sure that it is not pneumomia.
- Aim: which results truly contribute to estimation diagnostic probability. We need to estimate the
probability of pneumonia better.
o The less the better → weighing predictive value vs. measurement burden and costs
We want to make sure that we go from A to B → that we aim to treat and give antibiotics.
How are we going to do this? (Go from A to B?)
→ diagnostic testing!
- 24 years of age, male sex, athlete
- Antecedents: upper respiratory tract infections
- Dyspnoea/ cough/ sweating
- Fever: 39.2 celsius
- Morning sputum (‘somewhat yellowish’) → indication of pneumonia
- ‘some’ crepitations/ rhales left
- Tachycardia: 112/ minute → too fast heartbeat.
→ you remain in doubt. What do you do next? → we still are not sure whether it is pneumonia or not.
You need to send him to the hospital to do some additional testing.
Additional diagnostic test → are they available?
- Lung function?
- CRP? → C-reactive protein; indicator for pneumonia when too high.
- Chest X-ray?
- Something else?
What do we do in research? We measure the same as the GP, and we use a reference test (pathology,
imaging, microbiology).
- Commonly hierarchical process
o Potential determinants:
o History (H) + physical examination (PE)
o Simple lab (blood CRP) /urine)
o Imaging
o Functional tests/ extensive lab
Test research
So, we are doing a research in 200 patients with a reference test 40 people who actually have pneumonia.
- Using the reference test 40 out of 200 patients with suspected pneumonia actually have
pneumonia
o Pneumonia (culture positive) n=40
o Culture negative n=160
Pr(TD)= Pr(pneumonia)=0,20 (40/200), the chance of the patients that will have pneumonia, we need to be
sure that the tests are as close as possible to the reference tests.
In test research, we want to evaluate whether an index test usefully approximates the reference test.
, Two types of validity that we measure → sensitivity and specificity from the reference tests.
Sensitivity is how many on the basis of this positive test are we going to detect among all the diseases →
a/k1.
Specificity is how many of the true negatives are also stated as negative by the test, which is d/k 2.
Test characteristics: positive predictive value: if we look at the positive tests, what is the chance that you
actually have pneumonia? → a/r1.
Negative predictive value: c/r2.
As sensitive and as specific possible is needed.
Study design for diagnostic research
Cross sectional → no follow-up.
Determinants and outcome measured at ‘same’ moment → at one period of time, there needs to be a
diagnostic research.
Prognostic research
Prognosis → basis for medical treatment. It determines how aggressive the treatment will be.
If I have a certain disease/ risk factor, what is my diagnosis?
Do I need (extra, more aggressive) treatment?
The benefits of buying summaries with Stuvia:
Guaranteed quality through customer reviews
Stuvia customers have reviewed more than 700,000 summaries. This how you know that you are buying the best documents.
Quick and easy check-out
You can quickly pay through credit card or Stuvia-credit for the summaries. There is no membership needed.
Focus on what matters
Your fellow students write the study notes themselves, which is why the documents are always reliable and up-to-date. This ensures you quickly get to the core!
Frequently asked questions
What do I get when I buy this document?
You get a PDF, available immediately after your purchase. The purchased document is accessible anytime, anywhere and indefinitely through your profile.
Satisfaction guarantee: how does it work?
Our satisfaction guarantee ensures that you always find a study document that suits you well. You fill out a form, and our customer service team takes care of the rest.
Who am I buying these notes from?
Stuvia is a marketplace, so you are not buying this document from us, but from seller esrapolat85. Stuvia facilitates payment to the seller.
Will I be stuck with a subscription?
No, you only buy these notes for $15.10. You're not tied to anything after your purchase.
Can Stuvia be trusted?
4.6 stars on Google & Trustpilot (+1000 reviews)
66579 documents were sold in the last 30 days
Founded in 2010, the go-to place to buy study notes for 14 years now