Transcript Farmakoinformatika přednáška 3

Pharmacoinformatics
lecture 3
Mgr. Zdeněk Kučera, Ph.D.
kucera.xf.cz
Evidence Based Medicine - EBM
• Definition:
When good quality research evidence is used in
conjunction with professionals’ skills and users’/society’s
perferences to solve specific problems
“is the conscientious, explicit and judicious use of current
best evidence in making decisions about the care of
individual patients … integrating clinical expertise with
the best available external evidence from systematic
research”
Sackett et al (1996)
-----------Pharmacist – self-medication, consultations with patient
Influence factors
expectation
opinions
finances
evidence
(EBM – not all-powerful)
experience
politics
laws
time
Factors influencing the clinician´s decision
Why EBM?
• we are often asked for help! – we need to know, where
to find relevant information
• not possible to rely on more experinced colleague
• not good to rely on monographs
– By 1970, several studies had shown that thrombolysis (clot
dissolving) drugs reduced the number of deaths from a heart
attack.
– The clinical value of thrombolysis ... remains uncertain.”
Oxford Textbook of Medicine, 1987
•
•
•
•
postgraduate education – not systematic
increase of new information, commercial information
we have not time
etc...
Advantages EBM
• quick availability of new information
• teach the art of critical appraisal of the
literature
• systematic review – new methodology of
review
• objective guidelines
• assessment of new therapies
Criticism of the EBM
• EBM is „an academic exercise“ for medical students (3-4
questions/patient – do I have a time to search for evidence?)
• RCT – „greenhouse“ population – only because of ethics...?
• Governments, hospitals and HMO’s have used the jargon of EBM to
justify decisions, directives, or incentives that are seen by clinicians as
inappropriate.
• Misuse of EBM by pharmaceutical companies – „half-truths“
• treatment of the patient (co-morbidity, polypragmasia, aging, lactation,
gravidity, pediatrics)
• EBM is seen by some to promote “therapeutically nihilistic”
approaches to medical practice since critical appraisal of commonly
used therapies often concludes that they provide little or no benefit –
e.g. cough - Over the counter cough medicines for acute cough
cannot be recommended because there is no good evidence for their
effectiveness. BMJ 2002;324:329
• Guidelines – impersonal, schematic ...
• No evidence – no treatment???
• If I do not treat according to guidelines – see you in a court?
Evidence-based practice
5 STEPS:
• Formulate a good question
• Look for evidence
• Critically appraise what you find
• Act on the evidence
• Evaluate the results
Step one:
Formulating questions
1. Formulation of question
• A well formulated question allows you to
search in the medical database
The three/four part question
Who/what Intervention Comparison
(persons or
problem)
(treatment,
exposure,test
etc)
(comparison
intervention,
if relevant)
:
Outcome
(effect, result
diagnosis
etc)
What is a good question?
Who
Intervention Alternative Result
Infant
sleeping in
parents’ bed
acupuncture
Sleeping
alone
none
cot death
Media campaign
Information
leaflets
Prescription
at the
doctors
smoking
prevention
prevalence of
unwanted
pregnancy
Women
under
childbirth
Young
people
Girls
between
16-20
Free
contraception
from school
nurses
pain
1. Formulation of question - example
Component Question
Example
Patient or
Description of the group I have 55 old
problem
like my patient?
man, after
myocardial
infarction, feeling
„tired by life“...
Intervention The main intervention? antidepressants?
Comparison Alternative?
or not
intervention
Result
What I hope to achive? decrease in
mortality after
IM?
Pubmed – limits, MeSH, clinical queries
„Use of selective serotonin reuptake inhibitors in depressed patients who experience
an acute MI might reduce subsequent cardiovascular morbidity and mortality.“
Component Question
Example
Patient or
problem
Description of the group like
my patient?
Woman – back pain
Intervention
The main intervention?
ibuprofen
Comparison
intervention
Alternative?
diclofenac
Result
What I hope to achive?
what is better?
"Back Pain"[MeSH] AND "Diclofenac"[MeSH] AND "ibuprofen"[MeSH] AND
Randomized Controlled Trial[ptyp]
„Diclofenac-K 12.5 mg demonstrated superiority vs placebo on the primary
efficacy parameter and almost all secondary initial dose outcomes. With
respect to the initial dose, diclofenac-K 12.5 mg was also significantly superior
to ibuprofen 200 mg on SPID-3.“
Int J Clin Pharmacol Ther. 2003 Sep;41(9):375-85.
http://www.jr2.ox.ac.uk/bandolier/booth/painpag/Acutrev/An
algesics/Leagtab.html - Oxford league of acute pain Bandolier
Další odkaz – google.com - NNT ibuprofen diclofenac evidence based
http://www.aafp.org/afp/20050301/913.html - American Family Physician
What kind of question did you
formulate?
• How many people have a this problem ….?
• Why do some people get this problem..?
• How can we decide whether someone has this
problem…?
• What can we do to prevent or cure ..?
• What is the prognosis ……?
• How does it feel to ….?
• How can we understand …..?
Kinds of study design
SINGLE STUDIES
REVIEWS OF STUDIES
Systematic reviews
Meta-analysis
Hierarchy of evidences
- Systematic review
(meta-analysis)
----------------------------- Randomised clinical
trial
- Cohort study
- Case-control study
- Cross-sectional
study
- Case reports
Case reports
• Description of disease, treatment,
phenomenon in one selected patient
• Can be summarised into case-series
Czech Republic: http://www.zdn.cz/kazuistiky
Cross-sectional studies
Snap-shot in time’
Measure exposure and effect at the same time
time of study
conduction
population
random
selection
exposed with
disease
exposed
not exposed not exposed
without
with disease without
disease
disease
http://ucebnice.euromise.cz/index.php?conn=0&section=epidem&node=node1
Case-control studies
-Population are defined by those with disease (cases) and those without the
disease (control).
-The level of exposure to a health hazard(s) is measured in the two groups
and compared.
-What health hazard has caused the cases?
time
direction of observation
exposed
not exposed
cases – with
disease
population
exposed
controls –
without disease
not exposed
http://ucebnice.euromise.cz/index.php?conn=0&section=epidem&node=node1
Case control studies
Time
Inquiry direction
Exposed
Start with:
Nonexposed
Cases (people
with disease)
Exposed
Nonexposed
Controls
(people with
no disease)
Cohort studies
• The population of study is defined by
exposure to a health hazard
• They are followed-up over time to observe
incidence of disease in exposed and nonexposed
Cohort studies
Time and direction of
enquiry
Disease
No disease
Population
People
without
disease
Disease
Nonexposed
No disease
A cohort study
A cohort (or follow-up or
longitudinal) study is
when a group of people
exposed and a group of
persons unexposed to a
potential cause of
disease are followed-up
over time and the
incidence of the disease
in one group compared
with the incidence in the
other.
Randomised controlled trial
new treatment
group 1
Outcome
group 2
Outcome
population
control treatment
RCT – randomized clinical trial
Clinical - experimental study on patient
Controlled – experimental drug or procedure are compared
with another drug, procedure, or placebo
Randomized - patients are divided into groups randomly
Double blind – patient and doctor does not know what drug
is used
Intention-to-treat – calculating the results with the number
of patients in the beginning of the study
Check:
• Diagnostic or screening tests
• Treatment technique
• Effectiveness and safety of of drug treatment
RCT results - Odds Ratio
Best estimate
Confidence Interval
(wobble factor)
0.5
less than 1
1
2
more than 1
Numbers needed to treat
• Number of people on average needed to
receive treatment to produce one
additional beneficial outcome
• http://www.healthcare.ubc.ca/calc/clinsig.h
tml
Confidence Interval (CI)
= the ‘uncertainty’ factor, how sure are
we about the results?
- the shorter the CI the more certain we are
about the results
- if it crosses the line of 1 (no treatment effect)
the intervention might not be doing any good
and could be doing harm
The p-value
How often you would see a similar result by chance, when
actually there was no effect by the drug or treatment.
0
Impossible
1
Absolutely
Certain
p=0.001
Very unlikely
1 in 1000
p=0.05
Fairly unlikely
1 in 20
p=0.5
Fairly likely
1 in 2
p=0.75
Very likely
3 in 4
RCT - Results
RCT Increase
Number of evidences per day
Why bother with reviews?
• volume of literature is condensed
• “new” information is made accessible
• some reviews are extremely good and
take considerable time to find all the
information on one topic
Types of review:
Reviews
Systematic reviews
Meta-analysis
“The medical literature can be compared to a
jungle. It is fast growing, full of dead wood,
sprinkled with hidden treasure and infested
with spiders and snakes.”
Peter Morgan, Scientific Editor,
Canadian Medical Association
Systematic Review
• Review of all
literature
• On particular topic
• Systematically
identified, appraised
and summarised
Meta-analysis
• Statistical synthesis of results of several
studies, which dealt with the same
question
Odds Ratio Graph (Blobbogram)
LEFT
E
S
S
M
O
RIGHT
E
Line of no significance
0.5
less than 1
1
2
more than 1
Odds Ratio (Blobbogram)
0.5
less than 1
1
2
more than 1
Meta-analysis - blobbogram
!RCT – caution:!
• generalizability – not included children, aged, pregnant,
in lactation, with other drugs etc.
– e.g. RALES study – spironolactone – patients with serious heart failure
+ normal kidney function; practice – mild heart failure + kidny failure
Consequence – increased prescription of spironolactone caused many
deaths due to hyperkalemia!
• comparator – better is „usual care“ than placebo!
– e.g. PROVE-IT study – 80 mg of potent atorvastatinu vs. 40 mg of
pravastatin – surprised by results???
• surrogate end-points – „measurable“ outcome measures
– patients with cardiac arrhytmia were given encainide and flecainide,
perfect effect to arrhytmias, but increase of mortality!
• relative risk, absolute risk:
– group A – 1 case from 100, B – 2 cases from 100 = relative reduction:
50 %, NNT – 100.
– group C – 1 case from 1 000 000, D – 2 cases from 1 000 000 = relative
reduction 50 %, NNT 1 000 000. see
http://www.cebm.utoronto.ca/practise/ca/statscal/
– confidence intervals – 95% CI – the result is between the border for
95% certainty
!RCT – caution:!
• statistical vs clinical relevance
– statistical significance does not neccessarily mean clinical significance
(more patients – better statistical significance)
• Side effects „overlook“
– COX2 cardiotoxic, but Merck (rofecoxib) does not care for years, COX2
prevented 5 ulcers from 1000patients, but caused 6 cardiovascular events
Side effects with frequency 1:1000 statistically undetectable.
• combined measures in „primary“ outcome of study
– UKPDS – 1 „event“ = 21 different parameters (death, infarction,
nefropathy...)
• primary end-point not significant, secondary significant
– p = 0,05 – 1 from 20 repetition can be mistake and I admit this! If 20
results – 1 mistake very presumable
• class effect
– studie HOPE, PEACE, EUROPA – ACEI are not the same!
• influence of pharmaceutical companies
– They fund so they often adjudicate the publication or analyse the data!
Negative result publication bias... NEJM 352:2202-2210 – 50 % of CRO
accept companies advices for publication
Step two:
Finding Evidence
2. Finding the Evidence
EBM books
EBM journals
EBM databases
EBM Guidelines
Now
merged
with ACP
ACP Journal
Club
– see
bi.cuni.cz
ABC Index
EBSCO or
Evidence
Based
Medicine
Reviews
Cochrane collaboration
• International non-profit independent organisation
– main aim – development of Cochrane Library
• Archie Cochrane – founder
• Why? Too much information, is there the
possibility to sort? Aim – development of the
database of systematic review
• 90 countries – different centres – language
support
• Collaborative Review Groups – development
and maintenance of systematic reviews
• Mainly volunteers
• http://www.cochrane.org/index0.htm
Cochrane Library – main parts
The Cochrane Database of Systematic Reviews (Cochrane
Reviews)
A rapidly growing list of regularly updated summaries of the
best available evidence prepared by members of the Cochrane
Collaboration.
The Database of Abstracts of Reviews of Effects (DARE)
Other published reviews prepared by people outside of the
Cochrane Collaboration.
The Cochrane Central Register of Controlled Trials
(CENTRAL)
Where to look if you want to track down the original studies –
the largest single source of controlled trials.
http://www.nicsl.com.au/cochrane/guide_data.asp - bližší podrobnosti
http://144.32.150.197/scripts/WEBC.EXE/NHSCRD/start - DARE
• Clinical Queries – filters for selection of
reviews and clinical trials
TRIP database (Turning
Research into Practice)
• http://www.tripdatabase.com/
• TRIP searches a broad range of evidencebased health care-related websites
including Bandolier (good for allied health)
and Cochrane Library databases. Simple
search. View articles one by one
SUMSearch
“that was some search!”
• http://sumsearch.uthscsa.edu/
• Devised by the University of Texas Health
Sciences Center at San Antonio.
• Searches MEDLINE, DARE, an on-line textbook,
usually the Merck manual, etc.
• Uses MeSH thesaurus
guidelines
• Document „How to treat“ based on the newest
evidences
• Czech Republic:
– http://www.cls.cz/dp/index.htm CLS JEP
– http://www.svl.cz/default.aspx/cz/spol/svl/default/men
u/doporucenepostu/doporucenepostu SVL
----------• World:
– National Guideline Clearinghouse http://www.guideline.gov/ - USA
– NICE - http://www.nice.org.uk/ - UK – NHS
– GIN – www.g-i-n.net – International network
– The Agency for Quality in Medicine (AQUMED) –
www.leitlinien.de – BRD
– http://www.phac-aspc.gc.ca/dpg_e.html - Canada
– etc.....!
Step three:
Critical appraisal
3. Critical appraisal
• Case report < case-control study < cohort
study < RCT < systematic review (<
guideline)
• CASP – critical appraisal skills programme
http://www.phru.nhs.uk/casp/critical_apprai
sal_tools.htm
Step four:
Implementing evidence
Context sensitive practice
Professional
skills
Evidence
Context
Context
Senstive
Sensitive
Practice
Practice
Values,
resources,
policy
Patient
knowledge
and values
Step five:
Evaluating results
Evaluation
Practice can be evaluate through
• your own reflection
• quality assurance programmes
• further research
Drug Information Centres (DIC)
• http://www.napra.org/docs/0/95/157/164.a
sp - Canada - list
• http://www.uh.edu/~libn/drug.htm example of information sources in DIC
• http://www.shpa.org.au/docs/druginfo_int.h
tml - !international register! of Drug
Information Centres
DIC CR – see Solutio – www.medon.cz
•
•
•
•
•
Informace o farmaceutických přípravcích: Farmaceutický informační servis
(Phoenix)
Vinohradská 72, 618 00 Brno
tel. 548 135 171
[email protected]
Lékové informační centrum (LIC)
Lékárna FN U sv. Anny v Brně Pekařská 53, 656 91 Brno
tel. 543 182 175-7
tel./fax 543 211 429 [email protected] [email protected]
Lékové informační centrum
Katedra sociální a klinické farmacie, FaF UK v Hradci Králové Heyrovského 1203,
500 05 Hradec Králové 5
tel. 495 067 452
fax 495 512 266
[email protected]
Středisko vědeckých informací (SUKL)
Šrobárova 48, 100 41 Praha 10
tel. 272 185 829, 272 185 333
fax 272 185 756
[email protected], [email protected]
Lékové informační centrum při 3. LF
MUDr. Martin Votava
Ústav farmakologie 3. LF UK
Ruská 87
100 00, Praha 10
www.zdravcentra.cz –
[email protected]
formulář pro vložení dotazu
Drug agencies…
• Czech Republic
– www.sukl.cz
• EU
– www.emea.eu.int
• see „human medicines“
• requires JAVAscript!
• EPAR – better than SPC!
• USA
– www.fda.gov
• CDER
---------USA – see http://www.cdc.gov/ ! – description of different
diseases