1
Victor BabesUNIVERSITY OF MEDICINE AND
PHARMACY TIMISOARA

• DEPARTMENT OF
• MEDICAL INFORMATICS
• 2005 / 2006

2

• http//www.medinfo.umft.ro/dim/metodologie.htm

3

• Medical Scientific Research MethodologyMCS
• 2nd Module
• Course 1

4
1. Medical Information Object of the course
5
1.1. MEDICAL INFORMATION

• Pacient Physician Relation
• Elementary Cycle of Medical Activity
• Medical Information Used in Medical Practice
• DATA individual character facts
• studied by MEDICAL/HEALTH INFORMATICS
• KNOWLEDGE general character concepts
• studied by MEDICAL RESEARCH METHODOLOGY

6
1.2. ELEMENTARY CYCLE OF MEDICAL ACTIVITY
7
1.3. TYPES OF DATA

• QUALITATIVE Anamnesis (descriptive)
• NUMERICAL Laboratory investigations
• GRAPHICAL Biosignals (ECG, EEG)
• SOUNDS Phonocardiogram
• STATIC IMAGES X-Ray, NMR
• DYNAMIC IMAGES movies

8
1.4. TYPES OF KNOWLEDGE

• a) Explicit knowledge
• Formalized
• expressed in words, schemes, formulae
• Easy transmission (books, lectures)
• Education
• b) Tacit knowledge
• Personal experience
• skill
• Difficult transmission
• Training

9
1.5. Relation between knowledge and
diagnosis/treatment

• LOGICAL INFERENCE
• Classical Medicine
• STATISTICAL INFERENCE
• Evidence Based Medicine

10
1.6. Object of Medical Scientific Research
Methodology

• Production of explicit knowledge research
• Research methods
• Observational
• surveys
• clinical trials
• Experimental
• laboratory studies - specific
• Theoretical modeling/simulation

11
1.7. Research Activity Steps

• Documentation
• Study Organizing
• Clinical Studies
• Laboratory Studies
• Theoretical Research formalizing knowledge
• Data Processing
• Presentation of results
• Writing scientific papers

12
1.8. TYPES OF CLINICAL STUDIES

• Public Health Studies
• Health Services Studies
• Evaluation of a Diagnostic Procedure
• Evaluation of a Therapeutical Approach
• Risk and/or Prognostic Factors Studies

13

• 1.9. TYPES OF PAPERS
• original papers - book reviews
• case studies - theses
• reviews editorials
• Approach Modes
• Descriptive
• Analitical

14
I. TYPES OF STUDIESGeneral Characteristics
15
2. DESCRIPTION OF A HEALTH PHENOMENON
16

• 2.1. STEPS
• - first step data collection
• does not explain causes
• comparation with existing etalons
• 2.2 THE PROTOCOL
• Estabilishing target population
• Pathological state definition
• Choosing variabiles, scales, standards etc
• Epidemiological indicators computation
• Comparing with known indicators

17
2.3. Mandatory Criteria inDescriptive Studies

• Representativity of
• general (target) population
• studied pathology
• Comparability
• with results from other studies
• variables and their measurement
• Utility
• Medical and scientific pertinence
• Priorities
• Efficiency for patients

18
3. EVALUATION OF A DIAGNOSTIC PROCEDURE
19

• 3.1. PHASES
• 1. Determination of
• Method validity
• Fiability influenced by inter and
  intra-observational reproducibility
• 2. Qualitative Assessment of the Test (Procedure)
  
• Sensitivity / Specificity
• ROC Curve (Receiver Operating Characteristic)
• Sn f (1 Sp)
• Youden Index Y Sn Sp 1 (diagn.value)

20
CLASSIFICATION QUALITYASSESSMENT
3.2. Classification table Test
T T-
D N11 N12 R1 D- N21 N22 R2
C1 C2 N
Real
21
3.3. PARAMETERS

• false negative N12
• false positive N21
• sensitivity SN p(T/D) N11 / R1
• specificity SP p(T-/D-) N22 / R2
• inverse relationship between SN and SP
• positive/negative predictive values
• PPV p(D/T) N11 / C1
• NPV p(D-/T-) N22 / C2
• accuracy AC (N11 N22) / N
• classification error rate ER (N12 N21) / N

22

• 3.4. Test Assessment (practical)
• PPV, NPV AC, ER
• Test D (detectability AC2 / AC1)
• Test V (validity Y2 / Y1)
• 3.5. Index Evaluation
• CB Cost Benefit
• CE Cost - Efficiency

23
4. EVALUATION OF A THERAPEUTIC
PROCEDURECLINICAL TRIALS
24
4.1. Definitions and Terms

• Def CLINICAL TRIAL clinical research technique
  for estimating the effect of a therapy, within an
  extremely restrictive methodological frame
• Estimating EFFICIENCY
• Comparison of the treated group with a
  control group (placebo)
• Terms
• Controlled clinical trial with controls
• Uncontrolled clinical trial without controls

25
4.2. Methodological Principles in clinical
trials

• A) Exact Scope Definition
• Unilateral study (refer.), bilateral (2
  treatments)
• B) Patient Groups Definition
• Samples selection size, continuity, inclusion /
  exclusion criteria
• C) Comparison Modes
• Comparable groups , randomization, stratification
• Definition of treatment control type (no
  treatment, placebo, std)
• Mentaining results quality (systematic errors
  elimination)

26

• Defining the general comparing plan
• Parallel groups each group a scheme
• Cross-over automartor order by randomozation
• Disadv need of a stable pathology, objective
  evaluation criteria, wash-out periods
• Secquential patients recruited for a longer
  period stop at significant differences.
• D) Defining assessment criteria
• Main criterion conforming the objective
• E) Analysis mode and comparison interpretation
• Type of analysis, statistical tests, prognostic
  factors

27

• Comparability (avoid bias)
• Randomization random numbers
• Simple / Double Blind Studies
• simple blind the patient does not know
  his/her treatment, the physician does
• double blind neither the patient nor the
  physician know the allocation to a certain
  treatment group

28

• The ideal study
• (simple, cheap, good results)
• Controlled
• Randomized
• Double blind
• Parallel groups

29
4.3. Writing the Protocol of a CLINICAL TRIAL

• It should be detailed its precision is
  determining the quality of the study
• Introduction basic data describing the medical
  problem and the therapy under study
• Scope clear definition of the scope and of the
  objectives
• Method description of the general comparing plan
• Groups defining subjects/patients (inclusion and
  exclusion criteria), patient acceptance signature
  for each individual in the study

30

• 5. Sample size details on calculating should be
  included in the protocol
• 6. Selection randomization method when, how,
  type (even, stratified)
• 7. Treatments
• a) Studied treatment procurement method,
  physical-chemical properties, prescribing
• b) Placebo or reference treatment procurement,
  comparability (look, physical-chemical qualities
  etc)
• c) Associated treatments list of authorized
  substances, administration conditions, forbidden
  substances list

31

• 8. Initial balance list of necessary
  investigations for inclusion or prognostic
  factors for rejection, which will be studied in
  the trial
• 9. Evaluation criteria for the therapeutic
  effect details on data collection, main
  criterion, secondary criteria
• 10. Surveillance mode all subjects will be
  surveyed in the same manner, at the same
  intervals. The rhythm and content of control
  examinations shall be presented. Following
  features must be observed, and the conduct in
  such cases shall be detailed
• a) Secondary effects
• b) Therapy break-up
• c) Lost during the study

32

• 11. . Data analysis
• a) Statistical analysis methodology comprises the
  hazard evaluation, systematic errors and
  confusion factors, with alternate explanation for
  the recorded results
• b) The attitude on the lost shall be mentioned
  the following postulate is recommended once
  randomized mandatory analyzed. None of the
  subjects included in a study group shall be
  omitted in the analysis
• 12. Criteria for breaking-up the trial
• a) Clear benefic results in an intermediary
  phase, due to a highly significance of the tests,
  correlated with the clinical context
• b) When the treatment may be dangerous for the
  subject(s)

33

• 13. Attached documentsThe protocol has usually
  the following attached documents
• List of persons and their responsibilities
• Activities calendar
• Observation log
• Subjects identification data
• Inclusion data
• Results of successive balances
• The informative document given to patients
• List of acceptance signatures
• Ethical commission approval, containing details
  on the trial quality control strategy, during the
  study, for subjects / patients protection

34
Pause