1

Clinical Study: Clinical Study: Design and Design and

MethodsMethodsHail M. Al-Abdely, MDHail M. Al-Abdely, MD

Consultant, Infectious DiseasesConsultant, Infectious Diseases

King Faisal Specialist Hospital & King Faisal Specialist Hospital & Research CentreResearch Centre

2

““Systematic investigation towards Systematic investigation towards increasing the sum of knowledge”increasing the sum of knowledge”

(Chambers 20th Century Dictionary)(Chambers 20th Century Dictionary)

““an endeavour to discover new or collate an endeavour to discover new or collate old facts etc.old facts etc.

by the scientific study of a subject or by a by the scientific study of a subject or by a course of critical investigation.”course of critical investigation.”

(The Concise Oxford Dictionary)(The Concise Oxford Dictionary)

Definitions of ResearchDefinitions of Research

3

Where to Start?Where to Start?

A good clinical study starts with A good clinical study starts with a good questiona good question based on based on good good

hypothesishypothesis that is based on that is based on good and good and comprehensive reviewcomprehensive review of the available of the available evidence from pre-clinical and clinical evidence from pre-clinical and clinical datadata

Type of design depends on the Type of design depends on the question to be answeredquestion to be answered

4

Formulating a Research Formulating a Research QuestionQuestion

Focused and specificFocused and specific What is the prevalence of Hepatitis B surface Antigen in What is the prevalence of Hepatitis B surface Antigen in

Saudi Arabia? Saudi Arabia? Cross-sectional studyCross-sectional study What are the risk factors for hepatitis B infection? What are the risk factors for hepatitis B infection?

Prospective cohort or case-controlProspective cohort or case-control Is interferon a useful therapy for hepatitis B infection? Is interferon a useful therapy for hepatitis B infection?

Therapeutic clinical trialTherapeutic clinical trial Supported by available dataSupported by available data

Is vancomycin better than ceftazidime against gram Is vancomycin better than ceftazidime against gram negative organisms?negative organisms?

Not a replication of already established evidenceNot a replication of already established evidence Is smoking associated with lung cancer?Is smoking associated with lung cancer?

EthicalEthical AnswerableAnswerable

Methods, resources ….etcMethods, resources ….etc

5

ObjectivesObjectives

Specific aimsSpecific aims Clear and detailedClear and detailed

End point(s)End point(s) PrimaryPrimary

The main answer to the research questionThe main answer to the research question SecondarySecondary

Answer other related questionsAnswer other related questions

6

Study DesignStudy Design

Your questionYour question DescribeDescribe AnalyzeAnalyze

Your resourcesYour resources RetrospectiveRetrospective ProspectiveProspective

CommunityCommunity Acceptance of researchAcceptance of research

ObservationalObservational InterventionalInterventional

7

Clinical Study TypesClinical Study Types

Observational StudiesObservational Studies Cohort (Incidence, Longitudinal)Cohort (Incidence, Longitudinal) Case-Control Case-Control Cross-Sectional (Prevalence)Cross-Sectional (Prevalence) Case SeriesCase Series Case Report Case Report

Experimental StudiesExperimental Studies Uncontrolled TrialsUncontrolled Trials Controlled TrialsControlled Trials

8

Level I: N of 1 randomized trial (double-blinded, cross-over)Level I (A): Systematic reviews of randomized trialsLevel I (B): Single randomized trialLevel II (A): Systematic review of observational studies addressing patient-important outcomeLevel II (B): Single observational study addressing important outcomeLevel III: Physiologic studiesLevel IV: Unsystematic clinical observations (case-reports, anecdotal)

Levels of EvidenceLevels of EvidenceHierarchy of Strength of Evidence for Treatment Decisions

JAMA 2000; 284(10):1290-96

9

Observational study Observational study Clinical trialClinical trial

exposed

non exposedoutcome

ClinicalTrial

observationalstudydescribe as

occurring in nature

allocaterandomly

Ethics!

10

Important issues in Important issues in Study DesignStudy Design

Validity: Validity: TruthTruth External Validity: External Validity:

Can the study be generalized to the Can the study be generalized to the populationpopulation

Internal Validity: Internal Validity: Results will not be due to chance, bias or Results will not be due to chance, bias or

confounding factorsconfounding factors Symmetry Principle: Groups are similarSymmetry Principle: Groups are similar

11

Confounding: distortion of the effect of one risk factor by the presence of another

Bias: Any effect from design, execution, & interpretation that shifts or influences results Confounding bias: failure to account for the

effect of one or more variables that are not distributed equally

Measurement bias: measurement methods differ between groups

Sampling (selection) bias: design and execution errors in sampling

Important issues in Important issues in Study DesignStudy Design

12

IntroductionIntroduction Why this study is needed ?Why this study is needed ?

What is the purpose of this study? What is the purpose of this study? Was purpose known before the study?Was purpose known before the study? What has been done before and how does this What has been done before and how does this

study differ? study differ? inadequacies of earlier work or next step in an inadequacies of earlier work or next step in an

overall research projectoverall research project Does the location of the study have relevance?Does the location of the study have relevance?

13

Why doing a study?Why doing a study?

Alternative: Alternative: census: test every individual in the census: test every individual in the

populationpopulation use available data, e.g. hospitalsuse available data, e.g. hospitals

But:- data availability- data quality- cost- questions require specific type of data and circumstances

14

Types of observational Types of observational studiesstudies

CROSS - SECTIONAL STUDYCROSS - SECTIONAL STUDY

COHORT STUDYCOHORT STUDY

CASE CONTROL STUDYCASE CONTROL STUDY

CASE SERIES/CASE REPORTSCASE SERIES/CASE REPORTS

15

Characteristics of observational Characteristics of observational studiesstudies

No control over study units No control over study units need to clearly describe study individualsneed to clearly describe study individuals

Can study risk factors that have serious Can study risk factors that have serious consequencesconsequences

Study individuals in their natural Study individuals in their natural environment (>> extrapolation)environment (>> extrapolation)

Possibility of confoundingPossibility of confounding

16

Aims of observational studiesAims of observational studies

Evaluate the effect of a suspectedrisk factor (exposure) on an outcome(e.g. disease) define ‘exposure’ and ‘disease’

Describe the impact of the risk factoron the frequency of disease in apopulation

17

Cross - Sectional Cross - Sectional StudyStudy

18

Cross - Sectional Study Cross - Sectional Study (1)(1)

Exposure and disease measured once, i.e. Exposure and disease measured once, i.e. at the same point in timeat the same point in time

present futurepast

n

exposed ?diseased ?

19

Cross - Sectional Study Cross - Sectional Study (2)(2)

Random sample from populationRandom sample from population i.e. results reflect reference populationi.e. results reflect reference population

Estimates the frequencies of both Estimates the frequencies of both exposure and outcome in the exposure and outcome in the populationpopulation

Measuring both exposure and outcome Measuring both exposure and outcome at one point in timeat one point in time

Typically a surveyTypically a survey

20

Cross - Sectional Study Cross - Sectional Study (3)(3)

Can study several exposure factors and Can study several exposure factors and outcomes simultaneously outcomes simultaneously

Determines disease prevalenceDetermines disease prevalence Helpful in public health administration & Helpful in public health administration &

planningplanning QuickQuick Low cost (e.g. mail survey)Low cost (e.g. mail survey) Limitation: Limitation:

Does not determine causal relationshipDoes not determine causal relationship Not appropriate if either exposure or outcome is Not appropriate if either exposure or outcome is

rarerare

21

Cross-Sectional: Pediatrician-to-Child Cross-Sectional: Pediatrician-to-Child Ratio Ratio

Greg et al. (2001) Pediatrics.107(2):e18

0

5

10

15

20

25

30

35

40P

edia

tric

ians

per

10

00 C

hild

ren

Rural Urban

1981198619911996

22

Cross-Sectional: Risk Factors for Cross-Sectional: Risk Factors for SmokingSmoking

VariableVariable OROR 95% CI95% CI

No. friends who smoke:No. friends who smoke:

- all vs. none of them- all vs. none of them

- most vs. none of them- most vs. none of them- about half vs. none of about half vs. none of themthem- a few vs. none of thema few vs. none of them

36.536.5

18.418.4

7.57.5

2.12.1

9.3 – 9.3 – 142.8142.8

5.5 – 5.5 – 61.861.8

2.2 – 2.2 – 26.026.0

0.6 – 7.90.6 – 7.9

Any siblings who smoke: Y Any siblings who smoke: Y vs. Nvs. N

2.82.8 1.8 – 4.31.8 – 4.3

Mother smokes:Mother smokes:

Yes vs. NoYes vs. No

Have no mother vs NoHave no mother vs No1.91.9

3.53.51.3 – 2.91.3 – 2.9

0.8 – 0.8 – 15.015.0

23

Cohort StudiesCohort Studies

24

Cohort studiesCohort studies

Follow-up studies; subjects selected on Follow-up studies; subjects selected on presence or absence of exposure & presence or absence of exposure & absence of disease at one point in time. absence of disease at one point in time. Disease is then assessed for Disease is then assessed for allall subjects subjects at another point in time.at another point in time.

Typically prospective but can be Typically prospective but can be retrospective, depending on temporal retrospective, depending on temporal relationship between study initiation & relationship between study initiation & occurrence of disease.occurrence of disease.

25

Cohort Study (1)Cohort Study (1)

Individuals selected by exposure status and Individuals selected by exposure status and future occurrence of disease measuredfuture occurrence of disease measured

present futurepast

n

Exposed yes no

disease ?disease ?

n

Exposed yes no

disease ?disease ?

26

Cohort studies (2)Cohort studies (2)

More clearly established temporal More clearly established temporal sequence between exposure & sequence between exposure & diseasedisease

Allows direct measurement of Allows direct measurement of incidenceincidence

Examines multiple effects of a single Examines multiple effects of a single exposure (nurses’ health study, OC exposure (nurses’ health study, OC and breast, ovarioan cancers)and breast, ovarioan cancers)

27

Cohort studies (3)Cohort studies (3) Limitations: Limitations:

time consuming and expensive time consuming and expensive loss to follow-up & unavailability loss to follow-up & unavailability

of data of data potential confounding factors potential confounding factors inefficient for rare diseasesinefficient for rare diseases

28

Prospective Cohort Study Prospective Cohort Study

without outcome

Cohort

with outcome

with outcome

withoutoutcome

Exposed

Unexposed

TimeOnsetof study Direction of inquiry

Q: What will happen?

29

Prospective Cohort StudyProspective Cohort Study

Appropriate for frequent diseaseAppropriate for frequent disease Can examine only few risk factorsCan examine only few risk factors Usually expensiveUsually expensive RR = ‘relative risk’ = incidence rate RR = ‘relative risk’ = incidence rate

ratioratio AR = incidence difference AR = incidence difference

30

Case-Control Case-Control StudiesStudies

31

Case-Control Study (1)Case-Control Study (1)

RetrospectiveRetrospective Can use hospital or health register data Can use hospital or health register data

First identify casesFirst identify cases Then identify suitable controlsThen identify suitable controls

Hardest part: who is suitable ??Hardest part: who is suitable ?? Then inquire or retrieve previous Then inquire or retrieve previous

exposureexposure By interviewBy interview By databases (e.g. hospital, health By databases (e.g. hospital, health

insurance)insurance)

32

Case-Control Study (2)Case-Control Study (2)

Diseased and non-diseased individuals Diseased and non-diseased individuals are selected firstare selected first

Then past exposure status is retrievedThen past exposure status is retrieved

present futurepast

n

yes no

diseaseexposed ?exposed ?

33

Case-Control Study (3)Case-Control Study (3)

Good for rare disease (e.g. cancer)Good for rare disease (e.g. cancer) Can study many risk factors at the Can study many risk factors at the

same timesame time Usually low costUsually low cost Confounding likelyConfounding likely OR (not RR !!)OR (not RR !!)

34

Case-Control Study Case-Control Study DesignDesign

Cases

Controls

Exposed

Unexposed

Exposed

Unexposed

TimeDatacollection

Direction of inquiry

Q: What happened?

35

• Study subjects selected on basis of Study subjects selected on basis of whether they have (case) or do not whether they have (case) or do not have (control) a disease have (control) a disease

• Useful for disease with long Useful for disease with long latency periodlatency period

• Efficient in terms of time & costsEfficient in terms of time & costs• Particularly suited for rare Particularly suited for rare

diseasesdiseases• Examines multiple exposures to a Examines multiple exposures to a

single diseasesingle disease

Case-Control study (4)Case-Control study (4)

36

Case-control study (5)Case-control study (5)

Limitations: Limitations:

(1) susceptible to bias (particularly (1) susceptible to bias (particularly selection & recall) selection & recall)

(2) difficulties in selection of controls (2) difficulties in selection of controls

(3) ascertainment of disease & (3) ascertainment of disease & exposure statusexposure status

(4) inefficient for rare exposures (4) inefficient for rare exposures unless attributable risk is highunless attributable risk is high

37

Case Selection

• Define source population

• Cases– incident/prevalent– diagnostic criteria (sensitivity + specificity)

• Controls– selected from same population as cases– select independent of exposure status

38

Control Selection

• Random selection from source population

• Hospital based controls:– convenient selection– controls from variety of diagnostic groups other

than case diagnosis– avoid selection of diagnoses related to

particular risk factors– limit number of diagnoses in individuals

39

Characteristic

Cross -

Sectional

Case Control

Cohort

Sampling

Random sample: population

Purposive sample: diseased/non-diseased

Purposive sample: Exposed/non-exposed

Time One point Retrospective Prospective Causality Statistical

association Screening for many risk factors

Testing one (or few) risk factors

Frequency measure

Prevalence None Incidence

Risk parameter

Prevalence (risk) ratio, odds ratio

Odds ratio Relative risk, odds ratio

Summary of Observational Summary of Observational StudiesStudies

40

Clinical TrialsClinical Trials

41

Clinical Trials Clinical Trials – Drug – Drug DevelopmentDevelopment

BasicResearch

NovelCompounds

SafetyTesting

Drug Licensing& Release

In-VitroScreeningIsolated cells

& tissues

In-VivoScreening In A

nimals

Clinical Trials I - III

In Humans

42

Clinical trials in drug Clinical trials in drug development development (Any alternatives)(Any alternatives)

In-Vitro Tests Can Show Whether:• A compound has the desired effect on isolated

cells or tissues• There are adverse effects on those tissues

• In-Vitro Tests Cannot Show Whether:• The desired effect will occur in a complete

living system

• There will be any adverse effects in a complete living system

43

Animal Tests Can: Suggest which drugs are likely to be

effective in humans Indicate which drugs may not be

harmful in humans Animal Tests Cannot:

Predict with absolute certainty what will happen in humans

Clinical trials in drug Clinical trials in drug development development (Any alternatives)(Any alternatives)

44

Clinical trial vs. Cross-Clinical trial vs. Cross-sectionalsectional

Clinical trial:Clinical trial: Individuals selected Individuals selected

by entry conditionby entry condition Control over Control over

exposureexposure Exposure groups Exposure groups

fully comparablefully comparable Outcome measured Outcome measured

after allocating after allocating individuals to individuals to exposureexposure

Therefore: causal Therefore: causal association likelyassociation likely

Cross Sectional Study:Cross Sectional Study: Individuals selected Individuals selected

randomlyrandomly Exposure observed as Exposure observed as

occurring in nature occurring in nature (groups not ‘identical’)(groups not ‘identical’)

Exposure AND outcome Exposure AND outcome measured at one point measured at one point in timein time

No causal interpretationNo causal interpretation

45

Clinical Trials-PhasesClinical Trials-Phases

Phase I - Does it hurt the Patient?Phase I - Does it hurt the Patient? Usually in normal volunteers, small groups for safety Usually in normal volunteers, small groups for safety

testingtesting

Phase II - Does it help the Patient?Phase II - Does it help the Patient? On patients to confirm the effectiveness of the drugOn patients to confirm the effectiveness of the drug

Phase III - Is it any better?Phase III - Is it any better? Large groups of patients for statistical confirmation of effect and Large groups of patients for statistical confirmation of effect and

incidence of side-effectsincidence of side-effects

Phase IV - Does it work in the community?Phase IV - Does it work in the community? Post marketing studies. Fine tuning and new rare Post marketing studies. Fine tuning and new rare

findings from a very large populationfindings from a very large population

46

Clinical Trial: Study Clinical Trial: Study DesignDesign

UncontrolledUncontrolled ControlledControlled

Before/after (cross-over)Before/after (cross-over) HistoricalHistorical Concurrent, not randomizedConcurrent, not randomized RandomizedRandomized

47

Non-randomized Trials Non-randomized Trials May Be AppropriateMay Be Appropriate

• Early studies of new and untried therapies

• Uncontrolled early phase studies where the standard is relatively ineffective

• Investigations which cannot be done within the current climate of controversy

• Truly dramatic response

48

Advantages of Randomized Advantages of Randomized Control Clinical TrialControl Clinical Trial

1.1. Randomization "tends" to produce Randomization "tends" to produce comparable groupscomparable groups

2.2. Assure causal relationshipAssure causal relationship

3.3. Randomization produces valid statistical Randomization produces valid statistical teststests

49

Disadvantages of Disadvantages of Randomized Control Clinical Randomized Control Clinical

TrialTrial1.1. Generalizable Results?Generalizable Results?

Participants studied may not represent Participants studied may not represent general study population.general study population.

2.2. RecruitmentRecruitment HardHard

3.3. Acceptability of Randomization ProcessAcceptability of Randomization Process Some physicians will refuseSome physicians will refuse Some participants will refuseSome participants will refuse

4.4. Administrative ComplexityAdministrative Complexity

50

Clinical Protocol (1)Clinical Protocol (1)

Background/JustificationBackground/Justification--Where we are in the field--Where we are in the field--What the study will add that is --What the study will add that is importantimportant

ObjectivesObjectives--Primary hypothesis--Primary hypothesis--Secondary hypotheses--Secondary hypotheses--Other--Other

51

Study PopulationStudy PopulationSubset of the general population Subset of the general population

determined by the determined by the eligibility criteriaeligibility criteria

General populationGeneral population

Eligibility criteriaEligibility criteria

Study populationStudy population

EnrollmentEnrollment

Study sampleStudy sampleObservedObserved

52

Clinical Protocol (2)Clinical Protocol (2)

Study Design and MethodsStudy Design and Methods Type of study, comparisonType of study, comparison Inclusion and exclusion criteriaInclusion and exclusion criteria Description of intervention (what, how)Description of intervention (what, how) Concomitant therapyConcomitant therapy Examination procedures (baseline, follow-up, Examination procedures (baseline, follow-up,

outcome assessment)outcome assessment) Intervention assignment procedureIntervention assignment procedure Data collection sheetData collection sheet Informed consentInformed consent

53

Eligibility CriteriaEligibility Criteria(inclusion & exclusion)(inclusion & exclusion)

State in advanceState in advance ConsiderConsider

Potential for effect of interventionPotential for effect of intervention Ability to detect that effectAbility to detect that effect SafetySafety Ability for informed consentAbility for informed consent

54

Method Outlines (1)Method Outlines (1)

The independent (predictor) and dependent (outcome) The independent (predictor) and dependent (outcome) variables in the study should be clearly identified, defined, variables in the study should be clearly identified, defined, and Measured? and Measured?

How to choose subjects? How to choose subjects? Random or notRandom or not Are they going to be representative of the population? Are they going to be representative of the population? Random selection is not random assignment Random selection is not random assignment

Types of Blinding (Masking) Single, Double, Triple.Types of Blinding (Masking) Single, Double, Triple. Control group? How is it chosen?Control group? How is it chosen? How are patients followed up? Who are the dropouts?How are patients followed up? Who are the dropouts? How is the data quality insured? Reliability? How is the data quality insured? Reliability? Consider independent review of data? Compliance?Consider independent review of data? Compliance?

55

Methods outlines (2)Methods outlines (2)

Reference any unusual methods?Reference any unusual methods? Statistical methods specified in sufficient Statistical methods specified in sufficient

details details Is there a statement about sample size issues or Is there a statement about sample size issues or

statistical power?statistical power? ? multicenter study. Quality assurance ? multicenter study. Quality assurance

measures should be employed to obtain measures should be employed to obtain consistency across sites?consistency across sites?

56

Comparing TreatmentsComparing Treatments• Fundamental principle

• Groups must be alike in all important aspects and only differ in the intervention each group receives

• In practical terms, “comparable treatment groups” means“alike on the average”

• Randomization• Each participant has the same chance of receiving any of the

interventions under study• Allocation is carried out using a chance mechanism so that neither the

participant nor the investigator will know in advance which will be assigned

• Blinding• Avoidance of conscious or subconscious influence• Fair evaluation of outcomes

57

Patients and Clinicians Kept Blind To Treatment?

InvestigatorInvestigator Care takerCare taker

58

Monitoring and ManagementMonitoring and Management--Data and safety monitoring --Data and safety monitoring

--Adverse event assessment, reporting--Adverse event assessment, reporting

--Contingency procedures--Contingency procedures

--Withdrawal criteria--Withdrawal criteria

Methods outlines (3)Methods outlines (3)

59

Regular Follow-upRegular Follow-up

Routine Procedures (report forms)Routine Procedures (report forms) InterviewsInterviews ExaminationsExaminations Laboratory TestsLaboratory Tests

Adverse Event Detection/ReportingAdverse Event Detection/Reporting Quality AssuranceQuality Assurance

60

Compliance/adherence

Pill counts and computers

Diaries

Biological tests

61

Lipid lowering drugs after myocardial infarction

Mortality

clofibrate 18.2%

placebo 19.4%

Clofibrate Adherence

80 < 80%

18.2% 15.0% 24.6%

Overall

Clofibrate

62

StatisticsStatistics--Sample size--Sample size

--Stopping guidelines--Stopping guidelines

--Analysis plans--Analysis plans Participant protection issuesParticipant protection issues

Methods outlines (4)Methods outlines (4)

63

Sample SizeSample Size

The study is an experiment in peopleThe study is an experiment in people Need enough participants to answer Need enough participants to answer

the questionthe question Should not enroll more than needed Should not enroll more than needed

to answer the questionto answer the question Sample size is an estimate, using Sample size is an estimate, using

guidelines and assumptionsguidelines and assumptions

64

Contingency PlansContingency Plans

Patient managementPatient management Evaluation and reporting to all Evaluation and reporting to all

relevant persons and groupsrelevant persons and groups Data monitoring plansData monitoring plans Protocol amendment or study Protocol amendment or study

terminationtermination

65

Human Subjects ProtectionHuman Subjects Protection

• Institutional Review Board Institutional Review Board • Informed consentInformed consent• Different levels of riskDifferent levels of risk• Confidentiality as well as risk of new txConfidentiality as well as risk of new tx• Patient can refuse to participate w/o Patient can refuse to participate w/o

effecteffect• Path to exit study knownPath to exit study known• CompensationCompensation

66

SummarySummary Selection of design should be made on the Selection of design should be made on the

basis of the particular hypothesis to be basis of the particular hypothesis to be tested with consideration of current state of tested with consideration of current state of knowledgeknowledge

Consider available resources when deciding Consider available resources when deciding on a study designon a study design

A clear and organized study design leads to A clear and organized study design leads to successful resultssuccessful results

Observational studies are especially Observational studies are especially valuable in epidemiologyvaluable in epidemiology

Clinical trials carry the highest level of Clinical trials carry the highest level of evidence and should be pursued whenever evidence and should be pursued whenever feasiblefeasible