Study Design

Prof. Dr. Mohamed Zak Khalil, MD,

FACP, FACC, FESC, FAHA, FKSUProfessor of Medicine

Consultant Interventional Cardiologist

Objectives:

Performing a studyDefinitionsChoosing study designProposalAppraisalConclusions

Performing a study

Objective

•Aim Or Purpose

Methods

•Study design•Settings

Results

•Interpretations•Recommendations

Definitions

Observational studies

Cohort, cross sectional, and case-control studies are collectively referred to as observational studies.The investigator simply observes.No interventions are carried out by the investigator.

COHORT STUDIES

The best design for determining the incidence and natural history of a condition.The studies may be prospective or retrospective and sometimes two cohorts are compared.

Prospective cohort studies

A group of people is chosen who do not have the outcome of interest.The investigator then measures a variety of variables that interfere with the outcome.Over a period of time the people in the sample are observed to see whether they develop the outcome of interest

Internal controls:

In single cohort studies those people who do not develop the outcome of interest are used as internal controls.

External control:

Two cohorts are used:One group has been exposed to or treated with the agent of interest (subjects).The other group has not, thereby acting as an external control.

Retrospective cohort studies

The cohort is “followed up” retrospectively.The study period may be many years.The time to complete the study is as long as it takes to collect and analyze the data.

COHORT STUDIES

Cohort

Present

ProspectiveRetrospective

Sample group in cohort:

Each subject must have the potential to develop the outcome of interest.Sample population must be representative of the general population.

CROSS SECTIONAL STUDIES:

Primarily used to determine prevalence.All the measurements on each person are made at one point in time.Only one group is used.Data are collected only once.Multiple outcomes can be studied.

CASE-CONTROL STUDIES:Case-control studies determine the relative importance of a predictor variable in relation to the presence or absence of the disease.People with the outcome of interest are matched with a control group who do not.Case-control studies are retrospective for a rare outcome.The flexibility of the variables studied comes at the expense of the restricted outcomes studied.Can only look at one outcome, but useful for hypothesis generation.Liable to bias & confounding variables.

Experimental Studies:Experimental intervention is administered to subjects by the Investigator.Requires approval by ethical committee.Either prospective randomized controlled trial or meta-analysis of multiple trials addressing the same outcome retrospectively.

Randomized controlled trial:

Subjects are assigned by statistically randomized methods to two or more groups.All variables other than the proposed intervention are evenly distributed between the groups.Blinded to minimize bias.Multiple centers under supervision of steering committee.

Meta-analysis:Retrospective analysis of multiple randomized experimental trials.Data obtained by performing specific search in medical databases.Produce best available evidence for clinical practice.

Definitions

BiasThe inclusion of subjects or methods such that the results obtained are not truly representative of the population from which it is drawn.

Sampling bias

The patients with the disease may be a biased sample .The controls may be biased.

Observation and recall bias

Assessment of predictor variables retrospectively there is great potential for a biased assessment of their presence and significance by the patient or the investigator, or both.

Blinding

The process by which the researcher and or the subject is ignorant of which intervention or exposure has occurred.Single or double.

Cohort

Is a component of a population identified so that one or more characteristic can be studied as it ages through time.

Confounding variable

A variable that is associated with both the exposure and outcome of interest that is not the variable being studied.

Control group

A group of people without the condition of interest, or unexposed to or not treated with the agent of interest.

Choosing study design

Choose a design:

Aim: The prevalence of obesity among Saudi children.A) Cohort studyB) Cross sectional studyC) Case control studyD) Randomized prospective controlledE) Meta-analyses

The answer is B

Aim: The prevalence of obesity among Saudi children.A) Cohort studyB) Cross sectional studyC) Case control studyD) Randomized prospective controlledE) Meta-analyses

Cross sectional studies

Cross sectional studies are the best way to determine prevalenceAre relatively quickCan study multiple outcomesDo not themselves differentiate between cause and effect or the sequence of events

Cross sectional studies

Cohort

Cross section

Choose a design:

Aim: The incidence of breast cancer among Saudi women.A) Cohort studyB) Cross sectional studyC) Case control studyD) Randomized prospective controlledE) Meta-analyses

The answer is A

Aim: The incidence of breast cancer among Saudi women.A) Cohort studyB) Cross sectional studyC) Case control studyD) Randomized prospective controlledE) Meta-analyses

Cohort studiesCohort studies describe incidence or natural history.They analyze predictors (risk factors) thereby enabling calculation of relative risk.Retrospective cohorts where available are cheaper and quicker.Cohort studies measure events in temporal sequence thereby distinguishing causes from effects.

Confounding variables are the major problem in analyzing cohort studies.Subject selection and loss to follow up is a major cause of bias.

Choose a design:

Aim: Survival of patients with bronchial asthma who were infected with H1N1 flu.A) Cohort studyB) Cross sectional studyC) Case control studyD) Randomized prospective controlledE) Meta-analyses

The answer is C

Aim: Survival of patients with bronchial asthma who were infected with H1N1 flu.A) Cohort studyB) Cross sectional studyC) Case control studyD) Randomized prospective controlledE) Meta-analyses

Case-control studies

Case-control studies are simple to organiseRetrospectively compare two groupsAim to identify predictors of an outcomePermit assessment of the influence of predictors on outcome via calculation of an odds ratio

Case-control studies

Useful for hypothesis generationCan only look at one outcomeBias is an major problem

Overcoming sampling biasA convenience sample—sampled in the same way as the cases, for example, attending the same outpatient department.Matching—the controls may be a matched or unmatched random sample from the unaffected population.Using two or more control groups.Using a population based sample for both cases and controls.

Choose a design:

Aim: Establish guidelines for the management of multiple sclerosis.A) Cohort studyB) Cross sectional studyC) Case control studyD) Randomized prospective controlledE) Meta-analyses

The answer is E

Aim: Establish guidelines for the management of multiple sclerosis.A) Cohort studyB) Cross sectional studyC) Case control studyD) Randomized prospective controlledE) Meta-analyses

Lack of experimental studies in KSA

It requires intervention.Lack of volunteers.It requires major funding.Established research infrastructure.Multicenter fashion requires standardization of health care system throughout KSA.

Proposal

ProposalRequired for:Approval by ethical committee.Approval for fundingRecruitment of subjects.Monitoring the progress of the study.

Proposal

Manuscript writingAbstractAim of the studyContribution of the study to Health Methodology of the studyFunding expected to complete the studyStatistical analysis of the study

Appraisal

Appraisal

Study purpose:The aim of the study should be clearly stated.Did the conclusions clearly satisfy the aim of the study.Did the objectives of the study accomplished.

Sample:The sample should accurately reflect the population from which it is drawn.The source of the sample should be stated.The sampling method should be described and the sample size should be justified.Entry criteria and exclusions should be stated and justified.The number of patients lost to follow up should be stated and explanations given.

Control group:The control group should be easily identifiable.The source of the controls should be explained—are they from the same population as the sample?Are the controls matched or randomised—to minimise bias and confounding.

Quality of measurements and outcomes:Validity—are the measurements used regarded as valid by other investigators?Reproducibility—can the results be repeated or is there a reason to suspect they may be a “one off”?Blinded—were the investigators or subjects aware of their subject/control allocation?Quality control—has the methodology been rigorously adhered to?

Completeness:Compliance—did all patients comply with the study?Drop outs—how many failed to complete the study?DeathsMissing data—how much are unavailable and why?

Distorting influences:Extraneous treatments—other interventions that may have affected some but not all of the subjects.Confounding factors—Are there other variables that might influence the results?Appropriate analysis—Have appropriate statistical tests been used?

Statistical tests:

Relative risk reduction (RRR):

RRR = (CER – EER) / CERCER: control event rateEER: experimental event rate

The absolute risk reduction (ARR):

ARR = CER – EERThe number needed to treat (NNT):NNT = 1/ARRThe smaller NNT is the more cost effectiveness in clinical practice:NNT = 10 (treat 10 patients to save 1)NNT = 1000 (treat 1000 patients to save 1)

Statistical Assessment of the Data:

Disease True positive = a present absent False positive = b

Test + a b False negative = c - c d True negative = d

Prevalence = (a+c)/(a+b+c+d)Positive predictive value = a/(a+b)Negative predictive value = d/(c+d)Sensitivity = a/(a+c)Specificity = d/(b+d)Positive likelihood ratio = sensitivity/(1-specificity)Negative likelihood ratio = (1-sensitivity)/specificity

Critical appraisal:

Study has been reviewed for its validity and relevance to the physician and it is determined to have significant clinical applicability.

Can these results benefit the patient?

Conclusions

Conclusions

All medical research is the study of relationships among variables.Each study design serves certain objective.Observational studies need no intervention.Researchers require good grip of statistics.The main purpose of performing a study is to improve medical care and to serve our fellow humans in the best way.

Thank you