Category: Education

Presentation Description

No description available.


Presentation Transcript




OVERVIEW Epidemiology as basic concept Definition, Components, Types , Difference Experimental epidemiology Aims ,Advantage , Disadvantage , Uses, Animal & human experiment Ethical issues in intervention studies What is mean by RCT ? Steps in planning, designing and conduct of RCT

Slide 3: 

Types of RCT a) clinical trial Definition ,phases ,variant of designs b) field trial c) risk factor trial d) community intervention trial e) health services evaluation trial f) cessation experiment Non randomized trial Definition , Types




EPIDEMIOLOGY AS BASIC CONCEPT (Greek; Epi = upon, Demos = populations, Logos = scientific study). “The study of the frequency, distribution and determinants of diseases and health - related states and events in human populations” and the application of this knowledge in prevention, control and mitigation of these problems .



Classification Chart of Epidemiological Designs : 

Classification Chart of Epidemiological Designs

Difference between observational & experimental studies : 

Difference between observational & experimental studies



Slide 13: 

WHAT IS AN EXPERIMENT ? An experiment is set of observations, conducted under controlled circumstances ,in which scientist manipulates the conditions to ascertain what effect ,if any, such manipulations has on observations.

Slide 14: 

For epidemiologist the word , experiment usually implies that the investigators manipulates the exposure assigned to participants in the study (as subject for manipulations is human ). Experimental design it is a model (paradigm) scenario in epidemiology.


AIMS FOR EXPERIMENTAL STUDIES To provide scientific proof of etiological /risk factors which may permit modifications or control of those disease To provide a method of measuring the effectiveness and efficiency of health services for prevention , control, treatment of disease & improve health of the community


ADVANTAGES OF EXPERIMENTAL STUDIES Scientifically ideal method. Removes a large number of biases related to selection and measurement. Controls for confounding randomization it fulfils the basic dictum of research, which says “all other things being equal, it is the exposure of interest that has made the difference in the “outcome”. Ensures temporal relationship between exposure and outcome. Builds up “faith” in the findings of the study.


DISADVANTAGE OF EXPERIMENTAL STUDY In many situations, especially those which concern study of “risk factors” or “prognostic factors”, one can not “randomly” allocate human beings into two groups; The ethical issue; at times it may not be ethical to randomly divide, thus exposing the ‘exposed’ group to a potentially harmful treatment or procedure; or to deprive the ‘non exposed’ group of a potentially useful measures. Other problem like cost , feasibility


USES OF AN EXPERIMENTAL DESIGN Studying the efficacy of a therapeutic procedure (drug, surgical procedure etc.); Studying the efficacy of a preventive procedure (vaccine, sera, chemoprophylaxis, personal protective masure, exercice programme etc.); Studying the efficacy of a health care system or procedure (e.g. Domiciliary v/s. Institutional care).

Slide 19: 

Experimental studies can be conducted in animals human beings


ANIMAL STUDIES Applications – Confirming the etiological hypothesis Study of pathological phenomena by experimental reproduction of human disease in animals Testing efficacy of preventive &therapeutic measure Completing natural history of disease

Slide 21: 

ADVANTAGE Bred in labs & multiply rapidly Manipulated easily according to wishes of the investigators Enable to carry out certain experiment LIMITATIONS Not all human disease can be produce in animals Conclusions drawn from this not strictly applicable to human beings


HUMAN EXPERIMENT The first experiment (clinical trial ) in clinical medicine… Dr James Lind Vitamin C and scurvy


DIFFICULTIES & ETHICS Patients availability inclusion & exclusion criteria Willingness to participate Compliance Attrition

Slide 25: 

Data adverse effects influence of other factors random variation of effects between people Bias allocation bias assessment bias analysis Cost


ETHICAL ISSUES Interventions may be harmful, as well as helpful, and participants are asked to undergo potential hazards, discomforts, and expenditure of time, the question being addressed in any clinical trial must be important 'Clinical equipoise' (freedman 1987). That is, there must be uncertainty as to the usefulness of the intervention among those knowledgeable about the intervention.

Slide 27: 

Ethics addresses issues of conduct among members of any group in society. MEDICAL ETHICS is founded upon four principles Respect for autonomy refers to the individual’s right to self determination and respect for human dignity and freedom. This includes the need to tell the truth (veracity) and to be faithful to one’s commitments (fidelity). Non-maleficence refers to taking actions that will not result in harm, derived from the ancient medical maxim, primum non nocere (first, do no harm).

Slide 28: 

Beneficence refers to the need through one’s intended actions to do good, which members of the public health professions like to think is the main function of public health; although, sometimes we are viewed by others as “do-gooders,” interfering busybodies whose paternalist interventions are unwanted and sometimes resented. Justice refers to social and distributive justice, requiring fairness in the distribution of risks and benefits, and to the need for equity and impartiality across all members of the greater community.




RANDOMIZED CLINICAL TRIAL Ideally, an experimental design should have the three essential elements Randomization Controls Blinding This is what is called the “Randomized, Controlled, and Blinded Trial” (RCT).

Slide 31: 


Slide 32: 

This is scientifically the strongest design, because : The possibility that “subjects might have taken up the exposure due to natural selection factors which may also be related to the outcome”, is ruled out by random allocation. The absolute requirement of “temporality” for a “cause – effect” relationship is fulfilled. There is no possibility of recall ‘bias’ . the experimental design must be used answer the questions about “treatment (therapy)” or “preventive procedure since the subjects can be randomised into 2 groups, provided it is ethically correct do so.

: Strength of Evidence Based on Type ofEpidemiological Designs : 

: Strength of Evidence Based on Type ofEpidemiological Designs

Slide 34: 

Problem while studying “risk factors”, “markers” and “prognostic factors” it is impossible for the investigator to randomise the subjects into two groups – one getting the exposure and the other not. E.g. in a study of the association between cigarette smoking (exposure) and Lung CA (outcome), it is impossible for the investigator to “randomly allocate” the subjects into 2 groups, one group being told to smoke and other being told not to do so” LIMITATION

The first randomized trials… : 

The first randomized trials… Sir Ronald Fisher

Slide 36: 


Slide 37: 

Step 1 – Deciding Step 2 - Clearly state the research question and the variables of study Step 3 - Enumerate the inclusion and exclusion criteria Step 4 - Defining the populations Step 5 - Sample Size Calculation Step 6 - Detailed Descriptions of Measurement Protocols

Slide 38: 

Step 7 - Enrolling the Participants Step 8 – Randomisation Step 9 - Introduce the Intervention and Placebo Control Modalities Step 10 - Ensure Blinding Step 11 - Follow Up and Assessment Step 12 - Statistical Analysis

Step 1 – DECIDING - Is clinical trial really required? Should it be done? Can it be done? : 

Step 1 – DECIDING - Is clinical trial really required? Should it be done? Can it be done? One should carefully assess whether it is really necessary to undertake a clinical trial. Detailed review of literature should be undertaken Whether the trial is ethical ?

Step 2 – Clearly state the research question and the variables of study : 

Step 2 – Clearly state the research question and the variables of study The investigator should clearly defined research question Primary question – specified in advance Use for sample size calculation Secondary questions Are subsidiary to the primary question Help the investigator to understand the mechanism of action of intervention

Slide 41: 

Investigator should clearly list out as to all “variables” Broadly, four categories The exposure variable : This means the “intervention” under study. The complete details including the dosage, method of administration etc. should be clearly defined Co - Interventions : Co - interventions should be as clearly defined as primary exposure variable

Slide 42: 

The outcome variable : clearly define one “primary outcome” or the “major endpoint” variable around which the analysis and sample size calculations would revolve. “secondary outcome” or the “other endpoint” variables can be defined The confounder variables : No need to measure the confounder variables in a clinical trial (randomization ) . But in small size trials & to do a baseline comparison between the intervention and the control group in respect of important confounding variables collect data on considering this

Step 3 – Enumerate the inclusion and exclusion criteria: : 

Step 3 – Enumerate the inclusion and exclusion criteria: “Inclusion criteria” for entry into the trial “Exclusion criteria” – for those who will not be eligible to be included in the study.

Step 4 – Defining the populations : : 

Step 4 – Defining the populations : Reference population /Universe/Target population This is the very large collection of patients or subjects to whom the results of the study would be generalized.

Step 5 – Sample Size Calculation : : 

Step 5 – Sample Size Calculation : Determining sample size must be planned carefully If samples are too large may waste research time, resources, patient effort and money If samples too small may lead to inaccurate results

Slide 46: 

Sample size calculation depend on    1.    The difference in response rates to be detected    2.    An estimate of the response rate in one of the groups    3.    Level of statistical significance (α)     4.    The value of the power desired (1 - β)    5.    Whether the test should be one-sided or two-sided

Slide 47: 

For qualitative data Sample size is 4pq L2 for quantitative data sample size is 4 p q f(1.1) L2 X vnh P-prevalance of indicator , L-margin of allowable error , q- 1-p, f-design effect , v- proportion of target population , nh= average house hold size


SAMPLING METHODS Probability sampling Simple random sampling Stratified random sampling Multistage random sampling Multiphase sampling Cluster sampling Non probability sampling Incidental sampling Quota sampling Sequential sampling

Slide 49: 

RANDOMIZATION Reference population Study population New t/t Current t/t External validity /Generalizability Internal validity

Slide 50: 

EXTERNAL VALIDITY / GENERALISABILITY / REPRESENTATIVENESS extent to which results of investigation can be generalized to other samples & situations TYPES 1) population validity generalized results to population from which it was drawn 2) ecological validity generalized to other situation

Step 6 – Detailed Descriptions of Measurement Protocols : 

Step 6 – Detailed Descriptions of Measurement Protocols Develop the detailed protocols of clinical procedures, laboratory investigative procedures, as well as the details of questionnaire and interview protocols

Step 7 – Enrolling the Participants : 

Step 7 – Enrolling the Participants

Step 8 – Randomisation : : 

Step 8 – Randomisation : “Heart” of control trial Assigning a participant/subject/patient to a treatment group based on chance rather than choice Introduces unpredictability or randomness

Advantage : 

Advantage To have comparable groups similar in baseline characteristics Removes potential bias in allocation Requirement for most statistical tests to be valid Participants are enrolled in the same time period therefore, temporal trends in care or in the nature of the condition being studied are equal in the two groups. Also guarantees that statistical test of significance will be valid

Types of randomization : 

Types of randomization Fixed allocation procedures. e.g. if 20 participants are needed for study then simply by tossing the coins Blocked randomization equal no. of participant in the group is guaranteed after several is enrolled .with block randomization equal no. are ensured throughout enrolment period Stratified randomization balance between group A&B is not only in no. but in kind of participants Adaptive randomization procedures. the likelihood of randomization to one or another group changes based on occurrence of study

Step 9 Introduce the Intervention and Placebo ControlModalities : 

Step 9 Introduce the Intervention and Placebo ControlModalities one group- Trial modality other group- The control or the baseline Ensure ways to bring about compliance in both the groups. Ensure a placebo control

Step 10 – Ensure Blinding : 

Step 10 – Ensure Blinding Single blind trial Double blind trial Triple blind trial

Step 11 – Follow Up and Assessment : : 

Step 11 – Follow Up and Assessment : Losses to follow up, which may otherwise seriously bias the study results Keep the “stoppage rules”-- Evidence comes up in between against the intervention modality. Evidence of clearly high mortality or complication in the intervention group comes up.

Step 12 – Statistical Analysis : Statistical analysis is indeed : 

Step 12 – Statistical Analysis : Statistical analysis is indeed present the participant’s recruitment and flow chart, giving the actual data at each step. give a table showing baseline comparison between 2 groups clearly states primary exposure and outcome variables (E+; E-; /O+; O- ) make a 2X2 table

Slide 61: 

Calculate the incidence of outcome in the exposed and non exposed group as Calculate the risk ratio (RR) i.e. The “effect” of the intervention as : RR (effect) = IE/ INE. Calculate the 95% CI of RR Calculate the numbers needed to treat NNT= 1/ (IE- INE) (Where IE and INE are measured as proportions out of 1) Next step is to undertake probability testing procedures

Types of RCT : 

Types of RCT


CLINICAL(THERAPUTIC ) TRIALS The International Conference on Harmonisation defines a clinical trial as 'any investigation in human subjects intended to discover or verify the clinical ,pharmacological, and/or other pharmacodynamic effects of an investigational product(s), and/or to identify any adverse reactions to an investigational product(s), and/or to study absorption, distribution, metabolism, and excretion of an investigational product(s) with the object of ascertaining its safety and/or efficacy' (ICH 1996).

Slide 66: 

clinical trial may be defined 'a prospective study comparing the effects and value of intervention(s) against a control in human beings' (Friedman et al. 1998). The “unit of study” in a clinical trial are “patients” suffering with a given disease, the therapy of which is to be studied.

The first randomized trial in clinical medicine… : 

The first randomized trial in clinical medicine… Patients with pulmonary tuberculosis Streptomycin No treatment Medical Research Council. Streptomycin in tuberculosis trials Committee. Streptomycin treatment of pulmonary tuberculosis. BMJ 1948; ii: 769-82 Sir Austin Bradford Hill


PHASES OF CLINICAL TRAIL Drug discovery Exploratory development Full development Chemistry Preclinical Phase I Phase II Phase III Phase IV Laboratory Animals Humans Healthy volunteers Patients

Phase I studies : 

Phase I studies First in man In small no (20-80) Primarily done to establish the safety and tolerability of the new drug Healthy volunteers (history, clin exam, lab tests) Except cancer Toxic agents

What dose to give? : 

What dose to give? Dose toxic dose minimally effective dose therapeutic index NOAEL: No Observed Adverse Effect Level MABEL: Minimum Anticipated Biological Effect Level

Phase II studies : 

Phase II studies In large no of people (100-200) Looking for pharmacokinetic effects & pharmacodynamic effects of the drug, therapeutic efficacy determine the dosage range for future phase III studies safety of the drug

Phase III studies : 

Phase III studies The actual, classical stage of clinical trial Also known as the Randomised Controlled trial (RCT). Following phase III, the drug is marketed and simultaneously phase IV also starts.

Phase IV studies : 

Phase IV studies “Post Marketing Surveillance”. Data on the effect of the drug or procedure is collected from various agencies. Side effects which did not appear in phase - III, are detected in this phase and the drug may be withdrawn or its usage modified. Classical examples Thalidomide,

Variants in the Design of Clinical Trials : 

Variants in the Design of Clinical Trials Parallel design Cross-over design Randomized replication design Factorial design Latin square design

Randomized replication design : 

Randomized replication design

Randomized block design : 

Randomized block design

Factorial design : 

Factorial design

Latin square design : 

Latin square design

Efficacy and effectiveness of trails : 

Efficacy and effectiveness of trails Efficacy trial attempts to evaluate whether an intervention works under reasonably optimal circumstances i.e. if active drug is taken by all in the intervention group and almost no one in the control group takes the active drug will alter some clinical outcome Effectiveness trial allows for non adherence to the assigned treatment , it resemble what is likely to happen in actual practice Sometimes it is not possible to distinguished between this two

Slide 83: 

The efficiency and effectiveness and cost of a clinical trial depend on: • Response to each treatment • Influence of other factors such as age, gender or life style • Number of patients • How patients are selected for the trial • How patients are allocated to treatments • Type of trial: parallel or crossover • Compliance of patients to treatments • How data are recorded, analysed and interpreted


FIELD TRIAL The unit of study are healthy individuals, usually in the community. Usually undertaken in respect of a preventive procedure as a vaccine,sera, chemoprophylaxis, personal protective measures, e.g. Tuesday, January 18, 2011 polio vaccine trial


RISK FACTOR TRIAL Differ from preventive trial The intervention is not an actual physical administration but rather an abstract phenomena Asking a group of subjects (randomly selected) “regular physical exercise”, here, regular physical exercise is the “intervention” of interest which is not physically administrated (like a vaccine or drug) but is rather a “conceptual” procedure


COMMUNITY INTERVENTION TRIAL community intervention trial, where “random allocation” is done at the level of “communities” or “groups” of subjects, though the assessment of outcome is done at individual subject’s level. Such trials are of special interest for public health administrators for evaluating the effectiveness of environmental procedures, health educational measures, etc. e.g. north karelia project, stanford five city project


HEALTH SERVICES EVALUATION TRIAL The efficacy or effectiveness of health services or health policies; Architecture is the same as that of community intervention trials, with an added element of health economic analysis e.g. Evaluation of domiciliary t/t in T.B. ;


CESSATION EXPERIMENT An attempt is made to evaluate the termination of habit (or harmful factor is “removed” ) from the intervention group e.g. to study the role of cessation smoking in primary prevention



Slide 91: 

one may not be able to “randomly” allocate subjects into two groups for ethical administrative. At times the “placebo” part of control may be impossible (e.g. surgical v/s medical management of IHD). When disease frequency is low & natural history is long (Ca cervix) Are also called as “Quasi - experimental” studies


THE DISADVANTAGES OF QUASI - EXPERIMENTAL DESIGN High potential that the intervention and control group used for comparison are not similar to each other. So chance of confounding bias. Selection” factors may be operating;( e.g. pts who are taken into surgical treatment group for IHD may be in a much better state of cardiovascular function as compared to medical treatment group). Improvement noticed in a ‘before and after trial’ may simply be because other patient management techniques may also have improved recently; or else because the data collected earlier was incomplete


TYPES UNCONTROLLED TRIALS Trials with no comparison groups Even in these trials one may use historical controls (experience of earlier pt having same disease ) E.g. studies on cervical cancer screening pap smear is effective in reducing mortality


NATURAL EXPERIMENTS Some experimental studies are not possible in human . Natural circumstances may mimic as an experiment Populations involved in natural experiments comprise following groups – a) migrants b) religious & social group c)atom bombing in Japan d) earthquake E.g. John snow discovery of cholera as water borne disease

Slide 95: 

Lambeth company – supply from River Thames Sothwark & -sewage polluted water basin Vauxhall company


BEFORE AND AFTER COMPARISON STUDIES Two types Without control With control


BEFORE AND AFTER COMPARISON STUDIES WITHOUT CONTROL Events which took prior to the use of new t/t or preventive procedure used as standards for comparison i.e. experiment serves as its own control e.g. prevention of scurvy among soldiers by James Lind in 1750 by providing fresh fruits. death rate due to car accidents after introduction of compulsory seat belt legislation was compared, in Australia, with the death rates before such legislation.

Effect of adoption of compulsory seat belt legislation in Victoria , Australia : 

Effect of adoption of compulsory seat belt legislation in Victoria , Australia


BEFORE AND AFTER COMPARISON STUDIES WITH CONTROL In absence of control group may give misleading results Epidemiologist tries to use natural control group in e.g. cited seat-belt legislation in Victoria Australia ,compare with other states in australia


REFERENCES Park k. ‘Textbook of Preventive and Social Medicine’ 20 th edition 2009 Bhanot publications p.n. 76-82 Dr J.P. Baride Dr. A.P. Kulkarni.‘Textbook of community medicine’ 3rd edition Vora publications 2006,p.n. 205-207,139-143,148-152 ‘Oxford Textbook of Public Health’ 5th Edition by oxford press publications Lawrence M. Friedman and Eleanor B. Schron‘Methodology of intervention trials in individual page no.527-535 ‘Oxford Textbook of Public Health’ 5th Edition by oxford press publications Allan Donner Methodological issue in the design and analysis of community intervention trials’ Rothman, Kenneth Greenland, Sander; Lash, Timothy L ‘Modern Epidemiology’ 3rd Edition Lippincott publications 2008 ;. ‘Types of Epidemiologic Studies

Slide 102: 

Wallace/Maxcy-Rosenau-Last ‘public Health & Preventive Medicine 15 th edition Mc Graw Hill Publications 2008 Ethics and Public Health policy p.n. 29-31 Leon Gordis ‘Epidemiology’ 4th edition Saunders Elsevier publications 2009 Bhalwar Rajvir . ‘Textbook of Public Health And Community Medicine’ published by dept of community medicine , AFMC,pune 2009 ‘Experimental (Intervention) Studies’ p.n.151 Fletcher H., Wagner Edward ‘Clinical Epidemiology The Essentials’ 3rd edition p.n. 191, Brownson R. Diana b. Petitti ‘Applied Epidemiology Theory to Practice’ oxford university press 1998 page no 278 Dr. Bhaskara Rao. ‘Methods in Medical Research’ first edition 2002 Kothari C.R. ‘Research Methodology Methods and Technique’ second revised editions New age international publishers 2004

Slide 103: