Study designs in medical research

1

• Study designs in medical research
• Dr. Nashwa Radwan
• Professor of Public Health and Preventive
  Medicine

2

• ??? ???? ?????? ??????

3
Learning Objectives

• To understand the concepts of different study
  designs
• To learn about the advantages and disadvantages
  of several study design.
• To be able to choose the best study design to
  assess certain health problem.

4

• What is research?

5

• Research is the manner by which we attempt to
  solve problems or to conform the validity of the
  solutions to problems

6
Purpose

• The basic aim of clinical research is to advance
  knowledge of human health and illness .
• 1-Assess the health status or clinical
  characteristics of a well defined population
• the annual rate of suicide in teenager .
• The incidence of rabies in certain locality.
• 2-Probe the natural history of the disease
• The clinical course of retinopathy in diabetics
  over 10 years follow up.

7

• Examine clinical decision- making processes
• The best screening test for glaucoma in general
  population.
• 4-Determine and assess treatment outcomes .
• Tumor response of laryngeal cancer in patients
  who receive radiation .
• 5-Identify and assess risk factors
• The incidence of lung cancer in smokers relative
  to non smokers

8
Principles used in assessing such data

• Efficacy describes the true treatment or
  intervention effect under ideal condition
• Effectiveness describes the true treatment or
  intervention effect under routine condition .
• Reliability describes the reproducibility of
  the test results .
• Bias is a systematic error that is
  unintentionally made .
• Sensitivity describes test positives or true
  abnormal correctly identified .
• Specificity describes test negatives or true
  normal correctly identified .

9

• What the types of epidemiological studies?

10
Observational Experimental

• Use an existing phenomena in attempt to
  understand aspects of health or illness . The
  investigator neither control the population nor
  the factors to which the population is exposed .
• a-Case report study.
• b-Case series study.
• a-Cross sectional or prevalence study.
• b-Retrospective or case-control study.
• c-Prospective or incidence study.

• Test the effect of some intervention on a certain
  aspect of health or illness . The investigator
  can control both the population and the factors
  to which the population exposed .
• a-Single blind.
• b-Double blind.
• c-Trible blind.

11
Types of Epidemiologic Observational Studies

• based on
• type of sampling from population - based on
  Exposure /or Disease or neither
• temporal sequence of observation - one time
  point, forward, backwards

12
Types of Studies

• A. Experimental - study factor is manipulated by
  the investigator
• Types
• 1. Pure versus Quasi-experimental
• 2. Laboratory versus real world
• B. Observational - no manipulation of study
  factor by the investigator
• 1. Descriptive versus Analytic
• 2. Retrospective versus Prospective

13
A-Case report

• -Description
• Is a brief objective report of clinical
  characteristic or outcome from a single clinical
  subject or event
• -Study question
• It is commonly used to report unusual or
  unexpected events
• -Examples
• A report of advanced diabetic retinopathy in a
  patient with no other clinical evidence of
  diabetes .
• -

14
Methodology

• A single event first must be identified .
• Data collection generally is retrospective ,with
  a review and a descriptive summary of subject or
  event .
• No statistical analysis or comparative group is
  included
• -Strengths and limitation
• It provides the first report of unexpected event,
  hypotheses for testing and definition of issue
  for further study. But the results are rarely
  generalized

15
B-Case series report

• -Description
• An objective report of a clinical characteristics
  or outcome from a group of clinical subjects .
• -Study question
• Report new diseases or health related problems .
• -Examples
• Is one that identified several children with
  birth defects who were born to mothers who had
  taken thalidomide.During pregnancy.
• -.

16

• -Methodology
• Subjects must be identified with regard to the
  clinical characteristics
• Data collection may be retrospective or
  prospective , but a control or comparison group
  is not included
• Descriptive statistics are calculated
• - Strengths and limitations
• Generalization of the results is limited because
  the selection of study subjects is
  unrepresentative and lack of comparison group

17
Cross-Sectional Study Prevalence Study

• Description
• Cross-Sectional Studies measure existing disease
  and current exposure levels.
• This study analyzes data collected on a group of
  subjects at one time rather than over a period of
  time The method of research that simply looks
  with intense accuracy at the phenomena of the
  moment (what is happening? Right now) and then
  describes precisely what the researcher sees.

18

• The point should be clearly emphasized that
  looking or seeing is not restricted to perception
  through the physical eye. In research we have
  many ways of seeing examples
• -The physician looks at the heart through a
  stethoscope and electrocardiogram.
• -The educator and the psychologist look at
  intelligence, attitude, believes through IQ
  tests, attitude scales and other evaluation
  scales.
• -The epidemiologist looks in survey through
  interview and questionnaire.

19
(No Transcript)
20

• sample without knowledge of Exposure or Disease
• sample at one point in time
• Mostly prevalence studies/surveys

21

• -Study question
• Concern the occurrence of the disease and the
  rate of other events In a single point of time(
  the prevalence)
• -Example
• The rate of occurrence (prevalence) of smoking
  among high school students at a particular point
  of time.
• -

22

• Methodology
• The population for the study must be carefully
  chosen, clearly defined.
• Target population is surveyed at a single
  (particular) point of time (e g. one year)..
• The rate of occurrence of diseases and events
  among the target population in a particular point
  of time (e.g.the prevalence) is calculated.
• No statistical analysis is included.

23

• Strengths and Limitations
• -It is quick, cheap and easy
• -True rates are determined (the prevalence).
• The prevalence rate is used as a measure of
  health status in the population of interest and
  for comparisons between different populations and
  communities.
• Can study multiple exposure and multiple
  diseases.

24

• Impractical for rare diseases
• Not a useful for establishing causal
  relationships. It does not allow us to answer the
  question ,which came first (which caused which)
• No control over sample size for each exposure by
  disease subclass.
• Data are particularly susceptible to distortion
  through the introduction of bias into the
  research during sampling questionnaire and
  interviewing.

25
How Bias Enters the Research Design

• Example
• A researcher decides to use a city telephone
  directory as a source for selecting a random
  sample.He will open a page at a random, close his
  eyes, put the point of the pencil down on the
  page, and the names that comes to the pencil
  point will be selected.you can not get much more
  random than that. But the evil spirit of bias is
  there. HOW?

26
D- Case control study Retrospective Study

• -Description
• An observational study in which diseased and non-
  diseased (controls) subjects are identified after
  the fact and then compared regarding specific
  characteristics to determine possible association
  of risk for the disease .
• --

27

• Case-Control Studies identify existing disease/s
  and look back in previous years to identify
  previous exposures to causal factors.
• Analyses examine if exposure levels are
  different between the groups.

28
(No Transcript)
29

• Cases are those who have a disease.
• Controls are those without a disease.
• look back in previous years to identify previous
  exposures to causal factors.

30

• Study question
• Address the issue relating to risk factors for
  disease by comparing the effect of a particular
  factor in diseased population to that in
  non-diseased population .
• -Examples
• Comparing the rate of smoking in normal subjects
  and patients with lung cancer.

31
Methodology

• Find cases of the disease and obtain from them or
  about them the history of the assumed cause
  exposure.
• Find control persons similar to the cases, and
  similarly obtain histories of exposure from them.
  
• Statistical analysis is done between both groups
  to determine that the difference between two
  groups is real not to chance .
• Conclusion are useful for generation of
  hypotheses and for initial evidence of risk
  association

32
Strengths and limitation

• -Cases are easily available (usually hospital
  based) . Thus, its the method of choice for rare
  diseases.
• -Historical data are often available from
  clinical records of patients, So it is possible
  to use these data without obtaining further
  information from case or controls.
• It is quick and cheap as the disease of interest
  has already occurred.
• -It tends to support, but not prove, causal
  hypotheses by establishing associations of risk
  factor.

33

• -Bias in data collection
• - The study is un masked. ( since the presence or
  absence of disease is known to the subject and
  the observer).
• -Recall bias may also occur because exposure to
  risk factor is often dependant on memory of
  subjects.
• -Bias in the selection of subjects
• (Non-representative ness of cases) since a case
  control study is not population based such
  fallacy is called( Berkson Fallacy).
• -The occurrence of the assumed cause is obtained
  from the history of the selected cases and
  controls Thus we can not know which occurred
  first? (Which Caused Which) .

34
E- Cohort Prospective-Incidence Study

• -Description
• It is an observational study in which exposed and
  non exposed population are identified and
  followed prospectively over time to determine the
  rate of a specific disease or event .
• -Study question
• Address issues relating to risk factors by
  comparing population exposed to a factor under
  study to population that are not exposed .
• -Examples
• The incidence of lung cancer in smokers to
  nonsmokers .

35
(No Transcript)
36

• Exposed and non exposed population.
• Follow prospectively over time.
• Measure the incidence rate.

37
Methodology

• a-A population of exposed and non exposed must be
  determined and followed prospectively for a
  certain period of time(e.g. one year)
• b- Exposure to risk factor under study.
• c-The incidence rate of disease or an event for
  both exposed and non exposed must be calculated
  and the relative risk is measured .
• d-Statistical comparison is made between exposed
  and non exposed to assess that the observed
  difference is real not due to chance

38
Strengths and limitation

• -Costly and time consuming due to the need for
  large number of subjects and prolonged time of
  follow up
• -Potential bias in recall and selection is
  lessened
• -Allows study of rare diseases.
• -The results is considered more conclusive than
  case- control study
• -Allows for determination of a population based
  rate of new disease or event(e.g.the Incidence
  Rate) and the Relative Risk .

39

• How do we calculate the Incidence Rates(IR)?
• How do we calculate the Relative Risk(RR)?

40
F- Clinical trial

• -Description
• It is an experimental study used to assess the
  differences between two or more groups receiving
  different interventions or treatments .
• -Study question
• To compare outcomes between different
  intervention .

41
Examples

• A comparisons between chemotherapy to
  chemotherapy plus radiation for laryngeal
  carcinoma .
• Testing different types of vaccines on population
  of children to prove their efficacy in preventing
  the diseases.
• -Testing different forms of health service
  delivery.

42

• -Types
• -Single blindWhen the studied persons do not
  know whether they belong to the treatment or the
  control group.
• -Double blind When this knowledge is not known
  by both the studied persons and the experimenter.
• -Triple blindWhen this knowledge is not known by
  all the three parties the studied persons, the
  experimenter, and the statistician.

43

• -Methodology
• a-A population with a clinical characteristic
  requiring intervention must be determined .
• b-Subjects must be allocated , randomly , to each
  of the treatment intervention .
• c-Treatments are administrated in an identical or
  controlled manner .
• d-Treatment outcomes and other results ( side
  effect , costs , benefits ) must be measured .
• e-Statistical analysis is made to compare the
  rates of measured treatment outcomes .

44

• Strengths and limitations
• a-A clinical trial allows for control of all
  variables ,which can also affect outcomes under
  investigation .
• b- An insufficient number of subjects may lead to
  failure to detect true differences .

45
META_ANALYSIS STUDY

• -Description
• Meta- analysis seeks to combine the results of
  several clinical studies on the same issue to
  derive definite conclusion from the varied and
  sometimes contradictory results.
• Similar to a review article, a meta- analysis
  begins with a literature review identifying
  studies of a similar research question.
• Unlike a review article, meta- analysis attempts
  to analyze statistically the aggregate results to
  drive a single integrated conclusion.

46

• Study question
• Describe on the best clinical approach to a
  problem based on several related studies
• Examples
• -Fibrinolytic therapy for acute myocardial
  infarction.
• -Postoperative irradiation in breast cancer
  treatment.
• -Steroid treatment of acute hepatitis.

47

• D-Methodology
• Formulation of meta-analysis includes qualitative
  and quantitative techniques.
• -Qualitative aspects include
• a-Developing the research question to be
  analyzed.
• b-Reviewing the literature for relevant clinical
  studies.
• c- Evaluating the studies to identify
  similarities or to explain differences among
  them.
• d-Select the best clinical outcome.
• -Quantitative aspects include
• Statistical analysis of the results.

48
What type of study to chose depends on

• what is the research question/ objective
• Time available for study
• Resources available for the study
• Common/rare disease or production problem.
• Often there are multiple approaches which will
  all work
• Choosing an established design gives you a huge
  head start in design, analysis and eliminating
  biases

49
Ethics of clinical studies
50
Definition

• Ethics of clinical studies concern the balance
  between risks and benefits of individuals who
  participate as subjects in research studies.

51
Protection of human rights

• Any person who is requested to consent to
  participate as a subject in a research study has
  the right to
• Be informed of the nature and purpose of the
  experiment .
• -Be informed about the steps to be followed in
  the experiment .
• Be informed of any attendant discomforts and
  risks to be expected from the experiment .
• Be given an explanation of any benefits to the
  subject to be expected from the experiment .
• Be informed of medical treatment to the subject
  after the experiment if complication arise .

52

• Be given an opportunity to ask any questions
  concerning the experiment or the procedures
  involved .
• Be instructed that consent to participate in the
  medical experiment may be withdrawn at any time ,
  and the subject may discontinue participation in
  the experiment without prejudice.
• Be given a copy of a signed and dated written
  consent form when one is required
• Be given an opportunity to decide to consent or
  not to consent to a medical experiment without
  the intervention of any element of force .

53

• Problems

54
Problems 1

• For each case history described below, select the
  study design it most appropriately illustrates.
• A-Case report
• B-Case- control study
• C-Clinical trial
• D-Cohort study
• F-Case series report
• -

55
A

• A total of 300 newly diagnosed patients with
  laryngeal cancer are allocated to treatment with
  either surgical excision alone or surgical
  excision plus radiation treatment.

56
B

• Patients admitted for carcinoma of the stomach
  are age and sex matched with fellow patients
  without a diagnosis of cancer and surveyed as to
  smoking history to assess the possible
  association of smoking and gastric cancer

57
c

• -A 39 years old man who presents with a mild
  fever, sore throat, malaise, and headache is
  treated with penicillin for presumed
  streptococcal infection. He returns after a week
  with hypotension, fever, rash, and abdominal
  pain. He responds favorably to chloramphenicol,
  after a diagnosis of Rocky Mountain spotted fever
  is made.

58
d

• A total of 3500 patients with thyroid cancer are
  identified and surveyed by patient interviews
  regarding past exposure to radiation.

59
e

• A total of 3000 elderly people living in Tanta
  city were identified and followed up for one year
  to determine the frequency and risk factors for
  falls among elderly people.

60
Problem 2

• In a prospective study, 500 children were
  followed up for 20 years to assess the risk for
  thyroid cancer. 100 children were exposed to
  radiation of neck ,and from this group, 7
  subjects developed thyroid cancer. Also 3
  subjects developed thyroid cancer from the non-
  exposed group.Calculate the incidence rate(I R)
  and the relative risk (R R) for thyroid cancer.

61
Problem 3

• In Pediatric Department Tanta University, the
  staff decided to determine the percentage of
  medication compliance among epileptic children
  attending the out patient clinic for follow up
  and obtain their medicine. How would design a
  study to achieve this objective?

62
Problem 4

• There is evidence that lead exposure has a
  significant effect on children intellectual
  abilities and academic performance. How would you
  design a study to examine this effect of lead on
  a primary school children in Tanta city?

63

• Thank you