You are on page 1of 19

Study Design

Types of research
designs
A-Observational studies: in which the
investigator doesn’t have any control on the
factors under study, he just observes and
collect data, this category includes the
following designs:

1- ecologic
2- descriptive
3- cross sectional
4- case control
5- cohort
B- intervention / experimental
studies
All the factors related to the studied
relationship are controlled by the
investigator. They include

lab experiments -1
clinical trials -2
community interventions -3
:Ecologic studies
Purpose: generation of new hypothesis, evaluation of an
intervention and study of secular trends
It is used to develop hypothesis about possible causes of -
disease occurrence. The presence of a risk factor can be
measured in different population and compared to the
incidence of a particular disease
It is called ecologic because the analysis is at the level of -
the entire population rather than the level of individual
persons (the sampling and analytical units are groups not
(individuals
it is called also correlation studies because it seeks to-
determine the extent to which two characteristics
(risk factor and disease occurrence) are related
Ecologic studies
It helps to generate hypothesis about the link between - -
two diseases (ex, AIDS and TB) but it can not establish a
causal relationship between them, studies that are
designed to test the likelihood of a cause and effect
relationship between a risk factor and a disease are
termed hypothesis-testing investigations (case control,
(cohort

Ecologic fallacy: is the inability to infer a causal -


explanation from a correlation study. This occur when
a suspected risk factor and disease occurrence are
associated at the population level but not the
individual subject level
:Ecological fallacy
Pop. May have high incidence rates of
AIDS (risk factor) and TB (disease
occurrence) without the same
persons being affected by both
conditions
How to analyze ecologic
studies
geographic: maps outlining the distribution of -1
the exposure factor are compared with maps
describing the distribution of the suspected
disease or outcome

time trend: rates are plotted against time to -2


study the effect of an intervention, a risk
factor or any other phenomenon ( secular
trend of Tb in relation to socio-economic
changes and discovery of anti-
tuberculous drugs
How to analyze ecologic
studies
Correlation: average exposure levels are -3
plotted against disease rates in several
(groups ( breast cancer and fat consumption

multiple group comparisons: average exposure -4


levels of multiple agents and the suspected
disease rates are plotted against time ( study
of the association between lung cancer and
(coal, cigarette and fuel derivatives
:Advantages
short time -1
low cost -2
good source of a new hypothesis -3
best for the evaluation of an -4
intervention
:Descriptive studies
it is used for the description of a disease, risk factor, or -
health phenomenon in a pop., and for the generation of a
new hypotheses and there is no control group

:Types
Purposive
-1case study (report): there is no study groups, nor
statistical analysis, data collection is retrospective. It
is a brief report of a clinical characteristic from a
single clinical subject. It is the 1st evidence of
unexpected or unusual event
:Descriptive studies
:case series -2
data collected from a group of individuals selected-
for possessing a certain common characteristic, it
is most commonly used to describe clinical
characteristics such as signs and symptoms of
disease, patients having the signs were identified
and data collection may be retro or prospective

Frequency of certain factors or phenomena can be -


calculated ( you can not do comparison with
previous data because there is no random
selection
:Descriptive studies
Probability
:Surveys
e.g. for the determination of the prevalence of
hypertension in a certain community, a random sample
of the population will be selected and assessed for the
presence or absence of the disease. You can determine
the prevalence rate and you can compare with previous
reports rates because of random sampling
:Advantage
:Case report, case series
Short time -1
low cost -2
add to the core of the knowledge -3
might generate a new hypotheses -4

:Survey
important for description of the health and -1
disease phenomena in the community
might be used for comparison -2
Cross-sectional study
It is used for the study of the association between two -1
or more variables, and for the generation of a new
hypotheses

there is no a priori selection of the control group, -2


sample is randomly selected from the pop. Regardless of
the exposure or disease status, sample could also be
selected from records

exposure and outcome are measured at the same time in -3


the sample, it does not take into account the temporal
relationship between factors studied

individuals are classified according to exposure and -4


outcome status into four categories
population

sample

-+
++ +- --
ED
ED ED ED

- - + +
E= exposed D= diseased E= not exposed D= not diseased
Prevalence odds ratio
((POR
disease
- +
+ a b
exposure
C d
-

POR= ad/bc
Cross-sectional study
:prevalence rate
it is the number of existing cases in the -1
.population P=C/N
it can be point prevalence or period-2
.prevalence
Because it contain all cases so it is a measure-3
of the amount of the problem and can be used
to determine the health care needs of that
community
it is affected by the incidence and duration -4
(of illness (p= Ix D
Advantage of cross
:sectional studies
relatively short time -1
low cost -2
may provide data on prevalence -3
may serve as time zero for a cohort study -4
might generate a new hypotheses -5
:Problems
representativeness is the cornerstone for -1
generalization of data on prevalence

absence of temporal relationship between exposure -2


and outcome, since both were measured at the same
point in time

prevalence-incidence bias: dependence on prevalence -3


cases will miss all fatal cases and short course cases

weak evidence of association and worse for -4


causation

You might also like