Post on 05-Aug-2015
KNOWLEDGE FOR THE BENEFIT OF HUMANITYKNOWLEDGE FOR THE BENEFIT OF HUMANITY
PUBLIC HEALTH AND EPIDEMIOLOGY (HFS3063) Epidemiological Study Designs:
CASE CONTROL
Dr. Dr. MohdMohd RazifRazif ShahrilShahril
School of Nutrition & Dietetics School of Nutrition & Dietetics
Faculty of Health SciencesFaculty of Health Sciences
UniversitiUniversiti Sultan Sultan ZainalZainal AbidinAbidin
1
Topic Learning Outcomes
By the end of this lecture, students should be able to;
• describe case control study design.
• explain the advantages and disadvantages of case control study design.
2
Case-control studies • Purpose is to establish association between exposure to
risk factors and disease.
• Members of the population with the disease are selected into the study at the outset and risk factor information is collected retrospectively – Known as CASES
• A second group of individuals who do not have the disease is also included in the study – Known as CONTROLS
• Often used in the study of rare disease or preliminary study – Where little is known about the association between the risk
factors and disease of interest
3
(cont.) Case control studies • Case-control studies are prone to bias and confounding
• To minimize bias care must be taken in
– the selection of cases and control
– establishing definitions of disease, risk factors
– ensuring there are no confounding associations between
detection of disease and risk factors exposure.
6
Choice of cases • Care must be taken when choosing cases for the study.
• It is important to distinguish between stages or subtypes
of disease and to define a measure of health status
– E.g. when studying physical activity it is important to define what
is meant by physical activity in terms of types, nature and level.
• It is also important to establish whether interest is in
– incident cases, subjects entered into study on detection of
disease, OR
– prevalent cases, those who have been diagnosed as having the
disease prior to the study
7
(cont.) Choice of cases • Views, behavior and reports of exposure to risk factor
amongst incident cases and prevalent cases will tend to
differ
– Those diagnosed previously are likely to be more informed about
the disease and may have altered their behavior and attitudes
• Incident case design is preferred as it reduces recall bias
and over-representation of cases with long standing
base.
8
Choice of controls • Controls should
– come from the same population at risk of disease
– not have the disease
– be representative of the target population
• Selecting controls often proves harder than cases and
requires great care in the prevention of bias.
• A sampling frame of hospital patients is often used to
select controls
– however risk factors such as diet and smoking are commonly
linked to many diseases
9
(cont.) Choice of controls • Selecting controls from hospital patients sampling frame
might therefore over-estimate population exposure
– underestimation of association between disease and exposure
– to overcome this; use more than one control group
• Multiple controls can be used for each case
– Giving the study greater power
– Particularly where the number of cases is small due to the
disease being rare.
10
Exposure to risk factors and matching
• Exposure measurements are reliant on memory
(interviewed retrospectively) and/or medical records.
• Exposure estimates are vulnerable to recall bias
– Those with disease are more likely to remember exposure than
those without
• Exposure estimates are vulnerable to interview or
measurement bias
– Interviewer interviews or reports findings systematically
differently between cases and control and confounding factors
– Can be overcome by including blinding in the design so that they
do not know who is a case and who is a control at the time of
interview.
11
(cont.) Exposure to risk factors and matching
• Confounding factors must be identified prior to the start
of study.
• Individuals matched to controls where it is thought that
other factors, aside from those risk factors of interest,
might contribute to the development of disease
– E.g: age, sex,
• The factor upon which cases and controls are matched
can not be studied as risk factor.
• Alternative method of overcoming confounding is to
collect relevant information on them and adjust
statistically.
12
(cont.) Exposure to risk factors and matching
• Matching is used in case-control studies for;
– To permit allowance for confounders which are complex of
difficult to define
• e.g. by comparing within identical twin case/control pairs it is possible to
allow for ill-defined genetic confounders.
– To make allowance for confounders statistically more efficient.
Efficient analysis requires that there be a similar ratio of cases to
controls at each level of exposure to the confounding variable
– To reduce biases in the ascertainment of exposure.
• Egg. Data collected from deceased cases should be matched to deceased
control as information is collected from their family members.
13
Analysis of data
• Case control study measure the odd of exposure based
on disease
– Compared to cohort study which measure relative risk of disease
based on exposure
• Odds ratio has two components;
– The odds of exposure for cases
– The odds of exposure for controls
14
(cont.) Analysis of data
• The odds of exposure for cases
– Number of cases exposed / number of cases unexposed given
by Odds / cases = a / b
• The odds of exposure for controls
– Number of controls exposed / number of controls unexposed
given by Odds / controls = c / d
• The estimated Odds Ratio is then Odds = ad / bc
15
(cont.) Analysis of data
• The Odds Ratio is interpreted as
– OR < 1; Odds of exposure for cases are less than those for
control. Exposure appears to reduce risk of disease.
– OR = 1; Odds of exposure for cases are the same as those for
control. Exposure does not appear to be a risk factor.
– OR > 1; Odds of exposure for cases are more than those for
control. Exposure appears to increase risk of disease.
• A 95% confidence interval (95% CI) gives an indication
of the confidence we have in the estimated Odds Ratio
– E.g. if the entire 95% CI is above 1, it is concluded that exposure
significantly increases the risk of disease at the 95% level.
16
(cont.) Analysis of data
• Assessment of whether an observed association is likely
to be directly causal and not the results of unrecognized
confounding depends on;
– The size of the relative risk – higher relative risk are less likely to
be explained by unknown confounders.
– The presence of a dose-response relation – the observation of a
higher risk in subjects with a greater exposure to the risk factor
favors a direct causal relationship.
– The existence of plausible biological mechanism which might
explain a causal relationship.
17
Advantages of case control studies
• Quick
• Cheap
• Particularly suited to the study of rare diseases as the
diseased are selected at the outset of the study.
18
Disadvantages of case control studies
• Difficulties in overcoming potential bias and confounding.
• The successful selection of both cases and controls who
are representative of their respective population is often
difficult.
• An inability to infer causality and no information on the
chronology of disease and exposure.
• Inefficient in studying risk factors which are rare.
• Studies are often not population based, therefore it is
impossible to calculate incidence of disease.
19
Recapitulate In this lecture, you have been exposed to;
• definition of case-control studies
• choice of cases and controls
• exposure to risk factors and matching
• analysis of data for case-control studies
• advantages and disadvantages of case-control studies
20